Prolog thousand database
This file is a preliminary version of a database of applications of Prolog and related languages, containing over 500 entries. It is made available for the benefit of the Prolog community. It may be freely copied, but not published without the explicit permission of the maintainers below.
It does not contain the full addresses of the authors of the software. Anyone who wishes to follow up a reasonable number of entries should send an email message to Al Roth (alroth@cix.clink.co.uk) who will be happy to supply these details. 
If you haven't submitted an entry, a copy of the original submission form is given below.
-------------------------------------------------------------
Entries may be made either by the original authors or (by agreement) those in 
contact with them who have access to the information asked for (which includes 
the size of program and time taken). Entries are not limited to commercial 
programs - shareware and non-distributed and privately used programs are 
perfectly acceptable, but we are not interested in programming exercises. The 
results of research projects may be submitted if they have actually been used 
for their designed end.
An entry only takes a few minutes, so why not do it right away? Pass this file 
on to your friends and get them to enter too!
Send entries to:
	Al Roth, PO Box 137, Blackpool, Lancashire, FY2 0XY, U.K.    
	Email: alroth@cix.compulink.co.uk
	Fax: +44 253 53811 Telephone: +44 253 58081. 
	(Floppy disks for PC or Mac in text form are also welcome, 
		and paper entries may also be sent).
Queries may also be addressed to the people who conceived the project:
	Chris Moss (cdsm@doc.ic.ac.uk)
	Leon Sterling (leon@alpha.ces.cwru.edu)
Sponsored by:  ALP-UK, Dept. Computing, Imperial College, London SW7 2BZ
	Email: alpuk@doc.ic.ac.uk
------------------------------------------------------------
Entry Form and submission details
=================================
Put your details in place of the description below, leaving the field names in 
angle brackets. Do not  use angle brackets elsewhere, but free use of spaces and 
newlines is ok. Several entries may placed after each other in the same file.
<Program>  what it is conventionally called in the documentation
	<Purpose> (up to 20 words)
	<Developed by>   people mainly responsible
	<Organization>  directing or sponsoring the work
	<Country> in which developed
	<Domain>  the work area and activity to which this is relevant. 
		Try and be as specific as possible. e.g.: Electronic 
		Design, Financial Analysis, Legal Drafting, Medical  
		Diagnosis. General descriptions such as CAD or Decision 
		Support should be used only if the product is directly 
		useful for several work areas.
	<Contact> Name of someone who can answer questions, with 
	<Address> and<Telephone>     and<Email>   address (or fax number) for the contact (if known)
	<Description>   (up to150 words)
	<Prolog>  Name of interpreter/compiler(s) used
	
	<Platform>      Hardware / operating system(s) (development / delivery)
	<Other Languages>    or systems used, with approximate extent
	<Lines>   Approximate number of lines of Prolog
	<Predicates>    Approximate number of Prolog predicates
	<Development effort> Number of person-months for initial system
	<Maintenance effort> Number of person-months since, including enhancements
	<Current Status>     1=Prototype, 2=Evaluation, 3=Released, 
		4=No longer supported, 5=Dead
	<Use>     1=Personal use, 2=in-House use, 3=Free/public domain, 
		4=Licensed, 5=Sold product 
	<Users>   Approximate number
	<Date deployed> Month/year of inital practical use of program
	<Dated>   date to which this information refers
	<References>    Citations of published or available work, if any 
</Program>     This ends the entry.
------------------------------------------------------------

<Program> 3TAP
	<Number> 454
	<Domain> Hardware Verification, Natural Language Processing.
	<Purpose> Tableau-Based Automated Theorem Prover for arbitrary
finitely-valued first-order logics including the handling
of equality and sorts.
	<Description> Tableau-Based Automated Theorem Prover for 
arbitrary
finitely-valued first-order logics including the handling
of equality and sorts.
	<Developed by> Reiner Haehnle, Wolfgang Wernecke.
	<Organization> IBM Germany, University of Karlsruhe, DFG.
	<Country> Germany.
	<Prolog system> Quintus Prolog, runs on versions  2.4 and later.
	<Other languages> C / 5%
	<Lines> 17,500 incl comments.
	<Predicates> 910
	<Platform> Sun 3, Sun Sparcstations, IBM PS/2, IBM RS/6000 all Unix.
	<Type of Use> 
	<Current Status> Released
	<Date deployed> May 1992
	<No. of Users> 7 - 10
	<Information Date> Oct 1992
	<References> @TECHREPORT{Beckert92a,
		   AUTHOR = {Bernhard Beckert},
		   INSTITUTION = {IBM Germany, Institute for 
Knowledge Based Systems},
		   MONTH = {jan},
		   NUMBER = {208},
		   TITLE = {Konzeption und {I}mplementierung von 
		{G}leichheit f\"{u}r einen
				     tableau--basierten 
		{T}heorembeweiser},
		   TYPE = {{IWBS} {R}eport},
		   YEAR = {1992}}				
		@INPROCEEDINGS{BeckertHaehnle91,
		   AUTHOR = {Bernhard Beckert and Reiner H\"{a}hnle},
		   BOOKTITLE = {Proc.~$11^{th}$ Conference on  Automated Deduction {CADE}, Albany/NY},
		   EDITOR = {D.~Kapur},
		   PAGES = {507 -- 521},
		   PUBLISHER = {Springer Verlag},
		   SERIES = {{LNCS}},
		   TITLE = {An Improved Method for Adding 
Equality to Free Variable Semantic
 Tableau},
		   VOLUME = {607},
		   YEAR = {1992}}		    
		@TECHREPORT{Haehnle89,
		   AUTHOR = {Reiner H\"{a}hnle},
		   INSTITUTION = {Wissenschaftliches {Z}entrum, 
		{IWBS}, {IBM} {D}eutschland}
,
		   NUMBER = {136},
		   TITLE = {Spezifikation eines 
		{T}heorembeweisers f\"{u}r dreiwertige {F}ir
st--{O}rder {L}ogik},
		   TYPE = {{IWBS} Report},
		   YEAR = {1990}}		   
		@INPROCEEDINGS{Haehnle90,
		   AUTHOR = {Reiner H\"{a}hnle},
		   BOOKTITLE = {Proceedings Workshop on Computer 
Science Logic, Heidelberg},
		   PAGES = {248 -- 260},
		   PUBLISHER = {Springer, {LNCS} 533},
		   TITLE = {Towards an Efficient Tableau Proof 
Procedure for Multiple--Value
d Logics},						    
		@INPROCEEDINGS{Haehnle91,
		   AUTHOR = {Reiner H\"{a}hnle},
		   BOOKTITLE = {Proc. International Symposium on 
Multiple--Valued Logic, Vic
toria},
		   PAGES = {238 -- 245},
		   PUBLISHER = {{IEEE} Press},
		   TITLE = {Uniform Notation of Tableaux Rules 
for Multiple--Valued Logics},						
		@BOOK{Haehnle92f,
		   AUTHOR = {Reiner H\"{a}hnle},
		   MONTH = {To appear},
		   PUBLISHER = {Oxford University Press},
		   TITLE = {Automated Proof Search in Multiple--
Valued Logics},				     @ARTICLE{HaehnleSchmitt91,
		   AUTHOR = {Reiner H\"{a}hnle and Peter H. 
Schmitt},
		   JOURNAL = {Journal of Automated Reasoning, to 
appear},
		   TITLE = {The liberalized $\delta$--rule in 
free variable semantic tableau
x},
		   YEAR = {1993}}
		@TECHREPORT{Schoepke91,
		   AUTHOR = {Gisela Sch\"{o}pke},
		   INSTITUTION = {Wissenschaftliches {Z}entrum, 
		{IWBS}, {IBM} {D}eutschland}
,
		   MONTH = {October},
		   NUMBER = {188},
		   TITLE = {{M}\"{o}glichkeiten des {E}insatzes 
eines dreiwertigen {T}heorem
beweisers},
		   TYPE = {{IWBS} Report},
		   YEAR = {1991}}
		   YEAR = {1993}}
		   YEAR = {1991}}
		   YEAR = {1990}}		 
		@TECHREPORT{TAP92,
		   AUTHOR = {Reiner H\"{a}hnle and Bernhard 
Beckert and Stefan Gerberding an
d Werner Kernig},
		   INSTITUTION = {Wissenschaftliches {Z}entrum 
		{H}eidelberg, {IWBS}, {IBM} {
D}eutschland},
		   MONTH = {July},
		   NUMBER = {227},
		   TITLE = {{T}he {M}any--{V}alued {T}ableau--
		{B}ased {T}heorem {P}rover {\T
AP}},
		   TYPE = {{IWBS} {R}eport},
		   YEAR = {1992}}
</Program>

<Program> A Code Generator for Database Applications in Prolog.
	<Number> 297
	<Domain> CASE Tools
	<Purpose> Use of Prolog as a specification language for transactional 
systems and as a tool for automating software cycle.
	<Description> The project consists of the development of a source code 
generator (implemented in Prolog) for DB transactional applications.  The 
underlying Information System makes use of Prolog as a specification 
language to describe an extended Entity-Relationship model.  Two 
applications have been developed, producing source code in dBASEIV and RPG.
	<Developed by> S. Dulli - R. Sprugnoli - L. Veronese.
	<Organization> University of Padova
	<Country> Italy
	<Prolog system> Prolog - 2		   E.S.I.
	<Lines> 1,000  (130K memory)
	<Predicates> 300
	<Platform> P.C. MSDOS
	<Current Status> Prototype
	<Date deployed> Dec. 1990
	<No. of Users> 4
	<Information Date> 25/11/91
	<References> S. Dulli, R. Sprugnoli, L. Veronese "A Code 
Generator for Database Applications in Prolog"  Proceedings of 
Information Technology Interface, June 1991, Yugoslavia.
S. Dulli, R. Sprugnoli, L. Veronese "Un generatore di codice sorgente per 
applicazioni transazionali su basi di dati realizzato in Prolog"  Atti del 
VI convegno sulla programmazione logica, giugno 1991, Pisa Italy. 
</Program>

<Program> AALPS
	<Number> 198
	<Domain> Aircraft planning
	<Purpose> A load planning system for aircraft cargo
	<Description> The Automated Air Load Planning System provides a 
flexible spatial representation and knowledge base techniques to reduce the 
time taken for planning by an expert from weeks to two hours. It 
incorporates the expertise of loadmasters with extensive cargo and aircraft 
data.
	<Organization> SRI International, XVII Airborne Corps, Military Aircraft 
Command
	<Country> USA
	<Prolog system> Quintus
	<Other languages> C
	<Platform> SUN workstation
</Program>

<Program> ACACIA
	<Number> 243
	<Domain> Nuclear power
	<Purpose> a knowledge-based framework for the on-line dynamic synthesis 
of emergency operating procedures in a nuclear power plant
	<Description> It's hard to see the status of this
	<Organization> CEA, Electricite de France
	<Country> France
	
	<Prolog system> SPIRAL
ÇIF*** The text replacing this argument is too long ***È
	<References> A. 
Poujou, B. Papin, R. Soldermann: Dynamic Synthesis of Emergency operating 
procedures based on the generalized state approach. International Working 
Group on Machine-machine communication for emergency operation in Nuclear 
Power PLants, IAEA Schliersee, RFA, Oc 1988.
</Program>

<Program> ACAPS Loan Printing System
	<Number> 458
	<Domain> loan paperwork production.
	<Purpose> Expert system to create loan documentation from ACAPS main 
frame loan system.
	<Description> Took raw IBM mainframe data from a loan analysis & 
approval package and produced all the relevant paperwork, including 
attempted reconstruction of upper/lower case text. The system spooled 
was made fault tolerant and
controlled a bank of laser printers and fax modems. It replaced a 
(basically) manual system. 
	<Developed by> John Grant.
	<Organization> Security Pacific National Bank.
	<Country>  USA
	
	<Prolog system> Arity Prolog
	<Other languages> C - 5 lines
	<Lines> 10000
	<Predicates> 200
	<Platform> IBM PC - OS/2 v 1.3
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> March 1990
	<No. of Users> 100
	<Information Date> 1991
</Program>

<Program> ADAM
	<Number> 437
	<Domain> General purpose object-oriented database, with emphasis
on extensibility with new modelling constructs by using
metaclasses.  Comes with simple persistent store.
	<Purpose> Object-Oriented Database.
	<Description> Adds the ability to structure prolog programs and data 
using the object-oriented paradigm.
	<Developed by> Norman Paton.
	<Organization> Heriot-Watt University.
	<Country> UK
	
	<Prolog system> SICStus + Sepia (Quintus,MegaLog).
	<Other languages>  C (1000 lines)
	<Lines> 4000
	<Platform> Sun + UNIX
	<Type of Use> Free/Public Domain
	<Current Status> Prototype
	<Date deployed> March 1989
	<No. of Users> Sites = 4; People = 50
	<Information Date> April 1992
	<References> Paton, N.W. and Diaz, O., Object-oriented Databases 
and Frame-Based Systems: A Comparison, Information and Software 
Technology, Vol 33, No 5, 357-365, 1991.
Gray, P.M.D., Kulkarni, K.G. and Paton, N.W., Object-Oriented 
Databases: A Semantic Data Model Approach, Prentice-Hall, March, 
1992.
Paton, N.W., ADAM: An Object-oriented Database System Implemented In 
Prolog, Proc 7th British National Conference On Databases, Williams 
(ed), CUP, 147-161, 1989
Paton, N.W. and Diaz, O., Metaclasses in Object-Oriented Databases, in 
Object-Oriented Databases: Analysis, Design and Construction (DS-4),  W. 
Meersman et al.(eds), North-Holland, 331-348, 1991.
</Program>

<Program> ADAN
	<Number> 375
	<Domain> Aquisition of knowledge from data analysis.
	<Purpose> Conceptual clustering of data bases.
	<Description> A generic product for the conceptual clustering of 
relevant
database through a non supervised learning technique.
	<Developed by> Antoni Trias, Xavier Fustero.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.3 for Windows.
	<Other languages> Turbo C, Microsoft C.
	<Lines> 6235
	<Predicates> 534
	<Platform> Developed PC 386 / MS DOS,Delivery PC 386 or 486 / MS DOS and 
Windows.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed>  August 1992.
	<Information Date> 07/1992.
	<References> AIA/ Software product description.
</Program>

<Program> ADAS
	<Number> 106
	<Domain> Digital design
	<Purpose> Application-driven Design Automation for Microprocessor Design
	<Description> Submitted to PAC. Paper 194
	<Developed by> Iksoo Pyo et al
	<Country> USA
	<Prolog system> Subset of standard Prolog
	<Platform> VLSI Microprocessor
	<Information Date> 17/10/91
</Program>

<Program> ADES (ATP Design Expert System)
	<Number> 425
	<Domain> Logic Circuit Design.
	<Purpose> Automatic design of the logic circuits constituting an 
Automatic Train Protection (ATP) system, starting from the station 
topography.
	<Description> ADES (ATP Design Expert System) is an expert system 
for automatically designing the logic circuits which constitute an 
Automatic Train Protection (ATP) system 
of stations belonging to a certain class determined by the Italian 
Railway Station Authority. Starting from the drawing of the station 
topography, introduced through a 
graphic input interface, ADES determines the circuit design for the 
ATP system of the station, and finally presents the user the logic 
circuits through a graphic
 output interface. The user can adjust the resulting design through 
the interface, and finally plot it. Moreover, several intermediate 
documents - usually adopted by the Italian 
Railway Station Authority  for validating the project, are also 
automatically produced by ADES.				     
ADES inference engine has been implemented by using the Prolog 
language, and meta-interpretation techniques in particular. 
Implementing both the expert system and some of its environment tools 
in Prolog has led to the rapid prototyping of the overall system. 
Optimization techniques, and partial evaluation in particular, have also 
been applied in order to obtain efficient execution.
	<Developed by> Roberto Cremonini, Evelina Lamma, Paola Mello, Augusto 
Torchi.
	<Organization> SASIB S.p.a. Bologna, jointly with Dipartimento di 
Elettronica, Informatica e Sistemistica (University of Bologna).
	<Country> Italy.
	<Prolog system> Quintus Prolog.
	<Other languages> Fortran, Graphic Kernel System.
	<Lines> 5000
	<Predicates> 800
	<Platform> microVAX Digital  VMS
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 1990
	<No. of Users> 10
	<Information Date> January 1991
	<References> Cremonini, R., Lamma, E. Lamma, Mello, P., 
ADES: An Expert System for ATP Design
AI EDAM Journal, 1989, Vol. 3, N. 1, pages 1-21.
</Program>

<Program> Aditi Compiler
	<Number> 44
	<Domain> Deductive Databases
	<Purpose> Query and rule compilation of Aditi Prolog to the base 
relational language RL for the Aditi Deductive Database.
	<Description> The Aditi Compiler consists of three main stages; 
(purish) Prolog to
High Level Data Structure; High Level to Low Level Data Structure and 
Low Level Data Structure to relational operations.  
A number of different optimizations/transformations are available at each 
level; including magic set transformations; standard relational 
optimizations and peephole optimization.  The resulting relational code is 
directly executed by the Aditi deductive database.
	<Developed by> Peter Stuckey (the current version)
	<Organization> Department of Computer Science; University of Melbourne
	<Country> Australia
	<Prolog system> NU-Prolog
	<Other languages> Aditi relational engine is written in C.
	<Lines> 7000 -- lines of code (including blanks)
	<Predicates> 550
	<Platform> Silicon Graphics R3000 based multiprocessors / Unix System 5
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> June 90
	<No. of Users> 8
	<Information Date> Feb 92
	<References> J. Vaghani; K. Ramamohanarao; D.B. Kemp; Z. Somogyi 
and P.J. Stuckey.
Design Overview of the Aditi deductive database system. Procs. 
Seventh International Conference on Data Engineering. pages 240--247; 
1991.
J. Vaghani; K. Ramamohanarao; D.B. Kemp; Z. Somogyi and P.J. Stuckey. 
An introduction to the Aditi deductive database system. Australian 
Computer Journal 23(2); pages 37--52. 1991.
D.B. Kemp; D. Srivastava and P.J. Stuckey. Magic Sets and Bottom up 
computation of Well-Founded Models. Procs. International Symposium on Logic 
Programming. San Diego; pages 337--354; 1991.
</Program>

<Program> ADP (Arc Welding Diagnosis and Planning System.
	<Number> 309
	<Domain> Design Manufacturing Systems
	<Purpose> Diagnosis and planning of technological parameters for gas-
shielded metal arc welding applications.
	<Description> The ADP system consists of two distinct knowledge 
bases supplying welding engineers in the diagnosis and the planning 
of technological parameters for gas-shielded welding applications.  
Welding knowledge is modeled in terms of objects and rules.  For 
diagnosis purposes hypotheses have been implemented connecting system 
symptoms with a set of applyable error recovery
measures.  For planning a set of rules has been defined deriving the values 
of a welding schedule.
	<Developed by> Fraunhofer-Institute for Production Systems and Design 
Technology
	<Organization> Dep. Robot System Technology
	<Country> Germany
	<Prolog system> IF-PROLOG 4.0
	<Other languages> C - extensions
	<Lines> ~ 1 MByte source
	<Platform> VAX station 3100
VMS 5.4
	<Type of Use> In-house
	<Current Status> Prototype
	<Information Date> 11/05/92
	<References> IFAC/IFIP/IMACS   Symposium on Robot Control SYROCO Sept. 
16-18, 1991, Vienna, Austria.  Preprint pp. 39.
</Program>

<Program> ADRAM (after an ICL program of the same name).
	<Number> 406
	<Domain> Data Conversion.
	<Purpose> Syntax-directed data converter.
	<Description> Originally developed for porting files of 
Government and other economic data from ICL 1900, ICL 2900, and IBM 
tapes to VAX VMS. We wanted to change the file structures as we did 
this (e.g. swap fields within records), rather than
translating them verbatim. Therefore decided to write a general 
purpose converter, whose inputs were (1)syntactic description of file 
layouts at the record and field level, in a notation which looked 
like a mixture of Cobol data definitions and Pascal record 
structures; (2) a description of the machine architectures: word 
sizes, byte directions,
and so on. ADRAM generated from these an Algol68
program which when ran, did the actual conversion.
	<Developed by> Jocelyn Paine.
	<Organization> Institute of Fiscal Studies.
	<Country> England.
	<Prolog system> Poplog.
	<Other languages> Algol68.
	<Lines> 3000.
	<Predicates> 600.
	<Platform> VAX/VMS.
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 12/1987.
	<No. of Users> 1 (organisation).
	<Information Date> 3/1992.
	<References> None.
</Program>

<Program> ADW Knowledge Coordinator
	<Number> 215
	<Domain> Systems development
	<Purpose> Tools to help developers to formalize, design and implement 
information systems
	<Description> The Application Development Workbench uses various 
diagramming techniques to help users develop and view their systems. The 
information the tools develop is pooled dynamically in an encyclopedia, 
consistency being ensured through a Knowledge Coordinator; this information 
will be stored later on in the IBM Repository Manager.
	<Organization> KnowledgeWare, Inc
	<Country> USA ?
	<Prolog system> IBM Prolog for OS/2, IBM MVS Prolog
	<Other languages> C code for screen graphics
	<Lines> 130 KLOC
	<Platform> IBM PS/2 with OS/2, IBM 370
	<References> B. Robinet: Logic Programming at IBM from the Lab to the 
customer. Proc ICLP8. pp912-921. 1991
</Program>

<Program> AFARS - Asset Forfeiture advisor and reporting system
	<Number> 104
	<Domain> Criminal Justice; Law enforcement
	<Purpose> Assists federal, state and local law enforecement agencies in 
conducting effective asset forfeiture programs to process seized criminal 
assets
	<Description> AFARS is designed to assist law enforcement and 
prosecution elements at the federal, state and local levels in the 
asset forfeiture process. The system includes:
  - Asset forfeiture training modules
  - A knowledge based expert system advisor for both federal and 
local law enforcement
  - A reference library library of federal and state statutes 
(criminal and civil), US Department of Justice Guides and Federal 
Rules with search capability
  - A system for preparing asset forfeiture documents and court 
petititions
  - An intelligence system to assist in link analysis to match 
offenders, crimes, drugs of choice, habits and locations
  - A statistical capability to create reports identifying totals and 
percentiles of various types of seizures by number of occurrences and 
monetary value
 
Planned: Graphics capability to depict statistical information
Release planned for May 1992.
	<Developed by> Dennis M. Scott, Presdent and Project manager
	<Organization> Scott Enterprises
	<Country> USA
	<Prolog system> PDC Prolog
	<Other languages> Some Microsoft Quick Basic, which will be reprogrammed in 
PDC Prolog in the near future
	<Platform> IBM XT compatible, 640K RAM, Hard drive, CGA, EGA or SVGA 
Color 
monitor with 512K Graphics. Requires approximately 3MB disk space which 
will grow as data bases expand
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Information Date> 8 Apr 1992
</Program>

<Program> AGATHA
	<Number> 3
	<Domain> Electronic Diagnosis
	<Purpose> Expert System to test and diagnose complex printed circuit 
boards
	<Description> Agatha uses a suite of mini expert systems; each 
customized (optimized) to diagnose a particular kind of test -- this 
necessary due to the diversity and complexity of the various tests.  Agatha 
reasons about the test results as well as suggesting further tests to run. 
Agatha runs the tester directly.  It provides an improved; friendly 
interface to the user; using X windows.
	<Developed by> Daryl Allred; Mike Bennett; Chris Preist; Yossi 
Lichtenstein;
Ajay Gupta
	<Organization> Hewlett-Packard Co.
	<Country> USA and UK
	<Prolog system> Quintus Prolog
	<Other languages> C & YACC (13 K lines); scripts (830 lines)
	<Lines> 26 K lines  (all lines are UN-commented)
	<Predicates> 2650
	<Platform> HP 9000/370; HP-UX (Un*x)
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> Jan 1990
	<No. of Users> 3 sites internally
	<Information Date> Oct. 1991
	<References> Agatha: An Integrated Expert System to Test and Diagnose, 
Complex PC Boards  Innovative Applications of Artificial, Intelligence 3  
1991  AAAI Press"
</Program>

<Program> AIA-TRANS
	<Number> 381
	<Domain> Freight distribution.
	<Purpose> Automation of the information/management system for a 
transportation and freight company (packages distribution).
	<Description> Integral management system for freight companies working 
with packages for the company Jet Way. It was developed in Prolog to be 
executable in a local network with Novel Netware software. Includes export 
documents and sea tranportion as well.
	<Developed by> Antoni Trias, Patricias Llopis, Xavier Fustero, Paqui 
Cabezas and programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.31
	<Other languages> void
	<Lines> void
	<Predicates> void
	<Platform> Developed PC 386 /MS DOS
Delivery local Novel network.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1991
	<Information Date> 07/1992
</Program>

<Program> AiGerm
	<Number> 107
	<Domain> Database design
	<Purpose> A Logic Programming Front End for Germ
	<Description> Submitted to PAC. Paper 210
	<Developed by> Safaa H. Hashim
	<Country> USA
	<Prolog system> "Quintus, BIM, LDL"
	<Information Date> 17/10/91
</Program>

<Program> Air Pollution ControlSystem
	<Number> 264
	<Domain> Pollution Control
	<Purpose> To advise on air pollution levels of county districts
	<Description> The system handles data about the basic 
concentration of several industrial pollutants in
Budapest and other counties of Hungary with each county having 15-20 
districts. The system checks whether the air pollution of working or 
planned plants is below the permitted level.  If   permitted levels are 
exceeded the system recommends appropriate  filtering equipment. The system 
is designed for a variety of users (managers, research workers etc) 
	<Developed by> J. Bendl, Gy. Lugosi, Z. Markusz 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> I. Futo, F. Darvas, P. Szeredi (1978)
The application of Prolog to the development of QA and DBM systems.
In Logic and Databases, pp.347-376
H. Gallaire and J. Minker (eds)
J. Bendl, Gy. Lugosi, Z. Markusz (1979)
An interactive system for checking air pollution - the information 
system.
Infromacio-Electronica, XIV, No 1, 1979. pp. 55-58.
</Program>

<Program> Air Sheduler
	<Number> 483
	<Domain> 
	<Purpose> Automatisation of aerodrome air traffic control service.
	<Description> Independent software section for air traffic control 
system embeded real-time conflict detection, and optimum safe air traffic 
planning.
	<Developed by> Prof. Stanislav Timokhin. post-graduate Serge 
Ivanets
  post-graduate Michail Krylov.
	<Organization> 
  Moscow Engineering Physics Institute.
	<Country> Russia.
	<Prolog system> Turbo Prolog 2.0
	<Other languages> not used
	<Lines> 4300
	<Predicates> 175
	<Platform> MS DOS
	<Date deployed> 1992
	<No. of Users> State Institute "AirNavigation"
  Airport Mineralnye Vody (international code UUWW)
</Program>

<Program> AJAX
	<Number> 229
	<Domain> Payroll administration
	<Purpose> Administering payments for civil servants
	<Description> The Automated Job Assembly eXpert provides an 
intelligent PC-based front-end to a large mainframe payroll system. 
Given a specification of what is required, it generates all the 
necessary instructions for jobs to be run, with all the parameters 
correctly set.
A further product 'Coordinate' arose out of this project.
	<Developed by> Jonathan Souster
	<Organization> Telecomputing
	<Country> UK
	<Prolog system> LPA Prolog Professional
	<Other languages> flex
	<Platform> IBM PC 286 with 2MB
</Program>

<Program> ALBERTUS  
	<Number> 362
	<Domain> Computer based training of Physics.
	<Purpose> Training Physics in Highschool and College.
	<Description> A graphics program which can be used for training 
Physics.
Package contains students program, separate editor for making 
lessonwork and modifying lessonworks that come with package. Program 
has tree level help function for student in each question. Student 
can use graphs of mathematical functions and a calculator to solve 
problems. Program checks answers of student and tells reasonable 
precision for answers if student gives for example too 
many digits in his or her answer.
	<Developed by> Osmo Miettinen, Antti Miettinen and Tapio Okkola.
	<Organization> Taosan Ky, supported by Finnish School Department.
	<Country> Finland.
	<Prolog system> PDC Prolog 3.20 - 3.30
	<Other languages> Assembler.
	<Lines> 5831 + 2311
	<Predicates> 445 + 150
	<Platform> IBM-PC/AP/PS2 and compatibles, PC-DOS or MS-DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 3/1989
	<Information Date> 4/1/1992
	<References> Taosan Ky and MFKA OY.
</Program>

<Program> ALIEN
	<Number> 222
	<Domain> Equipment documentation
	<Purpose> Documentation aid for equipment
	<Description> 
	<Organization> Dassault Electronique
	<Country> France
	<Prolog system> Emicat
	<Current Status> Prototype
	<References> Dassault Electronique report: Les Techniques 
Informatiques avancees et leurs applications. NE 485 352, 30th Apr 1991.
</Program>

<Program> ALTES
	<Number> 446
	<Domain> Mainframe Tuning.
	<Purpose> Real Time Expert System Diagnosis and Help for tuning 
Bull/GCOS8/TP8 mainframe.
	<Description> Connected to the mainframe, ALTES collect in real 
time a lot of system informations about jobs, dispatching, buffers, 
files access, disks, response times ...  and analyse them to product 
diagnosis, alerts, journal and graphs.  A synoptic resume the session 
alerts, so the user could immediatly see what's previously append.  
The knowledge base contains a set of parameters which could be 
modified during the session (without recompiling)
.
The user can consult an historic, rebuild the context, replay the 
analyse and see why a diagnosis was product (facts values and rules 
used).  He can modify the knowledge base with ALTES compiler and test 
it with an old journalized context to verify new rules. This allows 
specifics uses.
Graphs and journal generated by ALTES are a real help for understanding and 
tuning the mainframe.
	<Developed by> Stephane Gratepanche.
	<Organization> GFI / CdFi
	<Country> France
	<Prolog system> PDC Prolog
	<Other languages> C, Pascal
	<Lines> More than 10 thousands.
	<Platform> 80386, 486 / MsDos
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1990
	<No. of Users> 10
	<Information Date> March 1992
</Program>

<Program> An Automated Legal Reasoning System on
Indian Governement's Reservation Policy for
Scheduled Castes and Scheduled Tribes
	<Number> 436
	<Domain> Legal Reasoning.
	<Purpose> Assists in the decision making on various issues related to 
Reservation Policy for Scheduled Castes and Scheduled Tribes.
	<Description> Scheduled Castes and Scheduled Tribes are the most
backward and economically weaker segments of the Indian society. 
Historically they have been neglected and discriminated against 
through the centuries.  Government of
India brought in a special legislation for their uplift and development. 
Orders and instructions have been issued from time to time promulgating 
concessions to encourage their employment in the civil services of the 
Government and to safeguard their interests while in service. A logic 
programming based expert system has been developed for giving advice on 
this piece of legislation , called Reservation Policy of Scheduled Castes 
and Scheduled Tribes.
	<Developed by> Srinivas Pavuluri, S.B.Yadavalli,K.K.Bajaj.
	<Organization> National Informatics Centre,Government of India, Lodhi 
Road, CGO Complex, New Delhi-3,India.
	<Country> India.
	<Prolog system> LPA Prolog Professional.
	<Other languages> Nil
	<Lines> 1000
	<Predicates> 7
	<Platform> DOS
	<Type of Use> Licensed
	<Current Status> Evaluation
	<No. of Users> All Ministries/Departments in Government of India.
	<Information Date> April 16, 1992
	<References> Communicated to Second Pacific RIM International
Conference On AI, Scheduled to be held in Seoul,Korea in September 1992.
</Program>

<Program> APACHE
	<Number> 321
	<Domain> Interactive Scheduling.
	<Purpose> Assignment of airplanes to gates at international airports.
	<Description> The APACHE  scheduling system is an interactive 
resource utilisation system for the assignment of airplanes to gates 
at international airports while respecting many of the operational 
constraints found in such airports.
The system allows to reschedule the assignments and tailor the automatic 
incremental scheduler.  The system connects to an airports conventional and 
operating database via CHIP'S SQL link.
	<Developed by> COSYTEC
	<Organization> CPL Software and Services
	<Country> France
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 3000
	<Platform> SUN/HP/IBM/DIGITAL workstations
PC  386/486		    UNIX    DOS
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Int. Conf. on Fifth Generation computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> APEX
	<Number> 440
	<Domain> Presentation for prospective customer of modular house.
	<Purpose> Automatic living-room layout system for modular house.
	<Description> To satisfy the requirements of each customer,APEX
automatically layouts rooms for a type of modular house 
within the constraints of the legal regulations.
	<Developed by> ISAC, Inc.
	<Organization> Sekisui Chemical Corporation.
	<Country> Japan.
	<Prolog system> K-Prolog.
	<Other languages> C (50%)
	<Lines> 50,000 lines.
	<Predicates> 5,000 predicates.
	<Platform> SONY NEWS; 4.2/4.3BSD (development / delivery).
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> April 28, 1990
	<No. of Users> 100 persons (at 50 sites).
	<Information Date> April 15, 1992
	<References> NIKKEI ARTIFICIAL INTELLIGENCE No.105, pp.2-3 May 7, 
1990.
</Program>

<Program> ARCHON
	<Number> 108
	<Domain> Information systems
	<Purpose> Cooperation in Industrial Systems
	<Description> Submitted to PAC. Paper 405
	<Developed by> Jennings
	<Country> UK
</Program>

<Program> ARIES
	<Number> 508
	<Domain> M.I.S.
	<Purpose> Using Prolog to access relational databases with natural 
language.
	<Description> Access to computer based MIS is typically gained via pre-
requested reports or SQL.  Many managers find these too restrictive.  An 
industry based project was started in 1988 to implement a natural language 
interface to management information.  The project demonstrated the 
feasibility and practicality of implementing NLIS to databases and revealed 
many interesting insights concerning the style and range of English used by 
computer users.
	<Developed by> Alan Burton.
	<Organization> City University.
	<Country> U.K.
	<Prolog system> Quintus Prolog.
	<Other languages> Keyling Database Interface.
	<Platform> Del Vax Ivms.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 1990
	<Information Date> Spring 1992.
	<References> Burton and Steward-A NLI to Management Information.
</Program>

<Program> ASIGNA
	<Number> 59
	<Domain> Chemical processes scheduling
	<Purpose> This program is meant to schedule chemical processes sharing 
resources. It handles disjunctive constraints and large time periods.
	<Description> Resource-allocation problems occur frequently in 
chemical plans. Different processes often share pieces of equipment 
such as reactors and filters. The program ASIGNA allocates equipment 
to some given set of processes. Each process is described as a 
sequence of stages. At each stage, a process uses some piece of 
equipment for a certain time period. 
The problems solved with our program resemble critical-path problems that 
can be solved with commercial packages. Unlike some of these packages, our 
program can handle disjunctive constraints. To handle such constraints, we 
use a generalization of the " domains" pioneered by the CHIP language. We 
have modified these domains to efficiently represent numerous elements, as 
happens in practical resource-allocation problems.
	<Developed by> David A. Rosenblueth
	<Organization> IIMAS-UNAM
	<Country> Mexico
	<Prolog system> Arity Prolog
	<Other languages> none
	<Lines> 1900
	<Predicates> 211
	<Platform> PC/MSDOS
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> 5/1992
	<Information Date> 3/1992
</Program>

<Program> ASK  (Associative System Kentek)
	<Number> 327
	<Domain> Industrial Data Analysis and Management Information Systems.
	<Purpose> Knowledge retrieval and data analysis.
	<Description> A hybride knowledge retrieval and analyses system with 
machine-learning and neural network simulation, rule based inference 
mechanism, dynamic graphical presentation of decision trees.
	<Developed by> K. van Goor; R. F. Jansen.
	<Organization> Kentek Expertise.
	<Country> The Netherlands.
	<Prolog system> PDC prolog 3.30
	<Other languages> Turbo-C Program shell.
	<Lines> 25000.
	<Predicates> 1000.
	<Platform> Windows 3.1, Ms-dos.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 01/06/92
</Program>

<Program> ASTROLIZA / Aastro Brain Base
	<Number> 328
	<Domain> Is used in Holland, Belgium, Germany, Austria and Switzerland 
by professional astrologs who use it both for analysis and interpretaion of 
birth charts.
	<Purpose> Astrological Expert System.
	<Description> Astrological expert system with built-in AI and word 
processing facilities.  Makes use of all kinds of different input files and 
produces a unique personal report for each chart.
	<Developed by> Ole Eshuis, Amsterdam.
	<Organization> E.O. Eshuis Astrological Software.
	<Country> Holland.
	<Prolog system> First developed with Turbo Prolog, now running under PDC 
Prolog 3.30.
	<Other languages> Astro Brain Base uses a calculation module written in 
Turbo C++.
	<Lines> 2420 lines of source code divided over 5 modules.
	<Predicates> 186 local predicates and 26 global predicates.
	<Platform> MS DOS 640 KB.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> Spring 1988.
	<Information Date> 07/08/92.
	<References> In the autumn of 1992 there will be an article on Astro 
Brain Base in the German magazine "Meridian".
</Program>

<Program> ATMI
	<Number> 109
	<Domain> NL
	<Purpose> A Natural Language Interface to Management Information
	<Description> Submitted to PAC. Paper 164
	<Developed by> Dr A. P. Steward
	<Country> UK
	<Prolog system> Quintus
	<Platform> Vax
	<Information Date> 19/10/91
</Program>

<Program> ATR Trainer
	<Number> 213
	<Domain> Training
	<Purpose> Authoring language for training
	<Description> In training mode the material appears to a trainee 
as  hypertext style browsing environment. The user is able to learn 
by examing information which is structured as a set of related 
topics. They can also request illustrative examples and ask to be 
tested on the subject matter, using tests devised by the system 
according to the fats in the rulebase. The system maintains a model 
of the performance of each trainee for assessing progress.
In the advisory mode, the system behaves like a conventional expert system, 
with an inference engine and explanation facility using text and pictures.
	<Developed by> Paul French
	<Organization> Kingston Polytechnic
	<Country> UK
	<Prolog system> LPA Prolog professional
	<Platform> PC AT with 640K 
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> AUNT
	<Number> 196
	<Domain> Electronics Design
	<Purpose> Universal hardware description language translator for 
electronics design
	<Description> This allows an electronic design expressed in any of six 
hardware design languages (SPICE, EXT, ABCD, NET, HILO, GEMINI) to be 
translated into any of the others.
	<Developed by> Peter Reintjes
	<Organization> Microelectronics Center of North Carolina
	<Country> USA
	<Prolog system> Quintus
	<Platform> Unix, VAX workstation
	<References> P.B. Reintjes: A VLSI Design Environment in Prolog. 
ICLP5, 1988, pp70-81
</Program>

<Program> Automatic Porting DNS -> CNS
	<Number> 424
	<Domain> Automatic translating of specific software.
	<Purpose> It's an automatic porting tool enabling the simultaneous
manufacture of the DNS and CNS communication software versions.
	<Description>  The tool translates intructions for Bull's Datanet 
system in
Bull's Microfep system. The tool can be used for automatic porting  of 500 
000 compiler instructions, reducing the maintenance costs on all the common 
functions.
	<Developed by> Franc,oise Barichello, Francis Jeandel & Ge'rard Simon.
	<Organization> Internal work for Bull S.A.
	<Country> France.
	<Prolog system> SP-Prolog  (Bull S.A. registered trademark).
	<Lines> 1500
	<Predicates> 104
	<Platform> DPX 5000 Ros 4.1
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1988
	<No. of Users> 5
	<Information Date>  March 1992
</Program>

<Program> AUTOPROLOG
	<Number> 262
	<Domain> Legal drafting
	<Purpose> 
	<Description> Generates an expert system for a specific set of rules 
corresponding to an interpretation of a legal statute.
	<Developed by> Charles S. Saxon, Layman E. Allen
	<Organization> Eastern Michican University, U of Michigan Law School
	<Country> USA
	<Prolog system> Turbo Prolog
	<Platform> IBM PC
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 1987
	<Information Date> 1988
	<References> Layman E. Allen, Charles S. Saxon: Multiple 
interpretation of the structure of legal rules: Impediment or boon to 
legal expert systems. ICLP5 pp1609-1623, 1988
Layman E. Allen, Charles S. Saxon: Automatic generation of a legal expert 
system of Section 7 (2) of the United Kingdom Data Protection Act 1984, in 
4th Congresso Internatzionale sui tem, Informatica e Regolamentazioni 
Giuridiche, Roma, May 1988.
</Program>

<Program> BAMBI
	<Number> 110
	<Domain> Intelligent front end
	<Purpose> Intelligent aid to submission of several hundred  
interdependant jobs which update a fares database on a MVS system.  
	<Description> BAMBI is a KBS running on a PC linked to a large foures 
database on an IBM mainframe MVS system.  It uses knowledge of the batch 
jobs, of their parameters and of their mutual dependencies to simplify and 
speed the preparation of run requests, and automates their submission.  It 
allows the users to edit and maintain most of the knowledge for themselves, 
and requires very little other maintenance.  It has enabled its users to 
handle an increased workload with no increase in time nor decrease in 
accuracy, but improvement in both.
	<Developed by> David Barnard
	<Organization> British Airways
	<Country> UK
	<Prolog system> LPA Prolog Professional
	<Other languages> Gresham Telecomputing's CONNEX software to link to 
mainframe.
	<Lines> 11000
	<Predicates> 736
	<Platform> IBM  PS2 + DOS
	<Date deployed> ~ June 1990
	<No. of Users> 3
	<Information Date> 92/9/16
	<References> 1.  Proceedings 1st International Conference on the 
Practical Application of PROLOG 2/3 April 1992.  Short paper by David 
Barnard,  title:  BAMBI- etc.
2.  Expert Systems User - Article by Dr. Toshio Nomura, 1991 approx.  Aug. 
or Sept.
</Program>

<Program>  BARBARA THE SYLLOGIZER.
	<Number> 464
	<Domain> Logic exercises in syllogistic (monadic) logic.
	<Purpose> Exercises in syllogistic reasoning.
	<Description> BARBARA provides exercises in syllogistic 
reasoning. The instructor can provide the examples that he or she 
wishes to use in drilling students; examples are kept in a separate 
ASCII file which the instructor can edit with a wordprocessor or 
editor.  This file contains the sentences which are the premises of a 
particular example, the logical form of these sentences, and the 
lexicon of terms which appear in the premises.
The student may choose to use either the traditional or the modern 
rules for the syllogism and can change which rules are being used at 
any time.  The student enters, in English, a conclusion which 
follows.  Barbara parses the student's sentence and identifies the 
sentence's logical form.  Barbara checks to see if that logical form 
follows by first-order logic
from the form of the premises, and responds accordingly,						 
seeking to give the strongest possible reply for both correct and incorrect 
answers.
	<Developed by> R. G. Wengert
	<Organization> University of Illinois and IBM.
	<Country> U.S.A.
	<Prolog system> Arity Prolog interpreter and compiler.
	<Other languages> None.
	<Lines> 924 Lines (goals).
	<Predicates> 112 Predicates.
	<Platform> IBM PCs and clones, 8088s and up.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> April, 1989.
	<No. of Users> Several hundred at this university, unknown at others.
	<Information Date> November, 1992.
</Program>

<Program> BATEX
	<Number> 295
	<Domain> Building Design
	<Purpose> It is an expert system which helps to make the preliminary 
studies for the offices buildings, at the stage of the draft.
	<Description> Batex is an expert system for global design, intended for 
initiators to avoid costly errors.  The recommendations of Batex relate to 
our conditioning, acoustics, outside walls , glazed bays, possible window-
blind, energy consumption, etc.  It takes into account parameters of 
different fields such as geographical and topographical environment, 
climate, pollutions, seismicity, ground lift, noise sources.  It works 
either with forward chaining or backward chaining.  Betex gives the 
explanations of its conclusions in clear language (regulations, rules of 
the art...).  Batex ensures that nothing important is forgotten and permits 
simulations.
	<Developed by> Charles MENET - Marc MORO (expert)
	<Organization> DIALEC
	<Country> France
	<Prolog system> Prolog II +
	<Other languages> Microsoft C
	<Lines> 6500
	<Predicates> 315
	<Platform> PC 386 - Windows 3
				     -MS- DOS
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 07/1988
	<No. of Users> 2 (SNCF)_ the french railway company.
	<Information Date> 30/03/92
	<References> Systems Expert No.51-23 March  1992 ( A Jour, publisher - 
11, rue du Marche - St. Homore, 75001, France).
</Program>

<Program> BEACON
	<Number> 202
	<Domain> Computer manufacture
	<Purpose> Configuration of computer systems
	<Description> The Browser/Editor and Automated CONfigurator is 
designed to configure hard disks and tape drives to the user's 
particular needs. The assembly of the configuration is tedious and 
error-prone and is governed by a complex set of rules.
It is based on a knowledge environment called KNET which is a semantic 
network with inheritance. The configurator performs an "ordered walk" 
through the knowledge base, where component configurations are defined by 
experts. At each stage the user can revise configuration specifications 
since the system is interactive. The system does not permit invalid choices 
regarding items and quantities. The system is also linked tothe products 
database which contains pricing and bundling information.
	<Organization> Unisys, Paoli Research Center
	<Country> USA
	<Prolog system> Quintus
	<Platform> Unisys 5000/70
</Program>

<Program> BEARING
	<Number> 42
	<Domain> Mechanical Engineering
	<Purpose> The BEARING program provides a mechanical engineer with 
computer-aided selection of rolling bearings.
	<Description> The BEARING program is a mechanical engineer's 
workplace. This program asks bearing working conditions and selects 
the most suitable bearings. Its database includes data of the widely-
used bearings in the world (FAG; NTN; SKF; TIMKEN). The BEARING 
program has graphical interface; mouse support and help features.
The BEARING program consists of 70 seperate bearing selection programs (one 
for each type). The engineer can select suitable bearing of the given bore 
diameter for defined load and working conditions.
	<Developed by> Andres Lomp
	<Organization>  J/V IE Software.
	<Country> Estonia
	<Prolog system> LPA Prolog Professional 3.0.
	<Other languages> C; dbase III
	<Lines> 40 000 lines in PROLOG 
	<Predicates> 150 new predicates
	<Platform> IBM PC
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1990
	<No. of Users> Mechanical engineers
	<References> Magazine"Design News"
</Program>

<Program> BinProlog
	<Number> 78
	<Domain> Programming Language Implementation
	<Purpose> Prolog Compiler
	<Description> BinProlog is a fast (180 KLIPS on Sparc ELC) and small 
(49K emulator) Prolog compiler, written in a side-effect-free subset of 
Prolog, based on the transformation of Prolog to binary definite clauses. 
The compilation technique is similar to the Continuation Passing Style 
transformation used in some ML implementations.
	<Developed by> Dr. Paul Tarau
	<Organization> Universite de Moncton
	<Country> Canada
	<Prolog system> BinProlog 1.24
	<Other languages> C (30%)
	<Lines> 2675
	<Predicates> 242
	<Platform> sparc,sun3 (SunOs), 386 with go32 DOS-extender
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 10/91
	<No. of Users> 200
	<Information Date> Mars 10, 1992
	<References> @inproceedings{Tarau90:PLILP,
title={{E}lementary {L}ogic {P}rograms},
author={Paul Tarau and Michel Boyer},
pages={159--173},
editor={P. Deransart and J. Ma{\l}uszy{\'n}ski},
booktitle={Proceedings of Programming Language Implementation and
Logic Programming},
year=1990,
month=Aug,
publisher={Springer},
series={Lecture Notes in Computer Science},
keywords={Logic Programming},
number=456}
		@inproceedings{Tarau91:JAP,
title={A  Simplified Abstract  Machine for  the Execution of
Binary  Metaprograms},
author={Paul Tarau},
pages={119--128},
booktitle={Proceedings of the Logic Programming Conference'91},
year=1991,
month=Sep,
publisher={ICOT, Tokyo}}
		@inproceedings{Tarau91:RU,
title={Program  Transformations  and  {WAM-support}  for  the
Compilation of  Definite Metaprograms},
author={Paul Tarau},
booktitle={Proceedings of the Russian Conference of Logic 
Programming},
year=1991,
month=Sep,
publisher= {Springer}}
</Program>

<Program> Biomarker
	<Number> 426
	<Domain> Geochemistry for Oil Exploration.
	<Purpose> Expert system to help geochemists determine the
depositional environment of a rock or oil sample.
	<Description> This is an expert system designed to help 
geochemists 
determine what environment, e.g. marine or delta, existed when the 
source rock for a sample was laid down.  This is based on gas 
chromatography/mass spectrometry data.  The system provides extensive 
hierarchical window-based
explanations.  The knowledge is represented in a strictly logical way;  
great use is made of meta-level representation and reasoning.
	<Developed by> Dave Wolstenholme (Computing), Nigel Goodwin 
(Geochemistry).
	<Organization> BP Research.
	<Country> UK.
	<Prolog system> LPA MacProlog.
	<Lines> 5000
	<Predicates> 252
	<Platform> Macintosh II
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed>  9/90
	<No. of Users> 10
	<Information Date> 4/92
	<References> Wolstenholme, D.E. and Goodwin, N., 
Identification of geological environment taking an 
explanation-intensive and meta-level approach.   In: Procs. Avignon 91 - 
11th International Conference on Expert Systems and their Applications.  
EC2, 1991.
</Program>

<Program> BIS/Estimator System
	<Number> 92
	<Domain> Data Processing management
	<Purpose> 
	<Description> The BIS/Estimator system provides estimates for all 
phases of development of a data processing application from the initial 
feasibility study down to the fielding of the implemented system. 
BIS/estimator allows the use of five different estimating methods, which 
can be divided into two main categories: soft and hard. Each phase of the 
project can be analysed in detail at different levels. This makes it 
possible to produce cost estimates for the whole project, the individual 
phases or any task within a particular phase. The output from the program 
is a set of reports which include high-level project summaries, estimates 
at project, phase and task levels and reports which contain the data on 
which the estimates are based. These estimates can be loaded directly into 
other systems such as Lotus 1-2-3 or Project Manager Workbench (PMW). There 
is a close relationship between the number of outputs the system produces 
and the effor required to produce these reports. This makes it possible to 
produce a high-level, soft, cost estimate before carrying out even a 
feasibility study. The hard estimates which allow easy comparisons between 
the new project and a selection of old projects can provide a useful 
framework for calibrating the costing of new projects. BIS/Estimator can 
explain its estimates using heuristic knowledge encoded in rules associated 
with each task. It also draws expertise from a number of other sources, 
including statistical knowledge, scheduling, and mathematical techniques 
such as the Function Point Analysis method. 
	<Developed by> BIS Applied Systems Ltd.
	<Country> UK
	<Prolog system> LPA Prolog Professional
	<Other languages> C
	<Platform> IBM PC
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> BUP
	<Number> 248
	<Domain> computational linguistics
	<Purpose> parsing of natural language
	<Description> A bottom-up parser for natural language.
GALOP was developed from this (see Fuchi, Furukawa, 1986) 
	<Developed by> Y. Matsumoto et al
	<Organization> ICOT
	<Country> Japan
	<References> Y. Matsumoto, H. Tanak, H. Hirakawa, H. Miyoshi, H. 
Yasukawa: BUP A Bottom-up parser embedded in Prolog. New Generation 
Computing. 
</Program>

<Program> Bus-guide
	<Number> 111
	<Domain> Transport information
	<Purpose> An interactive information terminal for Public Transport users
	<Description> Submitted to PAC. Paper 504
	<Developed by> Le Dizes
	<Country> France
	<Prolog system> Prolog II
	<Platform> "Sun, Mac IICX"
	<Information Date> 6/11/91
</Program>

<Program> BUT
	<Number> 112
	<Domain> PLAN
	<Purpose> A Prolog-Based Expert System for Underground Illumination in 
Coal Mines
	<Description> Submitted to PAC. Paper 251
	<Developed by> Burgard, Luttrinhaus, Plumer 
	<Country> Germany
	<Prolog system> LPA Prolog.
	<Platform> PC MS-DOS
	<Information Date> 25/10/91
</Program>

<Program> CAD tools in electronic engineering
	<Number> 289
	<Domain> 
	<Purpose> To apply logic programming for CAD tools in electronic 
engineering
	<Description> The system consists of two main modules:
- a module for supporting the design and checking of printed circuit 
boards;
- a module that synthesises circuits for given Boole-functions.
	<Developed by> K. Pastrone-Varga 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> K. Pastrone-Varga (1981)
A solution of a CAD problem in PROLOG.
In Workshop of Logic Programming for Intelligent Systems, 1981
Los Angeles,California,US.
</Program>

<Program>  CAD tools in mechanical engineering
	<Number> 287
	<Domain> CAD
	<Purpose> To provide a set of CAD tools for mechanical engineering
	<Description> The system explores PROLOG for the following 
submodules:
-modelling machine parts with the aim of helping classification of 
machine parts;
-supporting design of production control system: scheduling a shop-
floor level production control of an integrated manufacturing system;
-designing fixtures from a bounded set of elements.
	<Developed by> J. Farkas, J.Fileman, A. Markus, Z. Markusz 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> J. Farkas, J.Fileman, A. Markus, Z. Markusz (1982)
Fixture design by PROLOG.
MICAD-82,Paris,France.
B.E. Molnar, A. Markus (1981)
Logic programming in the modelling of machine parts.
Proc. of Compcontrol'81, Varna,Bulgaria.
</Program>

<Program> CADRAN
	<Number> 311
	<Domain> CAD.
	<Purpose> Transformation of CAD Data.
	<Description> The systems transforms data between different CAD 
systems.
Complex design projects often require the combination of several CAD 
systems (Design, simulation, Finite Elements).  The transformation of data 
between these CAD systems is crucial to the success of the design project.  
Due to the lack of standard-interfaces, especially in 3D Systems, an 
automatical transfer of data was not possible.  With the CADRAN system it 
is now possible to transfer data from one CAD system to another without the 
loss of information, even if both systems do not have the same 
functionality and complexity.  The core of CADRAN is a knowledge -based 
system written in PROLOG.  It incorporates transformation rules how to 
represent objects in various CAD systems.
	<Developed by> Dr.  Uwe Crenze.
	<Organization> Interface Computer GmbH.
	<Country> Germany.
	<Prolog system> IF/PROLOG.
	<Other languages> FORTRAN.
	<Platform> DEC 3100
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/90
	<No. of Users> 5
	<Information Date> 14/02/92
	<References> Wissensbasierter Datenaustausch zwischen CAD-
Systemen CIM 5/92.
R. Oldenbourg Verlag, Munich.
</Program>

<Program> Calculation of Physico-chemical Parameters
	<Number> 275
	<Domain> Pharmaceutical industry
	<Purpose> To calculate the physico-chemical parameters for drug design 
purposes
	<Description> The system calculates the physico-chemical parameters for 
drug design purposes. In computer aided drug design, a considerable part of 
the calculations is based on the so-called logP value of components, a 
value indicating their lipophillic character. 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> F. Darvas, I.Futo, P. Szeredi (1978) 
Some application of theorem proving based in QSAR: automatic 
calculation of molecular properties and automatic interpretation of 
quantitative structure-activity relationships.
Proc. of the symposium on Chemical Structure -
Biological activity: Quantitative approaches, Suhl, Germany, 
Akademie Verlag, Berlin, 1978, pp. 251-257.
</Program>

<Program> CAPS - Computer Aided process Scheduling
	<Number> 24
	<Domain> Intelligent Databases
	<Purpose> Shop Floor planning; scheduling and monitoring
	<Description> CAPS is a system for production shop floor planning. It 
operates as an intelligent front end to MRP/MPS production planning systems 
for Computer Integrated Manufacturing (CIM); with an  interface to shop 
floor data aquisition equipment. CAPS Represents a new development in 
production shop floor planning; where logical constraints on processes and 
resources are combined with rules from an expert system. These rules are 
suited for each production plant and represent specific knowledge of how to 
create and revise production plans.
	<Developed by> Kim Thorshoej; Lars Henrik Jacobsen
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Lines> >10.000  <100.000
	
	<Platform>  OS/2 PM
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 09/91
	<No. of Users> 10
	<Information Date> 01/92
</Program>

<Program> CAPTAINS
	<Number> 476
	<Domain> Scheduling.
	<Purpose> Pilot scheduling for airlines.
	<Description> Scheduling of Pilots, instructors and other critical 
resources, like simulators, on a medium to long-term scale. The program is 
capable of incorporating  large numbers of constraints, such as flight 
staff union rules, safety regulations and annual leave rights.
	<Developed by> Large team effort.
	<Organization> Syllogic.
	<Country> Netherlands.
	<Prolog system> BIM Prolog (OEMed by IBM).
	<Other languages> SQL (Oracle), Xwindows + motif, Syllogic Knowledge Sheet, 
C, Operations Research Lib.
	<Lines> 110.000
	<Predicates> 4.500
	<Platform>  IBM RISC System 6000.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> September 1992.
	<No. of Users> 10
	<Information Date> 25 november 1992.
</Program>

<Program> CARE-X25 (version included in RETAT)
	<Number> 512
	<Domain> Telecom Network Design & Planning.
	<Purpose> Propose minimal cost equipments for nodes and links an X25 
private network according to X25 needs and engineering constraints.
	<Description> A network study with CARE-X25 (included in RETAT 
tool) to:
-verify the X25 market equipments and data transmission links 
characteristics and tariffs,
-verify the engineering constraints : minimal equipment and 
estension, maximal load,
-choose minimal cost links and X25 switch equipments according to 
nodes and links loads,
-list links and X25 equipments proposed for the network with their cost.
	<Developed by> H. FREI
	<Organization> H. FREI
	<Country> France.
	<Prolog system> Prolog II+-Prolog 1A
	<Lines> 9400
	<Predicates> 680
	<Platform> MacOS-Syst7 on MacII (68030)
Windows 3.1 on PC-386
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 5/91
	<No. of Users> 50
	<Information Date> 2/93
	<References> L'echo des recherches No 141-3eme trimestre 1990
"Planification du reseau interne de France Telecom" Pages 51 -60.
DICET/SMG - Centre National d'Etudes des Telecommunications 38-40, rue du 
General Leclerc - 92 131 Issy-les Moulineaux.
</Program>

<Program> CAS/FPS - Computer-Aided  Synthesis  of  Flexible Production 
Scheduling.
	<Number> 488
	<Domain> Production Planning  and  Scheduling  of  Multiproduct  (Batch) 
Plants.
	<Purpose> Multicriteria  design  and  control  of  flexible  
multiproduct
(batch) plants by computer-aided synthesis of the production plans and 
schedules from the possible building elements.
	<Description> In the traditional methods of  synthesizing  the  
multicriteria
suboptimal production plans and schedules, the possible  variants are 
represented by the partial structures of a previously defined 
superstructure. The problem size is limited by the  combinatorial 
explosion accompanying the  heuristically  controlled  systematic 
search. Exact optimization is generally addressed  using  various 
methods of the operational research. The multicriteria situations 
should  be  described  by  a  single  aggregated  goal  function,
supplemented by optional limiting constraints.
     In the PROLOG  represented  structural  models  the  various 
solutions are synthesized from  the  "free"  active  and  passive 
elements of the structural model. The memory requirement and  the 
combinatorial complexity can be reduced by the use of prototypes, 
describing "blank" activities and variable  or  virtual  storage.
The multicriteria evaluation can be carried  out  without  any  a
priori  aggregation  or  ranking  of  the  individual  goals  and
constraints. A set of flexibly modifiable suboptimal variants can be 
synthesized and the subjective human  decisions  can  also  be taken 
into consideration.
     The structural models are described by dynamically  declared clauses 
determining the activities and storage, while the various tasks are 
executed by  the  general  kernel  definitions  of  the metainterpreter 
operating above these databases.  The  activities and storage functions can 
be created, modified and deleted either through  a  user-friendly  
interface  or  by  the  support  of  a full-screen editor. The core 
definitions can be  supplemented  by the user-defined includes.
	<Developed by> Bela Csukas, Zoltan Kozar, Rozalia Lakner, Gabor 
Wittinger
	<Organization>  University  of  Veszprem,  supported  by  the  Hungarian  
Basic Research Grant N OTKA-2550.
	<Country> Hungary.
	<Prolog system> TURBOPROLOG 2.0 ; PDC PROLOG 3.31
	<Other languages> Can also be implemented in object oriented languages.
	<Lines> 4000
	<Predicates> 500
	<Platform> PC/AT 386 
	<Current Status> Evaluation
	<Date deployed> 12/1989
	<No. of Users> 20
	<Information Date> 12/1993
	<References>  Csukas B., Kozar Z., Arva P.:  Multicriteria  
Evaluated  PROLOG
Synthesizing Algorithms, Comput. Chem. Engng., 13, 595-602 (1989).
Csukas B., Lakner  R.,  Wittinger  G.:  Multicriteria  Suboptimal
Design and Control of Flexible (Batch) Plants. In:  L.  Puigjaner
and  A.  Espuna  Eds.:  Computer  Oriented  Process  Engineering,
Process Technology Proceedings 10, Elsevier, Amsterdam-London-New
York-Tokio, 1991, pp. 341-346 
</Program>

<Program> CASEy
	<Number> 82
	<Domain> expert system, manufacturing
	<Purpose> Expert System that guides shop floor personnel in the
correct usage of electrical process specifications.
	<Description> Connector Assembly Specifications Expert  provides 
the detailed assembly instructions for terminating wire into 
electrical interconnects such as connectors, terminal lugs, splices, 
etc.  It provides a prioritized list of authorized tools, settings, 
materials and procedures (with graphics) needed to perform the task.
In practice the system reduces the time for an operator to assemble a set 
of instructions from 42 to 5 minutes.
	<Developed by> Bill McClay, G. Craig Johnson, R. Howard Mitchell,
Dennis Hegedus, Michael Easter, Steve Miller,
Max Robinson
	<Organization> Boeing Computer Services
	<Country> USA
	<Prolog system> Quintus 3.1
	<Other languages> 1,000 lines of C
	<Lines> 104,000 lines of Prolog
	<Predicates> approximately 25,000
	<Platform> Sun 4, VAX, RS6000
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> November 1986
	<No. of Users> Shop floor personnel
	<Information Date> March 18, 1992
	<References> "Harnessing Detailed Assembly Process Knowledge" 
presented at the Conference on Innovative Applications of Artificial 
Intelligence, March 1989, sponsored by the American Association of 
Artificial Intelligence
</Program>

<Program> Cast Component
	<Number> 210
	<Domain> Manufacturing design
	<Purpose> Optimise component design time in motor industry and check 
casting
	<Description> The system works on a 3-dimensional grid 
representation of a component to check manufacturing requirements. It 
simulates the component cooling, predicts possible casting defects 
which are diplayed in 3-D. The operator can then modify the casting's 
characteristics to eliminate the defects, or the expert system can 
indicate this automatically.
The original conventional computerised method required several days to run, 
but recent progress has reduced this to just a few minutes.
	<Organization> Renault Motor Co.
	<Country> France
	<Prolog system> Delphia
	<Other languages> 13,000 lines C
	<Lines> 12,000
	<Platform> work station
</Program>

<Program> CAT2
	<Number> 452
	<Domain> Fully automatic translation of informative texts;
emphasis has been on European Commission texts, as well as general purpose 
texts.
	<Purpose> Multilingual machine translation.
	<Description> CAT2 is a unification-based natural language 
processing (NLP) system, designed for analysis, generation and 
translation of natural language sentences.  It embodies a particular 
formalism for NLP, as well as a grammar development environment.  The 
formalism is similar to PATR-II, in which linguistic information is 
represented by sets of attribute-value pairs, in positive, negative, 
disjunctive, or implicative relation.  Translations are defined as 
tree-to-tree transductions.  As a development tool, it includes means 
for loading and unloading grammars, reading/writing 
texts/translations from/to terminal or file, creating linguistic 
objects as well as displaying,
deleting, storing, retrieving and comparing them.  Grammars have been 
written for English, German, French, Spanish, with experimental versions 
for Russian, Greek, and Japanese.
	<Developed by> Randall Sharp.
	<Organization> IAI (Institute for Applied Informatics).
	<Country> Germany.
	<Prolog system> SICStus 2.1
	<Lines>  Approximate number of lines of Prolog.
	<Predicates> 350
	<Platform> UNIX; system has been developed on SUN Workstations.
	<Current Status> Prototype
	<Information Date> October 26, 1992.
	<References> Sharp, R. (1988). "CAT2-- Implementing a Formalism
for Multi-Lingual MT".  Proceedings of the 2nd
International Conference on Theoretical & Methodological Issues in 
Machine Translation of Natural Language, Pittsburgh, PA.  
Sharp, R. (1991). "CAT2: An Experimental Eurotra
Alternative". Machine Translation 6:215--228.
</Program>

<Program> CATS/DIANA
	<Number> 230
	<Domain> Electronic diagnosis
	<Purpose> diagnosis of analog electronic circuits
	<Description> The system uses model based diagnosis, in which a 
model of the correct behaviour of different components of the system 
using a structural description of the system is compared with 
observations. Any conflicts between these two can be analysed, 
without needing explicit lists of possible faults.
CATS is a general conflict recognition program designed for 
continuous dynamic systems: values of physical quantities are 
represented by interval arrays to take into account both dynamic 
aspects and uncertainty. 
DIANA is the application to analog electronic circuits, providing a 
library of models of electronic components, an incremental 
description interface and a connection to CATS entry points.
Conclusive industrial experiments have been carried out on hybrid 
(digital/analog) boards and on the electric power subsystem of a satellite.
	<Organization> IBM Paris Scientific Center, Dassault Electronique
	<Country> France
	<Prolog system> IBM Prolog and IPW
	<Platform> 370, OS/2
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> CELLBUILDER.  It comprises several programs, the main one of 
which (and the only one using Prolog) was called cellmapper.
	<Number> 482
	<Domain> Communications
	<Purpose> To generate optimal cell layouts for the cell library of the 
Crosspoint CP20K family of field-programmable gate arrays (FPGAs), starting 
from a transistor- or gate-level schematic.
	<Description> Of me, or the program?
	<Developed by> Howard A. Landman
One translator program was written by John McDonald.
	<Organization> At time of program development: Crosspoint 
Solutions, Inc.
Currently: HaL Computer Systems, Inc.
	<Country> USA
	<Prolog system> Initial feasibility testing was done using public domain 
Prolog on an IBM PC.  When the limits of this systems were reached, I 
switched to CProlog on a SPARCstation.
ÇIF*** The text replacing this argument is too long ***È<Other 
languages> Somewhat over half the system was written in Perl.  The 
cellmapper program is structured as an "oracle machine", with a Perl 
top level communicating with a Prolog "oracle" through a pair of 
pipes.  This was a nice mix of languages: The fabulous I/O and text 
processing capabilities of Perl allowed me to ignore the rather lame 
I/O of Prolog to a large extent and focus the Prolog engine entirely 
on search problems.  Meanwhile, Perl took care of all the file 
munging, text
formatting, error handling, etc.
This division of labor not only enforced a clear modular structure, but had 
the minor advantage that on a multiprocessor, the Perl portion and the 
Prolog portion could run on different processors.  This gives a speedup 
which ranges from negligible (for large cells with few solutions, which are 
compute-bound in Prolog) to perhaps 25% (for small cells with many 
solutions, which exercise the Perl portion more).
	<Lines> A couple thousand each of hand-coded Perl and Prolog, plus a 
fair amount of Prolog "facts" which were generated automatically by Perl 
scripts from various databases.
	<Predicates>  A reasonably complete description of the system is 
available inside Crosspoint, but unfortunately it is proprietary at the 
moment.  I may try to write this up for publication soon.
	<Platform> SPARC + Sunos.
	<Type of Use> 
	<Date deployed>  First prototype, June or July 1991.
		  Replaced human cell designer, August 1991
		  First customer ship of cell library built with system, February 
1992
	<No. of Users> The author, and other members of Crosspoint's Design 
Technology group.
</Program>

<Program> Chemical Information System
	<Number> 272
	<Domain> Chemistry
	<Purpose> Provide logic based tools for the chemical domain
	<Description> The system used in combination with fortran programs that 
aid statistical calculations on chemical structures. Prolog is then used to 
supply inference based on the properties of these chemical and biological 
structures. The system provides an environment for uniform knowledge 
representation and information handling of structures and properties.
	<Developed by> F. Darvas, I. Futo, J.Szeredi,  J. Redei, J. Bendl, P. 
Koves. 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> F. Darvas, I. Futo, J.Szeredi, J. Bendl, P. Koves 
(1978)
A PROLOG based drug design system.
Proc. of the Conf. of Programming Systems '78.
Szeged (Hungary), 1978, pp.119-126.
F. Darvas, I. Futo, J.Szeredi,  J. Redei (1979)
A Logic based chemical Information System -
theoretical considerations and experiences.
Proc. of National Conf of Von Neumann Computer Science Society, 
Szeged (Hungary), 1979, pp. 92-96.
</Program>

<Program> Chemical resource allocation
	<Number> 257
	<Domain> Manufacturing planning
	<Purpose> resource allocation in chemical plants
	<Description> The system allocates equipment to some given set of 
processes. This performs orders of magnitude faster than some other 
constraint satisfaction techniques.
	<Developed by> David A. Rosenblueth
	<Organization> IIMAS ?
	<Country> Mexico
	<References> D.A. Rosenblueth: Intervals as Domains for solving 
Resource-Allocation problems in a chemical plant. Abstracts of poster 
sessions, Int. Conference on Practical Application of Prolog, London 1992.
</Program>

<Program> ChemiSafe
	<Number> 23
	<Domain> Intelligent Databases
	<Purpose> Handling of directions for use for hazardous chemicals
	<Description> ChemiSafe is used to produce directions-for-use and 
labels for hazardous chemicals according to the complex rules for 
standardized risk and safty marking.
	<Developed by> Henrik Zangenberg
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Lines> 10.000  <100.000??
	<Platform> DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 05/88
	<No. of Users> 100
	<Information Date>  01/92
</Program>

<Program> CK Expert Modeller
	<Number> 186
	<Domain> Design
	<Purpose> To interpret and manipulate data from the CK cognitive Modeller 
system
	<Description> 
	<Developed by> Rodney Myers
	<Organization> Hodgson Myers Associates
	<Country> UK
	<Prolog system> LPA Prolog Professional
	<Platform> IBM PC
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program>   CLARE/CLE
	<Number> 360
	<Domain> Natural Language Processing, particularly interactive.
	<Purpose> A research system for the interactive analysis and generation 
of natural language.
	<Description>  contact SRI for further details.
	<Developed by>  Researchers at SRI International Cambridge.
	<Organization>   SRI International, DTI, BP, BP and DRA.
	<Country> UK.
	<Prolog system> Quintus 3.1, 2.5.
	<Other languages> C (< 1%).
	<Lines> 100,000.
	<Predicates> 5000.
	<Platform> Sun and other.
	<Type of Use> Licensed
	<Current Status> Prototype
	<Date deployed> 6/1991
	<No. of Users> 3
	<Information Date> 4/1992
	<References> Alshawi, H. (ed.) "The Core Language Engine", MIT 
Press,
Cambridge, Mass. 1992.
</Program>

<Program> CLIENT RISK MANAGEMENT
	<Number> 389
	<Domain> Risk handling in Bank front office, middle office and central 
office.
	<Purpose> Minimize risk in Day-to-Day operations with client companies in 
banking institution.
	<Description> Expert system for global treatment of risk under 
development for Banc Sabadell.
	<Developed by> Xavier Fustero, Xavier Orteu.
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.31 for Windows.
	<Other languages>  C
	<Lines> unknown.
	<Predicates> unknown.
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS and Windows
	<Type of Use> Sold Product
	<Information Date> 07/92
</Program>

<Program> CLINT
	<Number> 113
	<Domain> Learning
	<Purpose> An Overview of the Interactive concept-Learner Theory Revisor
	<Description> Submitted to PAC. Paper 375
	<Developed by> Raedt, and Bruynooghe
	<Country> Belgium
	<Prolog system> LPA MacProlog
	<Platform> Apple Macintosh II
	<Information Date> 6/11/91
</Program>

<Program> Cobol Program Generator
	<Number> 279
	<Domain> Software applications
	<Purpose> To generate Cobol Programs for data processing
	<Description> The system generates Cobol Programs according to 
the standards of SZKI. The generated programs solve data processing 
problems of the following types:
-listing data files
-data maintenance
-merging two data files
-validation of primary input data
The structure of the input/output data maintained by the generated Cobol 
programs and the work to be done can be specified by uniform parameters.
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> I. Futo, J.Szeredi, J. Redei (1979)
Generating COBOL programs complying with the COLAMI standards.
User's manual.
SzKI report, 1979.
</Program>

<Program> COMANDOS
	<Number> 114
	<Domain> OO
	<Purpose> A Prolog Object Oriented Persistent Environment
	<Description> Submitted to PAC. Paper 244
	<Developed by> A. El Habbash et al
	<Country> Ireland
	<Prolog system> C-Prolog
	<Information Date> 25/10/91
</Program>

<Program> COMBAT
	<Number> 319
	<Domain> Simulation.
	<Purpose> Simulation of a ground-air battle.
	<Description> This application uses CHIP to assign missile launchers to 
attacking planes in a simulation of a ground-air battle.  A scenario is 
generated graphically by the user by placing radar units and defence 
batteries around a target.  Then, the user places attacking planes 
configurations with their anticipated strike paths.  The system is then 
able to simulate the attack scenario based upon this configuration.
	<Developed by> COSYTEC
	<Organization> CLP Software and Services.
	<Country> France 
	<Other languages> C  X11
	<Platform> SUN/HP/IBM/DIGITAL workstations
PC  386/486		 UNIX   DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 12/91
	<Information Date> 03/92
	<References> M. Dincbas, P. VAn Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Int. Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> COMMUNAL
	<Number> 115
	<Domain> 
	<Purpose> The COMMUNAL Project
	<Description> Submitted to PAC. Paper 82
	<Developed by> Tim F O'Donoghue
</Program>

<Program> Conceptual Graph Editor (CGE)
	<Number> 417
	<Domain> Knowledge Representation.
	<Purpose> To easily edit and build conceptual graph knowledge bases.
	<Description> CGE is a graphical interface to the Conceptual 
Graph Tools (see the CGT entry) for X Windows environments.
Multiple windows may be opened on the same graph
database. Each window is an editor which uses
simultaneously the graphical and linear notation. CGE makes use of 
the mouse, menus, buttons, dialog boxes,
etc. to provide the user *only* with the operations
supported by CGT, which are higher-level than primitive
operations like "draw an arrow". As such, at any given
time, the displayed graphs are canonical (in the sense of
Sowa's theory). CGE can be considered to be a "syntax-oriented" 
editor. 
CGE uses the Graph Widget which is responsible for the
actual display of the graphs. CGE only works with X
Prolog which provides declarative access to the X Windows
Toolkit through the Widget Description Language (widget =
window gadget; X Windows terminology for "graphical
object").
	<Developed by> Michel Wermelinger.
	<Organization> AI Centre/UNINOVA.
	<Country> Portugal.
	<Prolog system> Quintus X Prolog 2.4 
	<Other languages> C (used to program the Graph Widget and X Prolog).
	<Lines> 1700 (not including CGT code).
	<Predicates> 106
	<Platform> VAX, Ultrix 3.1
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> February 1991
	<No. of Users> 1
	<Information Date> February 1992
	<References> Salvador Abreu
ALPES X-Prolog Programming Manual,
Centro de Intelig\^encia Artificial, UNINOVA, 1989
Paulo Quaresma,
Graph Widget: A Tool for Automatic Data Visualization, Technical 
Report RT-6/91-CIUNL, April 1991
Centro de Informatica da Universidade Nova de Lisboa
see also CGT entry
</Program>

<Program> Conceptual Graph Tools (CGT)
	<Number> 416
	<Domain> Knowledge Representation.
	<Purpose> to have a partial implementation of Sowa's Conceptual 
Structures.
	<Description> As Sowa puts it, Conceptual Structures "are a 
system of
logic with a graph based formalism that aims for a very wide 
expressive power.  Its primary purpose is to serve as an intermediate 
language between natural language and other
formalisms including database query languages [...] and
predicate calculus."
CGT includes predicates to implement the most important
operations on conceptual graphs, like the canonical
formation rules and the propositional inference rules. CGT
reads and writes conceptual graphs using their linear
notation. It also provides facilities to manipulate graph
databases. A sample one comes with CGT.
	<Developed by> Michel Wermelinger.
	<Organization> AI Centre/UNINOVA
	<Country> Portugal.
	<Prolog system> Quintus Prolog 2.4 
(should work with any "Edinburgh" Prolog with minor 
modifications).
	<Other languages> none.
	<Lines> 3679 (including comments; 
	<Predicates> 193
	<Platform> VAX, Ultrix 3.1 (or any other having a fairly "standard" Prolog 
system)
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> February 1991
	<No. of Users> 1
	<Information Date> February 1992
	<References> John F. Sowa,
Conceptual Structures: Information Processing in Mind
and Machine, Addison-Wesley Publishing Company, 1984.
Michel Wermelinger,
GET: Graph Editor and Tools---The Incomplete Reference, Tecnical 
Report RT-3/91-CIUNL, January 1991,
Centro de Informatica da Universidade Nova de Lisboa.
Michel Wermelinger,
GET - Some Notes on the Implementation, Tecnical Report RT-4/91-
CIUNL, January 1991, Centro de Informatica da Universidade Nova de 
Lisboa
Michel Wermelinger, 
GET: An Implementation of Conceptual Graphs, Tecnical Report RT-5/91-
CIUNL, January 1991,
Centro de Informatica da Universidade Nova de Lisboa
Michel Wermelinger and Gabriel Pereira Lopes,
 A Tool for Knowledge Acquisition and Representation
based on Conceptual Graphs, Published in Portuguese in the 
Proceedings of the Eighth Brazilian AI Symposium,
November 1991.
</Program>

<Program> CONSULT
	<Number> 116
	<Domain> CAD
	<Purpose> Object-Oriented Interactive CAD/CAM Programming in Prolog
	<Description> Submitted to PAC. Paper 502
	<Developed by> Jarmo T Alander
	<Country> Finland
	<Prolog system> Turbo
	<Platform> PC
	<Information Date> 31/10/91
</Program>

<Program> Contralto
	<Number> 117
	<Domain> Transport
	<Purpose> Constraint Reasoning applied to Logistics of Transport 
Organisations
	<Description> Submitted to PAC. Paper 503
	<Developed by> Le Dizes
	<Country> France
	<Prolog system> Prolog III
	<Information Date> 6/11/91
</Program>

<Program> CONTRALTO
	<Number> 172
	<Domain> Transport planning
	<Purpose> To help freight transporters find a solution which best matches 
demand and usage constraints
	<Description> CONsTraint Reaasoning Appliged to Logistics for 
Transport Organisations
Demand constraints are specified by the physical characteristics 
(type, weight, height) and other characteristics such as delivery 
date, cost, origin-destination.
Bidding constraints include the loading of vehicles,  routing, as 
well as other constraints such as departure times, with respective 
delays, the compatibility of loads, the availability of material and 
personnel, regulations.
	<Developed by> C. Guimaraes, J-M Le Dizes
	<Organization> Cete-Méditerranée
	<Country> France
	<Prolog system> Prolog III
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> Coronary Network reconstruction
	<Number> 237
	<Domain> Medical investigation
	<Purpose> 3-D reconstruction and labelling of coronary networks from X-
ray projections
	<Description> The program reconstructs a three-dimensional image 
of coronary networks from two simultaneous X-Ray projections. The 
procedures in the reconstruction-labelling process deal with the 
correction of distortion, the detection of centre-lines and 
boundaries, the derivation of 2-D branch segments whose extremities 
are branching, crossing or end points and the 3-D reconstruction and 
display.
All algorithmic components of the reconstruction were written in the C 
language, whereas the model and resolution processes were represented by 
predicates and production rules in Prolog. The user interface, which 
includes a main panel with associtated control items, was developed using 
Carmen, the Prolog by BIM user interface generator.
	<Developed by> M. Garreau, C. Chardenon, A. Rougée, J.L. Coatrieux
	<Organization> General Electric, University of ?
	<Country> France
	<Prolog system> Prolog by BIM
	<Type of Use> In-house
	<Current Status> Prototype
</Program>

<Program> COTORRA
	<Number> 383
	<Domain> Train dispatching.
	<Purpose> Assistance for the operator in charge of trains line 
information in order to improve the service quality.
	<Description> System for assisting the dispatching information operator 
at the Ferrocarrils de la Generalitat Catalunya. System under installation.
	<Developed by> Xavier Orteu and programmers team.
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.3 for Windows.
	<Other languages> C
	<Lines> 5205
	<Predicates> 150
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 /MS DOS and WIndows
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 8/92
	<Information Date> 07/1992
</Program>

<Program> CPE
	<Number> 41
	<Domain> Knowledge Based Programming
Automation of Programming
Engineering Problems
	<Purpose> CPE is a knowledge programming environment. It supports 
different programming styles: logic programming; constraint 
programming;
object-oriented programming; database programming; etc.
	<Description> CPE (Conceptual Programming Environment) includes 
tools for representing; interpreting; intergrating and testing the 
different knowledge sources: graphical objects; deterministic 
constraints; production rules; data tables; online help; diagrams; 
reports; etc. The knowledge listed above are represented in PROLOG; 
interpreted by metaprograms and integrated by a blackboard.
CPE is being used in the J/V IE Software.
The conceptual programming method in PROLOG was developed in 1987. CPE was 
designed in 1988 and its prototype was implemented in 1989. Versions 1. and 
2. were implemented in 1990 and 1991. CPE was developed and it is supported 
by one programmer.
	<Developed by> Andres Lomp
	<Organization>  J/V IE Software joint venture between the Institute of 
Cybernetics (Estonia) and the TIW Systems (USA) 
	<Country> Estonia
	<Prolog system> LPA Prolog Professional 3.0.
	<Other languages> C
	<Lines> 15000 lines in PROLOG
	<Predicates> 150 new predicates
	<Platform> IBM PC; MS DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 02.04.1990
	<References> Andres Lomp; Peep Eomois. Object-Oriented Programming in 
CPE. In: Proceedings of the Second Symposium on Programming Languages and 
Software Tools. Pirkkala; Finland; August 21-23; 1991.
</Program>

<Program> CPLAN
	<Number> 317
	<Domain> Planning.
	<Purpose> Project management system.
	<Description> CPLAN is an intelligent decision support system for 
solving project managment tasks.  The system takes into account precedence 
and distance constraints between tasks as well as the complex resource 
constraints induced by scarce resources.
	<Developed by> COSYTEC
	<Organization> CLP Software and Services.
	<Country> France.
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 2000
	<Platform> SUN/HP/IBM/DIGITAL workstations
PC  386/486		    UNIX    DOS
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 01/92
	<No. of Users> 2 
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc.  of The Int. Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, November 1989.
</Program>

<Program> CPPRO
	<Number> 478
	<Domain> Test Engineering - diagnosis and repair recommendation of 
faulty PBX circuit boards.
	<Purpose> Circuit Board Diagnostic Troubleshooter Tool.
	<Description> CPPRO allows a circuit board test engineer to 
graphically
build a diagnostic tree composed of nodes and arcs to aid in the 
diagnosis and repair of PBX circuit boards.  Nodes of the tree 
specify diagnostic tasks a technician must perform on the board when 
it is plugged into a troubleshooting test set.  Nodes also contain 
automatic messages sent to the test set to exercise board circuitry.
Leaf nodes contain recommended board repair actions. Arcs emanating from 
nodes are labelled with possible answers to the questions asked at the 
nodes. When used by the technician, CPPRO executes the diagnostic tree, 
asking questions of the technician, specifying probe actions, graphically 
displaying valid scope traces and component and pin locations on the 
circuit board, and making repair action recommendations.
	<Developed by> Dan M. Wages, Russell G. Bulsis, Kevin B. Tanguay, Kent 
D. Fishman
	<Organization>  AT&T - Denver Works, Denver, Colorado.
	<Country> U.S.A.
	<Prolog system> Quintus Prolog Release 3.1.1
	<Other languages> Quintus ProWindows 1.2 - SunView Version.
	<Lines> 10000
	<Predicates> 1000
	<Platform> Sun SPARCstation 1+ running SunOS 4.1.1
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 9/90
	<No. of Users> 5
	<Information Date> 11/25/92
</Program>

<Program> CVE
	<Number> 203
	<Domain> Digital design
	<Purpose> A tool for symbolic verification of digital circuits
	<Description> The Circuit Verification Environment is a tool 
which applies symbolic techniques to the verification of a digital 
circuit. It is based on a new principles that make it completely 
reliable and much faster than conventional simulators. The 
verification procedure realised by CVE computes the output behaviour 
of a digital circuit as boolean functions of the inputs. Sequential 
circuits are treated as finite automata.
At present CVE can handle ASICs with up to several tens of thousands 
of cells - typical for the functional blocks in today's ASICs, or up 
to approximately 15,000 transistors in full custom circuits.
CVE includes a compiler for EDIF netlists and for the CN format of SIGRAPH-
EL as well as several cell library descriptions.
	<Developed by> R Schmid, E. Tidén
	<Country> Germany
	<Prolog system> SNI (Siemens)
	<Other languages> C
	<Platform> SICOMP WS30 workstation under SR 10
	<References> R. Schmid, E. Tidén. Verifying ASICs by Symbolic 
simulation. in European ASIC 90 Conference proceedings, IEEE order no. 
2066. Paris 1990.
</Program>

<Program> D- PROLOG
	<Number> 298
	<Domain> 
	<Purpose> Prolog interpreter with Edinburgh syntax
	<Description> D-Prolog is a prolog interpreter with Edinburgh syntax 
developed inside DCL to allow full interoperability among C, common Lisp, 
Clos and Prolog.
	<Developed by> Delphi S.p.A.
	<Organization> Delphi S.p.A.
	<Country> Italy
	<Other languages> C/CL/CLOS
	<Platform> Sun with Sundos 4.x
PC 386/486 with SCO/ODT 1.0
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1/1/91
	<No. of Users> 80
	<Information Date> 02/12/91
	<References> Article by Messrs. Saracco & Gaspari
</Program>

<Program> DACH (Datenverarbeitung am Computer zu Hause)
	<Number> 470
	<Domain>  Training
	<Purpose>  Tutorial for office communication and long distance training.
	<Description> 
	<Developed by> Dr Hans Juergen Schlaesser.
	<Organization> German Ministry of Education and Science.
	<Country> Germany.
	<Prolog system> PDC Prolog 3.3 (Windows).
	<Other languages> C
	<Lines> 10.000
	<Platform> 80286, MSDOS 4.0
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Information Date> 8/92
</Program>

<Program> DAMOCLES - Damage Monitoring and Control Expert System.
	<Number> 329
	<Domain> Damage control, fire-fighting, stability and buoyancy.
	<Purpose> Damocles is supposed to aid a damage control officer aboard a 
Standard frigate of the Royal Netherlands Navy.
	<Description> TNO Physics and Electronics Laboratory, in collaboration 
with the NBCD School of the Royal Netherlands Navy, has developed Damocles, 
a Damage Monitoring and Control Expert System.  The main purpose of the 
Damocles project has been the development of a prototype expert system 
which supports the damage control (DC) officer aboard Standard frigates in 
maintaining the operational availability of the vessel by safeguarding it 
and its crew from the effects of weopons, collisions, extreme weather 
conditions and other calamities.  Basically DC-management includes the 
classical command and control cycle:  status maintenance, situation 
assessment, planning, tasking and evaluation.  An important way of making 
the total DC-organisation more effective is to improve the quality of the 
decisionmaking process by providing automated decision aids to the DC-
officer in addition to the information processing and presentation 
facilities already available.  This applies especially to damage assessment 
and planning.
	<Developed by> Command and Control Group of the TNO Physics and 
Electronics Laboratory.
	<Organization> Sponsored by the Royal Netherlands Navy.
	<Country> The Netherlands.
	<Prolog system> Quintus Prolog 3.1, Quintus ProWINDOWS 1.2
	<Other languages> None.
	<Lines> 68,997
	<Predicates> 2,069
	<Platform> Sun
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 1988
	<Information Date> 07/92
	<References> Bruin, J., M. Perre and J.J.C.R. Rutten, Damocles, 
een relationeel expertsysteem, In:  Proceedings AI Toepassingen '88, 
w.pl., 1988, NGI-SIC, p.  217-219, Dutch.
Bruin, J., and J.J.C.R. Rutten, Intelligent Decision Support for 
Warship Damage Control, In:  Expert systems and their applications, 
Avignon, 1988, European Coordinating Committee for Artificial 
Intelligence.
Leeuwen, E.W.A. van, and J.J.C.R. Rutten, Beslissingsondersteuning 
met kennissystemen, In:  Syllabus studiedag 'Instrumentele 
procesbeveiliging', w.pl.,1991, NIRIA/Kivi, Dutch.
Leeuwen, E.W.A. van, DAMOCLES  hulpmiddel bij calamiteiten aan boord 
van schepen, In: TNO Magazine Toegepaste Wetenschap, w.pl., 1991, 
TNO, volume 7, p. 40-43, Dutch.
Leeuwen, Ing. E.W.A. van, and ir. J.J.C.R. Rutten, Het zwaard van 
Damocles gekeerd, In: i2- Procestechnologie, Amsterdam, 1991  
Ingenieurspers B.V., volume 10, p. V1-V6, Dutch.
Leeuwen BSc, Ed W.A. van, and Michael Perre MA, DAMOCLES: an expert 
system  for damage control management aboard standard frigates, In: 
Proceedings Artificial Intelligence and Knowledge-based Systems for 
Space, Noordwijk, 1991, ESA/ESTEC, volume1.
Leeuwen, Ed van, Kennissystemen om het hoofd boven water te kunnen 
houden, In: Kennissystemen, Rijswijk, 1992, Stam Tijdschriften BV, 
volume 7/8, p. 19-21, Dutch.
Perre, M., and J.J.C.R. Rutten, Damocles voor de 'damage control'  
organisatie op een marineschip, In Kennissystemen, Rijswijk, 1990, 
Stam Tijdschriften BV, volume 4, p. 15- 17, Dutch.
Perre, M., J.J.C.R. Rutten and D.L. Mols, Damocles: An expert system for 
damage control aboard Standard frigates, In: Proceedings Ninth Ship  
Control Systems Symposium, Bethesda (USA), 1990, p.  2.101-2.109.
</Program>

<Program> DART
	<Number> 7
	<Domain> Analysis of concurrent programs
	<Purpose> To facilitate the deadlock analysis of concurrent programs.
	<Description> Concurrent programs often have a huge number of 
states; a fact that makes their analysis extremely difficult and 
sometimes impossible. DART is a system that has been developed in 
order to reduce the effort required for a specific analysis problem; 
namely deadlock detection. The system transforms the concurrent 
program into a Petri net model and uses techniques from Petri net 
theory to reduce the size of the model. The new model of the initial 
program; contains the same deadlock information but has substantially 
less states. DART has been
implemented in Prolog and has been used to efficiently analyze programs 
that have several millions of states
	<Developed by> P. Rondogiannis and M.H.M. Cheng
	<Organization> University of Victoria
	<Country> Canada 
	<Prolog system> ALS-Prolog; version 1.01.
	<Lines> 1080
	<Predicates> 140
	<Platform> Sun3/280S; SunOS Unix; release 4.1.1
	<Type of Use> Personal
	<Current Status> Evaluation
	<Date deployed> July 1991
	<No. of Users> 2 (Has also been requested by another research 
institution).
	<Information Date> October 25; 1991 
	<References> P. Rondogiannis and M.H.M. Cheng; "Using Petri net 
reduction
techniques to detect Ada static deadlocks"; to appear in The 10th 
Annual
National Conference on Ada Technology; Virginia; February 1992.
P. Rondogiannis "Detecting Deadlocks in CCS Agents using Petri net
reduction techniques"; Master's thesis; Dept. of Computer Science;
University of Victoria; 1991. 
</Program>

<Program> DAS
	<Number> 224
	<Domain> Financial  statistics
	<Purpose> Performing statical analysis of financial information
	<Description> The Data Analysis system includes a visual programming 
interface that allows research analysts to build complete 'programs' for 
performing statistical analysis of financial information residing in 
historical databases. It incorporates a highly integrated programming 
environment including maths libraries, a statisical package and a 
spreadsheet. Its key component is a declarative graphicsal programming 
interface for adding new calculation components to the system during 
execution. This entails configuration screens, data translation mechanisms 
and module specification and viewing algorithms.
	<Organization> O'Connor Associates, Chicago
	<Country> USA
	<Prolog system> Prolog by BIM
	<Other languages> C, C++, Fortran maths libraries, S statistical package, 
Sybase, graphics package and spreadsheets
</Program>

<Program> Data flow in Parlog
	<Number> 218
	<Domain> Systems design
	<Purpose> To model and animate dataflow diagrams to produce a prototype 
system
	<Description> A prototype program can automatically be generated 
from an object model of a Data Flow diagram and translated into 
Parlog (a declarative parallel logic programming language).
The analyst can then reason about the proposed system using multi-window 
trace and debug facilities available in the Parlog environment.
	<Developed by> Kevin Steer
	<Organization> Artificial Intelligence Ltd (now Strand?)
	<Country> UK
	<Prolog system> Parlog
	<Platform> Unix
	<Type of Use> Personal
	<Current Status> Prototype
	<References> K. Steer: Testing Data Flow diagrams with Parlog. ICLP5, 
1988, pp96-110.
</Program>

<Program> DATIN
	<Number> 471
	<Domain> 
	<Purpose> Social scientific dataretrieval. Instead of filling data 
into
questionaeres it can be done direstly into the computer.
	<Description> The program can be used in any area where a person 
is filing
data on other persons (e.g. like a teacher does on the pupils) and 
works instead of a questionaere. After input and retrieval of data 
from a disk the data are ready for statistical analysis: ASCII data 
and command data are made for SPSS. The program only contains few 
features which on the other hand
makes it very stable and easy to use. No specific training is necessary. 
DATIN is e.g. being used in residential homes for elderly people.
	<Developed by> Hermann Vierke.
	<Country> Netherlands.
	<Prolog system> PDC Prolog (formaerly Turbo Prolog 2.0).
	<Lines> 1400.
	<Predicates> 150.
	<Platform> MS-DOS.
	<Current Status> Evaluation
	<Date deployed> August 1989.
	<Information Date>  August 1992.
</Program>

<Program> DBKUNDAR
	<Number> 352
	<Domain> Customer support and follow-up, for sales people.
	<Purpose> Customer database for logging sales of a commercial program.
	<Description> To log sales of a program, with separately sold 
accessories, enable searching in the database, and producing mailinglists 
and financial statistics.
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> Turbo Prolog 2.0 / PDC Prolog 3.21 + PDC Toolbox.
	<Other languages> None.
	<Lines> 700
	<Predicates> 16
	<Platform> MS-DOS.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 12/90
	<Information Date> 12/08/92
</Program>

<Program> DECIBAC
	<Number> 181
	<Domain> Expert systems
	<Purpose> A hybrid expert system tool
	<Description> The DECIBAC workshop allows a combination of 
hierarchically organized objects, multiple inheritance and reflexes 
(daemons) with a rulebase, which allows forward and backwards 
chaining.
It has a friendly user interface and X-system openness including the 
ability to lin Prolog, C or Assembler routines.
	<Country> France
	<Other languages> C, Assembler (including user modules)
	<Platform> PC with MS-DOS and Unix SCO, soon OS/2
</Program>

<Program> Declarative graphics
	<Number> 246
	<Domain> computer graphics
	<Purpose> a grammar-based approach to representing graphics declaratively
	<Description> A set of tools a programming environment for 
generating, combining and recognizing pictures. The recognition 
process uses constraint techniques. 
Tools include a window manager, an interactive picture editor, system for 
program animation.
	<Developed by> Richard Helm, Kim Marriott
	<Organization> Univ. of Melbourne, Dept Computer Science
	<Country> Australia
	<Prolog system> MU-Prolog
	<References> Richard Helm, Kim Marriott: Declarative Graphics. ICLP3, 
1986. pp513-527
</Program>

<Program> DedChem
	<Number> 84
	<Domain> Organic chemistry
	<Purpose> Synthesis planning based on name reactions in organic chemistry
	<Description> In organic chemistry, syntheses need to be planned 
carefully prior to being performed. A synthesis plan contains all the 
information that is needed to synthesize a given substance, i.e. the 
reactions involved, the order in which the individual reactions are 
to be performed, the intermediate substances which are synthesized as 
by-products, etc.
DedChem is a system for synthesis planning in organic chemistry. In 
DedChem, a deduction component implemented in Prolog computes 
synthesis plans from name reactions and substance classes stored in 
external databases. DedChem differs from other synthesis planning 
systems in two important respects:
- it is based on name reactions instead of concrete reactions, and
- it is designed as a coupled system which accesses a variety of 
external relational or higher databases.
DedChem is designed to be able to access a variety of external 
databases for practical and legal reasons.
The current version of DedChem features a graphical user interface, a 
simple synthesis tree browser, and user-friendly access to the reactions, 
substance class and superclass databases.
	<Developed by> Christoph Draxler (1)
Roland Kunz (2)
	<Organization> (1) Centre fuer Information- and Speechprocessing
Ludwig-Maximilians-University Munich
Leopoldstr. 139
D 8000 Munich
(2) Organic Chemistry Institute
University of Zurich
Winterthurerstr. 190
CH 8057 Zurich
	<Country> (1) Germany
(2) Switzerland
	<Prolog system> LPA MacProlog 3.5
	<Other languages> Database access languages: SQL and proprietary database 
languages
	<Lines> approx. 2000 lines incl. empty lines, one literal per line
	<Predicates> Database				20
User Interface		49
Synthesis Planning   7
Auxiliary				8
---
Total				   84
	<Platform> Macintosh 1 MB minimum
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> Oct 91
	<No. of Users> prototype versions delivered to
Organic Chemistry Institute Zurich (Kunz)
Polytechnic of North London		  (Solomonides)
University of California San Diego (Wibke)
Technical University of Vienna     (Gottlob)
	<References> [Draxler/Bauknecht 90]     C. Draxler, K. 
Bauknecht: Name Reactions in Organic Chemistry - A New Application 
Domain for Deductive Databases DEXA 90 Conf. Vienna, Springer Verlag 
1990
[Draxler 91] C. Draxler: DedChem v2.5 Short Reference Manual
CIS, LMU Munich, 1991
[Draxler 92] C. Draxler: DedChem - Synthesis Planning in Organic Chemistry 
Poster Session, Practical Applications of Prolog Conference, London 1992
</Program>

<Program> DEFINE BENEFIT STRUCTURE/MODULE of PENSION  FUND 
VALUATION 
PROGRAM
	<Number> 325
	<Domain> Pension Fund Variation.
	<Purpose> Selection and Definition of Pension Benefits.
	<Description> The PROLOG module interacts with an end-user to select 
benefits applicable for a category of members.  The benefits are defined by 
selecting and assigning values to a set of parameters based on information 
obtained by user interaction.  The resulting benefit structure is written 
to an Information database.  The PROLOG knowledge base is structured around 
a benefit classification tree  (frames).  The knowledge interpreter is a 
standard frames/rules inference engine with an extra layer of domain 
specific functionality added.
	<Developed by> Gavan Duffy, Andrew Collins, Andrew Langdon.
	<Organization> Bacon and Woudrow.
	<Country> U.K.
	<Prolog system> Interface Prolog  (IF)
	<Other languages> Informix 46L   50%
'C'		 25%
UNIX						 1%
	<Lines> 50,000
	<Predicates> 2200
	<Platform> Howlett Packard Unix Box
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> Not yet/ 07/92
	<No. of Users> 30
	<Information Date> 23/04/92
</Program>

<Program> DELBOLISM
	<Number> 282
	<Domain> Computer architecture design
	<Purpose> Evaluates and tests  high level architectures
	<Description> The system provides computerised support to the 
design process of language oriented computer architecture. It can be 
used for:
- measuring quantitative factors characterising the effectiveness of 
the given architecture;
-the experimental validation of the specified architecture by means 
of running test programs;
-measuring the dynamic statistics of the use of the source language.
	<Developed by> V. Kiss and G. Simor
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> V. Kiss and G. Simor (1979)
Evaluating the design and experimental testing of high level 
architectures -DELBOLISM.
SzKI report, 1979.
</Program>

<Program> DES
	<Number> 200
	<Domain> Electronic manufacture
	<Purpose> Diagnostic system to provide early detection of errors in 
integrated circuits
	<Description> The Diagnostic Expert System is used to provide 
early detection of errors because of the complexity of the diagnostic 
problem and the need to capture the expertise of highly experienced 
troubleshooters. It is built using Lirov and Yue's Semantic Control 
Paradigm and consists of identifier, goal selector and adaptor 
modules. The test engineer specifies the signal path, failure 
probabilities and test costs using the identifier. The goal selector 
derives the optimal diagnostic/replacement tree. Finally the adaptor 
traverses the diagnostic tree and recommends the measurements or 
replaces accordingly.
A color display is used to highlight errors of missing or failing parts of 
a signal path.
	<Organization> AT&T
	<Country> USA
	<Prolog system> Quintus
	<Other languages> ProWindows
	<Platform> Unix workstation
	<Type of Use> In-house
	<Current Status> Evaluation
	<References> Y. Lirov, O. Yue. Circuit Pack Troubleshooting via 
semantic control and goal selection. Proc. Int. Workshop on AI for 
Industrial Applications. pp118-122, 1988
</Program>

<Program> DESA - Desassembler of VSLI circuits
	<Number> 418
	<Domain> CAD of VLSI circuits.
	<Purpose> Structural abstraction of VLSI.  This circuit 
desassembler is fully written in prolog  and  it  is part  of  a set 
of programs that perform the abstraction and the
verification of the VLSI circuits.
	<Description> CAD verification tools employed in BULL make use of  
an  abstraction process that extracts the structural and functional 
specifications of a design from its layout level. A VHDL description 
can be obtained at the ending stage of the process.
The recognition of all the basic parts of the design such as  
inverters, switches, nand, nor, latches, and others is provided.  
Management  of  the  great  diversity  of  models  is   accuratly 
done   thanks to an independent base of models.
All the basic cells to be recognized are depicted in the  models. The  
syntax  of these models is the one of prolog and the mechanisms of 
recognition are closely linked to the  prolog  mechanisms of 
unification and resolution.
Models can be recursive and are applied on sequence until saturation 
on the transistor netlist.				    
The validity of this approach has been proved  as  realistic  and the  
program has been employed for several years on real projects in the 
company. An efficient  control  of the recognition process allows very 
large quantities of data to be  easily processed.				  
	<Developed by> J.B.Guignet.
	<Organization> BULL S.A.
	<Country> France
	<Prolog system> SP-PROLOG 
	<Other languages> C language (5000 lines).
	<Lines> 17000  lines well documented.
	<Predicates> 5000.
	<Platform> SUN3,SUN4 and all BULL UNIX platforms.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 1986
	<No. of Users> more than 30 person in 2 Internal projects
	<Information Date> 26/03/92
	<References> AFCET : "Les Utilisations Industrielles du Langage 
Prolog" 
		   PARIS FRANCE - 5 Apr 90
EDAC : "The European Conference on Design Automation"
						  BRUSSELS BELGIUM - 19 March 92
</Program>

<Program> DESBAT
	<Number> 175
	<Domain> Architectural design
	<Purpose> Architectural design system
	<Description> 
	<Developed by> J. Autran, M. Florenzano
	<Organization> Ministre de l"Equipment et du Logement
	<Country> France
	<Prolog system> Prolog II
	<Other languages> VERSO database developed by INRIA
	<Platform> Macintosh Plus, II, Unix 
ÇIF*** The text replacing this argument is too long ***È<References> J. 
Autran, M. Florenzano, N. Khayat-Mellah, J. le Maitre, C. Palisser. 
Realisation du prototype d'un systme d'aide ˆ la gestion de données pour 
l'architecture et le b‰timent. DESBAT - Application ˆ la rédaction du devis 
descriptif de b‰timent. Rapport final de recherce GAMSAU - IN.PRO.BAT, 
Juillet 1988
</Program>

<Program> Detect5.pro
	<Number> 54
	<Domain> Natural language processing
	<Purpose> Didactic tool for second-language teaching
	<Description> Detect.pro is an error detector and corrector of 
badly formed natural language sentences that are entered by a user in 
French. The parser is a top-down implementation of Definite Clause 
Grammar (DCG) rules. It can identify errors of three types: order of 
words, agreement errors and ambiguities. When the parser has detected 
an error, it sends an error message and provides a correction for 
that error. In the case of ambiguous sentences, the system gives all 
possible meanings of that sentence. It also gives the syntactic 
structure of the sentence (in parenthetic form), as well as its 
optional tree representation.
In second language learning, the student must not only find out where he 
made a mistake, but he must also understand why he made that mistake.
	<Developed by> André Linden
	<Organization> University of Geneva, Faculty of Humanities
	<Country> Switzerland
	<Prolog system> PDC Prolog (version 3.21) and SICstus Prolog
	<Lines> 1000
	<Predicates> 80
	<Platform> AST 386SX and 286 PC under DOS 5.0
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 7/91
	<Information Date> 2/92
</Program>

<Program> DFT Spectral Analysis.
	<Number> 467
	<Domain> Digital Signal Processing Analysis and Simulation.
	<Purpose> Compute Power Spectral Density of Complex Sequences.
	<Description> This program demonstrates the utility and 
expressive power
of the Prolog language for digital signal processing analysis and 
simulation. At its present state, the program  accepts a list of complex 
numbers (list of dotted pairs) and produces its power spectral density for 
a specified number of frequency points.  Underlying the implementation is a 
short, clear, and efficient recursive form of the  Radix-2 FFT that 
implicitly incorporates correct ordering of the spectral output.  Of 
significance is the fact that this Prolog system supports numerical 
computation efficiently  not only via is/2, but also by means of real and 
complex scalar and vector primitives. In fact, the efficiency of this 
program ranges from 0.3 to 0.6 compared to C, depending on how the 
arithmetic is implemented. 
	<Developed by> Sam M. Daniel.
	<Organization> Motorola.
	<Country> U.S.A.
	<Prolog system> ALS Prolog.
	<Other languages> None.
	<Lines> 200.
	<Predicates> 40.
	<Platform>  Motorola 88K Unix V Delta System.
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 12/87
	<No. of Users> 3
	<Information Date> 11/92
	<References>  DFT Spectral Analysis in Prolog, Motorola Project 
Memorandum, PM-70304-0130-01, May 5, 1990.
</Program>

<Program> DIANA
	<Number> 503
	<Domain>  Technical Diagnosis (Fault finding in technical 
installations).
	<Purpose> Diagnostics of technical installations.
	<Description> A knowledgebased application with the aim of 
educating in
technicalfaultfinding.
	<Developed by> Rainer Knauf, Thomas Abel.
	<Country> Germany.
	<Prolog system> PDC Prolog 3.30
	<Other languages> Top Speed Modula 2.0
	<Lines>  5300 lines of Prolog, 15300 Lines of Modula.
	<Platform> 486 AT, MSDOS 5.0
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> mid 1989 (in Turbo Prolog version)
</Program>

<Program> DIAVOLA
	<Number> 372
	<Domain> Mechanical Engineering:Failure Diagnosis,Product Training.
	<Purpose> Diagnosis of failed or damaged rolling bearings.
	<Description> Diavola permits a systematically structured 
summarizing of
symptomes and damages on rolling bearings and parts thereof, 
summarizing of
operating and assembling conditions. Evidence for failure is deducted 
from those symptoms. Evidence and additional questions which the 
system may ask are used to perform
hypothesis. Hypothesis belonging to the same field of problems are 
collected in a listing together with values for probability. 
Explanations as well as recommandations
to prevent from further damage are available on request together with 
extensively explained theorie to fiels of problem in question. 
Hypothesis are classified in accordance with a treasure value always 
being calculated depending on the values of all hypothesis kept in 
memory. Hypothesis
have to pass this treasure value in order to be monitored on screen. A 
context sensitive help system is provided. Some fotos ease the task to 
identify a symptom.
	<Developed by> Horst Koehler.
	<Organization> none.
	<Country> Germany(West).
	<Prolog system> PDC 3.2 / 3.21 / 3.3 and Toolbox 2.01.
	<Other languages> none.
	<Lines> About 1850 lines of PDC Prolog.
	<Predicates>  About 90.
	<Platform> MS-DOS 3.3 and 5.0,PC-AT,80286/80386DX,true compatible.
	<Type of Use> 
	<Current Status> Released
	<Date deployed> 1991
	<Information Date> August,16th,
1992.
	<References> Not yet published,but by the end of 1992 eventually.
</Program>

<Program> DINSA HPP and Stocks
	<Number> 387
	<Domain> Decision support for stocks management.
	<Purpose> Optimization of stock management.
	<Description> Expert system for sales forecasting and optimal stock 
management for a product distribution company. Project developed for a firm 
of the holding Banesto STA (Servicios de Technologia Avanzada).
	<Developed by> Regina Llopis, Frederic Casanovas and programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.31 for Windows.
	<Other languages>  C++, Microsoft C
	<Lines> 11113
	<Predicates> 377
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS and Windows
	<Type of Use> Sold Product
	<Information Date> 07/1992
</Program>

<Program> Direct Guide Insurance System
	<Number> 363
	<Domain> Insurance administration.
	<Purpose> Administration of private insurance policies issued
through banks.
	<Description> An insurance system which guides an employee into finding 
the best insurance for the client, calculates the premium, and if accepted 
transmits the data through a dial-up line to DanGarant, where the main 
database is kept.  The system does the collection of premiums.
	<Developed by> Rune Tendal Kock (primarily).
	<Organization> DanGarant international insurance brokers.
	<Country> Denmark.
	<Prolog system> PDC Prolog 3.3
	<Other languages> In a very few cases, C, Pascal, and 8086-assembler 
have 
been used.
	<Lines> 23,000
	<Predicates> 1,100
	<Platform> DOS and OS/2
	<Current Status> Released
	<Date deployed> September 1989
	<No. of Users> 200
	<Information Date> July 1st, 1992
</Program>

<Program> DME(Distance Measuring Equipment)SIMULATOR 
	<Number> 330
	<Domain> Real Time simulation.
	<Purpose> Simulation of single-channel air navigation aid (ATCsystem).
	<Description> Programs provide for control of DMA hardware simulator.  
Interactive PC interface (Prolog) controls 8052 board to simulate 
proprietary (antique) real-time bus.
	<Developed by> Jai Hakim, Duncan Campbell.
	<Organization> Canadian Marconi Corporation.
	<Country> Canada.
	<Prolog system> PDC Prolog.
	<Other languages> C,  BASIC.
	<Lines> 700
	<Predicates> 100
	<Platform> MSDOS
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 01/91
	<Information Date> 08/92
</Program>

<Program> DOFLEX (Dortmund Flexible Legal Expert System Environment)
	<Number> 310
	<Domain> 
	<Purpose> Exploration of techniques of knowledge based systems to the 
legal and planning domain, also hypertext.
	<Description> The shell DOFLEX has been developed to support creation 
and maintanence of legal knowledge bases that have to be updated 
frequently.  The knowledge base therefore is restricted to propositional 
logic.  It can be outputted in PROLOG.  An interface to JURIS lagerndan 
online full text database has been implemented as well as to ORACLE 
(amongst other features).
	<Developed by> Klans Witulski
	<Organization> University of Dortmund, Dept. of Urban Planning
	<Country> Germany
	<Prolog system> IF-PROLOG, ARCLY-P, C-P, QUINTUS-P
	<Lines> ~3000
	<Predicates> > 100
	<Platform> PC, Workstations
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> Nov. 1989
	<No. of Users> ~ 5
	<Information Date> 5/92
	<References> K. Witulski:  Entwickhing einer Experten system 
umgebang,						     Deutscher Universitats 
Verleg, 1989.  (English dissertation).
K. Witulski:  Improving Access to Legal Information in Document Database in 
:  M. Schader, Analyzing and Modeling Data and Knowledge, Springer Virlag, 
1992.
</Program>

<Program> DOKSI
	<Number> 269
	<Domain> Documentation support
	<Purpose> A documentation support system
	<Description> The system helps in the production of program 
documentation or other textual objects according to some preset standards. 
It provides means for handling various forms of requirements (e.g. on the 
form or content of the text) and the stepwise refinement of the structure 
of a document in accordance with the requirements.
	<Developed by> I. Fidrich
	<Organization> SzKI
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> I. Fidrich (1980)
User's Guide to program documentation system DOKSI.
SzKI Report, 1980.
</Program>

<Program>  Drug interactions system
	<Number> 273
	<Domain> Pharmaceutical Industry
	<Purpose> To predict drug interactions in medical treatment
	<Description> Modification of clinical effects may arise when drugs are 
given simultaneously. The so-called drug-interaction constitute an aspect 
of medical treatment. The system considers the physico-chemical, 
pharmacological and chemical properties of drugs and, starting from these, 
infers the possible drug-interactions.
	<Developed by> F. Darvas, I.Futo, P.Szeredi 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> F. Darvas, I.Futo, P.Szeredi (1976)
A program for automatic filtering of drug interactions.
Proc. of Cool. on the application of Computing in Medicine and 
Biology.
1976, pp 413-422.
F. Darvas, I.Futo, P.Szeredi (1978)
A logic-based system for predicting drug interactions.
International Journal of Biomedical Computing, Vol. 9., 1978, pp.259-
271.
</Program>

<Program> DtoP
	<Number> 234
	<Domain> Manufacturing design
	<Purpose> provide an integrated product description through all phases of 
manufacturing from conception to manufacture
	<Description> Design to Product was a demonstrator for AI 
techniques in all aspects of manufacturing. 
The total project used 200 man years and £9m funding.
	<Organization> Edinburgh University, Loughborough, GEC, Lucas CAV
	<Country> UK
	<Prolog system> POPLOG
	<Other languages> Common Lisp, Pop-11, C, Fortran
</Program>

<Program> DUALS
	<Number> 250
	<Domain> linguistics
	<Purpose> discourse understanding in natural language 
	<Description> A discourse understanding system based on situation 
semantics which can analyze short stories taken from a textbook for 
elementary school students and answer questions
	<Developed by> Yasukawa
	<Organization> ICOT
	<Country> Japan
	<Prolog system> DEC-10 Prolog, CIL
	<References> H. Yasukawa, H. Hirakawa, K. Mukai, H. Miyoshi, Y. 
Tanaka: The outline of discourse understanding system DUALS. Tech. 
Mem. TM-0118, ICOT, 1985 (in Japanese) 
K. Fuchi, K. Furukawa: The Role of Logic Programming in the Fifth 
Generation Computer Project. ICLP3, 1986. pp1-24.
</Program>

<Program> DUST-EXPERT
	<Number> 430
	<Domain> Design of Vents for explosion relief.
	<Purpose> Design and obtain the venting requirements for the relief of 
dust explosions.
	<Description> The use of explosion relief vents to provide the 
ultimate
protection against the effect of an explosion is very 
often recommended for the equipment within which the ignition of a 
flammable dust is possible.DUST-EXPERT provides most of the design 
methods recommended  in the I.Chem.Eng. guidelines for the venting of 
dust explosions.  The system provides completeexplanations when a 
method is applicable, and when it is not applicable.  It's novel 
features include the calculations that are actually used to obtain an 
answer, and a backward calculation feature that aids 
design.  The developmentstrategy adopted is also novel in that the experts, 
from the Health and Safety Executive, represented the methods and knowledge 
using a Prolog developed shell.  DUST-EXPERT also includes access and use 
of a database of dust properties.
	<Developed by> HSE,Salford University Business Services Ltd.
	<Organization> British Materials Handling Board, and HSE.
	<Country> UK
	<Prolog system> Expert Systems Ltd's Prolog-2 386
	<Other languages> Small procedures in Micro-Soft C.
	<Lines> over 10,000 lines (incl comments).
	<Predicates> ??
	<Platform> PC-386 under MS-DOS 4.0
	<Type of Use> 
	<Current Status> Evaluation
	<Date deployed> Under evaluation.
	<No. of Users> Currently being evaluated by about 16 users.
	<Information Date> February 1992
	<References> R. Santon et. al. 1991,
"A feasibilty study into the use of expert systems for
Explosion Relief Vent Design", Proc. of HAZARDS XI, I. Chem. Eng. No 
124.
S. Vadera, and S. Nechab, 1991,
"Are Expert System Shells and toolkits too general?",
Proc. IMACS International Workshop on Decision Support
Systems and Qualitative Reasoning, North-Holland.
</Program>

<Program> DYNAMIC - system
	<Number> 331
	<Domain> Locking and physical access control to buildings and 
facilities.
	<Purpose> Locking.
	<Description> DYNAMIC is a general purpose system for controlling 
the physical access to buildings.
The locks are computer based and off-line.  The smart card is used as 
the key and the communication media.  DYNAMIC -system also includes 
AI based software packages for computer aided system design and 
management.  The design software is for planning new installations 
and for maintaining the old ones.  The management software is used 
for key management and for controlling the communication with the 
locks.
The system is programmed in Prolog.  Abstract descriptions of the 
access rights and the knowledge distribution concept are based on 
ACL, an in-house AI language for access control.
DYNAMIC-system is a versatile and high performance solution for access 
control.  It also is inexpensive and easy to install and use.
	<Developed by> A. Saksa, T. Merinen, O. Koskimaa.
	<Organization> Traditor Ltd.
	<Country> Finland.
	<Prolog system> PDC Prolog 3.30 and home ACL inference mechanism.
	<Other languages> ACL  (20%), C++ (10%), ASM (5%)
	<Lines> 50,000
	<Predicates> 1000
	<Platform> DOS, Windows with embedded computers.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1992
	<Information Date> 20/07/92
</Program>

<Program> DynaPay Payroll Awards Interpreter.
	<Number> 505
	<Domain> Payroll.
	<Purpose> Calculates employees pays and allowances using user - specified 
rules.
	<Description> Calculation of employees gross pay and allowances 
is in many industries the most difficult and error-prone part of 
payroll calculation. In large organisations with employees engaged in 
diverse activities many different "payroll rules" (i.e. award 
Conditions) may be in use. There is also 
considerable scope for individual interpretation of the meaning of 
the conditions specified in various awards. The DynaPay Awards 
Interpreter enables an infinite number of sets of Award Rules to be 
specified; each set is allocated to one or more employees, The system 
takes as input each 
employees daily starting and ending times (optionally gathered 
directly from a time clock) and calculated for each employee all 
necessary ordinary, overtime and penalty payments together with any 
allowances and leave payments 
which may be required producing an explanation of the "Reasoning" used in 
the payroll calculations.
	<Developed by> Tony Burnell.
	<Organization> Dynamation Services Pyt Ltd.
	<Country> Australia.
	<Prolog system> PDC Prolog Version 3.20
	<Other languages> Data-entry functions are written in Microsoft C 
Version 6.0 - 
these represent about 20% of the system.
	<Lines> 12.000 lines of PDC Prolog, 1.500 lines of C.
	<Predicates> 800
	<Platform> MSDOS 3.30 and above.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> July 1989
	<Information Date> 8/2/93
</Program>

<Program> EDF
	<Number> 212
	<Domain> Power generation
	<Purpose> Real-time centralised control and maintenance of hydro-electric 
power plants
	<Description> Remote power plant data is measured by 
microcomputers and local transdusers and transmitted to the control 
room and stored in an Oracle database. The expert sytem continuously 
analyses any changes in these parameters and decides on changes to be 
made to the settings at the power plants. The instructions are then 
formulated and sent to the power plants via a remote data link.
The system also allows the operator to monitor the state of the power 
plants at all times and to use mimetic diagrams to monitor changes in 
parameters, or to formulate and send instructions directly.
Due to the system, power plants which were for the most part left unused in 
winter can be operated continuously at all times, and their operation 
optimized at other times.
	<Organization> Delphia Engineering Centre, L'Enterprise Industrielle
	<Country> France
	<Prolog system> Delphia
	<Other languages> ORACLE
	<Platform> work station
	<Type of Use> In-house
	<Current Status> Released
	<No. of Users> 2 (20 planned)
</Program>

<Program> EhLogico.
	<Number> 468
	<Domain> Automated Theorem Proving, Cooperative Interfaces for 
Knowledge-Based Systems.
	<Purpose>  Natural Deduction Automatic Theorem Prover for Knowledge-Based 
Systems with an Explanations-Generating Cooperative User Interface.
	<Description> A complete and consistent automatic theorem prover 
for classical logic. Designed to be the inference module for 
knowledge-based systems which use classical logic as the language for 
domain description. It produces Natural
Deduction proofs in the Normal form that can be translated to natural 
language explanations of the inferences. The prover is designed to 
produce direct proofs (easier to understand, since they do not use 
contradiction) whenever possible. 
When the prover is unable to reach a conclusion, because of lack of 
information it allows the identification of the pieces of information that 
may complete the proof, enabling the system to complete proofs by directing 
questions to the user, or to the world in the case of autonomous systems.
	<Developed by> Denise A. S. OLIVEIRA
	<Organization> Pontificia Universidade Catolica do Rio de Janeiro - 
PUC/RJ and Universidade Catolica de Petropolis - UCP.
	<Country> Brazil
	<Prolog system>  PDC Prolog (formaerly Turbo PROLOG 2.0
	<Other languages> None. C language may be used in the future in some parts.
	<Lines> 6000 lines.
	<Predicates> 400 predicates.
	<Platform> IBM PC-XT or compatible, with 640k memory.
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> May/1991.
	<Information Date> November 05, 1992.
	<References> On the generation of explanations based on Natural 
Deduction 
proofs: "Explanatory Text Planning in Logic-Based Systems"
Souza, C.S.; Nunes, M.G.V.
14th. International Conference on Computational Linguistics
Nantes, France, July 1992.> 
Unpublished:
"Um Provador de Teoremas em Deducao Natural Capaz de Complementar seu 
Conhecimento" (A Natural Deduction Theorem Prover Capable of 
Complementing its Knowledge) Oliveira, D.A.S.
Master degree dissertation, Depto. de Informatica, PUC/RJ, April 
1992.
Also, on generation of explanations:
"Planning Explanations from Natural Deduction based reasoning" Nunes, 
M.G.V.
Notas do ICMSC-USP (Brasil) no. 118, August 1992, 13p.
</Program>

<Program> Electronic Catalog Construction Kit for Windows v1.0
	<Number> 448
	<Domain> Electronic marketing and sales.
	<Purpose> Permits vendors to create their own interactive electronic 
catalog.
	<Description> Kit permits vendors to create their own electronic 
catalog and order form.  The output of the kit is an electronic catalog exe 
file and database that vendors can distribute freely.
	<Developed by> Mark Garber.
	<Organization> Cerebral Software.
	<Country> USA
	<Prolog system> PDC Prolog.
	<Other languages> 20% Multiscope Modula-2
	<Lines> 40,000
	<Predicates> 1000
	<Platform> Windows.
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Date deployed> OCT 1992
	<Information Date> 11 MAY 1992
	<References> None.
</Program>

<Program> Electronic Catalog Construction Kit v3.04
	<Number> 447
	<Domain> Electronic marketing and sales.
	<Purpose> Permits users to create their own electronic catalog ordering 
system.
	<Description> The electronic catalog construction kit allows user to 
create their own electronic catalog and ordering system.  The catalog 
generated is an exe program together with a database which the vendor can 
distribute freely.
	<Developed by> Mark Garber.
	<Organization> Cerebral Software.
	<Country> USA
	<Prolog system> PDC prolog.
	<Other languages> None.
	<Lines> 6600
	<Predicates> 800
	<Platform> MS-DOS/OCT 1991, WINDOWS/OCT 1992.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1 March 1990
	<No. of Users> 22
	<Information Date> 11 May 1992
	<References> None.
</Program>

<Program> ELKARLAN
	<Number> 380
	<Domain> Train maintenance centres and parts replacement floor shop.
	<Purpose>  Optimization, revisions and maintenance of scheduling of 
trains.
	<Description> Optimization of revision and maintenance planning for 
trains of the Basque railroad company. Project developed for Eusko 
Trenbideak.
	<Developed by> Antoni Trias, Frederic Casanovas and programmers team.
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.31
	<Other languages> Microsoft C
	<Lines> 2385
	<Predicates> 55
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS and MS Windows
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/199
	<Information Date> 07/1992
</Program>

<Program> ELMA
	<Number> 69
	<Domain> Library Management
	<Purpose> Organizing educational media (books, videos, films, etc) 
for
the planning of courses, seminars, and personal professional development
	<Description> A kind of hyperbase with many-to-many relationships 
between different topics and titles, making the retrieval of relevant 
educational media an intuitive search. Searches can be "hyper-like," or 
"conventional" using a search-language. Context sensitive help.
	<Developed by> Audun Toennesen, Bjoern Bolstad
	<Organization> STATOIL - The Norwegian National Oil Company
	<Country> Norway
	<Prolog system> PDC PROLOG
	<Platform> IBM PC/ DOS
	<Type of Use> Free/Public Domain
	<Current Status> Dead
	<Date deployed> June 1988
	<No. of Users> approx 5-10
	<Information Date> March 1992
</Program>

<Program> EMANON
	<Number> 118
	<Domain> CASE
	<Purpose> A CASE-tool for Developing Knowledge-Based Systems for 
Constructive Problems
	<Description> Submitted to PAC. Paper 396
	<Developed by> Peter Nordin & Sundin
	<Country> Sweden
	<Information Date> 6/11/91
</Program>

<Program> EMEX
	<Number> 63
	<Domain> Strategic Marketing.
	<Purpose> Assist in building single equation econometric models.
	<Description> EMEX is intended to assist someone with a good 
qualitative knowledge of a market to build a quantitative model of that 
market.  It supplies the model building expertise; the user indicates the 
basic structure of the market and supplies historical data. The system is 
composed of a Prolog program that interacts with the user, and which then 
issues commands to, and interprets results from, a FORTRAN statistics 
package.  The user is thus shielded from the complexities of the model 
building process.  The resulting model is used to quantify the relationship 
between factors operating in the market; eg. a 1% rise in price will reduce 
sales by 2%, or  an advertising campaign produced a 2% increase in sales. 
It can also be used for making forecasts of future behaviour.
	<Developed by> Ted Walker, Simon McAllister, Mike Greenwell.
	<Organization> Expert Systems Ltd, Henley Centre for Forecasting,
initially Alvey funded.
	<Country> UK
	<Prolog system> Prolog-2, ESL.
	<Other languages> FORTRAN for statistical operations
	<Lines> 25300
	<Predicates> 977
	<Platform> PC/DOS, SUN/SunView.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1987
	<No. of Users> 100+
	<Information Date> 24 March 1992
</Program>

<Program> EMICAT
	<Number> 55
	<Domain> CASE Tool
	<Purpose> KBS development environment
	<Description> EMICAT is an object-oriented language extensionof 
prolog for the acquisition of complex knowledgeusing mechanisms such 
as frames, inheritance, demons and production rules. 
In addition EMICAT facilitates the way this knowledge is used thanks 
to the original methods of applying rules and mechanisms of 
hypothetical reasoning such as state memorisation.
EMICAT is used not only to develop expert systems or to generate 
them, but also to develop "conventional" programs while taking 
advantage of the object representation facilities. EMICAT is used 
both for prototyping and for implementation of operational systems.
EMICAT is sufficiently integrated within Prolog that all the advantages of 
logic programming can be used. It is a basic software program that has been 
marketed as such, and has brought the cost of developing KBS in industry 
down considerably.
	<Developed by> Dassault Electronique
	<Organization> DASSAULT ELECTRONIQUE
	<Country> France
	<Other languages> C
	<Lines> 30000
	<Predicates> >1000
	<Platform> SUN, HP, IBM RS/6000 under UNIX
PS40S2
IBM S390/VM
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1986
	<No. of Users> 200
	<Information Date> 27/2/92
	<References> P. TAILLIBERT "EMICAT: un environnement de 
developpement de systemes experts" 10eme journee Francophone sur 
l'informatique, Janvier 1988
 A. VIVANCOS "Utilisation d'une extension objet de Prolog pour le 
developpement de systems experts aerospatiaux". 55eme symposium AGARD/AVP 
sur 4 Genie logiciel et son application Avril 1988
</Program>

<Program> EMS
	<Number> 57
	<Domain> Manufacturing Scheduling
	<Purpose> Dynamic Scheduling of Flexible Manifacturing Systems
	<Description> The modular program is designed for dynamic scheduling of 
flexible manifacturing systems. The system is developed on the basis of 
blackboard concepts which incorporate several functional modules (Prolog) 
in conjunction with different algorithmic interface (Fortran) for support. 
Constraints on resources allocation are formulated in the search 
strategies.
	<Developed by> Vincent K.M. CHING
	<Organization> Imperial College London Center for Robotics and Automated 
Systems
	<Country> UK
	<Prolog system> QUINTUS PROLOG
	<Other languages> FORTRAN as foreign language module for interface/support
	<Lines> 7000
	<Predicates> 4500
	<Platform> DEC VAX/VMS    min-VAX 2000
(QUINTUS PROLOG)
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 1/7/89
	<No. of Users> 2
	<Information Date> 13/4/92
</Program>

<Program> ENDELON
	<Number> 500
	<Domain>  Project management.
	<Purpose> ENDELON helps a project manager in estimating and planning 
projects of any kind, assuming that a methodology is used.
	<Description> ENDELON produces global indicators for the project: 
man-days required for project completion: estimated delivery time: 
the best staff size for each one of the project phases: the GANTT 
schedule for project activities with resources automatically 
levelled: the profiles of the manpower to be used in the project; the 
cost and the economic risk of the project; the planned quality 
scores; automatic cancellation of
unapplicable activities. ENDELON is delivered with a set of knowledge bases 
that can be easily modified by the user. They contains the following 
information's: activities dictated by the methodology: finish to start 
relationships among activities; questions to    be answered by the user; 
formulas; risk and quality parameters; profiles of the manpower: There is a 
function for exporting the project planning to MS Project for Windows.
	<Developed by> Maurilio Vehturini.
	<Organization> Engineering - Ingegneria Informatica spa.
	<Country> Italy.
	<Prolog system>  PDC Prolog 3.30 Windows.
	<Lines> 9200
	<Predicates> 500
	<Platform> Development : PS2 L40SX 80 Mbyte disk, 4 Mbyte RAM, DOS 5.0 
andWindows 3.1; Delivery: any platform running Window 3.0, Windows 3.1 or 
OS/2 2.0 (thru WIN-OS2).
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> November 1990
	<Information Date> January 1993
</Program>

<Program> ENZYME SEQUENCER
	<Number> 290
	<Domain> Biochemistry
	<Purpose> To search for analogous sub-structures of enzyme sequences
	<Description> The system serves as a means of finding the substructures 
of enzymes with known sequences and a similar functioning mechanism. The 
system is suitable for the search of analogous primary sequence units of 
any size and  any number of error points.
	<Developed by> G. Matrai
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> G. Matrai (1979)
The application of PROLOG for search of similar substructures of 
enzyme sequences.
MTA SZBK report, 1980.
</Program>

<Program> EPSILON
	<Number> 119
	<Domain> Software design
	<Purpose> Communication with a Relational DBMS in a Logic Programming 
Framework
	<Description> Submitted to PAC. Paper 286
	<Developed by> Nurcan et al
	<Country> France
	<Prolog system> BIM Prolog + Informix
	<Platform> standard UNIX.
	<Information Date> 6/11/91
</Program>

<Program> ERGOPRO
	<Number> 70
	<Domain> Occupational Medicine
	<Purpose> Decision support in the prevention of musculo-skeletal 
disorders
in the work environment
	<Description> Users enter data about the work environment. The 
program
enhances the information to give the ergonomist a qualified advice about 
where and HOW to itervene. The cost/benefit issue is of particular 
interest. Context sensitive help.
	<Developed by> Audun Toennesen, Terje Salbo
	<Organization> STATOIL (The Norwegian National Oil Company)
	<Country> Norway
	<Prolog system> PDC PROLOG
	<Platform> IBM PC/ DOS
	<Type of Use> Free/Public Domain
	<Current Status> Dead
	<Date deployed> May 1989
	<No. of Users> 70
	<Information Date> March 1992
</Program>

<Program> Erlang
	<Number> 120
	<Domain> PAR
	<Purpose> Use of Prolog for developing a new programming language
	<Description> Submitted to PAC. Paper 130
	<Developed by> Joe Armstrong et al
	<Country> Sweden
	<Prolog system> "NV Prolog ported to Quintus Prolog, ALS Prolog, SICStus 
Prolog"
	<Platform> "VAX 11/750, PC/AT, Sun 3/60, SPARCS"
	<Information Date> 25/10/91
</Program>

<Program> ERS-1 Payload system
	<Number> 225
	<Domain> Aerospace planning
	<Purpose> Expert system for planning the payload of the ERS-1 earth 
observation satellite.
	<Description> This is a prototype expert system which is intended 
to be generalised into planning of other missions by the European 
Space Agency, Noordwijk, Netherlands. Emphasis was placed on the 
efficiency of computation of the plans and the flexibility of 
building and maintaining the user interface of the system.
This will be further developed into an operational system in collaborative 
fashion between ESTEC and BIM.
	<Organization> BIM
	<Country> Belgium, Netherlands
	<Prolog system> Prolog by BIM
</Program>

<Program> ESAP
	<Number> 496
	<Domain> Safety and Risk analysis (nuclear stations, transport etc.)
	<Purpose> Fault tree and Event tree analysis programs.
	<Description> ESAP is a program for graphically entering fault 
trees and
event trees to calculate risks. All analysis modules including the cutset  
generator are written in Prolog. It is expected that EPSAP will be 
developed to run under MS-Windows.
	<Developed by> Steve Lympany.
	<Organization> Electrowatt Engineering.
	<Country> England.
	<Prolog system> PDC Prolog.
	<Other languages> None.
	<Lines> 8000
	<Predicates> 400
	<Platform>  MS-DOS on IBM PC.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> August 1992
</Program>

<Program> ESFA
	<Number> 216
	<Domain> Electronic design
	<Purpose> Help design integrated circuits
	<Description> ESFA analyzes the program that describes the logic 
of a circuit in a high-level language. The technique used is a static 
analysis of the code (which is not executed), just as would be done 
during a manual inspection. ESFA is not only  capable of 
automatically detecting certain faults, but proposing better 
optimized alternatives. It uses a set of declarative rules written by 
experts ina pseudo natural syntax, thus allow very easy maintenance.
 In 1990, ESFA was used over 4,000 times in practice, detected 20,000 
serious errors and proposed 140,000 alternative optimizations.
	<Organization> IBM La Gaude laboratory
	<Country> France ?
	<References> B. Robinet: Logic Programming at IBM from the Lab 
to the customer. Proc ICLP8. pp912-921. 1991
B. Bertrand: La Conception de Circuits Intégrés assistée par un 
Systme Expert. Les Utilisations Industrielles du Langage Prolog, 
Afcet, Paris, April 1990.
E. Levy: ESFA: An extended Static Flow Analysis. 3rd Productivity and 
Process Tools Symposium. Thornwood, NY, Sept. 1989.
</Program>

<Program> ESTA
	<Number> 192
	<Domain> Expert system
	<Purpose> En expert system shell for the construction of Advisory systems
	<Description> ESTA provides facilities to design complicated sets 
of rules as found in taxation and social security, allow rule-based 
form-filling, adminstrative routines such as ordering, investment 
alternatives, use of software and process control systems.
It provides a dialogue with user-friendly interface, explanations of 
why a question is asked or answer given, repetition of advice given 
earlier. It enables the user to break off or reorder the dialogue. It 
provides a number of tools such as drawings of the dialogue tree and 
error-checks to facilitate development.
The system is available in a version called Esta Plus which can generate 
consultation-only versions of the system which may be distributed to end-
users without any further licensing arrangements.
	<Developed by> PDC
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Platform> IBM PC
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> EUODHILOS
	<Number> 40
	<Domain> Symbolic proofs
	<Purpose> A logic-independent proof editor and constructor that 
allows users interactively to define the language and derivation 
rules of a logical system relevant for the universe of discourse 
under consideration; and to construct proofs in the defined system. 
	<Description> A logic-independent proof editor and constructor 
that allows users interactively to define the language and derivation 
rules of a logical system relevant for the universe of discourse 
under consideration; and to construct proofs in the defined system. 
Used at ICOT; Fujitsu Laboratories; Australian National University; Algonne 
National Laboratory
	<Developed by> H. Sawamura; T. Minami; K. Ohashi and K. Yokota
	<Organization> Fujitsu Laboratories
	<Country> Japan
	<Prolog system> ESP 
	<Lines> <Program Size> 1.7MB
	<Predicates>  2073 methods
	<Platform> Personal Sequential Inference Machine PSI-II and PSI-III
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> September 1991
	<No. of Users> 4
	<References> [1] H. Sawamura; T. Minami; K. Yokota and K. 
Ohashi: A Logic Programming Approach to Specifying Logics and 
Constructing  Proofs; Proc. of the Seventh International Conference 
on Logic  Programming; edited by D. H. D. Warren and P. Szeredi; The 
MIT  Press; pp. 405-424; 1990. 
[2] Ohashi; K.; Yokota; K.; Minami; T.; Sawamura; H. and Ohtani; T. :  
An Automatic Generation of a Parser and an Unparser in the  Definite 
Clause Grammar; Transactions of Information Processing  Society of 
Japan; Vol. 31 ; No. 11; pp. 1616-1626; 1990 (in  Japanese).
[3] H. Sawamura; T. Minami; T. Ohtani; K. Yokota and K. Ohashi: A  
Collection of Logical Systems and Proofs Implemented in  EUODHILOS I; 
IIAS-RR-91-13E; Fujitsu Lab.; 1991.
[4] H. Sawamura; T. Minami and R. K. Meyer: Representing a Logic in 
EUODHILOS; IIAS-RR-92; Fujitsu Lab.; 1992 (in preparation).
[5] H. Sawamura; T. Minami and K. Ohashi: Proof Methods based on Sheet of 
Thought in EUODHILOS; IIAS-RR-92; Fujitsu Lab.; 1992. 
</Program>

<Program> EUREX
	<Number> 37
	<Domain> Law
	<Purpose> Expert System for the application of import/export regulations
	<Description> EUREX is an expert system that supports the 
decision procedures about importing and exporting sugar products. It 
is based on about 100 pages of European regulations and it is 
designed in order to help the administrative staff of the Belgian 
Ministry of Economic Affairs in filling in forms and performing other 
related operations.
EUREX knowledge base is composed of rules written in a specific language; 
documented in a dictionary and compiled into pure Prolog. EUREX is front-
ended by various powerful tools including a window-based user interface; 
tracing and explanation facilities. Moreover; a knowledge editor allows to 
update the rule base as well as the dictionary and thus to extend EUREX to 
other regulations.
	<Developed by> Fabienne Balfroid; Jan de Rijke; Geert Van Nevel
	<Organization> BIM sa/nv
	<Country> Belgium
	<Prolog system> ProLog by BIM
	<Other languages> Prolog/C interfaces to XView; Xlib
	<Lines> EUREX - core  : +-  26.000 lines 
EUREX - rules : +-   2.900 lines
EUREX - dict  : +-   9.400 lines
Total after merging: +/- 38.300 lines
	<Predicates> +- 900
	<Platform> Sun3; Sun4; SPARCstations
	<Current Status> Evaluation
	<Date deployed> March 1992
	<Information Date> January 1992
	<References> "An Expert System for the Application of Import and 
Export 
Regulations" G. Van Nevel; F. Balfroid and R. Venken
Proceedings of the Third International Conference on Artificial 
Intelligence and Law; Oxford UK; June 91
</Program>

<Program> EVA
	<Number> 376
	<Domain>  Automatic aquisition of knowledge in order to build and expert 
system.
	<Purpose>  Decision making process support.
	<Description> A generic product for decision support with the 
capability of
supervised learning in evaluation and classification problems.
	<Developed by> Antoni Trias, Xavier Fustero and programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.31.
	<Other languages> Turbo C.
	<Lines> void
	<Predicates> void
	<Platform> Developed PC 386 / MS DOS
Delivered PC 386 / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1990
	<Information Date> 07/1992
	<References> AIA Software product description.
</Program>

<Program> EVAL
	<Number> 52
	<Domain> Financial Analysis and Evaluation
	<Purpose> Support users in developing economical evaluations
	<Description> Application which supports users in defining each 
alternative in terms of costs and benefits, derives all financial 
indicators and assists in comparing alternatives; estimation is based on 
consultating specialised knowledge bases for most frequent items (man 
power, software development etc); knowledge bases are written by using EVAL 
knowledge representation language and can be easily updated by domain 
experts; design application is object oriented, supported by IPW (an 
object-oriented environment for IBM PROLOG).
	<Developed by> Francesco Costa
	<Organization> IBM Semea spa
	<Country> Italy
	<Prolog system> IBM PROLOG 370
	<Other languages> IPW - IBM PROLOG WORKBENCH
	<Lines> 100000
	<Predicates> 500
	<Platform> IBM 3090 (OS VM/SP)
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 3/92
	<No. of Users> 1000
	<Information Date> 1/3/92
	<References> Costa, F. "EVAL, A Knowledge-based Advisor for 
Economical Evaluations" -ICLP'91 -
Workshop on Expert System and Software Engineering and Applications
</Program>

<Program> EXCAP
	<Number> 233
	<Domain> Process planning
	<Purpose> A Kowledge-based system which automates process planning for 
computer aided design of turning operations in numerically controlled 
machining
	<Description> Process planning has proved difficult to 
computerize using conventional techniques and can be a lengthy 
procedure for a skilled operator. WOrking from the IGES description 
of a 2D CAD model, EXCAP produces a sequence of machining operations 
in reverse order, by a process of "adding metal" Playback then allows 
the operator to wathc a graphical simulation of machining operations 
in correct order, using POPLOG's windowing facility. The operator can 
interrgoate EXCAP about the process plan.
	<Organization> UMIST
	<Country> UK
	<Prolog system> POPLOG
</Program>

<Program> EXEGETE
	<Number> 332
	<Domain> Education in the Humanities.
	<Purpose> To provide a teaching and testing environment for students 
studying college level courses that allows hypertext commentary upon, 
interactive testing of documents in the humanities.
	<Description> EXEGETE provides a non-deterministic hypertext 
environment which allows users to highlight passages from a text and obtain 
commentaries on the passages.  The hypertext database in EXEGET is used not 
only to provide interpretation of tests, but to advise the student on the 
results of computerized tests, and to serve as a style and content checker 
for written assignments.
	<Developed by> Dr. Gearge Teschner.
	<Organization> Christopher Newport University, Newport News Virginia, 
23606.
	<Country> USA
	<Prolog system> PDC Prolog 3.30
	<Other languages> None.
	<Lines> 20,000+
	<Predicates> 500+
	<Platform> IBM and compatibles.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 10/89
	<Information Date> 09/09/92
	<References> Computer Implementation of Critical Thinking in the 
History of Philosophy, published in Proceedings from the Eight Annual 
Conference on Non-Traditional/Interdisciplinary Programs, ed. Sally 
J. Reithlingshoefer, Gearge Mason University.
Courseware Alternative to the Classroom in Philosophy in Teaching 
Philosophy, Spring 1991.
</Program>

<Program> Expert system for production planning
	<Number> 226
	<Domain> Process Planning
	<Purpose> providing an integrated planning and scheduling functionality 
for complex chemical processes
	<Description> The system was built around a  highly interactive 
graphics-oriented planning tool. This tool served initially to replace the 
manual tools used by the planning experts. Subsequently an increasing 
amount of 'intelligence' was embedded in the tool to assist the expert 
planners. Finally a fully-fledged planning algorithm, based on the 
expertise of human planners, was implemented and integrated in the system.
	<Organization> BIM, Sandoz, Switzerland
	<Country> Belgium
	<Prolog system> Prolog by BIM
</Program>

<Program> EXPOD
	<Number> 315
	<Domain> Chemical Design.
	<Purpose> EXPOD is an Expert System for Polymer Design.
	<Description> The development of a new , custom-designed polymer 
is a laborious and time-consuming task.  It involves extensive 
research by means of ccessing the literature and by experimentation.  
EXPOD is an expert-system conceived specifically for polymer 
scientists designing linear homopolymers.  It was created with two 
objectives in mind:
 
1.  Increased efficiency in polymer design.
     EXPOD's primary function is to predict the fundamental 
properties of a polymer from its molecular structure.  Using the 
optional reverse-inference module, it is also possible to generate a 
polymer structure from a set of required properties.
 2.  Smooth management of polymer data.
		  EXPOD comes with a knowledge base containing rules for the 
prediction of basic properties.  It's database contains 80,000 items 
covering over 2,400 known polymers.  Yet EXPOD has been designed as an 
open-ended, user-extendible system; both the knowledge base and the 
database may be edited and augmented.
	<Organization> Mitsubishi Research Institute.
	<Country> Japan.
	<Prolog system> IF/PROLOG
	<Other languages> C, X Windows.
	<Platform> SUN, SONY, HP.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 09/89
	<No. of Users> 50 sites.
	<Information Date> 03/92
</Program>

<Program> FAIRCHILD prototyping system
	<Number> 201
	<Domain> Electronics design
	<Purpose> Rapid prototyping of chip design
	<Description> This system consists of a number of independent 
programs built in a phased manner to prototype aspects of the design 
of a high-performance microprocessor known as the Clipper Chip. 
A prototype timing analysis program was written to analyze timing and 
identify conflicts or bottlenecks. A custom ROM assembler was developed to 
improve efficiency of developing microcoded instructions. A design 
simulator was written to analyze alternative designs for the instruction 
pipeline capabilities of the Clipper chip. A test vector generator was 
developed to test floating point instructions to conform with IEEE 
standards.
	<Organization> National Semiconductor, Fairchild division
	<Country> USA
	<Prolog system> Quintus
	<Type of Use> In-house
	<Current Status> No longer supported
</Program>

<Program> Fantasy Basketball
	<Number> 47
	<Domain> Sports statistics
	<Purpose> Maintains Routisserie/Fantasy Basketball leagues; keeps weekly 
NBA statistics in a Prolog database; ranks NBA players and teams.   
	<Description> Fantasy (or Rotissiere) Basketball leagues are a 
growing hobby among avid basketball fans.  These leagues consist of 
made-up teams whose players are in the NBA; and features competitions 
between teams based on the actual stats of their NBA players.  The 
primary job of this program is to download the weekly NBA stats; 
convert them into a Prolog clause database; and compute the weekly 
standings for teams in the league.  The resulting NBA statistics 
database is also used to rank NBA players on a variety of statistics 
to predict future performance.  Prolog has been an excellent vehicle 
for this program due to its built-in support for databases; sorting; 
and parsing.
This program has been used to produce the weekly standings for the Bay Area 
Fantasy Basketball League for two years.
	<Developed by> John Dowding
	<Organization> none
	<Country> USA
	<Prolog system> Quintus Prolog versions 2.5-3.1
	<Other languages> none
	<Lines> 1500
	<Predicates> 200
	<Platform> Sun Sparc-2
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 10-90
	<No. of Users> 1
	<Information Date> 3-7-92
</Program>

<Program> FARGETZT
	<Number> 353
	<Domain> Programming utility.
	<Purpose> To test different colour combinations, to find nice colours to 
use in my windows.
	<Description> Uses arrows to change colourattributes of both the frame 
and the pane in a small test window, until a satisfactory combination is 
reached.  The actual attributes can then be noted, and used in a program.
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> Turbo Prolog 2.0 (now PDC Prolog).
	<Other languages> None.
	<Lines> 13
	<Platform> MS-DOS.
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 1990
	<Information Date> 12/08/92
</Program>

<Program> FAST PHONEBOOK
	<Number> 301
	<Domain> Office Productivity
	<Purpose> Allows the user to store and retrieve data based on two 
fields  
either of which can be key or value.
	<Description> Allows the user to store, edit, and retrieve data in key-
value pairs of arbitrary size.  Either member of the pair can be used as 
the key for retrieval and all hits are displayed in a browser window.  Has 
"friendly" windowed interface.  Searches on substrings of keys.
	<Developed by> Gregory Bourassa
	<Country> Canada
	<Prolog system> TURBO PROLOG   V 2.0
	<Lines> ~ 200
	<Predicates> ~ 12
	<Platform> IBM  PC
DOS or OS/2
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> June 1989
	<No. of Users> 30
	<Information Date> Jan.  1992
</Program>

<Program> FAULTY
	<Number> 354
	<Domain> Fault isolation aid.
	<Purpose> To log how to fix common faults on equipment, to aid 
unexperienced technicians.
	<Description> The program is a shell.  An experienced technician 
builds a database of how to fix the few tens of common errors that a 
machine usually has.  The database is built vey intuitively, NO RULES 
ENTERED.  When building the database you mimic the step you yourself 
would normally take to isolate the fault, and repair the equipment.
This can be an enormous help to unexperienced technicians.  Just finding 
the fuse can be a problem with new equipment, when it is in a fusebank in 
the next room.
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> Turbo Prolog 2.0 (Now PDC Prolog).
	<Other languages> None.
	<Lines> 1000
	<Predicates> 40
	<Platform> MS-DOS.
	<Type of Use> In-house
	<Current Status> Dead
	<Date deployed> 1989
	<Information Date> 12/08/92
</Program>

<Program> FAUST
	<Number> 236
	<Domain> Electricity supply diagnosis
	<Purpose> on-line diagnosis of faults in the elctricity supply on the 
National Grid
	<Description> is this POP++ or Prolog?
	<Organization> ERDC Capenhurst, Thames Polytechnic
	<Country> UK
	<Prolog system> POPLOG
</Program>

<Program> FELIX - FEL and Logica Interactive Expertsystem
	<Number> 333
	<Domain> Intelligent Computer Assisted Instruction, Computer Based 
Training.
	<Purpose> To investigate the applicability and value of Intelligent 
Computer Assisted Instruction.
	<Description> The central activity in the FELIX project was the 
development of a demonstrator in order to research the applicability of 
ICAI for (military) training purposes.  The teaching domain concerns the 
processing of incoming and outgoing message forms by clerks of the Signal 
Service.  The demonstrator was evaluated during 1990 involving three draft 
classes.  Compared to traditional classroom instruction the evaluation 
showed better performance on tests and a reduction in learning time.
	<Developed by> TNO-FEL and Logica.
	<Organization> Sponsred by the Ministry of Defence.  (Directorate-
General Research and Development
	<Country> The Netherlands.
	<Prolog system> Quintus Prolog 3.1, Quintus ProWINDOWS 1.2.
	<Other languages> C.
	<Lines> 15000
	<Predicates> 750
	<Platform> Sun.
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 03/92
	<No. of Users> 2
	<Information Date> 07/92
	<References> De Niet, M., FELIX:  een intelligent computer 
ondersteund onderwijssysteem, In:  Proceedings, AI Toepassingen '89, 
Dutch.
Kuiper, H., FELIX:  een onderwijzend expertsysteem, In:  Kennissystemen.  
tijdschrift over intelligente software, nr 11, jaargang 5, Dutch.
</Program>

<Program> FEREL-FERTIGUNGSREGELUNG
	<Number> 326
	<Domain> CAM
	<Purpose> Shop floor control system - line capacity scheduling.
	<Description> Shop floor control system for highly automated flexible 
manufacturing systems.  The system has to plan and realise the production 
tasks in view of given schedule and capacity.
	<Developed by> Dr. Kupec, Simon.
	<Organization> Institute of Germany
	<Country> Germany.
	<Prolog system> IF PROLOG 4.1
	<Other languages> C, X-WINDOWS, MOTIF,
SQL, PASCAL, GKS
	<Lines> ~13,000
	<Predicates> ~700
	<Platform> Decstation 5000
ULTRIX 4.2
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 06/88
	<Information Date> 04/06/92
	<References> Intergration of Autonomous Mobile Robots in Flexible 
Manufacturing Systems.  Proceedings of IAS-2 Amsterdam, 11-14/12/89.
</Program>

<Program> FIM
	<Number> 385
	<Domain>  Front office sales of investment funds.
	<Purpose> Help the sales force of investment funds for a bank holding by 
defining clients' profiles.
	<Description> Expert system for building Personal Investment Plans on 
funds based on supervised machine learning of clients profiles. Developed 
for Dinerval of the Banesto's Holding. System being installed.
	<Developed by> Xavier Orteu
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.21
	<Other languages> C
	<Lines> 12000
	<Predicates> 717
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 01/1992
	<Information Date> 07/1992
</Program>

<Program> FLiPSiDE 
	<Number> 28
	<Domain> Distributed AI
	<Purpose> Financial Logic Programming System for Distributed 
Expertise.
Blackboard system to coordinate distribute heterogenous knowledge sources.
	<Description> A prolog-based blackboard system to integrate and control 
distributed heterogeneous knowledge sources.
	<Developed by> David G. Schwartz
	<Organization> Case Western Reserve University
	<Country> USA
	<Prolog system> BIM ProLog
	<Other languages> C
	<Lines> ~3000
	<Predicates> ?
	<Platform> Sun/3 and SPARC
	<Type of Use> In-house
	<Current Status> Prototype
	<References> "The FLiPSiDE Blackboard: A Financial Logic 
Programming System for Distributed 
Expertise"; First International Conference on Artificial Intelligence 
Applicati
ons on Wall Street; Oct. 1991.
"Metaprograms: The Glue to Integrate and Control Blackboard Knowledge 
Sources";
Center for Automation and Intelligent Systems Research technical 
report; 1991"
</Program>

<Program> Floor Plan Design
	<Number> 121
	<Domain> Architectural Design
	<Purpose> An Incremental Prolog Systems Development for Floor Plan Design 
by Dissectioning
	<Description> Submitted to PAC. Paper 86
	<Developed by> Laszlo Bela Kovacs
	<Country> Denmark
	<Information Date> 6/11/91
</Program>

<Program> FMOL
	<Number> 122
	<Domain> Software production
	<Purpose> Graphical Tools in an Automatic Prototyping Environment
	<Description> Submitted to PAC. Paper 292
	<Developed by> I Ferrandiz et al
	<Country> Spain
	<Prolog system> BIM
	<Platform> Sun
	<Information Date> 31/10/91
</Program>

<Program> FOM
	<Number> 178
	<Domain> Hotel management
	<Purpose> Expert system for hotel managers
	<Description> The Front Office Manager (FOM) uses skill and 
experience to predict critical parameters affecting the occupancy 
level of a hotel, which ultimately determinges whether a hotel is or 
is not economically viable. The system advises on overbooking levels, 
reports projected occupancy and recommends whether rate categories 
should be opened or closed.
May not be Prolog!
	<Organization> DCM Data Products, East India Hotels (Oberoi) group
	<Country> UK
</Program>

<Program> FORMALL
	<Number> 394
	<Domain> Form processing.
	<Purpose> Fills in, processes and generates 26 pre-printed forms commonly 
used in small businesses. FORMALL creates an electronic environment for 
form processing in a small business.
	<Description> An inexpensive, easy to use program that lets you 
fill and
generate common business forms right from your PC keyboard! FORMALL, 
The Business Forms Processing System - a total solution to your form 
processing needs. FORMALL reduces the time and effort that is 
required to complete your standard pre-printed forms, envelopes, 
labels and other forms used you process. Features include: single 
input database system; automatic data transfer; 
automatic formatting; cursor movement control; ASCII export; complete 
mathematic function and much much more. Price is $ 79.95.
	<Developed by>  Randolph Krofick, Chief Developer; Patrick Conte.
	<Organization> North Winds.
	<Country> USA.
	<Prolog system> PDC Prolog (90%)
	<Other languages> Borland C (10%).
	<Lines> 10.000
	<Predicates> 60
	<Platform> MSDOS 2.0, 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> January 1, 1989.
	<Information Date> 9/1/92
	<References> PC Magazine Article, PCM Magazine.
</Program>

<Program> FORMZ
	<Number> 393
	<Domain> Form Processing.
	<Purpose> To create, fill in, print, and process pre-printed forms. FORMZ 
creates an electronic environment for form processing.
	<Description> An inexpensive, easy to use program that lets you 
fill in 
pre-printed forms right from your PC keyboard! FORMZ, The Intelligent 
Forms Processing System - a total solution to your form processing 
needs. FORMZ reduces the time and effort that is required to complete 
your standard pre-
printed forms, envelopes, lables and other forms you process. Features 
include: single input database system; automatic data transfer; direct 
connections to Lotus 123 and DBase III; automatic formatting; curser 
movement control; ASCII mailmerge; ASCII export; complete mathematic 
function; string and logical IF statements; editing and search features and 
much much more. Price is $ 149.95.
	<Developed by> Randolph Krofick, Chief Developer; Patrick Conte.
	<Organization> North Winds.
	<Country> USA.
	<Prolog system> PDC Prolog (85%).
	<Other languages> Borland C (15%).
	<Lines> 22.000
	<Predicates> 1000
	<Platform> MSDOS 2.0, 512 K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> October 1, 1987.
	<Information Date> 9/1/92
	<References> PC Magazine Article, PCM Magazine.
</Program>

<Program> FRENCHIP
	<Number> 220
	<Domain> Electronic design
	<Purpose> synthesis of logic circuits
	<Description> The designer uses a very high-level definition 
(VHDL) language to construct the project. The design approach is top-
down using hierarchical block diagrams. The designer can intervene to 
modify options or supply parameters in the course of the synthesizing 
process. The high level of abstraction of the VHDL models allow the 
designer to focus on the design and validation of the function and on 
any other aspect which he judges critical. This also leads to 
increased productivity and control of time and costs. In addition the 
system includes rewriting rules at the same high level so that VHDL 
can be extended.
More than 70 projects have been developed, mostly in the range 30-60 
thousand gates, but some with more than 100,000.
	<Organization> Dassault Electronique
	<Country> France
	<Prolog system> Delphia
	<Lines> 60,000
	<Platform> Unix workstations with X-windows
	<Type of Use> Sold Product
	<Current Status> Released
	<No. of Users> 60+
</Program>

<Program> FuPlan
	<Number> 504
	<Domain> Engineering of Automated Building Management Systems.
	<Purpose> CAE . System for Configuration of Control System.
	<Description> In a building management system, a substation is 
one 
fundamental component. A substation enables processes to be Directly 
Digitally Controlled ( DDC ) by means of its Controlling Program, 
which must be configured and parameterized for different 
applications. The configuration 
is represented as a Command List in a special language. FuPlan fills 
the gap between the plant schematic and this list. One module is for 
drawing functional diagrams, which are automatically compoled to the 
Command List. Other modules 
permit management of projects, installations, substations, plant 
devices, sheets,signals, function modules, terminals and macros. The 
inherent relationships between these objects are guaranteed to be 
consistent: some dependencies, e.g. form substation type, are partly 
data-driven. Soem data 
may be extracted form plant schematics drawn by an other CAE-System. 
Documentation of various aspects may be printed. Input facilities, semi - 
graphics and data-structure give high efficiency in interaction, time and 
space.
	<Developed by> Hans Grob.
	<Organization> Fr. Sauter AG, CH - 4016  Basel.
	<Country> Switzerland.
	<Prolog system> PDC-Prolog.
	<Lines>  12.000
	<Predicates> 1.200
	<Platform> DOS.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 8.1990
	<Information Date> 2.1993
	<References> User*s Guide
</Program>

<Program> GALOP
	<Number> 249
	<Domain> computational linguistics
	<Purpose> parsing of natural language
	<Description> A bottom-up parser for natural language developed 
from BUP (see Fuchi, Furukawa, 1986)
Uses a formalism now called GDL0
	<Developed by> Y. Matsumoto et al
	<Organization> ICOT
	<Country> Japan
	<Prolog system> DECsystem-10 ?
	<References> Y. Matsumoto, H. Tanak, H. Hirakawa, H. Miyoshi, H. 
Yasukawa: BUP A Bottom-up parser embedded in Prolog. New Generation 
Computing. 
K. Fuchi, K. Furukawa: The Role of Logic Programming in the Fifth 
Generation Computer Project. ICLP3, 1986. pp1-24.
</Program>

<Program> GARP (General Architecture for Reasoning about Physics)
	<Number> 450
	<Domain> GARP is not restricted to a specific domain.  Qualitative 
prediction of behaviour can in principle be done on any system. However, 
physical systems, in particular thermo dynamics, are mostly used.
	<Purpose>  A qualitative reasoning shell.
	<Description> GARP is an integrated approach to qualitative 
prediction of behaviour. Given the description of a system (usually a 
physical system) GARP predicts the states of behaviour that the 
system will go through in qualitative terms (the latter
as opposed to quantitative models).
	<Developed by> Bert Bredeweg.
	<Organization> University of Amsterdam.
	<Country> NL
	<Prolog system> SWI-Prolog (see J. Wielemaker, 1992, University of
 Amsterdam)
	<Other languages> none.
	<Lines> 10000 lines of Prolog.
	<Predicates> ???
	<Platform> Sun workstations.
	<Type of Use> 
	<Current Status> 
	<Date deployed> January 1990
	<No. of Users> About 15 research institutes used tapes of ftp to 
get the program (it is unknown how many people actually
 use GARP)
	<Information Date> May 25, 1992
	<References> - B. Bredeweg and B.J. Wielinga (1988), 
Integrating qualitative reasoning approaches, Proceedings of ECAI--
88, Munich, Page 195-201.
- B. Bredeweg (1989),
Introducing meta-levels to qualitative reasoning, Applied Artificial 
Intelligence, New York, Vol. 3-2, Page 85-100.
B. Bredeweg (1992),
Expertise in qualitative prediction of behaviour, PhD thesis, University of 
Amsterdam.
</Program>

<Program>  GASMAD
	<Number> 386
	<Domain>  Maintenance decision support in gas distribution strategies.
	<Purpose>  Optimize global and daily maintenance in decisions 
repair/replace a gas distributions network.
	<Description> Expert system for the maintenance of a Gas 
Distribution networks.
 Project in development for Gas Madrid (Gas Natural S.A.).
	<Developed by> Regina Llopis, Rosa Fors, Frederic Casanovas and 
programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.31 for Windows
	<Other languages> C++
	<Lines> 26909
	<Predicates> 421
	<Platform>  Developed PC 386 / MS DOS
Delivey PC 386 / MS DOS and Windows.
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed>  Sep. 1992
	<Information Date> 07/1992
</Program>

<Program> GemTest
	<Number> 412
	<Domain> Mineral and precious stone (gemstone) identification.
	<Purpose> To assist in fast, reliable mineral (Gemstone)
identification. 
	<Description> The fast , reliable identification of stones is 
supported, based on measurements of hardness, specific gravity, light 
refraction, colour, pleocroism, u.v. reaction, etc. Missing measurement 
data can be substituted by others. If there os no one selection 
alternatives are suggested.
	<Developed by> CIM Research Laboratory.
	<Organization> Computer and Automation Institute, Hungarian 
Academy 
of Sciences.
	<Country> Hungary.
	<Prolog system> Turbo Prolog.
	<Lines> 400
	<Predicates> 90
	<Platform> PC XT/AT, MS-DOS
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> September 1986.
	<Information Date> March 1992
	<References> Publication: Kovacs G.L. et al: GemTest - Expert System 
for Mineral Identification; 5th IFAC AMMM Symposium, 22-26 Aug. 1986, 
Tokyo. Proc No. 10, pp. 501-503.
</Program>

<Program> Gene Sequence Alignment
	<Number> 98
	<Domain> Biological Research
	<Purpose> 
	<Description> The Genome Project is a major project funded by the US 
Department of Energy to produce a map of the human genome - the genetic 
'map of mankind'. The original code for this projet was written in C but a 
Strand88 harness has been developed which permits the production system to 
be run on a wide variety of multiple processor hardware, demonstrating the 
benefits of portability and scalability for multiple language applications.  
(see Computational Algebra for comments)
	<Developed by> Argonne National Laboratory
	<Country> USA
	<Prolog system> Strand88
</Program>

<Program> Generalized Query System
	<Number> 85
	<Domain> database front-end, system integration
	<Purpose> Provides a high-level general data access abstraction
mechanism using a client - server architecture for access to
diverse sources of data and knowledge.
	<Description> An access mechanism for either a Prolog program or 
a C program that provides a data abstraction service layer.  External 
data access and interpretation are encapsulated in the conceptual 
views and meta data of the Generalized Query Server.  The "core" 
Generalized Query System software provides a query language and 
network access layers so that application developers can access any 
of the encapsulated data sources.  Data from flat files, databases, 
object-oriented databases or even expert systems can be encapsulated.  
The query language makes the physical data access mechanisms and 
network layers transparent.
The query language is much like Prolog, but has a number of additional 
features such as second order logic, abstract data types and object-
oriented syntax.
	<Developed by> Bill McClay, Roger Speigle, George Hendrickson, 
Dennis Hegedus,
Michael Easter, Steve Miller, Jerry Gerstmann, Max Robinson
	<Organization> Boeing Computer Services
	<Country> USA
	<Prolog system> Quintus 3.1
	<Other languages> 9,000 lines of C
	<Lines> 20,000 lines of Prolog 
	<Predicates> approximately 3,000
	<Platform> Sun 4, VAX, RS6000
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> January 1992
	<No. of Users> Other applications
	<Information Date> March 1992
ÇIF*** The text replacing this argument is too long ***È<References>
	First mentioned in the "Harnessing Detailed Assembly Process 
Knowledge" presentation by Bill McClay and John Thompson at the Conference 
on Innovative Applications of Artificial Intelligence, March 1989, 
sponsored by the American Association of Artificial Intelligence.  It is 
the main theme of the paper, "A Query Server for Diverse Sources of Data 
and Knowledge" to be presented at the First International Conference on 
Practical Applications of Prolog.
</Program>

<Program> Generator of Cobol programs for Data Validation
	<Number> 278
	<Domain> Software applications
	<Purpose> To generate  Cobol code that is used for validating data
	<Description> The system generates ANSI COBOL programs to be used for 
checking input data. The generated COBOL program outputs the valid data on 
a file and prints the erroneous data (indicating the cause of error). The 
structure of the files is maintained by the generated COBOL program and the 
aspects of validation are defined by parameters specified by the user or in 
a file.
	<Developed by> I. Lang 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> I. Lang (1978)
On the generation of data processing ANSI COBOL programs in PROLOG.
Proc. of the Hungarian Conf. Programming Systems'78, Szeged, pp 364-368.
</Program>

<Program> geons
	<Number> 65
	<Domain> Computer vision
	<Purpose> To recognise the class of a 3-D volumetric primitive object in 
an image description which consists of curve properties and relations.
	<Description> In an area of computer vision research, classes of 
volumetric primitives are used as a kind of 'alphabet' in the 
recognition of complex 3-D objects. Two-dimensional images of 3-D 
volumetric primitives are 'input' in the form of facts about curves, 
lines and their properties and relationships (e.g. curved/straight, 
connectivity).
This information could in principle be derived automatically from a 
grey-level image.  The logic program models, in a qualitative way, 
(a) the 3-D objects, (b) the model-scene projection relationships, 
and (c)the image structure.
These declarative models constitute a 'parser' for the input curve data 
which is analysed by the program to recognise the appropriate category of 
geometric primitive.
	<Developed by> Roger Munck-Fairwood
	<Organization> University of Surrey
	<Country> UK
	<Prolog system> Poplog (others also)
	<Lines> 800 (with comments, etc.)
	<Platform> Sun4
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> 1988
	<No. of Users> 2
	<Information Date> 1992
	<References> "Recognition of Generic Components Using Logic-
Program Relations of Image Contours", 
R C Fairwood, Image & Vision Computing, vol 9 (2), pp 113-122 (April 1991)
</Program>

<Program> GESER
	<Number> 211
	<Domain> Ship design
	<Purpose> To check that a ship design satisfies all applicable 
regulations
	<Description> 200 pages of regulations are encoded into the 
system, which enables a person using the expert system to check a 
ship's design in a few minutes, after entering the necessary data 
which takes less than half a day. This compares with four to five 
days for an expert for each manual verification, which has to be 
performed several times during a design. Due to the exhaustive nature 
of its knowledge bases, GESER avoids errors that could otherwise have 
dramatic consequences if discovered at a later stage.
The system also has capabilities to help experts translate new 
regulations into the knowledge base. It can handle "meta-knowledge" 
such as "except if rule 1.A.34 applies..." and can handle bilingual 
(French and English) knowledge bases.
	<Organization> Delphia
	<Country> France
	<Prolog system> Delphia
	<Other languages> Emicat
	<Platform> work station
	<Date deployed> 1990
</Program>

<Program> GET - Graphic Editor and Tools
	<Number> 123
	<Domain> Software production
	<Purpose> An X-Windows Toolkit for Knowledge Acquisition and 
Representation based on Conceptual Structures
	<Description> Submitted to PAC. Paper 275
	<Developed by> Wermelinger & Lopes
	<Country> Portugal
	<Prolog system> X-Prolog
	<Information Date> 6/11/91
</Program>

<Program> GEX-Tool 
	<Number> 97
	<Domain> electronic assembly
	<Purpose> an environment for developing large diagnostic expert systems
	<Description> Graphical Expert System Creation Tool presents a full 
colour graphic display of a circuit pack, highlighting suspected 
components, asking for additional information and suggesting replacements. 
GEX-Tool handles both single and multiple faults and will trouble-shoot 
multiple circuit packs simultaneously. GEX-Tool is a true expert system 
using three kinds of knowledge to diagnose circuit problems:    (1) 
experiential heuristics    (2) general purpose tactics and troubleshooting 
techniques    (3) connectivity model based reasoning GEX-Tool consists of 
four modules: automatic rule generation, manual rule editing, rule 
compilation and a rule exception and operator interface module. The first 
module takes files describing component and circuit topology and also 
replacement and testing costs and prepares data for the second module which 
allows an engineer to edit and optimise the diagnostic procedures. The 
third module generates rules in the C language which are then compiled to 
form the knowledge base used by the shop operator. This cycle can be 
repeated as often as necessary to deal with changing conditions. Most 
expert systems are notoriously slow, but GEX-Tool uses proprietary 
compilation techniques to deliver remarkable performance while it supports 
one of the largest, most complex knowledge bases of any industrial expert 
system.  
	<Developed by> A T & T
	<Country> USA
	<Prolog system> Quintus
	<Other languages> C
	<Platform> Unix workstation
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> Git'N GO P.O.S & Inventory System
	<Number> 334
	<Domain> Business Application.
	<Purpose> To allow multi-user Point of Sale / Inventory Control for 
bonded warehouses.  Automate US Customs paperwork.
	<Description> The system allows multi-user point of sale operation with 
live updating of inventory information.  Daily sales are posted to a 
secondary program which handles a portion of the paperwork required by US 
customs regulating bonded warehouses.  Data may be exported in ACSII format 
for analysis by other programs.  The system is currently installed in three 
sites in upstate NY and runs on PC's using DOS 5 and Lantastic NOS.
	<Developed by> Tom Unger.
	<Organization> RAM CONTROL, Champlain, NY.
	<Country> USA.
	<Prolog system> PDC Prolog 3.21
	<Other languages> None.
	<Lines> Approx. 5000
	<Predicates> Approx.  400
	<Platform> IBM PC clones running DOS 5 and Lantastic NOS.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 12/91
	<Information Date> 07/92
</Program>

<Program> GOLDFINDER
	<Number> 296
	<Domain> Geology
	<Purpose> To advise an exploration geologist on the best location to site 
a drill hole to find gold.
	<Description> Goldfinder is a knowledge-based system which assesses the 
potential of a gold prospect and gives advice on the best locality to site 
a drill hole.  The geologist's map of the prospect is the primary source of 
input data.  Knowledge is stored partly as semantic networks but mainly in 
a production rule formalism.  By interaction with the user the system 
establishes a dynamic database containing, at a sophisticated level, 
information about the main lithologies, the nature of any mineralisation or 
alteration, and the structural features of the prospect.  Using a method of 
intersecting loci, the system locates a drill site with a precision 
governed only by the scale of the map.
	<Developed by> D.D. Hawkes
	<Organization> Micro Prof
	<Country> England
	<Prolog system> LPA Mac Prolog
	<Lines> 275 K
	<Predicates> 280 relations
	<Platform> Foppy disc with Macintosh, hard disc and 2 Mb RAM
	<Date deployed> June 1992
	<No. of Users> new release
	<Information Date> 17 June 1992
	<References> Hawkes, D D 1992 Goldfinder: a knowledge-based system for 
mineral prospecting.  Jour. Geol. Soc. London 149, 465-471.
</Program>

<Program> Grammar Development Tool
	<Number> 421
	<Domain> Natural language processing.
	<Purpose> A tool for development of grammars by linguists.
	<Description> The Grammar Development Tool (GDT) is a tool for
development, test, and translation (into Prolog) of grammars. It is a 
multilingual tool for development of both analysis and generation 
grammars for different languages. Linguists use this tool to develop 
the grammars in grammar
formalisms and then automatically translate them  into a DCG format or 
Prolog. Users of the tool can view the grammar rules, change existing 
rules, and add new rules. Grammars can be tested by running test examples 
one at a time, or in batch. GDT also allows tracing of the grammar rule 
applications for debugging purposes.
	<Developed by> NLP Department, IBM Nordic Laboratories.
	<Organization> IBM
	<Country> Sweden.
	<Prolog system> IBM Prolog.
	<Other languages> REXX
	<Lines> Around 5000 (comments not counted).
	<Predicates> Not counted.
	<Platform> VM/XA for IBM Mainframe.
	<Type of Use> In-house
	<Date deployed> December 1989
	<No. of Users> 10
	<Information Date> March 1992
</Program>

<Program> Graph?
	<Number> 124
	<Domain> Software design
	<Purpose> Managing Networks of Persistent Data in Prolog
	<Description> Submitted to PAC. Paper 229
	<Developed by> Martin Schnuch
	<Country> Germany
	<Prolog system> IF Prolog.
	<Information Date> 27/1/91
</Program>

<Program> Graphical Front Ends
	<Number> 17
	<Domain> Electronic design; data modelling
	<Purpose> Provide a graphical front end to internal ECAD design tools
	<Description> This project consists of a series of workstation based 
tools that display and analyize the output of internal VLSI design tools; 
many of which reside on mainframes.  At this point Prolog is used as a 
rapid prototyping language with a high level X interface.  Since design 
needs and technology change quickly; this is a major feature of the 
project. The first tool that is near release displays clock tree 
connectivity and computes the steiner net wiring. Design data is modelled 
using frames; some of the interface is modelled with graphical objects. 
Future work may incorporate simple constraints and rules.
	<Developed by> Greg Munger
	<Organization> IBM
	<Country> USA
	<Prolog system> Quintus
	<Other languages> awk
	<Lines> 1000-2000 per tool; 2 written so far
	<Predicates> 300
	<Platform> IBM Risc System 6000; AIX 3.1
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 10-91
	<No. of Users> 20
	<Information Date> 11-91
</Program>

<Program> GraphicKnowledge
	<Number> 479
	<Domain> Geographic Information Systems (GIS).
	<Purpose> Extract structure from unstructured graphic files
(e.g., "un-WYSIWYG" a 2-d line map).
	<Description>  GraphicKnowledge provides a framework for 
extracting
structure from unstructured graphic files.  The problem is similar to 
that of restructuring a document into SGML or similar mark-up, given 
only the output typeset form of the document (that is, we are "un-
WYSIWYG"ing the files).  In the case of the graphic files, we have 
lines, pieces of text, and symbols ("cells").  Based on contextual 
information, we reconstruct these into objects (e.g., a street with 
the associated name, block, and connectivity to other streets).  The 
first version of this program is being used to restructure
the City of Edmonton's water maps (1700 maps, containing
300,000 objects).  The cost of the automated conversion will be about 1/3 
of the cost using conventional techniques and will reduce elapsed time from 
2 years to about 6 months (including programming and quality checking).
	<Developed by>  Peter Ludemann.
	<Organization> Quintus, Intergraph Canada, City of Edmonton (Alberta).
	<Country> USA
	<Prolog system>  Quintus.
	<Other languages>  MDL (for data extract).
	<Lines> 5000 (plus use of Quintus Prolog library).
	<Predicates> 900
	<Platform> SUN, Clipper / SunOS, CLIX / SunOS, CLIX.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> 12/92
	<No. of Users> N/A
	<Information Date>  09 Dec. 1992.
</Program>

<Program>  GROG - General Recognition Of Gates in VLSI circuits
	<Number> 419
	<Domain> CAD of VLSI circuits.
	<Purpose> Structural and Functional abstraction of VLSI. The prototype of 
this expert system is built upon Prolog. It performs the hierarchical 
abstraction of full custom VLSI circuits.
	<Description> CAD verification tools employed in BULL make use of  
an  abstraction process that extracts the structural and functional 
specifications of a design from its layout level.
As the company has gained a great experience in VLSI abstraction, the 
main goal of this new  project is  to give  the users all the power,  
the  ease  of use and the maintainability of a rule-based expert 
system.
In former applications the circuit models could  only  be  
implemented by specialized PROLOG programmers.		    Now the 
syntax of the rules has been made so that it can be easily  employed  
by non programmers.
Abstraction is mainly based upon the trial and  error  resolution of  
hypotheses.  Such  a  mechanism is therefore locally provided within 
the premise part of each production rule. Action part of rules leads  
to  the  classification  of  all  the recognized  parts  of  circuits  
(usual  classes  are  inverters, switches, latches, registers, ...).				  
Backward and forward chaining allow to control  the  hierarchical 
abstraction of the circuits that made up the VLSI. Abstraction of 
complex parts of design  such  as  registers,  adders,  ram,  can 
therefore be easily provided.
Today, the validity of such an approach has been proved as feasible on real 
subsets of design. Full  developpement  may  be  considered.
	<Developed by> J.B.Guignet.
	<Organization> BULL S.A.
	<Country> France
	<Prolog system> SP-PROLOG (contains OOP capabilities).
	<Lines> 2000 
	<Predicates> 1500
	<Platform> SUN3,SUN4 and all BULL UNIX platforms.
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 18/12/91
	<Information Date> 26/03/92
</Program>

<Program> GRONPAR
	<Number> 335
	<Domain> Genetic research, parallel programming.
	<Purpose> MS-DOS operating system extensions for PROLOG interprocess 
communication over serial ports between programs running in parallel.
	<Description> Operating system extensions have been designed to give 
the equivalent of the interprocess communication as possible under UNIX or 
OS/2 to MS-DOS computers running in parallel.  The present system uses 
serial ports for communication and is based on a ring topology.  The system 
is designed to be used for parallel computation of genetic pedigree 
analysis (see the entry GRONLINK).  Two applications are foreseen:  one to 
use spare capacity of networked MS-DOS machines running under OS/2, and one 
to build a parallel computer from single board PCs.
	<Developed by> Erick Meijer, Rene de Geus and Gerard J. te Meerman PhD.
	<Organization> Department of Medical Genetics, University of Groningen.
	<Country> The Netherlands.
	<Prolog system> PDC Prolog 3.20 and later
	<Other languages> C, Assembler.
	<Lines> 1000
	<Predicates> 10
	<Platform> IBM PC, MSDOS 3.2 and higher, OS/2 machines.
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 06/91
	<Information Date> 17/02/92
	<References> Erick Meijer and Rene de Geus, Mimicking 
Transputers on MS-DOS systems, unpublished Thesis for the higher 
Informatics Degree, Groningen State Polytechnic.
</Program>

<Program> GULICO
	<Number> 125
	<Domain> Medical information
	<Purpose> Prolog Based Medical Data Manipulating System for Cardiologists
	<Description> Submitted to PAC. Paper 296
	<Developed by> V. Meskhi
	<Country> Russia
	<Prolog system> Turbo
	<Platform> PC AT
	<Information Date> 31/10/91
</Program>

<Program> GUNGA CLERK
	<Number> 336
	<Domain> Legal expert systems.
	<Purpose> Substantive legal knowledge-based advisory system in New York 
State Criminal Law, advising on sentencing, pleas, lesser included offenses 
and elements.
	<Description> GUNGA CLERK is designed to accept key facts of a criminal 
case and provide guidance to attorneys and judges as to statutory rules 
affecting sentence parameters, regulation of plea bargaining, 
identification of lesser included offenses, and offenses chargeable based 
on designated conduct.  Logs of interactive sessions may be captured to 
disk for editing and insertion in word processing reports.  Explanations 
include citations to legal authority and display of chains of legal 
inferences.
	<Developed by> David E. Woodin, Esq.
	<Organization> Due Process Software.
	<Country> USA.
	<Prolog system> Turbo Prolog 2.0 (Now PDC Prolog)
	<Other languages> None.
	<Lines> 3227 plus Toolbox.
	<Predicates> 220 plus Toolbox.
	<Platform> IBM-PC and compatibles, MS DOS 2.0 or later.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 06/88
	<Information Date> 30/07/92
	<References> Woodin, David E., Design and Implementation of 
Substantive Systems in Criminal Law:  Beyond a Court Management 
Perspective; Proceedings of the International Conference on Database 
and Expert Systems Applications (DEXA '90), Vienna, Austria, 
Springer-Verlag/Wien, New York 1990, p388.
Woodin David E., Design and Implementation of Gunga Clerk:  A Substantive 
System in New York Criminal Law, The Defender, (Journal of the New York 
State Defenders Association) October 1989, p35.
</Program>

<Program> HAMP
	<Number> 194
	<Domain> Shipping scheduling
	<Purpose> Dynamic voyage scheduling for tankers etc.
	<Description> HAMP generates optimal voyage schedules for maximum 
vessel utilization, based on financial parameters and customized 
rules for vessel deployment. In the dynamic scheduling process the 
system gives full overview of the stock/shipment situation and sets 
alarms for stock overflow situations.
Financial calculations can be simulated for trades and shipments based on 
updated information of market supply and demand, fuel prices, available 
transport resources, etc. This provides decision makers with quick 
calculation of economic key parameters. The cost and demurrage calculator 
also shows consequences of delays and changes in financial parameters, to 
enable management to adjust shipping plans accordingly.
	<Organization> H.J. Holst
	<Country> Denmark
	<Prolog system> PDC 
	<Platform> 286/386 based PC with 8MB RAM, VGA or EGA monitor, OS/2 v 1.1 
or later, any OS/2 compatible network
</Program>

<Program> HAPPS
	<Number> 441
	<Domain> Production management system.
	<Purpose> Parts selection for a type of modular house.
	<Description>  For each customer's housing plan,
1) enter the plan using GUI operation,
2) solve the constraints between the parts,
3) select the adequate parts to satisfy the plan.
	<Developed by> ISAC, Inc.
	<Organization> Sekisui Chemical Corporation.
	<Country> Japan.
	<Prolog system> methodLog (object-oriented prolog, embedded in K-Prolog)
	<Other languages> Graphic library for the X window system written in 
C.(10%)
	<Lines> 150,000 lines (including parts database).
	<Predicates> 3,000 methods (equivalent to 10,000 predicates).
	<Platform> SONY NEWS; 4.2/4.3BSD (development / delivery).
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> April 18, 1989
	<No. of Users> 60 persons (at 10 sites, including 7 plants).
	<Information Date> April 15, 1992
	<References> NIKKEI ARTIFICIAL INTELLIGENCE No.72 p.3
December 19, 1988(bi-weekly newsletter)
</Program>

<Program> HAPPS, TAPPS and MAPPS
	<Number> 435
	<Domain> 
	<Purpose> Parts selection for modular housing.
	<Description> Plant operators entry each customer's housing plan with 
almost only mouse operation based on graphical use interface output: ascii-
coded parent parts data for parts selection which is appropriate for host 
computer (IBM mainframe).
	<Developed by> ISAC, Inc.
	<Organization> Sekisui Chemical Corporation.
	<Country> Japan.
	<Prolog system> MethodLog (object-oriented prolog interpreter,
compiled into K-Prolog interpreter).
	<Other languages> raphic library on X window written by C.
	<Lines> Over 100,000 lines.
	<Predicates> 3,000 methods (equivalent to 10,000 predicates).
	<Platform> Ews (Sony News) ; 4.2/4.3BSD (development / delivery).
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> June 1, 1989.
	<No. of Users> 60 persons (at 10 sites, including 7 plants).
	<Information Date> April 9, 1992.
	<References> NIKKEI ARTIFICIAL INTELLIGENCE No.72 p.3
December 19, 1988(bi-weekly newsletter) ditto; No.97 Apendix p.27,January 
15, 1990.
</Program>

<Program> HATE
	<Number> 35
	<Domain> Natural Language Processing
	<Purpose> Semiautomatic extension of a thesaurus
	<Description> HATE builds up thesaurus relations based on an existing 
thesaurus and syntactic relations between words
	<Developed by> R. Wirth
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Lines> 950
	<Predicates> 270
	<Platform> VM
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 12/84
	<Information Date> 1/92
</Program>

<Program> HESS (House Estimating System)
	<Number> 58
	<Domain> Construction Estimating
	<Purpose> Prepares accurates estimates for the construction of houses 
using typical methods of construction
	<Description> HESS is used by a builder to simplify the 
preparation of estimates for housing construction.
The user sketches the plan of the house to scale in a graphical window 
within the program. HESS analyses the structure of the plan and prepares 
list of quantities of materials used in the building. Cost data for 
materials, labour and overheads are then used to calculate the recommended 
cost price of the house.
	<Developed by> Robin Drogemuller, John Smith
	<Organization> James Cook University, University of Central Queensland
	<Country> Australia
	<Prolog system> LPA PROLOG, MACPROLOG
	<Other languages> none
	<Lines> 4500
	<Predicates> 350
	<Platform> IBM PC- MSDOS,
Machintosh
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 11/91
	<No. of Users> 1
	<Information Date> 3/91
	<References> Drogemuller, R., Smith, J., "Estimating Housing 
Construction Costs" Proc. of the 1st International Conference on Practical 
Applications of Prolog (1992)
</Program>

<Program> HESYOD
	<Number> 126
	<Domain> Program Synthesis
	<Purpose> Prototypes as Specifications for Automatic Synthesis of Logic 
Programs
	<Description> Submitted to PAC. Paper 234
	<Developed by> A. Ja. Dikovsky
	<Country> Russia
	<Information Date> 19/10/91
</Program>

<Program> HOSROS
	<Number> 404
	<Domain> Health.
	<Purpose> Optimizing a roster of Doctors in a hospital.
	<Description> A roster of hospital doctors was scheduled as a generate 
and test paradigm with testing intertwined with generation.  Each of the 
doctors (about 20) is asked for his preferences for weekdays and weekend 
days as prefers, can do and cannot do.  First the prefers then the can and 
finally the notcannot slots are filled.  If more than one person is a 
candidate for a slot at a given preference level, then a probabilistic 
conflict resolution scheme is used.  Constraints are applied early:  number 
of free days between duty periods, balanced number of weekend and weekdays 
for each person and no duty allocation on cannot do days.  If the roster is 
not successfully completed (a rare occurrence) after say N attempts, then 
the constraints are relaxed by the program until a solution is found.  The 
program outputs:  The complete roster, individual letters and statistics 
including an individual"index of satisfaction".
	<Developed by> Stewart Rosenberg.
	<Organization> Bourges Psychiatric Hospital.
	<Country> France.
	<Prolog system> Turbo Prolog.
	<Other languages> UNIX / SB - Prolog     +
80386 / LPA - Prolog ++
	<Lines> 850
	<Predicates> 105  (non-system).
	<Platform> 8088/DOS 3.0
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> July 1988
	<No. of Users> unknown.
	<Information Date> 06/10/92
	<References> Programming medical rosters in Prolog, Medical 
Informatics (1988), Vol. 13, 187-198.
</Program>

<Program> HPLus
	<Number> 308
	<Domain> Computer Based Training
	<Purpose> Diagnosis of errors during execution of PROLOG programmes and 
generation of alternatives.
	<Description> HPLus is a system of tools for static programme analysis 
and dynamic debugging facilities.  If an error occurs, a three level 
diagnosis process is started that includes opportunistic help on it's 
lowest level, analysis of the call-context on the second level and 
references to the history on the third level.  This analysis proceeds 
depending on the derived results or user interaction.  A user model is used 
for presenting results and alternatives queries.
	<Developed by> Karlhorst Klotz
	<Organization> Technische Universitat Munchen
	<Country> Germany
	<Prolog system> IF/PROLOG
	<Other languages> MOTIF (Interface)
	<Lines> 20,000
	<Platform> HP, SUN, VAX
UNIX
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 5/91
	<No. of Users> 30
	<Information Date> 4/92
</Program>

<Program> HPSG-PL
	<Number> 444
	<Domain> Computational linguistics, natural language parsing.
	<Purpose> Designing grammars for parsing natural language.
	<Description> HPSG-PL is a Prolog implementation of Head Driven Phrase 
Structure Grammar.  The system consists of a lexical compiler, constraint 
processor, chart parser and a module for linking the parser to a graphic 
interface.  Using this system, a user can examine the properties of the 
HPSG formalism itself, and can investigate characteristics of specific 
grammars that utilize the formalism.  A sample grammar which covers a 
fragment of English is provided.
	<Developed by> Fred Popowich, Sandi Kodric, Carl Vogel.
	<Organization> Natural Language Lab, Simon Fraser University.
	<Country> Canada.
	<Prolog system> Quintus Prolog / Sicstus Prolog.
	<Other languages> Has an interface to the TreeTool system, a graphic
interface written in C running under Sunview and X view
Has also been used with a Lisp-based program for
converting natural language queries into SQL which is then passed onto an 
Oracle Database.
	<Lines> Total number of lines (blank and comment lines included) 
comes to approx. 1000.
	<Predicates> about 110
	<Platform>  SUN / UNIX
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> October 1991
	<No. of Users> 20
	<Information Date> Apr 22, 1992
	<References> @TechReport(Pop91d,
		  author =		  "Popowich, Fred and Carl Vogel",
		  title = "The {HPSG-PL} System",
		  institution =   "School of Computing Science,
		   Simon Fraser University, Burnaby, B.C.",
		  year =  "1991",
		  number =		  "CSS-IS TR 91-08",
		  key =   "Popowich and Vogel")		     
		@InCollection
		  (Pop91a,
		  author =		  "Popowich, Fred and Carl Vogel",
		  title = "A Logic Based Implementation of Head-Driven Phrase 
Structure
Grammar",
		  year =  "1991",
		  editor = "C.G. Brown and G. Koch",
		  booktitle = "Natural Language Understanding and Logic 
Programming, III"
,
		  publisher = "Elsevier, North-Holland",
		  pages= "227-246",
		  key =   "Popowich and Vogel")								
		@InProceedings(Pop92a,
		  author = "Popowich, Fred and Paul McFetridge and Dan Fass and
						Gary Hall",
		  title = "Processing Complex Noun Phrases in a Natural 
Language
						Interface to a Statistical Database",
		  booktitle = "14th International Conference on
						Computational Linguistics",
		  organization =  "Nantes, France",
		  year =  "1992",
		  key =   "Popowich et. al.")
</Program>

<Program> HRM SIMULATOR
	<Number> 300
	<Domain> Human Resourse Management
	<Purpose> Model the impacts of human resource management policies within 
organizations.
	<Description> Permits the user to model policy matrices of arbitrarily 
complex organizations, as a combination of attribute states for persons and 
multiple rule-bases and daemons attached to attribute state changes.  
Simulates progress of the organization with these policies over extended 
periods.  Allows policy change during runs.  Has extensive user interface 
and custom reporting facilities. 
	<Developed by> Gregory Bourassa, Patrick Cunningham
	<Organization> The Cody Group (under contract to Department of National 
Defence.
	<Country> Canada
	<Prolog system> PDC PROLOG  (OS/2)  ver. 3.21
	<Other languages> PROLOG meta-interpreter written in PDC PROLOG.  Less 
than 
1/5 of code but does all user rule interpretation.
	<Lines> ~5000
	<Predicates> ~700
	<Platform> IBM Compatible     386, 486, machines running OS/2
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> March 1991
	<No. of Users> 10
	<Information Date> Jan. 1992
	<References> User guide and system reference manual available from 
(above) contact.
</Program>

<Program> Human Rights Hypertext System
	<Number> 405
	<Domain> International Relations, Education.
	<Purpose> Hypertext System for eight international human rights 
documents.
	<Description> This program provides access to eight international 
human rights documents, the American Declaration of Human Rights, the 
Universal Declaration of Human Rights, the European Convention on 
Human Rights, the European Social
Charter, the International Covenants on Civil, Political, Social, 
Economic and Cultural Rights, the American Convention of Human Rights 
and the African Charter.  These documents are interlinked by concepts 
and specific rights
permitting quick access and comparison.  Documents can also be accessed in 
sequence and individual articles may be printed out.
	<Developed by> Michael Niemann.
	<Organization> Trinity College.
	<Country> USA.
	<Prolog system> PDC Prolog.
	<Other languages> none.
	<Lines> ??
	<Predicates> 500
	<Platform> MS DOS Computers.
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> September 1991
	<No. of Users> 10
	<Information Date> March 1992
	<References> none.
</Program>

<Program> Hypertext Toolbox
	<Number> 193
	<Domain> Hypertext system
	<Purpose> To provide the building blocks to develop a hypertext system
	<Description> There are three basic building blocks in the 
toolkit.
 HAM - the Hypertext Abstract Machine. The bottom-layer database 
manager with a collection of predicates for updating nodes and links, 
supporting full version control.
  Editor - a special object-oriented hyper editor with full mouse 
support.
  Object-oriented prpgramming tools to build mouse-based user 
interface for an application, and a set of user interface tools.
The toolkit comes with the full source code of a hypertext system 
which can be used as the basis of the user's system.
  
	<Organization> PDC
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Platform> IBM PC
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> IBM SAA LanguageAccess
	<Number> 423
	<Domain> Natural language processing.
	<Purpose> Natural language interface to relational databases.
	<Description> LanguageAccess is a multilingual natural language 
query interface to relational databases which gives users access to 
the information from  databases without requiring any specific 
knowledge of the database organization or of SQL.  It translates 
users' queries into SQL and also paraphrases them back for 
confirmation.  LanguageAccess is founded on a solid natural language 
processing technology developed at
IBM Nordic Laboratories in cooperation with IBM  research organizations 
around the world.  It is generally available for English and German  and 
working prototypes exist for several other  languages.
	<Developed by> NLP Department, IBM Nordic Laboratories.
	<Organization> IBM
	<Country> Sweden.
	<Prolog system> IBM Prolog, Quintus Prolog, Arity Prolog.
	<Other languages>  C, Assembler.
	<Lines> Around 80000 (comments not counted).
	<Predicates> Not counted.
	<Platform> VM/XA, MVS/XA, and MVS/ESA for IBM mainframe,  AIX for 
RS/6000, 
and OS/2 for IBM PS/2.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> December 1990
	<No. of Users> Information not available.
	<Information Date> March 1992
	<References> (1) International Business Machines Corporation 
(1990): IBM SAA LanguageAccess  General Information (Publ. No. SH19-
6680),  Query Interface User's Guide (Publ. No. SH19-6681),   
Customization Tool User's Guide (Publ. No. SH19-6683),  Application 
Programming Guide (Publ. No. SH19-6687), etc. 
(2) Sanamrad, M. and Bretan I. (1992):  "IBM SAA LanguageAccess: A 
Large-Scale Commercial Product Implemented in Prolog",  Proc. 1st 
Int. Conf. on Practical Applications  of Prolog.
(3) Jonsson, G. (1991): "The Development of IBM SAA LanguageAccess: An 
Experience Report",  Proc. 7th Int. Conf. on Data Engineering.
</Program>

<Program> ICEBERG ELECTRIC
	<Number> 390
	<Domain> Electrical dispatching.
	<Purpose>  Support for the train tracks electrical network management.
	<Description> Expert system that is associated with tools for 
editing the
topological properties of the train tracks electrical networks and linking 
to the knowledge base for management of the electrical installation.
	<Developed by> Jose A. Marques and programmers team.
	<Organization> AIA 
	<Country> Spain.
	<Prolog system> PDC Prolog 3.21
	<Other languages> void.
	<Lines> 35000
	<Predicates> 1500
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed>  05/1992
	<Information Date> 07/1992
</Program>

<Program> IF/Man
	<Number> 324
	<Domain> Text analysis.
	<Purpose> Automatic conversion of the IF/PROLOG programming manual into 
hypertext.
	<Description> The program converts the input data-a programming manual-
into Hypertext.  Starting with files prepared for output on a Kyocera laser 
printer (this allows wysiwyg browsing) the first two passes of the program 
system reconstruct the objects (words, lines,boxes etc.) on each page and 
recognise headlines, tables and normal paragraphs based on formal criteria 
such as typical words or fonts.  In the next step a sophisticated algorithm 
reduces the words to their stems and weights them by significance.  Only 
sufficiently important terms are finally used to create links between their 
referencing and defining occurrences between paragraphs related to a common 
topic.
	<Developed by> Norbert Rossler, Thomas Schonauer.
	<Organization> Interface Computer GmbH, Munich.
	<Country> Germany.
	<Prolog system> IF/PROLOG  Version 4.1
	<Other languages> C for printer emulation  (~ 5000 l.o.c.)
	<Lines> 7200
	<Predicates> 380
	<Platform> Any UNIX system.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 06/91
	<No. of Users> 4
	<Information Date> 04/92
	<References> Conference Proceedings:
-IF/PROLOG User Day 1991  (also:  Newsletter Vol. 6)
-GUUG 1991, Wiesbaden.
</Program>

<Program> IMAGIA
	<Number> 221
	<Domain> Electronic manufacture
	<Purpose> Visual inspection system for detecting faults in VLSI circuits
	<Description> The system is used to detect significant 
differences between an observed and reference images of a portion of 
a VLSI chip. An analysis of the grey scales of the image leads to a 
hypotheses about the materials present. The system uses its knowledge 
from the inspection and of the technology to decide about the nature 
and gravity of the defects.
	<Organization> Dassault Electronique
	<Country> France
	<Prolog system> Emicat
	<References> Dassault Electronique report: Les Techniques 
Informatiques avancees et leurs applications. NE 485 352, 30th Apr 1991.
</Program>

<Program> INDICAID (Intelligent diagnostic computer aid)
	<Number> 338
	<Domain> Trouble-shooting in technical installations.
	<Purpose> Correlation of faults detected by built-in test equipment in 
the sensor, weapon and command system on the Dutch Multipurpose-frigate.
	<Description> INDICAID is able to correlate BITE detected faults.  It 
will provide all possible explanations in terms of sub-systems, ordered 
according to likelyhood.  Knowledge acquisition can be done in run-time, 
that is, the system learns from experiences.  However, the knowledge 
representation is very structured, and model-based diagnosis like 
algorithms are used.  The programme contains facilities to test and monitor 
three different types of system modelling.  Based on the results it will be 
possible to select the best of these in an operational version.
	<Developed by> Paul van Oordt.
	<Organization> Sponsored by the Royal Netherlands Navy.
	<Country> The Netherlands.
	<Prolog system> Quintus Prolog.
	<Other languages> None.
	<Lines> 6000
	<Predicates> 300
	<Platform> Sun.
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 06/92
	<Information Date> 07/92
	<References> Van Oordt, H.P.  INDICAID:  zeer hoog nivo model-
gebaseerde diagnose voor het relateren van foutmeldingen, In:  
Kennissystemen, Rijswijk, 1992, Stam Tijdschriften BV, volume 6, P. 15-18, 
Dutch.
</Program>

<Program> Infologics Faultfinder System
	<Number> 90
	<Domain> Service Engineering
	<Purpose> 
	<Description> FaultFinder is a diagnostic tool which combines expert 
systems and multimedia techniques to enhance the productivity of service 
engineers. It provides on-line assistance in the form of a knowledge base 
which helps the service engineer in diagnosis and maintenance. This is 
backed up by a Knowledge Management System which is used by product 
specialists in creating and maintaining the knowledge bases. The product is 
a domain oriented tool, not a general purpose expert system shell. 
FaultFinder incorporates both fundamental and heuristic knowledge which 
Infologics says are both important when troubleshooting technical 
equipment. Fundamental knowledge is typically that provided by the 
manufacturers, including design features, test procedures, etc. Heuristic 
knowledge is based on rule of thumb, usually gained from years of hands-on 
experience. FaultFinder knowledge bases can be located at the support 
centre, in field service or at the repair shop. The support centre is 
normally the first contact between a customer and the organisation. 
FaultFinder can be used here to diagnose faults and, if the job is to be 
handed to a field engineer, pass on information about appropriate spare 
parts and tools. Repair shops typically have several service engineers 
maintaining a broad range of products. 
	<Developed by> Infologics
	<Country> Sweden
	<Prolog system> LPA MacProlog
	<Platform> Macintosh
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> Information retrieval system processing data on pests and 
pesticides
	<Number> 281
	<Domain> Retrieval Systems
	<Purpose>  Retrieves and processes data about pests and pesticides
	<Description> The system examines the results of applying 
different pesticides under certain given situations. The following 
factors are considered by the system:
- diseases, pests, etc detrimental to a culture;
-insecticides, pesticides, etc, against given diseases;
-cultures, application areas of given insecticides, pesticides.
	<Developed by> I. Futo, F. Darvas, P. Szeredi
	<Country> Hungary
	<Information Date> 1984
	<References> I. Futo, F. Darvas, P. Szeredi (1978)
The application  of Prolog to the development of QA and DBM systems.
In Logic and Databases 
H. Gallaire and J. Minker (eds)
</Program>

<Program> Initial growth model for radiata pine in the Central
North Island
	<Number> 6
	<Domain> Forest management
	<Purpose> To represent a mathematical model of the response of 
young
Radiata pine trees to weed control; ripping; discing; and 
fertilisation
on a range of sites.
	<Description> A mathematically represented model of the survival and 
growth of Radiata pine between ages 0 and 5 in the Central North Island of 
New Zealand.  The model is sensitive to variation in site quality and to 
changes in management strategy.  The program has a graphical user interface 
which allows managers to visually compare the outcomes of alternative 
strategies.  ASCII file output is also provided; and will be accessed by 
other models of growth at older ages.
	<Developed by> E.G. Mason
	<Organization> Forest plantation establishment
	<Country> New Zealand
	<Prolog system> PDC Prolog 3.2
	<Other languages> None
	<Lines> 3000 approx
	<Predicates> 150 approx
	<Platform> MS-DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> not yet deployed to managers
	<No. of Users> none yet
	<Information Date> 24 October 1991
	<References> E.G. Mason; (in prep); PhD thesis at the University 
of
Canterbury.
Mason; E.G.; D.J. Geddes; B. Richardson; & N.A. Davenhill;
1991; Application of Knowledge-based programming techniques to
cost-effective selection of herbicides in forestry; IN: Proc. of the 
1991
Australian and New Zealand Institutes of Forestry Conference on
; edited by J.C. Allen
and A.G.D. Whyte; pp 292-300
New
Directions in Forestry: Costs and Benefits of Change
</Program>

<Program> INTERACT
	<Number> 239
	<Domain> medical prescription
	<Purpose> to give information about possible pharacon/drug interactions
	<Description> The knowledge base contains information about how 
to avoid the clinically manifested symptons, how to treat the patient 
etc. Pharmaceutical preparations are classified according to the ATC 
code developed by the Nordic Council on Medecines and recommended by 
WHO.
The system covers clinically proven interactions of 1500 pharmacons and 
drugs
	<Organization> Hungarian Institute of  Pharmacy, IQSOFT
	<Country> Hungary
	<Prolog system> MProlog
	<Lines> 1800 clauses (KB), 900  (maintenance) 800 question
	<Platform> PC 286, MS-DOS
	<References> K. Konzné Zar‡ndy, K. Moln‡r: INTERACT: consulting system 
for drug interaction. in A. G‡bor (ed) Expert Systems 88 - Knowledge based 
information processing, SzŒmalk, Budapest, 1988. pp213-231 (in Hungarian).
</Program>

<Program> IPHONE
	<Number> 457
	<Domain> Office Automation.
	<Purpose> On-line electronic corporate phone book.
	<Description> IPHONE is an interactive program which exploits the
database, non-determinism and pattern-matching capabilities of Prolog 
to allow telephone operators to query a corporate phone book. The 
phone book information, including the addresses, departments, and 
managers of employees are stored as Prolog terms. Queries on any of 
the
fields are supported allowing regular-expressions and fuzzy matches. The 
application is built with a client-server application so that a single 
(large) program has complete control over the data-base, while light-weight 
menu programs provide an interface for users. Updates are supported through 
a priviledged version of the user interface.
	<Developed by> Peter Reintjes.
	<Organization> Intergraph Corporation.
	<Country> USA
	<Prolog system> Quintus Prolog
	<Other languages> none
	<Lines> 553
	<Predicates> 131
	<Platform> Quintus Platforms
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 12/90
	<No. of Users> 10
	<Information Date> 11/92
	<References> ``Prolog for an Afternoon on the Phone''
Intergraph internal document
</Program>

<Program> ISCN Expert
	<Number> 87
	<Domain> Health
	<Purpose> Interpret chromosomal abnormalities
	<Description> The expert system allows geneticists to better reference 
and interpret chromosomal abnormalities such as those which result in Down 
Syndrome, mental retardation or physical disabilities. It interprets the 
International Human Cytogenetic Nomenclature, which is the standard 
notation used to represent human chromosomal abnormalities. These 
notations, each representing a person's genetic layout, are maintained in a 
computerized registry for reference and comparison against each other. 
Inconsistencies may arise, however, because each chromosomal variation can 
be written in a number of formats and thus be difficult to retrieve from 
the registry. 
	<Developed by> Dr. Glen Cooper,  Dr. J.M. Friedman
	<Organization> British Columbia Health Care Research Foundation, 
Ministry of Advanced Education, Training and Technology of Canada
	<Country> Canada
	<Prolog system> Arity Prolog
	<Platform> IBM PC AT with 640K RAM and hard disk
	<Type of Use> Sold Product
	<Current Status> Released
	<References> Reference: G. Cooper, J.M. Friedman. Computers and 
Biomedical Research 23, 153-164 (1990)
BC Discovery May/June 1990, Vol 6-1. 
</Program>

<Program> (IT) 2  Information-theoretic induction tools
	<Number> 304
	<Domain> Process optimization
	<Purpose> Induction of rules from examples with real-valued attributes 
used for the classification of semiconductor films
	<Description> Based on given pre-classified examples the system 
generates rule-based classifiers.  By now, it contains two induction 
algorithms, ID3 (originally developed by Quinlan) and ITRULE (Goodman 
& Smith).  Both methods were extended to handle real-valued 
attributes efficiently.
Additional parts realize an interpretation and test environment for the 
induced classifiers.
	<Developed by> Erik-Kamsties, Gunter Seidelmann
	<Organization> Hahn-Meitner-Institut, Berlin GmbH
	<Country> Germany
	<Prolog system> IF/PROLOG
	<Other languages> None
	<Lines> ~ 4000
	<Predicates> ~ 200
	
	<Platform> DECstation
ULTRIX
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> October 1991
	<No. of Users> 3
	<Information Date> April 1992
	<References> J.R. Quinlan, Learning efficient classification 
procedures and their application to chess end games.  In:  Machine 
Learning:  An Artificial Intelligence Approach, Tioga Press, 1983.
P. Smyth, R.M. Goodman.  Rule induction using information theory.  
In:  Knowledge Discovery in Databases, G. Piatetsky-Shapiro, W. 
Frawley (eds.), The MIT Press, 1990.
</Program>

<Program> JONSWAP 3D Wave Force Analysis
	<Number> 99
	<Domain> Nautical Research
	<Purpose> 
	<Description> The joint North Sea wave analysis project was established 
at Newcastle University and Imperial College, London University to develop 
facilities for measuring the forces exerted by waves on structures in the 
North Sea. The original code was written in Fortran and, as with the Genome 
Project, a Strand88 harness was written which allows the application to run 
on a wide range of different multiple processor hardware.  (see 
Computational Algebra for comments)
	<Developed by> Newastle University, Imperial College, London
	<Country> UK
	<Prolog system> Strand88
</Program>

<Program> JUMAN
	<Number> 438
	<Domain> Natural Language Processing
	<Purpose> Morphological analysis for Japanese language.
	<Description> A Japanese sentence is segmented into a lattice
structure of words.  The dictionary transfomation and look-up part are 
written in C.
	<Developed by> Yuji Matsumoto.
	<Organization> Deptartment of Electrical Engineering, Kyoto University.
	<Country> Japan.
	<Prolog system> SICStus Prolog
	<Other languages> C programs for the interface of the dictionary.
	<Lines> 500
	<Predicates> 35
	<Platform> SUN SPARCstation.
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> February 1992
	<No. of Users> 20
	<Information Date> 20 April 1992
</Program>

<Program> KALIPSOS
	<Number> 232
	<Domain> User interfaces
	<Purpose> converting natural language sentences into a rigorous formal 
representation in order to allow deduction and information retrieval
	<Description> Kalipsos is a shell system that converts French 
texts into  Sowa's Conceptual Graph model. 
It includes a morphological analyzer containing 40,000 lemmas which 
can recognize all the inflected forms in a French disctionary; a 
syntactic botoom-up analyzer which applies a context sensitive 
grammar of about 350 rules; a semantic analyzer which generates 
conceptual graphs for sentences using the syntactic trees and the 
semantic lexicon describing the meanings of words, including defaults 
for unrecognized words.
There is also an information retrieval and deduction component which 
applies expert system rules to interpret the conceptual graphs when simple 
pattern matching is not enough. Finally there is a natural language 
generation component for answers.
	<Organization> IBM Paris Scientic Center
	<Country> France
	<Prolog system> IBM Prolog
	<Platform> 370 (VM/SP) and OS/2
</Program>

<Program> KAMIIH		 (Kasmoc  Models programming environment
	<Number> 303
	<Domain> Engineering drawing in 2D and 3 D
	<Purpose> Graphic definition and management of objects in 2D and 3D 
domains
	<Description> KAMIIH is an integrated system for graphic processing of 
quoted engineering drawing.  The modeling environment provides an 
homogenous methodology for construction and manipulation of both BI and 
three dimensional graphic entities, together with their topological and 
design properties.  The system allows the joint of data and functionalities 
, and the visualization of object as well as of the coding correspondingly 
to the single geometric component.  
	<Developed by> SUSI DULLI - VITALIANO MILANESE
	<Organization> University of Padova and University of Udine
	<Country> Italy
	<Prolog system> LPA PROLOG 3.0
	<Lines> 800
	<Predicates> 200
	<Platform> Apple Macintosh II FX
Soft:  KASMOC
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> January 1989
	<No. of Users> 4
	<Information Date> Nov. 25, 1991
	<References> Milanese V.		  "KASMOC:  A clausal language for 
cad modeling systems with morphological constraints".  Computer 
Graphics Forum.  Vol. 9,1,39-51, 1990
Dulli Susi-Milanese V.  "A Database Design for Graphical Models".  
ACM Sigplan Notices, Vol. 25, N.12, 1990.
Dulli S. -Milanese V.  "A Graphic Programming Environment  based on KASMOC 
to appear in Journal of Computer Graphics Forum.
</Program>

<Program> KANAL (KArlsruhe NArrowing Laboratory)
	<Number> 367
	<Domain>  Logic-functional programming.
	<Purpose> Narrowing laboratory.
	<Description> Narrowing can be used for equation solving in 
equational
theories defined by canonical rewriting systems. KANAL implements different 
narrowing strategies. KANAL offers statistical tools for their 
comprehension and evaluation.
	<Developed by> Stefan Krischer, Ralf Dieter Bauer, Andreas Werner.
	<Organization> University of Karlsruhe, partly supported by the Deutsche 
Forschungsgemeinschaft (SFB 314, project S2).
	<Country> Germany.
	<Prolog system>  C-Prolog  (original version by S. Krischer)		    
KA-Prolog (translated by R. Bauer)
LOPSTER   (LOgic Programming with SubTypes and Extended   Resolution, 
order-sorted extensions by A. Werner).
	<Lines>  circa 9400
	<Predicates>  circa 580
	<Platform> UNIX, SUN workstations.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> 1990
	<No. of Users> 5
	<Information Date> August 12th, 1992
	<References>  S. Krischer: Vergleich und Bewertung von 
Narrowing-Strategien, Diplomarbeit, Fakultaet fuer Informatik, 
University of Karlsruhe, 1990.
 A. Bockmayr, S. Krischer, A. Werner: An Optimal Narrowing  Strategy for 
General Canonical Systems, Proceedings of the 3rd International Workshop On 
Conditional Term Rewriting Systems, to appear in the LNCS series (Springer 
Verlag), 1992.
</Program>

<Program> KANN (KArlsruhe Normal Narrowing)
	<Number> 369
	<Domain> Used as declarative language for general purposes like PROLOG 
itself.
	<Purpose>  Compiler for the logical-functional language KANN which is an 
extension of the KA-Prolog compiler.
	<Description>  A logical-functional KANN program consists of a 
set of con
ditional term rewriting rules. A KANN query consists of a set of 
equations to be solved by the strategy of left-to-right basic normal 
narrowing.  The KANN compiler is used to translate KANN programs and 
queries into an extended WAM code in order to realize their 
interpretation by means of narrowing and normalizing. By an extension 
and modification of a given Prolog compiler it was possible to 
implement the compiler very quickly.  The numbers below are refering 
to the whole in KA-Prolog written compiler. The indications in 
brackets are describing that part of the compiler, which makes the 
additional functionality of the 
KANN compiler available in relation to the original (in source code given) 
Prolog compiler. 
	<Developed by> N. Lindenberg, C. Scharnhorst.
	<Organization> University of Karlsruhe, partly supported by the 
Deutsche 
Forschungsgemeinschaft (SFB 314, project S2).
	<Country> Germany.
	<Prolog system> KA-Prolog (WAM based).
	<Lines> 7500 (500)
	<Predicates> 400  (20)
	<Platform> UNIX-OS, SUN workstations.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> does exist since 12 months.
	<No. of Users> 5
	<Information Date> August 11th, 1992.
	<References> M. Gollner, C. Scharnhorst: "Integration logischer 
und fun
ktionaler  Sprachkonzepte mit dem normalisierenden Narrowing-System 
KANN",
Proc. of the 13th Workshop "Sprachen fuer KI-Anwendungen"
Bad Honnef 1992, University of Muenster.   C. Scharnhorst: "Entwurf 
eines uebersetzenden Narrowing-Systems
basierend auf der WAM", University of Karlsruhe, 
Fakultaet fuer Informatik, Interner Bericht 9/91.
</Program>

<Program> Karyotype parser
	<Number> 206
	<Domain> Genetics Research
	<Purpose> 
	<Description> This is one of a suite of programs for analysing 
linguistically forms of the DNA molecule generated as part of the Human 
Genome project. Karyotypes are specified with a nomenclature governed by 
the International System for Human Cytogenetic Nomenclature. The parser is 
able to determine from the number and type of aberrations specified by the 
cytogeneticist whether the overall chromosome number given is accurate and 
whether the karotype is balanced. It can also be used to answer questions 
about ploidy of specific chromosomes.
	<Developed by> D. B. Searls
	<Organization> Paoli Research Center, Unisys Corp.
	<Country> USA
	<References> D. Searls. Prolog and the Human Genome Project. 
Proc. Int. Conf. on Practical Application of Prolog. London, 1992.
D.B. Searls: Investigating the Linguistics of DNA with Definite clause 
Grammars. North American Conference on Logic Programming 1989. pp189-208.
</Program>

<Program> KASA
	<Number> 81
	<Domain> Software Engineering
	<Purpose> To analyse the structure of large (Ada) software systems.
	<Description> Information about the structure of a software 
system
is obtained from an Oracle database. This information
is manipulated to form various graph structures. There are a number 
of views based on these graphs that a software
analyst can use to determine more information about
the software, thereby assisting them to locate potential
"hot-spots" in the code. Queries can be made to highlight
packages with certain properties and trial
modifications to the software structure can be made.
The information can help the analyst determine
suitable changes to the structure of the system to eliminate
the hot-spots.
	<Developed by> Philip Dart
Rudi Vernik
Gina Kingston
	<Organization> Defence Science & Technology Organisation
	<Country> Australia
	<Prolog system> Quintus Prolog with Quintus Prowindows and
Quintus ProDBI.
	<Other languages> Oracle database for information retrieval, and 
results
storage. An Ada parser is used to put the initial information
into the database.
	<Lines> 3500 (lines in current in-use system)
	<Predicates> 326
	<Platform> Sun SPARC OS 4.1.1
	<Type of Use> Personal
	<Current Status> Evaluation
	<Date deployed> Jan 1991 - Information and views were in use as 
they
we added to the system. Text files of the information
were used before any views were available.
	<No. of Users> 3 (directly) + 3 (use results, screen dumps ...)
	<Information Date> 17 March 1992
	<References> Paper in progress.
</Program>

<Program> KBTDA
	<Number> 127
	<Domain> DDB
	<Purpose> a Knowledge-Based Database Transaction Design Tool Implemented 
in Prolog
	<Description> Submitted to PAC. Paper 233
	<Developed by> Wang, Gray & Fiddian
	<Country> UK
	<Prolog system> Poplog
	<Platform> Sun
	<Information Date> 6/11/91
</Program>

<Program> KCM Help- Desk
	<Number> 128
	<Domain> Computer Systems
	<Purpose> The KCM / Help- Desk Project
	<Description> Submitted to PAC. Paper 317
	<Developed by> Richard Bland et al
	<Country> UK
	<Prolog system> SEPIA
	<Platform> KCM
	<Information Date> 31/10/91
</Program>

<Program> KNOWLEDGE EDITOR 3.20
	<Number> 340
	<Domain> Computer Based Training.
	<Purpose> General Purpose Training Program.
	<Description> The user writes question/answer pairs to be learned 
and enters them.  The program presents these tests from time to time.  
The intervals between the presentations of the same test varies from 
some seconds to months or years depending on the previous performance 
of the learner.
Tests can use DOS commands.  Therefore questions or answers may include 
graphics and/or sound.  Was used to learn German shorthand, Thai alphabet, 
HAM-radio abbreviations.
	<Developed by> Albrecht Muller
	<Country> Germany
	<Prolog system> PDC Prolog 3.20
	<Other languages> DOS batch language for graphic and sound application.
	<Lines> void
	<Predicates> void
	<Platform> DOS 4.01, AT 1 MB RAM, 20 MB Disk.
	<Date deployed> ca.  1988 (including previous version).
	<No. of Users> 1
	<Information Date> 23/06/92
	<References> void.
</Program>

<Program> Knowledge-based Image analysis system
	<Number> 228
	<Domain> quality control
	<Purpose> visual quality control in the production of complete parts
	<Description> Images of bread loafs produced by a line-scan camera are 
converted and analysed by a knowledge based system for acceptance or 
rejection. Particular aims are the provision of a highly flexible interface 
for defining the acceptance and rejection rules and a sophisticated 
explanation mode for treating dubious cases.
	<Developed by> BIM, Saab Instruments
	<Country> Belgium, Sweden
</Program>

<Program> KONE Predicate library
	<Number> 129
	<Domain> Software Develpment
	<Purpose> Prolog Predicate Library: A Practioners Approach
	<Description> Submitted to PAC. Paper 209
	<Developed by> Mr Kari Hakkarainen
	<Country> Finland
	<Information Date> 17/10/91
</Program>

<Program> KONFIRM
	<Number> 355
	<Domain> Database for church use.
	<Purpose> Database to log candidates for confirmation.
	<Description> The program logs candidates for confirmation, produces 
address-stickers to mail them, or their parents, produces namelists etc.  
It lets you search for any info, and make reports like:  "who are all the 
boys that sing in a choir".
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> PDC Prolog 3.21 + PDC Toolbox.
	<Other languages> None.
	<Lines> 1500
	<Predicates> 37
	<Platform> MS-DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 07/91
	<Information Date> 12/08/92
</Program>

<Program> KONFIX
	<Number> 130
	<Domain> PLAN
	<Purpose> a Configuration System For SOS-Emergency Services Systems
	<Description> Submitted to PAC. Paper 505
	<Developed by> Peter Nordin
	<Country> Sweden
	<Prolog system> "Prolog II, Arity, SICSTUS Prolog,  LPA Prolog, 
MacProlog"
	<Platform> "PC386, SUN Sparcstation 1, Mac"
	<Information Date> 6/11/91
</Program>

<Program> KonText
	<Number> 12
	<Domain> Text Analysis
	<Purpose> Text Analysis
	<Description> KonText provides a range of text analysis facilities 
including concordancing;collocation; wordlists and indexes.  A range of 
options allows the KonTextto operate in a multi-lingual group environment; 
scanning text files withpowerful wildcard facilities.
	<Developed by> Paul Holmes-Higgin
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog
	<Other languages>  Quintus ProWindows
	<Lines> 3000
	<Predicates> 330
	<Platform> Sun 4 / SunOS4
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1991
	<No. of Users> 5
	<Information Date> Oct 1991
	<References>  Ahmad; Fulford; Griffin & Holmes-Higgin; 1990; "The 
Translator's Workbench";In (Ed.) C.Picken; 'Translating and the Computer 
II'; pp. 394-403. ASLIB; London.
</Program>

<Program> KORE/IE
	<Number> 219
	<Domain> Expert systems
	<Purpose> A fast implementation of a production system in Prolog
	<Description> The Knowledge-Oriented Reasoning Environment/Inference 
Engine uses a speedy refutation mechanism, partial evaluation techniques 
and fast searching for heads of clauses to provide a efficient 
implementation of production systems which compares in efficiency with 
compiled OPS5 when compiled with Quintus Prolog.
	<Developed by> Toramatsu Shintani
	<Organization> IIAS-SIS Fujitsu Ltd
	<Country> Japan
	<Prolog system> C-Prolog, Quintus
	<Platform> VAX 11/780, SUN 3/52
	<References> T. Shintani: A Fast Prolog-based Production System 
KORE/IE. ICLP5, 1988, pp26-41
</Program>

<Program> KORNPLAN
	<Number> 19
	<Domain> Logistics
	<Purpose> Minimization of costs of transporting and placing grain.
	<Description> Before and during the harvest; planners in grainfirms 
have to minimize the cost of transporting and placing different types of 
grain. The grain can be placed in a number of different locations; with a 
lot of restrictions on placement.  The "KORNPLAN" program helps the planner 
optimize placement by calculating a suggested placement of the total amount 
of grain. As the problem is P-complete this suggestion is not the best 
conceivable solution; but the planner can interact with the program  and 
reach a satisfactory solution. The plan can be evaluated in a number of 
ways; including a graphical representation of transports.
	<Developed by> Lars Bo Steffensen
	<Organization> AHTS; Rosenoerns Alle 1; Frederiksberg C.
	<Country> Denmark
	<Prolog system> PDC Prolog 3.21
	<Other languages> None
	<Lines> 10000?
	<Predicates> 500?
	<Platform> IBM comp. PC; DOS 3.3 or higher.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1/10 1990
	<No. of Users> 10
	<Information Date> 6/1 1992
</Program>

<Program> L-Form
	<Number> 403
	<Domain> Form Processing.
	<Purpose> L-Form is used in corporate or government markets to automate 
form processing. Generate a laser graphic form.
	<Description> The L-Form program series generates froms on 
Hewlett Packard compatible laser printers. The L-Form program is 
designed to process both individual forms or complete form series. 
The L-Form systems permits any unlimited number of forms to be added 
to the self-generating menu. Features
include, graphic forms generated on blank paper, builds databases of 
information, form image entry screen on question and answer, automatic math 
and logic calculations, single entry system, your company logos graphic and 
forms, prints multiple copier with routing, look up tables, loads your 
databases, currency conversion, automatic numbering, ascii accumulation and 
transfer.
	<Developed by> North Winds.
	<Organization> North Winds & Hoffmann-La Roche, Inc.
	<Country> USA.
	<Prolog system> PDC Prolog
	<Other languages> None.
	<Lines> 15000
	<Predicates> 600
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser, 
printers hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> September 15, 1992.
	<Information Date> September 14, 1992.
	<References> N/A
</Program>

<Program> LanguageAccess
	<Number> 30
	<Domain> cross industry
	<Purpose> Natural language query of relational databases
	<Description> LanguageAccess is a multilingual natural language 
query interface to IBM relational databases DB2 and SQL/DS. It 
notably lowers the complexit threshold faced by users who need access 
to information from databases in IBM mainframes.
  English and German modules have been announced and prototyping has been 
performed for French, Spanish, Italian and Swedish.
	<Developed by> M. Sanamrad; G. Jonsson; H. Lehmann; M. Zoeppritz; B. 
White
	<Organization> IBM Nordic Lab; IKBS; T.J.Watson RC
	<Country> Sweden; Germany; USA
	<Prolog system> VM/Prolog; Quintus Prolog; BIM, Arity, Sicstus
	<Other languages> C; Assembler
	<Lines> 60000
	<Predicates> 10000
	<Platform> MVS;VM
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 12/91
	<Information Date> 1/92
	<References> IBM Corp (1990) IBM SAA LanguageAccess General 
Information. Publ. no. SH19-6680
IBM Corp (1990) IBM SAA LanguageAccess Query Interface User's Guide. 
Publ. no. SH19-6681
IBM Corp (1990) IBM SAA LanguageAccess Customisation Tool User's 
Guide. Publ. no. SH19-6683
IBM Corp (1990) IBM SAA LanguageAccess Application Programming Guide. 
Publ. no. SH19-6687
M.A. Sanamrad, I. Bretan. IBM SAA Languae Access: A large scale 
commercial product implemented in Prolog. 1st. Int. Conf. on 
Practical Application of Prolog, London. 1992.
</Program>

<Program> Laser Blankrupter
	<Number> 396
	<Domain> Legal forms processing.
	<Purpose> Completely processes all the Federal Bankruptcy Forms for 
chapter 7, 11, 12 & 13 bankruptcy filings.
	<Description> Laser Blankrupter processes and generates on laser 
printers complete documents for chapter 7, 11, 12 & 13 for the U.S. 
Bankruptcy Court. Sold nationwide through Julius Blumberg, Inc. 
distribution channel, this program has new capabilities never 
available in bankruptcy software. These include laser generated 
Blumberg forms, the creation of dynamic forms (ie. the amount of 
entry determines the number of forms produced) and a new file
achieving system designed for law firms. Price: $ 400.00.
	<Developed by> North Winds and Julius Blumberg Inc.
	<Organization> North Winds and Julius Blumberg Inc.
	<Country> USA.
	<Prolog system> PDC Prolog (100%)
	<Other languages> none.
	<Lines> 18.000
	<Predicates> 600
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/1/91
	<Information Date>  9/1/92
	<References>  Article: National Law Review by Richard Gertler.
</Program>

<Program> LDM
	<Number> 268
	<Domain> Software development tools
	<Purpose> A logic based development method
	<Description> A logic based development method (LDM) is designed which 
comprises of an extended logic programming language, a methodology for 
software development and a program to support both the language and the 
methodology.  The LDM system makes possible efficient interactive testing 
of specification by providing a compiler from LDM to Prolog.
	<Developed by> K. Balogh, Z. Farkas, E. Santane-Toth, P. Szeredi 
	<Organization> SzKI
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> K. Balogh, Z. Farkas, E. Santane-Toth, P. Szeredi 
(1979)
Logic based program design
(Hungarian) Proc. of first National Conf. of von Neumann Computer 
Society, Szeged 
1979. pp 36-45
K. Balogh, Z. Farkas, E. Santane-Toth, P. Szeredi (1979)
The LDM system - a design outline.
SzKI report, SOFTECH D46, SZAMKI, 1979.
Z. Farkas (1981)
User's manual of the LDM system
SzKI report, 1981.
Z. Farkas, P. Szeredi, E. Santane-Toth (1982)
LDM -a program specification support system.
SzKI report, 1982.
P. Szeredi, K.Balogh, E. Santane-Toth, Z. Farkas (1980)
LDM - a logic based development method.
Reprints of Logic Programming Workshop, Debrecen(Hungary) 1980.
</Program>

<Program> Leonardo
	<Number> 177
	<Domain> Database interface
	<Purpose> Intelligent front end to free text database
	<Description> Leonard provides an intuitive menu driven interface to 
information of critical importance to the nuclear power industry. This must 
be accessed for years to come by people wholly unfamiliar with the database 
products and query language. The system also gives advice to the user on 
improving the nature of the query, which in turn gives more meaningful 
question to the database to produce the most effective response.
	<Organization> Quintec systems Ltd
	<Country> UK
	<Prolog system> Quintec
</Program>

<Program> LES - Laboratory Expert System
	<Number> 88
	<Domain> Biomedical Research
	<Purpose> Expert System for Designing PC Boards
	<Description> Rockefeller University's Biomedical Research Laboratory 
produces "Labmaster", a PC board designed for biomedical research. 
Labmaster monitors outside data and converts electronic signals to digital 
signals that the computer can read. The specific configuration of of the PC 
board required for a given experiment varies depending on the research and 
types of data required. Determining an apropriate board for the experiment 
can be a time-consuming and error-prone task. To minimize the time and cost 
involved, Arity Prolog was used to develop the Laboratory Expert System 
(LES). A research technician utilizing LES simply inputs the criteria for a 
given experiment. LES, run on a PC, then analyses the requirements and 
generates a list of possible board designs that meet the criteria.
	<Developed by> Rockefeller University
	<Country> USA
	<Prolog system> Arity Prolog
	<Type of Use> In-house
	<Current Status> Released
</Program>

<Program> LEX0
	<Number> 31
	<Domain> Legal expert system
	<Purpose> Expert System on German traffic law
	<Description> Cases involving illegal absentation from the scene of an 
accident are interactively described and solved by the expert system 
	<Developed by> S. Grundmann; W. Schoenfeld
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Lines> 1000
	<Predicates> 200
	<Platform> VM
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 6/84
	<Information Date> 1/92
</Program>

<Program> LEX1
	<Number> 32
	<Domain> Legal expert system
	<Purpose> Expert System on German traffic law
	<Description> Cases involving illegal absentation from the scene of an 
accident are described in natural language and interpreted and solved by 
the expert system
	<Developed by> B. Blaeser; H. Lehmann; W. Schoenfeld; M. Zoeppritz
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Other languages> PL1
	<Lines> 10000
	<Predicates> 2000
	<Platform> VM
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 9/87
	<Information Date> 1/92
</Program>

<Program> LEX2
	<Number> 33
	<Domain> Legal expert system
	<Purpose> Expert System on German traffic law
	<Description> Cases involving illegal absentation from the scene of an 
accident are interactively described and solved by the expert system 
	<Developed by> B. Blaeser; H. Lehmann; W. Schoenfeld; M. Zoeppritz
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Other languages> PL1
	<Lines> 1000
	<Predicates> 200
	<Platform> VM
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 12/87
	<Information Date> 1/92
</Program>

<Program> LIMBO
	<Number> 195
	<Domain> Airline planning
	<Purpose> Aircraft maintenance planning
	<Description> Limbo provides an on-line graphical environment to 
prevent LIne Maintenace BOttlenecks. It can take information from mainframe 
databases and provide a state of the art user interface on a PC 
workstation. It makes sure that no maintenance job is delayed or omitted 
without notice, checks for bottlenecks, integrates material consumption in 
the planning, optimizes job selection so jobs are not performed more often 
than necessary, an checks flight and maintenance plans for consistency.
	<Organization> PDC
	<Country> Denmark
	<Prolog system> PDC 
	<Platform> 386 based PC with 12MB RAM, VGA or EGA monitor, OS/2 v 1.2 or 
later. 3270 and LU6.2 communications currently implemented
	<Current Status> Released
</Program>

<Program> LIMBO - Line Maintenance Bottleneck Detection
	<Number> 21
	<Domain> Airline Fleet Management
	<Purpose> Integrated Flight and Maintenance Planning
	<Description> LIMBO is an on-line graphical environment for 
aircraft maintenance management. The coordination of flight plans and 
mandatory maintenance jobs is an overwhelming task for a large 
aircraft fleet. With LIMBO this task is highly automated: Maintenance 
opportunities are extracted from the flight plans and maintenance 
jobs are allocated according to the complex rules of aircraft 
maintenance and resource availability. LIMBO provides information 
processing of data from large mainframe databases on a PC workstation 
with state-of-the-art user interface and scheduling
 features; hence providing an efficient and flexible tool for information 
management. With LIMBO; maintenance planners can detect possible problem 
situations and resource bottlenecks in a user friendly  point-and-click 
manner; allowing for timely action and rescheduling.
	<Developed by> Lars Malmgren Hansen; Lars Henrik Jacobsen; Mikael 
Nielsen
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Other languages> C
	<Lines> >10.000  <100.000
	<Platform> OS/2 PM
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 03/92
	<No. of Users> 25
	<Information Date> 01/92
</Program>

<Program> Linear Objects (LO)
	<Number> 15
	<Domain> Computer systems
	<Purpose> Prototype of an interpreter for a process based concurrent 
programming language (called LO).
	<Description> The language LO is a logic based concurrent 
programming language.
It is a tool for coordinating the behaviors of computational processes; in 
an open world environment. It has an object oriented flavor; in the sense 
of actors languages. In the prototype implementation; concurrency is only 
simulated; using the coroutining capabilities of Prolog.
	<Developed by> JM. Andreoli and R. Pareschi
	<Organization> European Computer Industry Research Center
	<Country> Germany
	<Prolog system> Sepia
		  - compiles to some extended WAM code
	<Other languages> none
	<Lines> 1200
	<Predicates> 150
	<Platform> Sun workstation; Unix
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> around June 91
	<No. of Users> 4
	<Information Date> November 91
	<References>   Among others
		  "LO and behold! Concurrent structured Processes"
		  JM. Andreoli and R. Pareschi
		  proc. of OOPSLA'91 (Phoenix; Az)
</Program>

<Program> LMT
	<Number> 204
	<Domain> Machine translation
	<Purpose> Machine translation system for English to German
	<Description> The Logic-based Machine Translation system is based 
on a grammatical formalism called Modular Grammars based on slot 
filling techniques developed by the author, which includes some 
automatic semantic translation and handling of metagrammatical rules.
The principle aim of the system is translate computer manuals from English 
into German, and testing was done on the IBM CMS Editor, XEDIT. It is 
capable of translating around 6 sentences per second on an IBM 3081.
	<Developed by> M. McCord
	<Organization> IBM Yorktown Heights Laboratory
	<Country> USA
	<Prolog system> VM/Prolog
	<Lines> 1,950 clauses, plus 1,750 for lexicons
	<Platform> IBM 3081
	<References> M.McCord. Design of a Prolog-based machine 
translation system. Proc 3rd Int. Conf. Logic Programming, Springer-
Verlag, Berlin, 1986. pp350-374
M.C. McCord. Using slots and modifiers in logic grammars for natural 
language. Artificial Intelligence, vol 18 pp. 327-367, 1982.
</Program>

<Program> LMTX
	<Number> 205
	<Domain> Machine translation
	<Purpose> Multi-target machine translation system
	<Description> LMTX is a reorganization and subsystem of the 
earlier LMT system which provides a shell for English-to-X language 
translation. Versions have been started for French, Danish, Spanish 
and Portugese and the German version is being further improved. One 
can switch languages in the same session by simple command.
The analysis and synthesis phases are treated by different techniques with 
no attempt at providing symmetry. It uses a syntactic transfer but with 
semantic information in the transfer tree. 
	<Developed by> Michael McCord
	<Organization> IBM Yorktown Heights Laboratory
	<Country> USA
	<Prolog system> VM Prolog
	<Platform> IBM 3033
	<References> M. C. McCord: A multi-target Machine Translation System. 
Proc. Int. Conf. on Fifth Generation systems, 1988. pp1141-1149.
</Program>

<Program> Loganair Scheduler
	<Number> 185
	<Domain> Airline scheduling
	<Purpose> Help in planning and calculation of flight crew duty rosters 
for airline
	<Description> Reduces the time for calculating duty rosters from a 
couple of days to two hours. Allows adherence to requirements laid down by 
the Civil Aviation Authority, including mandatory rest periods and minimum 
period of notification. Used to schedule the rosters of around 200 flight 
crews.
	<Organization> Hoskyns
	<Country> UK
</Program>

<Program> LogicBase
	<Number> 131
	<Domain> KBS
	<Purpose> "LogicBase, A Knowledge Base Program"
	<Description> Submitted to PAC. Paper 368
	<Developed by> Ruggles
	<Country> UK
	<Prolog system> PDC Prolog
	<Platform> any IBM XT
	<Information Date> 6/11/91
</Program>

<Program> LOGICLANDIA 
	<Number> 433
	<Domain> Education, more precisely the teaching of predicates logic.
	<Purpose> Supporting the teaching of predicates logic in Italian middle 
and upper secondary schools.
	<Description> LOGICLANDIA is a Prolog shell to create, edit and 
query simple databases, which represent a formalisation of a given 
problem in terms of classes of objects and relations among them. The 
language used to interact with the software is  a formal language, 
which includes the logical connectives (and, or, not, implies) and 
the logical quantifiers (existential and universal). LOGICLANDIA also
includes a facility that translates into Italian the student
formalisation of a query, to facilitate the students  understanding 
of their mistakes. LOGICLANDIA's user interface is implemented in C 
language.
LOGICLANDIA  belongs to a courseware aiming at  promoting the learning of 
propositional and predicate logic through a problem oriented approach. 
Besides the software, the courseware includes a student workbook, teacher's 
notes and assessment material. The courseware has been field tested in 
Italian schools.
	<Developed by> Donatella Persico, Carlo Innocenti.
	<Organization> Consiglio Nazionale Ricerche - Istituto Tecnologie 
Didattiche
	<Country> Italy.
	<Prolog system> Arity Prolog
	<Other languages> Microsoft C 6.0
	<Lines> 30000
	<Predicates> 1000
	<Platform> 286 micro + DOS OS
	<Current Status> Evaluation
	<Date deployed> September '92
	<No. of Users> 400 during evaluation, hopefully many more when sold.
	<Information Date> April 92
	<References> Persico D. (1990)
Teaching logic: why and how
in: Roy Winterburn (ed) Aspects of Educational and Training 
Technology, Vol. XXIV, Kogan Page, London, pp.111-116
Persico D. (1990)
Guided discovery learning of logic
in: Education and Application of Computer Technology, De Blasi M., Luque 
E., Scerri E. (eds), Third book of the biennal meeting of the Community of 
Mediterranean Universities on Microcomputers and their Applications, 
Edizioni Fratelli Laterza, Bari, pp.53-63,
</Program>

<Program> LOPSTER (LOgic Programming with SubTypes and Extended 
Resolution)
	<Number> 370
	<Domain> Used as declarative language for general purposes like 
PRO
LOG itself.
	<Purpose> Compiler for the order-sorted logical language LOPSTER 
which is an extension of the KA-Prolog compiler.
	<Description>  LOPSTER is an extension of PROLOG by order-sorted 
resolution.  The main benefits offered by sort-restricted deduction 
are shorter programs and hence smaller search spaces (part of the 
deduction  is effected during unification) and built-in inheritance 
mechanisms. In order to implement dynamic typing the LOPSTER compiler 
generates extended WAM code which can manage the order-sorted 
unification on runtime. If possible, the compiler tries to simplify 
statically the greatest lower bound operation on the sort lattice.  
By an extension and modification of a given Prolog compiler it was 
possible to implement the compiler very quickly. The numbers below 
are refering to the whole in KA-Prolog written compiler. The 
indications in brackets are describing that part of the compiler, 
which makes the additional functionality of 
the LOPSTER compiler available in relation to the original (insource  code 
given) Prolog compiler. 
	<Developed by>  A. Biere, C. Hintermeier, N. Lindenberg, T. Weinstein.
	<Organization> University of Karlsruhe, partly supported by the 
Deutsche 
Forschungsgemeinschaft (SFB 314, project S2).
	<Country> Germany.
	<Prolog system> KA-Prolog (WAM based).
	<Lines> 9500 (1000)
	<Predicates> 500   (40)
	<Platform> UNIX-OS, SUN workstations.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> does exist since 2 years.
	<No. of Users> 10
	<Information Date>  August 11th, 1992.
	<References> H. Huber, I. Varsek: "Extended Prolog with Order-
Sorted Re
solution" (1987), Proc. 4th. Symp. Logic Programming, San Francisco 
1987.
C. Hintermeier, C. Brzoska, A. Biere, I. Varsek, T. Weinstein, A. Werner: 
"Lopster: Sprachdefinition", University of Karlsruhe, Fakultaet fuer 
Informatik, Interner Bericht 18/91.
</Program>

<Program> Loqui
	<Number> 132
	<Domain> NL
	<Purpose> A new approach to natural language interfaces
	<Description> Submitted to PAC. Paper 440
A multilingual natural language interface (English, German, French, 
Dutch) to databases.  It offers a large linguistic coverage, a fully-
fledged discourse manager and an object-oriented representation of 
the domain of discourse. 
It offers response time of less than 1 second for complex queries.
	<Developed by> BIM
	<Country> Belgium
	<Prolog system> BIM
	<Lines> 30,000
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> LORSMS
	<Number> 494
	<Domain> Risk Calculation for optimising Safety.
	<Purpose> Transport Risk Analysis.
	<Description> Performs automatic event tree generation and 
consequence
analysis to calculate the risks to passengers, the public and employees on 
Rail Systems.
	<Developed by> Steve Lympany.
	<Organization> British Rail.
	<Country>  England.
	<Prolog system> PDC Prolog.
	<Other languages>  None.
	<Lines> 2000
	<Predicates> 200
	<Platform> MS-DOS on IBM PC.
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Information Date> Feb 1993.
</Program>

<Program> LTUTOR
	<Number> 34
	<Domain> Legal
	<Purpose> Teaching a Law student about German Civil Law
	<Description> Interactive solution of cases involving expert system 
techniques
	<Developed by> S. Grundmann
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Lines> 1500
	<Predicates> 300
	<Platform> VM
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 12/86
	<Information Date> 1/92
</Program>

<Program> MacCLINT
	<Number> 18
	<Domain> Machine Learning; Inductive Logic Programming 
	<Purpose> Interactive Concept-Learning; Theory Revision; Belief Updating
	<Description> MacCLINT is a user-friendly interactive concept-
learner and theory revisor that combines several interesting 
features: it constructs most of its own examples; uses knowledge; 
handles integrity constraints; performs abduction; shifts its bias; 
postpones examples and converges to knowledge bases in the limit. 
This unique combination of features makes MacCLINT significantly 
differe
nt from other approachess in Inductive Logic Programming. MacCLINT 
contributes to the removal of the knowledge acquisition bottleneck; can be 
used for intensional knowledge base updating and as a module for learning 
apprentices.
	<Developed by> Luc De Raedt; Wim Van Holder; Maurice Bruynooghe
	<Organization> K.U. Leuven
	<Country> Belgium
	<Prolog system> LPA MacPROLOG 3.5
	<Lines> 9000
	<Predicates> 525
	<Platform> Apple Macintosh 
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> June 1991
	<No. of Users> 15
	<Information Date> Dec 17 91
	<References> De Raedt; L. and Bruynooghe; M.; Towards friendly 
concept-learners. In Proceedings of the 11th International Joint 
Conference on Artificial Intelligence; 1989; p849-856; Morgan 
Kaufmann;
De Raedt; L. and Bruynooghe; M.; An Overview of the Interactive 
Concept-Learner and Theory Revisor Clint. To appear in Inductive 
Logic Programming; S. Muggleton; Academic Press; 1991;
De Raedt; L. and Bruynooghe; M.; On interactive concept-learning and 
assimilation. In Proceedings of the 3rd European Working Session On 
Learning; Sleeman; D.; Pitman; p167-176; 1988;
De Raedt; L. and Bruynooghe; M.; Interactive Concept-Learning and 
Constructive Induction by analogy. To appear in Machine Learning Journal; 
vol. 8; no. 2; p107-150; 1992.
</Program>

<Program> Machiavelli
	<Number> 182
	<Domain> Sales planning
	<Purpose> Sales planner
	<Description> An extension to Prolog which offers
	<Developed by> Dermot Bradley
	<Country> UK
	<Prolog system> LPA MacProlog
	<Platform> Macintosh
</Program>

<Program> Machine Assisted Terminology Elicitation Environment.
	<Number> 510
	<Domain> Translation.
	<Purpose> Technical translation of large volumes of technical data.
	<Description> M.A.T.E. provides a multi lingual text processing 
facility including a term bank, building tools, grammar, style and 
spell checking facilities.
It consists of a set of tools that allow graphical display of term bank.  
The term publisher allows a number of multi-lingual and monolingual 
dictionary formats.
	<Organization> University of Surrey-A.I. Group.
	<Country> U.K.
	<Prolog system> Quintus Prolog.
	<Platform> Unix.
	<Current Status> Evaluation
	<No. of Users> 5
	<Information Date> Jan 1992.
	<References> Ahmad,K, Fulwood,H; Griffin, S; and Holmes, P; 
(1991)
"Text Based Knowledge Aquisition - A language for special purposes 
Perspective"  In (tds) I.M. Graham & R.W. Milne, R & D in E.S. V11, pp 148-
162.  Cambridge University Press.
</Program>

<Program> MacLogic
	<Number> 188
	<Domain> Teaching system
	<Purpose> Teaching of systems of natural deduction in logic
	<Description> Allows both top-down and bottom-up proof construction of 
propositional logic proofs. It includes a special font for representing 
logical constants ad a full on-line help system.
	<Developed by> Dr. Roy Dyckhoff
	<Organization> University of St. Andrews
	<Country> UK
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> Macmillan Rehabilitation ?
	<Number> 184
	<Domain> Disabled computing
	<Purpose> To aid disabled people to configure and maintain appropriate 
settings on computer input devices
	<Description> 
	<Developed by> Rod McGill
	<Organization> U. Toronto with Hugh Macmillan Rehabilitation centre
	<Country> Canada
	<Prolog system> LPA MacProlog
	<Other languages> 100KB compiled Think Pascal
	<Lines> (170KB)
	<Platform> Macintosh IIci
</Program>

<Program> MailerMate II
	<Number> 398
	<Domain> Form processing.
	<Purpose> Used to fill and print self-mailer forms using a micro computer 
and printer.
	<Description> MailerMate II is a friendly, easy way to fill in 
your self-
mailer forms using your micro computer and to print them error free 
in minutes. You type information into an electronic image of the 
form. The form can be edited and printed immediately or saved to a 
batch for later printing. 
MailerMate II remembers information about your clients and 
transactions. It does all the math calculations and pertinent data 
from each form is saved in a transaction register. Its ASCII 
mailmerger feature allows you to merge 
your databases into your mailer forms. It also creates mailing lists, has 
macros, uses almost any printer.
	<Developed by> North Winds.
	<Organization> North Winds and Transkrit Corporation.
	<Country> USA.
	<Prolog system> PDC Prolog.
	<Other languages> None.
	<Lines> 12.000
	<Predicates> 500
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> January 1, 1991.
	<Information Date> September 1, 1992.
	<References> Literature.
</Program>

<Program> ||MAPLE|| (speak: parallel Maple)
	<Number> 27
	<Domain> Symbolic mathematics
	<Purpose> System for Parallel Symbolic Computation
	<Description> ||MAPLE|| is a System for parallel Symbolic 
Computation where  logic programming provides parallelism and 
imperative programming provides efficiency. 
 It is built as a combination between the parallel logic programming 
language Strand and the sequential Computer Algebra system Maple.
 Since ||MAPLE|| can check for arbitrary algebraic properties in 
guards; it may also used as constraint logic programming language.
	<Developed by> Kurt Siegl
	<Organization> RISC-Linz; University of Linz
	<Country> AUSTRIA
	<Prolog system> STRAND88
	<Other languages> Maple computer algebra system: 
400 KB object code C written kernel; 8 MB sequential library. Interface 
Kernel: 300 lines C-code
	<Lines> 250 lines Strand code
	<Predicates> one guard and two body calls.
	<Platform> 16 Transputer system under Helios; 
delivered: All where STRAND88 is available.
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 1.quarter 1992
	<No. of Users> 2
	<Information Date> 2. 12. 91
</Program>

<Program> Marvin
	<Number> 8
	<Domain> Knowledge Based Systems
	<Purpose> Knowledge Based Systems Development Environment.
	<Description> Marvin provides a unified framework based on frames 
for developing Knowledge Based Systems through three main levels of 
representation:frames; rules and tasks.  Various forms of forward and 
backward inferenceare supported as well as Variable Precision Logic.  
Its X-windows developmentenvironment makes full use of windows and 
graphical representation of theknowledge structures.
	<Developed by> Paul Holmes-Higgin
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog
	<Other languages> Quintus ProWindows
	<Lines> 3600
	<Predicates> 1180
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 1988
	<No. of Users> 10
	<Information Date> Oct 1991
</Program>

<Program> MASS-C (Mechanical Assembly Sequence
		  Satisfiability-Checker)
	<Number> 365
	<Domain> Mechanical Assembly Sequence Planning.
	<Purpose> Verifies and generates all assembly sequences based on the task 
precedence ordering specifications written as a logic program of 
implemented temporal predicates.
	<Description> An assembly constraint language based on linear time 
temporal logic of Manna and Pneuli. The temporal operators are implemented 
as logic predicates so that any temporal formulae, asserting the acquired 
assembly constraints, can be directly coded as a logic program of the 
respective predicates. Auxiliary predicates are implemented to verify or 
synthesize all assembly sequences that satisfy the logic program of 
assembly constraints.
	<Developed by> KiamTian Seow (Mr) and R.Devanathan (Dr).
	<Organization> Division of Instrumentation and Control School of 
Electrical and Electronic Engineering Nanyang Technological University.
	<Country>  Singapore.
	<Prolog system> Quintus Prolog Release 3.1.1 for Sun-3 and Sun-4.
	<Other languages>  Nil
	<Lines> 145
	<Predicates> 32
	<Platform> IPC SUN SPARC Workstation, Unix operating system.
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> March 1992
	<No. of Users> N.A
	<Information Date> 15th July 1992
	<References>  [1] KiamTian%Seow and R.%Devanathan, ``Temporal 
logic formulation of assembly sequence properties, '' Proceedings of 
the IEEE International Conference on Robotics and  Automation, May 
1992. Nice, France.
  [2] KiamTian%Seow and R.%Devanathan, ``A temporal logic framework 
for assembly sequence planning,'' Proceedings of the IEEE/RSJ 
International Conference on Intelligent Robots and Systems, July 
1992. North Carolina, USA.
  [3] KiamTian%Seow and R.%Devanathan, ``An extended logic framework for 
mechanical analysis of assembly sequences,'' To appear in Proceedings of 
the Second International Conference on Automation, Robotics and Computer 
Vision, Sept 1992. Singapore.
</Program>

<Program> Master Project Coordinating Program
	<Number> 461
	<Domain>  Intelligent Computer-Aided Building Design.Computer Integrated 
Construction using Component-Based Building Systems.
	<Purpose> The MPCP is a Blackboard System based on the Edinburgh BBS. It 
is used to maintain a 3D solid model of a building for intelligent design 
purposes.
	<Description> The MPCP underpins a collection of computer-aided 
building design programs which have been written to use components from a 
well defined set of construction elements. The MPCP maintains the 3D model 
of the building and filters out changes which will cause problems if the 
building is ever constructed. It also looks for a given set of problems 
such as clashing components, typically "silly" architectural details, and 
so on. It flags the problem and generally has a few good ideas about what 
to do about it.
	<Developed by> Colin Bridgewater
	<Organization> Construction Robotics Research Group, Dept of Civil 
Engineering, Imperial College.
	<Country> UK
	<Prolog system> Sussex Poplog PROLOG, v14.1.
ÇIF*** The text replacing this argument is too long ***È<Other languages>
	 The building model is in ASCII text format on either PC-DOS machines 
or the SPARCstation. The expert systems that have been written use either 
AutoCAD's implementaion of LISP, K&R C, TurboPascal or LOKI (the Virtual 
Reality interface script which has been developed by Mike Griffin at 
Reading University - guess what, we can do "walkthroughs" of the MPCP 
buildings using Mike's VR kit...).
	<Lines> The disk usage is ~ 250 Kb. I have no idea how many lines that 
is.
	<Predicates> 150 MPCP-specific predicates, 240 BBS-specific 
predicates.
	<Platform> SUN Microsystems SPARCstation 1, running SunOS 4.1.1 (Unix).
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed>  April 1992.
	<No. of Users>  2 (one in Construction Robotics, the other in VR).
	<Information Date>  Current.
	<References> 1.  A New Approach to Automating Construction, Proc 
6th Intl Symp on Automation and Robotics in Construction, 
Construction Indus Inst, San Fransisco, June1989.
2.  MPCP: Computer Infrastructure for A Component-Based Building 
System, Proc 7th Intl Symp on Autom and Rob in Constr, Bristol Poly, 
Bristol, June 1990.
3.  On Automation in the Construction Industry, Proc Intl Symp on 
Building Econ and Constr (CIB 90), 6, Sydney, 1990. 
4.  An Integrated Approach Towards Automating Construction, Proc 18th 
Intl    Association for Housing Science World Congress - Buildings in 
the Future, Rio de Janeiro, 1990.
5.  Design for Automation: An Operational Framework, Proc CIB W74 + 
W78 Symp on Comp Int Constr, Architectural Institute of Japan, Tokyo, 
1990, B22.
6.  Design for Automation: An Operational Framework, in Computer 
Integrated Construction, ed. H. Wagter, Elsevier Science Publishers, 
London, 1992.
7.  Modular Building Systems and the Role of Construction Automation 
Systems and Robotics, in The Practice of Site Management, 4, ed. P.A. 
Harlow, Chartered Institute of Building, Ascot, 1992.
8.  A New Approach to the Design of Buildings for Automated 
Construction, PhD thesis, Dept of Engineering, University of Reading, 
Reading, 1992. 
9. Component-Based Building Design Using Artificial Intelligence 
Techniques, paper submitted to EuropIA-93 conference, to be held in 
1993.
(all of these papers - with the exception of [8] - were written 
jointly by all of the members of the Construction Robotics Research 
Group. These were Prof BL Atkin of the Dept of Construction 
Management & Engineering, Mr P Atkinson of the Dept of Engineering, 
Snr J Ibanez-Guzman and myself
who were jointly owned by these two Departments of Reading University). 
</Program>

<Program> MATE
	<Number> 133
	<Domain> NL
	<Purpose> The Machine Assisted Terminology Elicitation Environment : Text 
and Data Processing and Management in Prolog
	<Description> Submitted to PAC. Paper 428
	<Developed by> Paul Holmes - Higgin
	<Country> UK
	<Platform> PC + Windows
	<Information Date> 6/11/91
</Program>

<Program> Mathias
	<Number> 49
	<Domain> Mathematics
	<Purpose> Animation of Discrete Mathematics
	<Description> Rules for Discrete Maths: to generate combinatoric 
objects like sets and set operations; partitions of integers (sets; 
sequences; bags); sub collections of sets; sequences and bags; 
vectors and matrices; cartesian products of various kinds; functions 
(total/partial; into/onto; oneone; manyone) and relations and 
operations on them (closure; domain restriction; etc);polynomial 
manipulation(not much); some useful system-independent input/output 
rules and higher order rules for rules which have an infinite number 
of solutions(count; sigma; memoization; while; reduce); standardizes 
arithmetic functions (eg mod and div on negative integers; floor; 
ceiling; power function; factorial;gcd; random numbers).
Works on a set of about 25 foundation rules called "underware" (for 
easy portability) and provides a prolog-system-independent way of 
writing standard Prolog with fixed meaning and guaranteed built-in 
predicates.  Uses built-in predicates where available and renames 
them for portability. Takes about half to 1 day to port the underware 
then the whole of Mathias runs identically on any system. Includes 
complete on-line help (with key words) and on-line Manual of about 90 
pages (generated by the system). Has been used on Safety Critical 
systems and to animate Z specifications and network protocols; object 
oriented set theory.
  On top of the basic Mathias system described here; there are other 
modules in preparation.
	<Developed by> Dr Ron Knott
	<Organization> University of Surrey
	<Country> UK
	<Prolog system> Quintus; LPA MacProlog; Arity; CProlog; SDProlog; 
SICStus; NIP; PrologIX
	<Other languages>  
	<Lines> 3415 Mathias +400 (under)+ 3980 (Graphs)
	<Predicates> 262(Mathias+Underware)
	<Platform> Vax; Sun; PC; Apple Mac; Prime
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 1988; latest version 1991
	<No. of Users> Software developers; Z users; mathematicians; education.
	<Information Date> 9 March 1992
	<References> "Making Discrete Maths Executable on a Computer" 
Proceedings of the IMA
Conference on "The Mathematical Revolution Inspired by the Computer" 
held
in April 1989; Clarendon Press ; Oxford (1991) pgs 127-144; ISBN 0 19
853658 5.
"Using Prolog to Animate Mathematics" in 'Logic Programming: New 
Frontiers';D R Brough (Ed);Intellect Books (1992) ISBN 1-8715116-25-
0.
"The Implementation of Z Specifications using Program Transformation 
Systems: The SuZan Project" in I.M.A. Conference Proceedings Series 
"The Unified Computation Laboratory" (Editors: C M I  Rattray; R G 
Clark); Oxford University Press; 1992. 
"Animating Set-Theoretic Specifications using Prolog (Collected Papers)" 
Technical Report; Department of Mathematical and Computing Sciences; 
University of Surrey; 1989. This contains the 4 Alvey Reports on Animating 
Z together with 3 papers on Prolog and the animation of discrete 
mathematics.
</Program>

<Program> MATRUSKA
	<Number> 71
	<Domain> Language Research
	<Purpose> For creating specialized Russian-English/vice versa 
dictionaries
	<Description> Translation of words, phrases and grammatical 
terms. Many-to-many relationships permit synonyms and alternative 
translations. The program "automatically" translates both way by 
sensing whether Latin or Cyrillic text is entered at the keyboard.
Use intended for special language domains (Petroleum technology, Marine 
biology, Law, etc), or for students of Russian. The system built-in text 
predicates are rewritten, e.g. "russ_fronttoken," "russ_upper_lower," etc. 
Context sensitive help.
	<Developed by> Audun Toennesen
	<Organization> Public Domain
	<Country> Norway
	<Prolog system> PDC PROLOG
	<Other languages> Requires a Russian Character-driver
	<Platform> IBM PC/ DOS
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> May 1991
	<No. of Users> Unknown.
	<Information Date> March 1992
</Program>

<Program> Max
	<Number> 134
	<Domain> Decision Support
	<Purpose> Knowledge-based Decision Support : A Practical Application of 
Logic Programming and Prolog
	<Description> Submitted to PAC. Paper 278
	<Developed by> Bhargava, Bieber, Kimbrough
	<Country> USA
	<Platform> Mac
	<Information Date> 6/11/91
</Program>

<Program> MCDS ?
	<Number> 256
	<Domain> manufacturing systems
	<Purpose> Manufacturing Cell Design System
	<Description> The configuration of the manufacturing cell is performed 
by an expert system which selects appropriate resources for the given 
manufacturing task. For layout purposes a special version of AUTOCAD system 
is used. The scheduling, based on the previous results, is done either by a 
traditional scheduler or by the simulation module itself.
	<Developed by> George Kov‡cs-Istv‡n, Mezg‡r-S‡ndor Kop‡csi
	<Organization> Hungarian Academy of Sciences, Computer and Automation 
Institute
	<Country> Hungary
	<Prolog system> CS-Prolog
	<Other languages> C, GSS*CGI graphics package
	<Type of Use> In-house
	<Current Status> Prototype
	<Information Date> Apr 92
	<References> George Kov‡cs-Istv‡n, Mezg‡r-S‡ndor Kop‡csi: A Prolog 
Based Manufacturing Cell Design System. Abstracts of poster sessions, Int. 
Conference on Practical Application of Prolog, London 1992.
</Program>

<Program> McExpert
	<Number> 135
	<Domain> Decision Support
	<Purpose> A logic programming model of multiple criteria decision making
	<Description> Submitted to PAC. Paper 151
	<Developed by> David McSherry
	<Country> UK
	<Prolog system> MacProlog
	<Platform> Mac
	<Information Date> 31/10/91
</Program>

<Program> Medical Expert Diagnostic System
	<Number> 96
	<Domain> Medical - GP
	<Purpose> 
	<Description> The Imperial Cancer Research Foundation developed this 
system in association with the Oxford University Press to investigate the 
feasability of providing a fully fledged medical expert diagnostic system 
for experimental deployment in the surgeries of GPs. The system was 
originally implemented in Prolog on a workstation but reimplemented on a PC 
using Quintec Prolog using a database of medical knowledge and rules of 
inference supplied by experts. The delivered system was able to provide a 
rapid and comprehensive response to aid in the the GP's own process of 
diagnosis, verification and elimination. The two phases took approximately 
one person year, but the developers estimate it would take two months to 
repeat the work. The system has been evaluated at several UK and European 
sites and the methodology is now being developed in an oncology project 
which includes facilities for the recommendation of effective treatment and 
drug therapy.  
	<Developed by> Imperial Cancer Research Foundation
	<Country> UK
	<Prolog system> Quintec
	<Lines> 20,000  facts and rules
	<Platform> PC
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> MedPro
	<Number> 342
	<Domain> Medical records and billing.
	<Purpose> Record patient information and bill for services provided by a 
Medical Practitioner operating in Ontario, Canada.
	<Description> A menu driven program which allows a physician to enter 
and retrieve information on patients.  Billing and reconciliation of 
payments is also handled.  The system supports multi-user access so that 
the database may be viewed and updated by more than one person 
simultaneously.  Locking is done at the patient record level.
	<Developed by> Steve Luck.
	<Organization> J.C.L. Enterprises Ltd.
	<Country> Canada.
	<Prolog system> PDC Prolog 3.21
	<Other languages> RT Link used for overlayes and run under VM 386 for 
multi-station access.
	<Lines> 9642 source lines, i.e. including blank lines and comments 7795 
lines original 1847 lines taken from Prolog Toolbox.
	<Predicates> 334 made up as follows:
73 global; 229 local;32 global data base.
	<Platform> IBM PC running DOS or DOS network.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 06/91
	<Information Date> 08/92
</Program>

<Program> MEES.
	<Number> 408
	<Domain> Expert System.
	<Purpose> Economic Simulator.
	<Description> An expert system for quantitative simulation of the 
American economy, at the level of "if quantity 1 increases, then quantity 2 
decreases". A rapid prototype, enabling the user to test various sets of 
rules before building a more comprehensive system (in KEE, I believe). 
	<Developed by> Jocelyn Paine.
	<Organization> Mike Salant (private).
	<Country> England.
	<Prolog system> Poplog.
	<Other languages> None.
	<Lines> 500.
	<Predicates> 50.
	<Platform> VAX/VMS.
	<Type of Use> Personal
	<Current Status> Dead
	<Date deployed> 12/1984.
	<No. of Users> 1 (person).
	<Information Date> 3/1992.
	<References> Referred to briefly in "The Guide to Expert  Systems" by 
Alex Goodall, Learned Information, 1985.
</Program>

<Program> MEGALOG
	<Number> 244
	<Domain> Database implementation
	<Purpose> a persistent logic programming environment
	<Description> Megalog provides a persistent store with database 
transparency, database functionality, graphic systems, debugger and 
profiler
	<Developed by> Michael Dahment
	<Organization> ECRC
	<Country> Germany
</Program>

<Program> Member Risk System
	<Number> 509
	<Domain> Financial Analysis.
	<Purpose> To analyse the credit assessments of customers to assess risk 
analysis.
	<Description> The Risk Assessment System analyses and advises on 
risk value of customers.
The system receives a credit description focussing on key factors such as 
level of financial resources, credit rating of members parent company etc, 
together with an overall intuitive rating of members risks.  These details 
are logically analysed to derive useful risk info and this is then used in 
managing members risk.
	<Developed by> London Clearing House.
	<Organization> Martyn Holmes.
	<Country> U.K.
	<Prolog system> Quintus Prolog
	<Platform> Unix.
	<Type of Use> In-house
	<Current Status> Released
	<No. of Users> 720
	<Information Date> Summer 1991
</Program>

<Program> MessageQ
	<Number> 420
	<Domain> Business and information systems.
	<Purpose> Phone Message tracking system.
	<Description> Quintus MessageQ is a client/server tracking system 
which has a Prolog database process acceptiong connections from 
multiple client front ends.  The client is a Prolog process which 
lets the user make database queries using a Motif graphical user 
interface to give an example
against which to search.  Receptionists use this system to take telephone 
messages for employees, and a copy of the message is automatically sent to 
the employee by email. Reports of messages can be created graphically and 
emailed or faxed electronically. 
	<Developed by> Georges Saab.
	<Organization> Quintus Corporation.
	<Country> USA.
	<Prolog system> Quintus Prolog 3.1.1.
	<Other languages> C, 5%
	<Lines> 1000 unique lines.
	<Predicates> ?
	<Platform>  Sun, Dec, IBM RS6000, HP, Intergraph.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> March 1991
	<No. of Users> 100
	<Information Date> March 1992
	<References> none.
</Program>

<Program> MESSE
	<Number> 240
	<Domain> medical treatment
	<Purpose> consulting system for advice on special gastroenerological 
examination
	<Description> The Medical Expert System for Special Examinations 
aids in choosing the most appropriate special examination for a 
patient, checks the indications and contraindications, provides 
information including the time of the examination for the doctor, 
nurse and patient.
It covers 7 types of examinations, 16 illnesses, 35 indications, 25 
contraindications, 35 information data groups.
	<Organization> St. Margaret Hospital, Budapest, IQSOFT
	<Country> Hungary
	<Prolog system> MProlog 2.3 and MProlog Dialog
	<Lines> 1500 clauses
	<Predicates> 610
	<Platform> PC 286, MS-DOS
	<Type of Use> In-house
	<Current Status> Prototype
</Program>

<Program>  MINOTAURO 1000
	<Number> 378
	<Domain> Train dispatching and control.
	<Purpose> Decrease the operator resolution time of failures on trains on 
the track lines by means of dialoging expert system.
	<Description> Expert system for the control of failures on the series 
1000 trains. System developed in Prolog and installed on a PS/2 IBM in the 
Control Room of Metro de Barcelona (Subway) in Sagrada Farnilla. Operating 
since 1989.
	<Developed by> Antonio Trias.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> Turbo Prolog 2.0 (Now PDC Prolog).
	<Lines> 3500
	<Predicates> 280
	<Platform> Developed PC 386 / MS DOS
Delivery PS/2 IBM / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 08/1989
	<Information Date> 07/92
</Program>

<Program> MINOTAURO 3000
	<Number> 382
	<Domain> Train dispatching and control.
	<Purpose>  Decrease resolution time of failures on trains on the track 
lines by means of a dialoging expert system.
	<Description> Expert system for failures control on the 3000 and 4000 
series of trains. Project developed and now being installed at Metro de 
Barcelona.
	<Developed by>  Xavier Fustero and programmers team.
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.31
	<Other languages> void
	<Lines>  3780
	<Predicates> 287
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Information Date> 07/1992
</Program>

<Program> MIS
	<Number> 252
	<Domain> Program synthesis
	<Purpose> program synthesis from examples
	<Description> The Model Inference System provides a tool for 
inductively synthesizing programs from examples. The method involves 
specifying a set of facts and an initial program which is a set of clauses 
and progressively debugging the program by use of the PDS algorithm.
	<Developed by> Ehud Y. Shapiro
	<Organization> Yale University
	<Country> USA
	<References> E. Y. Shapiro: Algorithmic Program Debugging. MIT Press, 
1982.
</Program>

<Program> MISTRAL
	<Number> 474
	<Domain> Dam safety.
	<Purpose> Filtering, intepretation and explanation of alarms coming from 
automatic monitoring systems of dams.
	<Description> MISTRAL is an expert system for evaluating, 
explaining and filtering alarms generated by automatic monitoring 
systems of dams.
MISTRAL belongs to a family of systems developed within the project 
DAMSAFE for managing dam safety.  The aims of MISTRAL are to provide 
on-line interpretation of the behaviour of a dam and to reduce the 
requests of expert intervention.  MISTRAL uses geometrical and 
physical knowledge, heuristic and engineering judgement based on 
experience. It deals with significance and reliability of the  
instruments to assess the safety of the dam. The result of the 
evaluation is the definition of the state of the dam and its main 
components, and the identification of the physical processes going 
on.  An explanation of the state of each component and
process is generated.
Both the evaluator and the explainer are written in Prolog, while the 
communication mechanisms with the monitoring system and the interface 
manager are external.
	<Developed by> Marco Lazzari.
	<Organization> ISMES
	<Country> Italy.
	<Prolog system> ESLPDProlog2.
	<Other languages> Microsoft Visual Basic.
	<Lines> 1500
	<Predicates> 120
	<Platform>  PC 386 / MS-DOS.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> October/1992.
	<No. of Users> 3
	<Information Date>  November/1992.
	<References> J.B. Comerford, M. Lazzari, P. Salvaneschi et al.,
"The role of AI technology in management of dam safety:
the DAMSAFE system", Dam Engineering, 3(4), 1992.
M. Cadei, M. Lazzari, P.Salvaneschi,  "Safety management of 
egineering structures using knowledge-based systems",
Proc. of the 3rd Int. Conf. on Industrial & Engineering
Applications of Artificial Intelligence & Expert Systems
(IEA/AIE 90), Charleston, S.C., July 15-18, 1990, ACM Press, New York, 
N.Y., 618-627.
</Program>

<Program> Mixtus 0.3.3
	<Number> 14
	<Domain> CASE Tool
	<Purpose>  An automatic partial evaluator for full Prolog
	<Description> Mixtus is an automatic partial evaluator for full Prolog.  
Given a Prolog program and a query it will produce a new Prolog program 
specialized forall instances of that query.  The program produced is meant 
to execute more efficiently than the original program.  No annotations are 
needed;and Mixtus is guaranteed to terminate for all input programs and 
queries.
	<Developed by> Dan Sahlin
	<Organization> SICS
	<Country> Sweden
	<Prolog system> SICStus Prolog
	<Other languages> none
	<Lines> 3776
	<Predicates> 460
	<Platform> Unix (whatever SICStus runs on)
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 24 May 91
	<No. of Users> 100
	<Information Date> 31 Oct 91
	<References> "An Automatic Partial Evaluator for Full Prolog" by Dan 
Sahlin;	SICS Dissertation Series 4; 1991
</Program>

<Program> MM4 Weather Modelling System
	<Number> 100
	<Domain> Weather Forecasting
	<Purpose> 
	<Description>   MM4 is a  mesoscale (medium-scale) weather modelling 
system developed at Penn State University and the National Center for 
Atmospheric Research in Boulder, Colorado.. It gives relatively detailed 
short-term forecasts of meteorological conditions in the continental USA. 
It is used particularly to model pollutant dispersion (e.g. acid rain 
deposition). The original program, which comprised 16,000 lines of Fortan, 
was harnessed with Strand88 and tenfold speedups have been obtained with 15 
processors.
	<Developed by> Penn State University, National Center for Atmospheric 
Research,Boulder, Colorado
	<Country> USA
	<Prolog system> Strand88
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> MMI2
	<Number> 38
	<Domain> Multimodal interfaces / Expert systems
	<Purpose> Multimodal interface for man machine interaction with knowledge 
based systems
	<Description> The project aims at making considerable advances in 
the domain of interface technology by offering users a wide range of 
interaction facilities within the same interface system:different 
communication modes are integrated up to a point where the user can 
mix modes in the course of a single interaction;an advanced dialogue 
manager controls the interaction process; making decisions as to 
user/system 
initiative; appropriate response mode; context driven interpretation; 
etc.;user modelling facilities provide an appropriate range of 
interactions relative to the user's level of competence.
The demonstration domain for the interface is computer network 
design. An expert system is built offering both design and technical 
and commercial analysis functionalities. The result is a cooperative; 
multi-modal dialogue between system and user.
The MMI2 system is a highly modularly built interface system and is 
destined to be portable to other kinds of knowledge based applications.
	<Developed by> Esprit II Consortium
	<Organization> BIM sa/nv; RAL;  INRIA; ISS; Leeds Univ.; EMSE; CRISS
	<Country> Belgium; England; France; Spain; England; France; France
	<Prolog system> ProLog by BIM
	<Other languages> Prolog/C interfaces to SunView; C
	<Lines> +- 121.000 lines
	<Predicates> +- 4700
	<Platform> Sun3; Sun4; SPARCstations
	<Information Date> October 1991 
	<References> "Architecture of a multimodal dialogue interface 
for knowledge-based systems" Binot J-L.; Falzon P.; Perez R.; Peroche 
B.; Sheehy N.; Rouault J. and Wilson M.D.  In Proceedings of 
Esprit'90 Conference;   Kluwer Academic Publishers     
 "Multimedia and Multimodal Systems: Architectures for  Co-operative 
Dialogue"  Wilson M.D. & Falzon P.  In Proceedings of the ERCIM 
Workshop on multimedia system;  Lisbon; November 1991
"An Architecture for Multimodal Dialogue"  Wilson M.D.; Sedlock D.; 
Binot J-L.; Falzon P. In Proceedings of the second Vencona Workshop 
on  Multi-Modal Dialogue; 1991 M.M. Taylor; F. Neel & D.G. Bouwhuis 
(Eds)
"Description of NEST; a Network design Expert SysTem" Balfroid F.; Jouve C. 
Esprit project 2474; deliverable d45; September 1991
</Program>

<Program> MODEL OODB
	<Number> 392
	<Domain> Object-Oriented Databases.
	<Purpose> Add routines to make Prolog into a distributed		     
object-oriented deductive database.
	<Description> Adds immediate and delayed assignments to Prolog. 
e.g.
		   b:=integer.  /*Integer type is readOnly*/
		   b#value:= 10 .
		   c(2):=b+b.  /* Assign 20 to array element 
c[2].*/
		   f(X)@=X+X.  /* Define function f(X).
				May add or union, depending on type of X 
*/
		   cd f.		 /* Change directory to f to
				look at f's rules*/
The slots and methods are consulted automatically if not currently in 
memory.  The files will first be ftp'd if not available locally.  
Appropriate editors are called when cd'ing to image, animated drawings, 
etc.
	<Developed by>  Prof. Dan Buehrer.
	<Organization> Institute of Computer Science and Information Engr.
National Chung Cheng University.
	<Country> Taiwan, Republic of China.
	<Prolog system> Sicstus.
	<Other languages>  csh
	<Lines> 1000
	<Predicates> 20
	<Platform> Sparc, Sun 3, Sun 4, Dec Ultrix.
	<Type of Use> Free/Public Domain
	<Current Status> Prototype
	<Date deployed> Oct. 1992.
	<Information Date> Sept. 23, 1992.
</Program>

<Program> MODGEN 2.0-model generator for chemical processes.
	<Number> 484
	<Domain> Dynamic Simulation.
	<Purpose> Dynamic simulation of chemical engineering process units  
by  a discrete  structural  model,  generated  automatically   in   
the
knowledge  of  the   user-defined   decomposition   pattern   and 
prototypes. 
	<Description> MODGEN represents a completely  new  approach  in  
the  dynamic simulation of process units in chemical engineering.  
Using  this tool, instead  of  the  many-fold  transformation  of  
the  model through the mathematical construct and the  numerical  
algorithm, the process unit is automatically decomposed into  an  
executable Prolog program. Starting from the knowledge of  the  user-
defined decomposition pattern and  from  the  actual  prototypes  of  
the physical and chemical changes, the respective balance element and
elementary transitions are algorithmically generated in the form of 
properly classified and ranked PROLOG clauses. The calculation itself is 
simply carried out by the consecutive execution of  the ranked elementary 
transitions. The user can communicate with  the system either through a 
user-friendly interface or  directly  via the generating database. The core 
definitions of  the  generating and executing metainterpreter can be  
supplemented  by  the  user defined prototypes of the given field.
	<Developed by> Bela Csukas, Karoly Varga, Rozalia Lakner.
	<Organization> University  of  Veszprem,  supported  by  the  Hungarian  
Basic Research Grant N OTKA-2550.
	<Country>  Hungary.
	<Prolog system> TURBOPROLOG 2.0, PDC PROLOG 3.1 CSPROLOG 3.25 
(ALL-
MULTILOGIC Ltd., Budapest)
	<Other languages> Can also be implemented in object oriented languages.
	<Lines> 3000
	<Predicates> 350
	<Platform> PC/AT 386; transputer and work station versions in preparation.
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 9/1990
	<No. of Users> 10
	<Information Date> 12/92
	<References> Csukas, B., Lakner, R., Varga, K., Jambor, L.:  
Intelligent
Dynamic Simulation by Automatically  Generated  Prolog  programs.
In: L. Puigjaner and A. Espuna  Eds.:  Computer-Oriented  Process
Engineering,  Process  Technology   Proceedings   10,   Elsevier,
Amsterdam-London-New York-Tokio, 1991, pp. 41-46
</Program>

<Program> MODGEN/DMS Dynamic Modular Simulation of Technological  
Systems.
	<Number> 491
	<Domain> Dynamic Simulation of Technological Plants.
	<Purpose> Dynamic simulation of  complex  technological  plants  using  a 
dynamic  modular  technique  comprising  the  quasi-parallel   or parallel 
execution of the  generated  structural  models  of  the process units.
	<Description> In  the  conventional  methods   used   for   the   
simulation
(flowsheeting) of a process containing recycle loops an iterative 
calculation is applied. In the enhanced methods  instead  of  the 
direct parameter estimation a  two-level  calculating  scheme  is 
used, where the upper level linear model calculates  the  streams 
from the estimated efficiency factors, next the individual models 
determine the new proposition for the efficiency parameters.
     In the Dynamic Modular Simulation this complicated procedure is 
replaced by a completely new method.  The  simulation  of  the 
individual process units is solved by the MODGEN model  generator 
that makes possible:
-tthe modeling of each unit with the same algorithm, and
-tthe step-by-step saving of the data  determining  the  temporal 
state of the models.
     The simulation is controlled by the supervisory module  that in 
the knowledge of the appropriate generating  databases  starts the  
generation  of  the  structural  models  and  organizes  the 
communication between them. Then the calculation is  carried  out 
quasi-simultaneously or effectively  in-parallel,  while  between the 
time steps the processes  communicate  with  each  other  vis 
messages.
     The  method  can  also  be  executed   in   a   conventional
monoprocessor machine, however, using a  multiputer  machine  the whole 
system can the simulated simultaneously.
	<Developed by>  Bela Csukas, Karoly Varga, Eszter Pozna.
	<Organization> University of Veszprem, supported  by  the  Hungarian  
Research Grant N OTKA-2550
	<Country> Hungary .
	<Prolog system> TURBOPROLOG 2.0, PDC PROLOG 3.31, CSPROLOG 3.25; 
(ALL-
MULTILOGIC Ltd., Budapest.
	<Lines> 1500
	<Predicates> 200
	<Platform>  PC/AT 386, PC/AT 386 with transputer; work station  version  
in preparation .
	<Type of Use> In-house
	<Current Status> 
	<Date deployed> 4/1992
	<No. of Users> 5
	<Information Date> 12/1993
</Program>

<Program> MODGEN/HPLC -  Modeling  Preparative  High  Performance  Liquid 
Chromatography.
	<Number> 490
	<Domain> Simulation of Preparative HPLC.
	<Purpose>  Dynamic simulation of preparative HPLC (High Performance 
Liquid Chromatography) columns  by  an  automatically  generated  PROLOG 
structural model. 
	<Description> The traditional modeling  technique  of  the  High  
Performance Liquid Chromatography columns needs the solution of a  
system  of second order partial differential equations containing  
nonlinear inhomogeneous functions  and  associated  with  rather  
difficult initial and boundary  conditions.  In  addition,  in  
preparative regime   the   columns   are   often   overloaded,   
consequently
discontinuities are formed and passed along the column.
     Using the model generator, the process unit is automatically 
decomposed into the structure of the primitive  balance  elements and 
elementary transitions  existing  between  them.  The  actual problem 
can be defined by the composition of the liquid,  by  the components 
to be adsorbed, as well as by the  parameters  of  the approximate 
multicomponent equilibrium function. The process unit is described by 
its geometrical data, by the  characteristics  of
the adsorbent and by the process parameters including the  liquid 
flow rate,  the  mixing  characteristics,  etc.  The  calculation 
itself is simply carried out by the consecutive execution of  the 
generated elementary transitions.
The generated structural model can  advantageously  be  used
for the simulation of the various regimes of the column including change of 
the solvent and/or the flow direction.  The  robustness of the method makes 
possible its use  for  the  modeling  of  the special techniques (e.g. 
displacement chromatography, etc.).
	<Developed by>  Bela Csukas , Karoly Varga , Sandor Kovats  , Antal 
Aranyi
	<Organization> University of Veszprem  and Pharmacenticals Gedeon 
Richter Ltd.
	<Country> Hungary.
	<Prolog system> TURBOPROLOG  2.0, PDC PROLOG 3.31,   CSPROLOG   
3.25; 
(ALL-MULTILOGIC   Ltd., Budapest. 
	<Other languages> Can also be implemented in object oriented languages.
	<Lines> 1500
	<Predicates> 200
	<Platform> PC/AT 386; transputer and work station versions in  
preparation.
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 3/1992
	<No. of Users> 4
	<Information Date> 12/1993
	<References> In preparation.
</Program>

<Program> Mondel Development Environment
	<Number> 429
	<Domain> Object-oriented specification development: Compiling, 
execution, and validation.
	<Purpose> Supports the formal specification language MONDEL: compiling, 
execution, and validation.
	<Description> Mondel Development Environment supports the 
specification language Mondel. Mondel is an executable object-
oriented specification language with a formally defined semantics. It 
has been developed for the specification and modelling of 
applications in the field of real-time systems.It supports 
persistancy and concurrency. The development environment supports 
four tools:
1- Compiler: checks the static semantics and generates the internal 
code for the interpreter.
2- Interpreter: executes the given specification by interpreting the 
internal code which is generated by the compiler
3- Verifier: constructs the reduced reachability graph of the Mondel 
specification.
4- Syntax analyzer: checks the context-free syntax of the given 
specification.
	<Developed by> Prof. Gregor von Bochmann, Prof. Michel Barbeau, Normand 
Williams, and Aiwu Liu.
	<Organization> Centre de recherche informatique de Montreal 
(CRIM).
Sponsered by: Bell Northern Research  (BNR)
	<Country> Canada.
	<Prolog system> Quintus Prolog
	<Other languages> C language, YACC and LEX. Extent : 20%
	<Lines> Approximatly 15000 lines of Prolog.
	<Predicates> Approximatly 1000 Prolog predicates.
	<Platform> SUN workstations/UNIX
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> January 1991
	<No. of Users> 15
	<Information Date> March 1992
	<References> G.v. Bochmann et al., "Mondel: an Object-Oriented 
Specification Language", technical report, CRIM, 1991.
M. Barbeau, G.v. Bochmann, "Formal Semantics and Formal Verification of 
Object-Oriented Specifications Based on The Colored Petri Net Model", 
submitted to IEEE Transactions on Software Engineering. 
</Program>

<Program> MOP2 
	<Number> 76
	<Domain> Project management, manpower planning and recording
	<Purpose> MOP2 (Managing Our Projects, Monitoring Our Programme) is 
a program and database to aid NPL DITC Software Engineering Group
in its project management, planning, and monitoring.
	<Description> The current Software Engineering Group programme 
has many different work packages, and many staff are working on 
several. MOP2 is a Prolog program and relational database which 
enable us to monitor the effort on each work package, and plan staff 
time so that collaboration is possible and convenient. Full user 
documentation is available (see reference below) 
MOP2 provides various reports:
-- Person-based which summarise what a member of SEG has done (or is 
planning to do),
-- Project-based which summarise the effort performed (or planned) on 
a project,
-- Checks that the data is consistent, for example to show that the total 
recorded amount of time for each member of SEG is correct.
	<Developed by> Roger Scowen
	<Organization> National Physical Laboratory
	<Country> UK
	<Prolog system> SICStus 2.1 #3
	<Other languages> None
	<Lines> 1500 (320 code, 350 database facts, 770 person/effort/project 
facts)
	<Predicates> 33 (14 database facts, 19 program)
	<Platform> Sun 3/50, Sun OS Release 4.1
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> February 1992
	<No. of Users> 2
	<Information Date> March 1992
	<References> MOP2 (Managing Our Projects, Monitoring Our Programme) 
Internal NPL DITC SEG report, SEG N12, March 1992.
</Program>

<Program> Morphological Analysis of Hungarian Texts
	<Number> 265
	<Domain> Natural language
	<Purpose> To analyse the morphology of Hungarian texts.
	<Description> The system analyses the morphology of Hungarian 
texts. The analysis is based on heuristics and it is done in two 
stages:
a- the morphological analysis of verb forms
b- the morphological analysis of tagged nouns.
	<Developed by> Z. Kiss,  G. Proszeki, L. Toth
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> Z. Kiss,  G. Proszeki, L. Toth (1979)
Morphological Analysis of Hungarian Texts (in Hungarian).
MTA NYTI report, SOFTTECH D41, SZAMKI, 1979.
G. Proszeki,  Z. Kiss,  L. Toth (1982)
Morphological and morphonological analysis of Hungarian word-forms by 
computer.
Computational Linguistics and Computer Languages,
No. XV., Budapest, 1982.
</Program>

<Program> Morse
	<Number> 501
	<Domain> Education.
	<Purpose> Learn and practice receive morse code with sound or light .
	<Description> Morse is program developed with its main purpose to 
learn to 
program in PDC Prolog and side purpose: to make a training program to 
train persons to receive and decode, morse code from sound or light. 
(Sound - for persond working with radio communications; light - 
persons working at navy).
Speeds available: from 5 char/sec to 100 char/sec morse code are 
organized in lessons in which 2 to 4 char (codes) are learned. There 
are 18 lessons. Inside lessons characters are in random sequence.		 
Program can read and code 
also ascii text files and characters / clauses can be typed from keyboard.  
It is possible to build simple hardware interface to standard devices for 
coding (oscilator. transmiter).
	<Developed by> Vojko Kalan.
	<Country> Slovenija.
	<Prolog system> PDC Prolog 3.20 /3.30 + PDC Prolog Toolbox 2.10
	<Other languages>  No.
	<Lines> 1500
	<Predicates> 100
	<Platform> PC/XT/AT, MS DOS, DS/DD
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> 12/1991
	<Information Date> 2/2/93
</Program>

<Program> Mosaic, 1.A.0
	<Number> 453
	<Domain> Strategy Board Game (with graphic interface.)
	<Purpose> Mosaic is a strategy board game for the Macintosh.
	<Description> Mosaic is a strategy board game for the Macintosh. 
The
game involves the placement of colored square tiles under various 
constraints.
The goal of a player is to build large regions that player's color.
	<Developed by> Lindsey Spratt.
	<Organization> (none).
	<Country> USA.
	<Prolog system> LPA MacProlog 3.5
	<Other languages> none.
	<Lines> 4000
	<Predicates> 800
	<Platform> Macintosh / MacOS Systems 6 and 7
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> June/1992
	<No. of Users> 2 registered (160 downloads).
	<Information Date> 10/30/1992
	<References> None.
</Program>

<Program> MPSSS
	<Number> 136
	<Domain> SIM
	<Purpose> Power Systems Protection Simulator
	<Description> Submitted to PAC. Paper 268
	<Developed by> Peter Kadar
	<Country> ?
	<Prolog system> Turbo Prolog/ Clipper/ C and dBase
	<Platform> IBM PC @ 286
	<Information Date> 30/10/91
</Program>

<Program> MTN
	<Number> 270
	<Domain> Teleprocessing Networks
	<Purpose> To generate models of teleprocessing networks
	<Description> The system generates the generation of simulation models 
of data processing networks. A simulation model is first generated in the 
language GPSS. The description of the network (the topology, the line 
control algorithms, the type of channels, etc) is entered into the system 
from a user. The generated model traces the throughput of the system, the 
waiting queues at the nodes and summarises other simulation parameters of 
interest.
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
</Program>

<Program> MULTI
	<Number> 456
	<Domain> Software Development, Electronic CAD Tool development.
	<Purpose> Builds complementary parsers and generator from a
single specification language.
	<Description> MULTI accepts a DCG-like grammar specification for 
a language and produces a parser and a generator for that language. 
These parser/generators can be built statically (as YACC does) for 
separate compilation, or constructed ``on-the-fly'' in programs which 
are language-independent, but need to load a language front-, and 
back-end at run time. When used with PLEX (Lex for Prolog) the 
MULTI/PLEX system provides an environment
in which a single language description acts as a source-file
for a complete system of tokenizing/parsing/generating software.
MULTI/PLEX grammars have been developed for VHDL, Verilog, Motive, 
Synopsys, and AT&T's FPDL language.
	<Developed by> Peter Reintjes.
	<Organization> Quintus Corporation.
	<Country> USA
	<Prolog system> Quintus Prolog
	<Other languages> none.
	<Lines> 881
	<Predicates> 183
	<Platform> Quintus Platforms
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> N/A
	<No. of Users> 3
	<Information Date> 11/93
</Program>

<Program> Multipurpose Substation  Simulator Shell (MPSSS)
	<Number> 53
	<Domain> Power Systems Training Analysis
	<Purpose> Simulation of the operation of the protection devices in the 
Power System Substations
	<Description> The aims of the simulator are:
-to tutor and train the stuff in the substations,
-to drive at real-time the telemechanical channels of the SCADA 
system.
The system represents the heuristic protection philosophy of the 
substations.
There are 15 built-in protection scenarios in the system.
The simulator uses pattern-matching and black-board techniques.
	<Developed by> Department of system Planning
	<Organization> EROTERV - Power plant and Network Engineering Company - 
Budapest
	<Country> Hungary
	<Prolog system> Turbo PROLOG
	<Other languages> Microsoft C4
	<Lines> 8000
	<Predicates> 270
	<Platform> IBM PC 286/386
DOS System
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Date deployed> 6/91
	<No. of Users> 4
	<Information Date> 1/2/92
	<References> Peter Kadar "A multipurpose Substation Simulator 
Shell"
2nd Conference on AI, Budapest, 24-26 January 1991
Peter Kadar  "An Expert System Application in the Hungarian Power 
System"
Eng. Appl. of AI Vol.5 No.1 1992
</Program>

<Program> N Derivatives of Composite Real Functions
	<Number> 266
	<Domain> Mathematics
	<Purpose> To compute the N derivatives of composite real functions
	<Description> The program generates the first N derivatives of a 
composite real function by symbolic differentiation. The function may have 
several variables of arbitrary complexity. Possible simplifications are 
done at the symbolic level. The output of the program is a FORTRAN 
subroutine for computing substitution values of the derivatives. Common 
subexpressions are evaluated only once. The program is based in the state-
space concept.
	<Developed by> V. Kofalusi, E. Halmai 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> V. Kofalusi, E. Halmai (1981)
State-space sets, state-space graphs and prefix expressions.
ACM SIGSAM Bulletin, 1981/2
V. Kofalusi, E. Halmai (1981)
State-space based data structures.
Hungarian Conference on  Programming Systems'81,
Szeged (Hungary), 1981, pp 342-351.
</Program>

<Program> Name matching
	<Number> 183
	<Domain> Office procedures
	<Purpose> Matching of non-European people's names
	<Description> Demonstrator for Automatic name matching, developed for 
the Immigration and Nationality department of the British Home office. This 
is particularly designed for non-European names.
	<Developed by> Graeme Smith
	<Organization> BBN, Edinburgh
	<Country> UK
	<Prolog system> LPA MacProlog
	<Platform> Macintosh IICX
</Program>

<Program> NeoMail
	<Number> 343
	<Domain> Communications.
	<Purpose> Electronic Mail Interface.
	<Description> Augmented interface useable for both Cyrillic and 
Roman Characters.  Introduces mechanisms to dynamically configure 
email groups as a system function.  Interface categorizes mail 
according to date/(source/destination)/subject, is highly visual, and 
requires minimal computer knowledge and keystrokes:  DOS and UNIX 
interfaces indistinguishable.  File transfer and data management 
facilities integral to the interface.
Communications between DOS and UNIX provided via a fault-tolerant 
streaming protocol across modem or RS232/422 terminal link.  
Dedicated user shell provided for call-ins from DOS machines.
Fully asyncronous email system minimizes hardware investment, and off-loads 
processing from UNIX server to DOS machines.
	<Developed by> Andrew Petukhov, Duncan Campbell.
	<Organization> NeoText SophtWear.
	<Country> Canada / Russia.
	<Prolog system> PDC Prolog.
	<Other languages> C
	<Lines> 2000
	<Predicates> 400
	<Platform> UNIX, MSDOS.
	<Current Status> Evaluation
	<Date deployed> 06/92
	<Information Date> 07/92
	<References> None.
</Program>

<Program> NEST
	<Number> 137
	<Domain> Network design
	<Purpose> a Network design Expert SysTem
	<Description> Submitted to PAC. Paper 324
	<Developed by> Christine Jouve & Fabienne Balfroid
	<Country> France
	<Prolog system> BIM
	<Information Date> 31/10/91
</Program>

<Program> Nestor
	<Number> 66
	<Domain> Corrective maintenance of telecom switching equipment
	<Purpose> Repair expert system for electromechanical telephone exchanges
	<Description> Portable repair expert system for Swedish Telecom field 
staff, used in the maintenance of A-204 exchanges. The system also includes 
a substantial amount of A-204 technical documentation; this information is 
packaged into an Oracle database callable from the fault-finding system.  
Another subsystem provides access to a central Oracle server where recorded 
fault events in the A-204s are stored.
	<Developed by> Annica Andersson, Rabbe Fogelholm, Nils Hagner,
Theo Kanter, Maria Lennman, Mats Lordin, Ann-Christine Nilsson,
Ingela Olsson, Mark Plotnick, Johan Svedberg, Anders Tunevi,
Dag Westling
	<Organization> Infologics AB, Sollentuna, Sweden
	<Country> Sweden
	<Prolog system> Prolog-2 
	<Other languages> Subsystem for retrieving technical documentation
implemented by means of Microsoft C and an Oracle RDBMS. Subsystem 
for
inspecting fault records implemented by means of Microsoft C and 
Oracle
SQL*Net for dial-up access to a central Oracle server.
	<Lines> 4500 lines of program code, plus 84000 lines of Prolog 
clauses
that make up the diagnostic knowledge of the system
	<Predicates> 376
	<Platform> IBM AT compatibles, MS-DOS 3.3 or higher
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> April, 1991
	<No. of Users> 30
	<Information Date> 12th December, 1991
	<References> "Hela stationen i en PC", Televarlden no 17, 1991, 
p 14,
in Swedish.
</Program>

<Program> Network Management
	<Number> 95
	<Domain> Retail network management
	<Purpose> 
	<Description> A sophisticated technology demonstrator for a retail 
network management system. The system which makes extensive use of graphics 
and object-oriented programming to model a network of point-of-sale retail 
terminals interrogated by a central controller. The network information can 
be altered graphically by the operator Ñ by clicking and dragging nodes and 
filling in forms. Previous polling activity can be simulated by 'playing' a 
log of polling events against the network which then uses colour to reflect 
the changing status. The user-interface is further enhanced by the novel 
use of video cassette recorder-style icons to control the event log play 
back.  
	<Developed by> Quintec Systems Ltd
	<Country> UK
	<Prolog system> Quintec
	<Platform> VAXstation running DEC Windows
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> NORMALIZER
	<Number> 261
	<Domain> Legal drafting
	<Purpose> 
	<Description> Provides possible variant readings of a legal text 
from a drafting provided by an analyst. It outputs the results in 
arrow diagrams, outlines and/or a normalized version. 
The output from Normalizer can be used in AUTOPROLOG.
	<Developed by> Charles S. Saxon, Layman E. Allen
	<Organization> Eastern Michican University, U of Michigan Law School
	<Country> USA
	<Prolog system> Turbo Prolog
	<Platform> IBM PC
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 1985
	<Information Date> 1988
	<References> Layman E. Allen, Charles S. Saxon: Multiple 
interpretation of the structure of legal rules: Impediment or boon to 
legal expert systems. ICLP5 pp1609-1623, 1988
Layman E. Allen, Charles S. Saxon: Computer aided normalizing and 
unpacking. in C. Walter (ed) Computer Power and legal reasoning. West Publ. 
Co. 1985 pp495-572
</Program>

<Program> NTM (Nurse Teaching Manager)
	<Number> 502
	<Domain> Administration
	<Purpose> Planning and documentation of the education of nurses at German 
hospitals.
	<Description> Planning and documentation of the education of 
male/female
nurses at nurse schools in Germany.
	<Developed by> Dr. Raimund F_hnrich.
	<Organization> Dr. F_hnrich & Partner.
	<Country> Germany.
	<Prolog system> PDC Prolog 3.2/ 3.21/ 3.30 (formerly Turbo Prolog 
1.0/2.0)
	<Other languages> There has been attempts to mix with C but they have 
been cancelled.
	<Lines> 7250
	<Predicates> approx. 1200
	<Platform> MS-DOS 3.3 / 5.0 & DR-DOS 6.0. In addition to PharLap DOS 
extender and OS/2.
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> OBIS
	<Number> 475
	<Domain> Education configuration.
	<Purpose> Configuration of an educational curriculum based on 
professional requirements.
	<Description> Product serves to improve the correspondence 
between professional practice and education.  Productivity gain 
achieved approx. 500%
Product improves quality by ensuring consistence  and completeness of the 
model that is being constructed.
	<Developed by> 5 person company & customer team.
	<Organization> Syllogic.
	<Country> Netherlands.
	<Prolog system> SWI-prolog.
	<Other languages>  SQL (Oracle), Xwindows + Motif, Syllogc Knowledge Sheet.
	<Lines> 65.000
	<Predicates> 2.100
	<Platform> IBM RISC System 6000.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed>  july 1990.
	<No. of Users> 30
	<Information Date> 25 november 1992.
	<References> None.
</Program>

<Program> Objective-Prolog
	<Number> 138
	<Domain> OO
	<Purpose> Programming by step-wise refinement of data using Objective-
Prolog
	<Description> Submitted to PAC. Paper 363
	<Developed by> Mattias Waldau
	<Country> Sweden
	<Prolog system> Objective Prolog
	<Information Date> 6/11/91
</Program>

<Program> Oceanic
	<Number> 214
	<Domain> Insurance systems design
	<Purpose> To provide all the programs for introducing a new insurance 
product within a few days
	<Description> Oceanic provides the experts (actuary, product 
manager) with an interactive system under OS/2 that allows them to 
express themselves in a language similar to their everyday insurance 
language in which they can define the characteristics of the product 
in detail - from basic formulae to office proedures.
When he is finished, OCEANIC automatically generates the executable 
code which will allow the insurance agent to actually handle the 
product: entry screens, accounting mechanisms, record management, 
historical rollback etc.
Both the initial and generated systems are entirely in IBM Prolog which can 
be run either under OS/2 (on PS/2) or MVS (on a 370).
	<Organization> SOCS (French insurance software house)
	<Country> France
	<Prolog system> IBM Prolog for OS/2, IBM Prolog for VM
	<Lines> 150,000
	<Platform> IBM PS/2 with OS/2, IBM 370
	<Type of Use> Sold Product
	<Current Status> Released
	<No. of Users> 3 (currently)
	<References> B. Robinet: Logic Programming at IBM from the Lab to the 
customer. Proc ICLP8. pp912-921. 1991
</Program>

<Program> OF
	<Number> 72
	<Domain> Activity Planning
	<Purpose> Activity planning and tracing
	<Description> A Manager's tool. Gives information about activity, 
deadline,
who's responsible, details about the activity - like current status, etc. 
Features a non-erasable trace, e.g. about delays, activity abortion or 
success. Creates end-of-year reports. Context sensitive help.
	<Developed by> Audun Toennesen
	<Organization> STATOIL - The Norwegian National Oil Company
	<Country> Norway
	<Prolog system> PDC PROLOG
	<Platform> IBM PC/ DOS
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> November 1990
	<No. of Users> 12
	<Information Date> March 1992
</Program>

<Program> OLP
	<Number> 139
	<Domain> TOOL
	<Purpose> A set of Prolog Programming Tools
	<Description> Submitted to PAC. Paper 343
	<Developed by> Cristina Ruggieri, Sancassani
	<Country> Italy
	<Prolog system> OLP+ALPES
	<Information Date> 6/11/91
</Program>

<Program> OMAR (operative management of aircraft bouting).
	<Number> 140
	<Domain> Aircraft Scheduling.
	<Purpose> A Scheduling of the Alitalia Fleet, i.e. assigning tail number 
of aircraft to flights.
	<Description> An interactive system designed for the predictive 
scheduling of the Alitalia Fleet.
The salving strategy combines network consistency and tree search 
techniques.
	<Developed by> Momigliano, Paltrinieri, Torquati
	<Organization> Bull Italia-Direzione sistemi esperti.
	<Country> Italy
	<Prolog system> Quintus Prolog.
	<Other languages> 5QL - Database Interface
C++/Windows - Graphical Interface
	<Lines> 20000
	<Predicates> 2500
	<Platform> Bull X-20 Minicomputer
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> June 1991
	<No. of Users> 3
	<Information Date> March 1992
	<References> "A Constraint Satisfaction Approach To Operative 
Management Of Aircraft Routing", Third Int. Conf. of Industrial and 
Engineering Applications of Artificial Intelligence and Expert 
Systems, Charlotte, NC, 1990-ACM Press.
"Scheduling Of An Aircraft Fleet".  First International Conference On The 
Practical Application of Prolog, London, UK, April, 1992.
</Program>

<Program> ONE-LEVEL WORKSHOP PLANNER
	<Number> 285
	<Domain> Computer-aided design
	<Purpose> Plans one level workshop buildings using prefabricated panels
	<Description> The system plans one-level workshop built with 
prefabricated panels. The ground plan is rectangle. The building is to be 
constructed from columns, beams and ceiling panels. The data of the 
available prefabricated elements (geometric size, net weight, supporting 
strength) are given as an initial database.  The system determines the 
ground plan (the distribution of the ceiling panels) and chooses the 
elements appropriate from the viewpoint of the geometric and statical 
conditions.
	<Developed by> Z. Markusz 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> Z. Markusz (1977)
The application of PROLOG for panel house design.
Informacio-Electronika XII, No. 3., 1977. pp.124-230.
</Program>

<Program> Operational Rations Advisor  (ORA)
	<Number> 294
	<Domain> Expert system for U.S. Army field operations.
	<Purpose> Automatically prepare ration menus from given food base, 
nutritionally balanced according to stated goals and environment.
	<Description> Uses Arity window system but with menubar replaced by 
multi-page bar.  Histogram bars show carbo-hydrate, fat and protein content 
of choice so far, and next choice increments before commitment.  Buttons 
provide next best choice or full completion.  Required components are 
enforced or automatically inserted.  Repetition constraints are enforced.  
Forbidden choices display reason.  Food base changes are reflected in 
dialog menus automatically.  Interrupted work can be saved and restored. 
	<Developed by> Ray Reeves  (Contractor).
	<Organization> U.S. Army Natick Research Labs.
	<Country> U.S.A.
	<Prolog system> Arity Prolog.
	<Other languages> None
	<Lines> 1300
	<Platform> PC
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 1989
	<No. of Users> Unknown
	<Information Date> May 1992
	<References> None.
</Program>

<Program> Opium
	<Number> 2
	<Domain> CASE tool
	<Purpose> An automated and extensible debugger for Prolog
	<Description> Opium is the preliminary result of a research project on 
automated debugging.  The aim is to have Prolog programs help to debug 
other Prolog programs. The debugging strategies are primiraly based on 
trace analysis.  Opium consists of a kernel which provides a handful of 
debugging primitives; connecting the debugged and debugging Prolog 
processes. Those primitives are partly implemented in C.  On top of the 
kernel a number of extensions have been built; for example to automate the 
debugging of non-terminating or failing programs. An extension handler 
(fully implemented in Prolog) ensures that all extensions have a consistent 
user interface and an up-to-date documentation. Prolog is the command 
language of the tracer; together with the tracing primitives this provides 
a query language which is more general and user-friendly than the usual 
programmable conditional breakpoints.
	<Developed by> Mireille Ducassé and Anna-Maria Emde
	<Organization> European Computer-Industry Research Centre
	<Country> Germany
	<Prolog system> Sepia (ECRC's Prolog)
	<Other languages> C
	<Lines> 21000 lines of commented Prolog; 17000 lines of plain Prolog
	<Predicates> 3000
	<Platform> Sun3/4; Unix
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> December 90
	<No. of Users> 5
	<Information Date> November; 11th 1991
	<References> M. Ducassé and A.-M. Emde. Opium user manual. 
Technical report TR-LP-60; ECRC Arabellastrasse 17; D-8000 Munich 81; 
Germany. May 91.
M. Ducassé.  Abstract views of prolog executions in Opium.  In V. 
Saraswat and K. Ueda; editors; Proceedings of the International Logic 
Programming Symposium. ALP; MIT Press; October 1991.
M. Ducassé and A.-M. Emde.  Opium: a debugging environment for Prolog 
development and debugging research.  ACM Software Engineering Notes; 
16(1):54--59; January 1991.  Demonstration presented at the Fourth 
Conference on Software Development Environments; SIGSOFT'90.
M. Ducasse'.  Opium+; a meta-debugger for Prolog.  In Proceedings of 
the European Conference on Artificial Intelligence; pages 272--277; 
Munich; August 1988. ECCAI.
M. Ducasse'.  Analysis of failing Prolog executions.  In Proceedings 
of the ICLP'91 Pre-conference Workshop on Logic Programming 
Environments; Paris; June 1991.  Proceedings published as Technical 
Report LIFO N 91-61; University of Orleans; France.
A.-M. Emde and M. Ducasse'.  Automated debugging of non-terminating Prolog 
programs.  In S. Bourgault and M. Dincbas; editors; Actes du Se'minaire de 
programmation en Logique; pages 89--103. CNET; Lannion; May 1990.
</Program>

<Program> Optimist
	<Number> 60
	<Domain> Oil exploration
	<Purpose> To assist geologists in oil exploration
	<Description>     The Optimist system was developed at the Turing 
Institute to assist geologists in assessing the probability of 
finding oil at prospective sites. Reasoning in this domain is 
characterised by the existence of different arguments for different 
conclusions, and experts frequently disagree. Unlike typical domains 
where expert systems have been developed, there is no single, agreed 
on, provably correct answer about the likelihood of finding oil.
 To assist experts in appraising oil likelihood, Optimist maintains 
separate models of experts' opinions stored both as rules and as 
records of specific oil prospect appraisals. The system and user 
`argue' to interactively assess a new prospect, Optimist assisting 
the  expert by arguing for consistency with previous decisions of 
this and other users. The importance of these approaches - 
personalised knowledge and an `argumentative' style of interaction - 
lie in the system's ability to  construct, faithfully remember and 
replay models of expert's reasoning and to highlight previous 
relevant appraisals the expert may be unaware of.
The system has been in commercial use for over two years.
	<Developed by> Turing Institute Ltd., Glasgow, UK
	<Organization> written for Enterprise Oil plc., London, UK
	<Country> UK
	<Prolog system> Quintus Prolog (V3)
	<Other languages> C (<5%), HyperNeWS (direct manipulation user interface)
	<Lines> 15000
	<Platform> (Current) Sun Sparc / UNIX
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> October 1989
	<No. of Users> 5
	<Information Date> 24th March 1992
	<References> Clark P. (1990) "Representing Knowledge As 
Arguments: Applying Expert System Technology to Judgemental Problem-
Solving", in "Research and Development in Expert Systems VII", 147-
159, 1990, Cambridge Univ. Press, Eds: T. R. Addis and R. M. Muir 
(also available as TIRM-90-043 from Turing Institute, 36 N Hanover 
St, Glasgow, UK)
Clark P. (1991) "A Model of Argumentation and its Application in A 
Cooperative Expert System", PhD Thesis, Dept. of CS, Univ. of Strathclyde, 
Glasgow, UK.
</Program>

<Program> OPUSPC
	<Number> 39
	<Domain> Airline Planning
	<Purpose> Real-Time Monitoring and decision support for airline Movement 
Control Center and placing grain.
	<Description> System for Monitoring a complete Airline fleet. The 
system communicates with an IMS system on an IBM mainframe via 3270-
communication. OPUSPC receives in real-time all the changes made to the 
traffic plans; and shows location bottlenecks etc. The system is a 
decision-support tool for the planners of the SAS fleet.
	<Developed by> Andrew Parrett; Finn Groenskov
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Other languages> none
	<Lines> >10,000
	<Platform> IBM comp. PC; DOS 3.3 or higher.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> April 1988
	<No. of Users> 100
	<Information Date> 01/92
</Program>

<Program> OSM
	<Number> 208
	<Domain> Medical information systems
	<Purpose> a comprehensive information management and decision support 
system for general practitioners
	<Description> OSM supports a number of knowledge and information 
retrieval functions, providing the user with rapid access to textual 
information from text sources, knowledge bases or patient database. 
Graphical display and ineraction capabilities (diagrams, images, 
numerical plots, body charts) are under development. 
The current knowledge base is only a fragment of the eventual 
requirement for general practice, being about 16,000 facts covering 
joint pain, nausea, vomiting and breathlessness. Assuming a full 
knowledge base of around 10 million facts, a storage requirement of 
200MB is indicated.
The system also incorporates the compact disk version of the Oxford 
Textbook of Medecine, a 300-author general medical reference work, 
and uses its indexes to retrieve text.
Other aspects include decision support, case reporting, patient notes and 
the Bordeaux Oncology Support System.
	<Developed by> John Fox, et al.
	<Organization> Imperial Cancer Research Fund
	<Country> UK
	<Prolog system> Prolog 2?
	<Other languages> C
	<Platform> PC AT or better
	<Type of Use> In-house
	<Current Status> Prototype
	<References> John Fox, A Glowinski, C. Gordon, S. Hajnal, M. O'Neill. 
Logic Engineering for knowledge engineering: design and implementation of 
the Oxford System of Medecine. Artificial Intelligence in Medecine 2, 
pp323-339. 1990
</Program>

<Program> OTAS
	<Number> 217
	<Domain> Financial
	<Purpose> a decision support system for analysis of stock options 
investment stratgies
	<Description> The Options Trading Analysis System automatically 
generates and analyzes investment strategies based on standard 
vertical option combinations. Its main elements are:
  a numeric database containing stock market data
  a symbolic database cntaining rules describing standard options 
combinations
  a numeric routine for the valuation of individual options
  a linear algebra module for the analysis of options combinations
  a driving module that generates options positions and triggers the 
analysis
  an input-output module that provides a spreadsheet-like interface
  a portfolio maintenance module that creates and updates portfolios and 
provides expert recommendations on follow-up options
	<Developed by> T. Huynh, Catherine Lassez
	<Organization> IBM T.J. Watson Research Center, Yorktown Heights
	<Country> USA
	<Prolog system> CLP(R) compiler
	<Platform> IBM RT/PC and VM
	<Current Status> Evaluation
	<References> T. Huynh, C. Lassez: A CLP(R) Options Trading 
Analysis System. Proc. ICLP5, pp59-69. 1988.
C. Lassez, K. McAloon, R. Yap: Constraint Logic Programming and Options 
Trading. IEEE Expert Vol 2,n.3, Fall 1987.
</Program>

<Program> P/FDM
	<Number> 141
	<Domain> OODB
	<Purpose> Using Prolog to Provide Access to Metadata in an Object-
Oriented Database
	<Description> Submitted to PAC. Paper 128
	<Developed by> Suzanne Embury
	<Country> UK
	<Prolog system> Prolog/FDM
	<Information Date> 30/10/91
</Program>

<Program> PAGAI
	<Number> 25
	<Domain> Intelligent Databases
	<Purpose> Bibliographic database system with intelligent retrieval using 
term relationship networks
	<Description> PAGAI is system for large-size databases. It uses  a term 
relationship network (a semantic network and thesaurus in one) with all the 
relevant terms for the domain in question. The network is used for 
automatic reformulation of search queries; by narrowing in or broaden a 
given search expression. The network is also used in search process.
	<Developed by> Henrik Zangenberg    
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Lines> >10.000  <100.000
	
	<Platform> DOS  OS/2
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 05/89
	<No. of Users> 10
	<Information Date> 01/92
</Program>

<Program> PAPAIN
	<Number> 507
	<Domain> Genetic Research.
	<Purpose> Protein sequence analysis.
	<Description> PAPAIN provides intelligent assistance in manipulating 
and interating diverse sources of information in a manner that will permit 
experimentation with hypothesis formation and reasoning styles.  QP has 
been used in the development of formal techniques for modelling strategic 
reasoning constraint propagation and the management of uncertainty in 
protein sequence analysis and structure prediction.
	<Developed by> Imperial Cancer Research Foundation.
	<Country> U.K.
	<Prolog system> Quintus Prolog.
	<Other languages> Quintus ProWindows.
	<Platform> Unix (Sun).
	<Type of Use> In-house
	<Current Status> Released
	<Information Date> April 1992.
</Program>

<Program> PARAFLOG
	<Number> 267
	<Domain> Dataflow Modelling
	<Purpose> Emulate the dataflow language PARAFLOG
	<Description> An implementation of PARAFLOG, a high level 
applicative dataflow language. PARAFLOG
is a definitional language without control structures, variables or 
assignment. It is very easy to learn and use in a parallel 
environment, without the need of explicit parallel program 
constructs. The system consists of two programs:
-a compiler that converts PARAFLOG programs to a dataflow graph
-an interpreter that simulates the dataflow execution
	<Developed by> A. Doman
	<Organization> SzKI
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> A. Doman (1981)
An applicative language for highly parallel programming.
SzKI Report, 1981
A. Doman (1982)
Cellural dataflow parallel processor for high speed computation.
(Hungarian) PhD Dissertation, 1982
</Program>

<Program> PC-PARLOG and MAC PARLOG
	<Number> 302
	<Domain> Parallel programming
	<Purpose> Programming language systems for PARLOG on the IBM-PC and 
Macintosh.
	<Description> These are full implementations of the PARLOG language, 
including 'deep guards' and simulated and- and or-parallelism.  The 
compiler generates fully debuggable code and the envirinment provides 
advanced tracing facilities, including channel spypoints.  Has over 100 
primitives.
	<Developed by> Tom Conlon and Steve Gregory
	<Organization> Parallel Logic Programming Ltd
	<Country> UK
	<Prolog system> LPA - Professional/MacProlog
	<Other languages> PARLOG - about 5%
	<Lines> 200K Source approx.
	<Platform> IBM + MS DOS
Mac + System 6 or 7
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> October 1989
	<No. of Users> 100+
	<Information Date> December 1991
	<References> 1.  'Programming in PARLOG'.  Addison- Wesley 1989, 
by Tom Conlon
2.  'Parallel Logic' Programming in PARLOG'  Addison-Wesley 1987, by 
Steve Gregory
3.  BYTE, International Section, pp80-5 to 80-14, October 1989.
</Program>

<Program> PCM
	<Number> 61
	<Domain> Program development,  configuration management
	<Purpose> support Prolog program development, rapid prototyping,
configuration management, software re-use, program
visualisation
	<Description> PCM is a repository for Prolog, C and SQL 
procedures,
with a collection of tools to support Prolog program
development and testing, software reuse, type and mode
checking, configuration management, automated building,
call dependency browsing, dialect conversion etc.
	<Developed by> System Engineering Research Group
	<Organization> Computer Science Dept., Keele University
	<Country> UK
	<Prolog system> Quintus Prolog 3.X
	<Other languages> C and SQL (tightly coupled to an Oracle V6 RDBMS 
kernel
via Keylink's ProDATA interface)
	<Lines> 25,000 lines of code including comments
	<Predicates> 1,000 (plus another 15,000 in libraries)
	<Platform> Sun 4 (SPARC) under SunOS 4.1
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> November 1988
	<No. of Users> 10
	<Information Date> 6th April 1992
	<References> Technical reports available from "Contact"
</Program>

<Program> PCUBE or P3
	<Number> 142
	<Domain> Production and other scheduling.
	<Purpose> Prolog-like Surfaced Expert System for an Overall Support of 
Parallel and Concurrent Process Systems
	<Description> P3 gives every such possible schedulings of a system of 
parallel and concurrent processes that satisfy the user given goals of the 
system.  These schedules are expressed in the form of time-tables.  The 
system itself is given by the permitted steps of the processes.  The user 
need not but may, of course, connect any of the steps to specified 
processes.  Besides a wide scale of SIMULA 67-like process interaction - 
communication instructions resource handling is also available.  The 
implementation tool of the system is not PROLOG but on its user level it 
behaves as if it was written in PROLOG taking the style of problem 
description (set of clauses) and its back -tracking mechanism into 
consideration.  The system is implemented in a hierarchical way.  The first 
implementation level is a list processing language and this is implemented 
presently in FORTH but we plan to change FORTH for C or C++.  Due to this 
second implementation level PCUBE is able to actually control the users 
process system according to the time-table found.  The system is available 
on IBM XT/AT.
	<Developed by> Katalin Szenes & Peter Forro
	<Organization> Hungarian Telecomm../ Ctr. of Information.
	<Country> Hungary
	<Other languages> FORTH programming language (base implementation)
	<Platform> IBM PC		XT/AT    HW
none- SW
	<Information Date> 20/03/92
	<References> K. Szenes:  Planning the activity schedule of 
process systems by the means of an AI based system.
Procds. of the 27th International MATADOR Conf., 20-21 Apr. , 1988, 
Manchester,  ed. :  B. J. Davies,  UMIST, MACMILLAN Education Ltd. , 1988. 
, 139 - 144.
</Program>

<Program> PDM
	<Number> 414
	<Domain> Used to define an intelligent model construction 
assistant to
decision support systems designed to support production and distribution 
planning.
	<Purpose> Assists in the formulation of Linear Programming
models in the Production Planning Domain.
	<Description> 
	<Developed by> Ramayya Krishnan.
	<Country> USA
	<Prolog system> C Prolog
	<Lines> 1000
	<Platform> Developed on Vax 11/780, Ported to a DEC 3100 .
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> 12/87
	<No. of Users> 1
	<Information Date> 5/88
	<References> Two  papers in the Internation Journal of Decision
Support Systems,  DSS, Vol. 6, pp. 123-152; Vol. 7, pp. 301-314.
</Program>

<Program> PDS - diagnosis system
	<Number> 253
	<Domain> Program debugging
	<Purpose> a declarative debugging system for a subset of Prolog
	<Description> Program faults are divided into three classes: incorrect 
solutions, missing (uncovered) solutions and non-termination. For each of 
these different algorithms are provided which will generally pinpoint the 
clause in error, with varying efficiency, by use of an oracle (the user) 
who knows the status of ground instances of all intended relations. The 
system is limited to ground queries in pure Prolog.
	<Developed by> Ehud Y. Shapiro
	<Organization> Yale University
	<Country> USA
	<Prolog system> Wisdom Prolog, Dec-10 Prolog
	<References> E. Y. Shapiro: Algorithmic Program Debugging. MIT Press, 
1982.
</Program>

<Program> PEP2
	<Number> 143
	<Domain> NL
	<Purpose> Parallel Expert Parser: A Rule- Based Spreading-Activation 
Model for Natural Language Understanding
	<Description> Submitted to PAC. Paper 327
	<Developed by> Bourgois et al
	<Country> Germany
	<Prolog system> Flat Concurrent Prolog
	<Information Date> 6/11/91
</Program>

<Program> Peptides Model Prediction System
	<Number> 263
	<Domain> Pharmaceutical research
	<Purpose> Calculation of parameters predicting biological activity for 
peptides
	<Description> The system generates a family of structure activity 
models and at the same time facilitates prediction of the biological 
activity on the basis of these models.It infers chemical structure 
units (substructures, fragments) from the aminoacid (i.e., chemical) 
composition of peptides and assigns numerical parameters to the 
relevant units. This is done by Prolog programs. The relationship 
between the parameters and the biological activities of peptides
is obtained by running Fortran programs.
	<Developed by> F. Darvas, A. Lopata, Gy. Matrai
	<Organization> EGYT (Pharmaceutical works)
	<Country> Hungary
	<Prolog system> MPROLOG
	<Other languages> FORTRAN
	<Platform> ORDRA 1305/GEORGE OS
	<Date deployed> 1980
	<Information Date> 1984
	<References> F. Darvas, A. Lopata, Gy. Matrai (1980)
A specific QSAR model for peptides. 
In Quantitative Structure Activity Analysis.
F. Darvas (Eds), Akademiai Kiado, Budapest, 1980.
</Program>

<Program> Perfect Pitch
	<Number> 103
	<Domain> Sports medecine
	<Purpose> Sports surfaces advisor
	<Description> Expert system to advise the safest and most 
economical sports surface to use for a given sport. Will also advise 
formal specification to aid in the tender process.
Suitable for local government, engineers, architects, etc.
	<Developed by> Michael Brown
	<Organization> Entretec Pty & Sports Technology Pty
	<Country> Australia
	<Prolog system> PDC Prolog v 3.21
	<Lines> 4600
	<Predicates> 550
	<Platform> DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> June 1990
	<Information Date> March 1992
</Program>

<Program> PERSONAL CASHFLOW
	<Number> 449
	<Domain> Personal financial analysis.
	<Purpose> Home budgeting program.
	<Description> PERSONAL  CASHFLOW  applies modern  business 
techniques to the  serious business of managing your own money 
including:
- preparing regular financial statements.
- preparing budgets and measuring your actual performance against 
them.
- analysing your cash flow.
- forecasting and planning for the future.
To  make  it  easy to use,  it  has  the familiar look and  feel  of  
a spreadsheet. It is not however just like any spreadsheet. It's a 
set of "hot-linked"  or  integrated spreadsheets  and data screens - 
when  you make  an  entry  in  CASHFLOW  it ripples instantly through  
all  these
spreadsheets to update them.				  
PERSONAL CASHFLOW does not apply artificial intelligence as such but 
is a good example of how a high level language like PDC Prolog can be 
used to produce a succesfull retail programme.
PERSONAL  CASHFLOW for the PC costs just 69 pounds sterling or 5 pounds for 
a demo disk.
	<Developed by> Ken Tyson.
	<Organization> Majengo Software Ltd.
	<Country> United Kingdom.
	<Prolog system> PDC Prolog 3.21
	<Other languages> None.
	<Lines> 1,897
	<Predicates> 142
	<Platform> IBM PC and compatibles.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed>  February 1991
	<No. of Users> 500
	<Information Date> May 15, 1992
</Program>

<Program> Personal Notekeeper
	<Number> 361
	<Domain> Intelligent Text-retrieval for multilingual Documents; Secure 
Electronic Notebook.
	<Purpose> Integrated, intelligent, multilingual, Fulltext Database 
for 
documents up to 64 KB, providing enhanced security features as high-speed 
encryption and access control. Specially suited for sensitive text's and 
notes on laptop's.
	<Description> Personal Notekeeper is an intelligent text-
retrieval system 
designed as an "easy-to-use" electronic Notebook that handles an 
almost unlimited number of documents with sizes up to 64KB. Each 
database can optionally be locked with different access-control 
levels, and specific privileges can be granted for as many users 
desired. Additional mechanisms can be "armed" in order to protect a 
DB from unauthorized "break-in attempts". Each document in a DB can 
be encrypted with a new and efficient encryption method, making 
sensitive information virtually unaccessible to unauthorized persons. 
Documents can be read from disk and/or edited in the integrated 
editor. Each
document has its own "Title-Sheet", where a title and additional 
information has to be specified. This way the documents can be classified 
and logically grouped like in a conventional database. An optional 
Expiration-Date enables the program to automatically remove "expired" 
documents in order to prevent the DB from growing unnecessary. With the 
complex search capabilities of Personal Notekeeper it is possible to 
quickly find any requested document or set of documents. Multilingual 
searches with automatic keyword translation between the five major EEC 
languages (English, German, French, Spanish and Italian) make finding of 
foreign language documents easy. Also documents with Synonyms of searched 
keywords will be found, if requested.
	<Developed by> Stefan Lago.
	<Organization> none.
	<Country> Switzerland.
	<Prolog system> PDC Prolog 3.30 for DOS
	<Other languages> 10 % C
	<Lines> app. 25.000
	<Predicates> app. 800
	<Platform> IPDC Prolog 3.30 for DOS
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> June 1992
	<No. of Users> void
	<Information Date> July 1. 1992
</Program>

<Program> PESKY
	<Number> 344
	<Domain> Quarantine, Management of Naure Reserves, Farming.
	<Purpose> To assess the risk of damage to plants or animals caused by 
introduction of potential pests.
	<Description> PESKY uses a special format for the rules and accesses 
databases continuing meteorological data as well as the output from models 
which predict the suitability of pest populations under the local 
conditions.  This data, together with input from the user is used to assess 
the potential damage to biological systems to be expected from the pest at 
any location in a particular season.
	<Developed by> W. Bottomley and R.W. Sutherst.
	<Organization> Cooperative Research Centre for Tropical Pest Management.
	<Country> Australia.
	<Prolog system> PDC Prolog.
	<Other languages> None.
	<Lines> 1500
	<Predicates> 125
	<Platform> PC using MSDOS.
	<Current Status> Evaluation
	<Date deployed> 1991
	<Information Date> 07/92
</Program>

<Program> PFT
	<Number> 318
	<Domain> Chemical industry.
	<Purpose> Production Formulisation Tool.
	<Description> The Production Formulisation Tool is a decision support 
system used in the chemical industry to provide scenarios for the 
production of chemical products taking into account formulation, raw 
material volumes and costs, and production constraints.
	<Organization> CPL Software and Services.
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 7000
	<Platform> SUN/HP/IBM/DIGITAL workstations
PC  386/486		UNIX   DOS
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 02/92
	<No. of Users> 1.
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> PHARGEST.
	<Number> 75
	<Domain> inventory management (drugs), decision support
	<Purpose> Management of the central medical store of the Tchadian 
Public
Health services (inventories, orders, deliveries, follow-up of
consumption).
	<Description> PHARGEST offers all the usual facilities of a 
classical inventory management system (commodities database, 
inventory control, orders, deliveries, clients and suppliers follow-
up, and so on)maybe with some special features.
1. It is centered around the "batch" defined as "drug+expiry date". 
Prevention of wastage by exceedind expiry dates is one of the 
objective of the system.
2. It is multi-stock: because different development agencies 
contributes to the funding of the Central Medical Store, it was 
necessary to adopt a "multi-stock" management system without 
networking facilities (the Tchad is one of the less-developped 
country in the world).
3. Due to the fact that each funding agency has its own management 
rules, PHARGEST has to provide for very different management systems, 
specially with regard to deliveries.
4. Because some managers are in charge of the distribution of drugs 
to all the public health services of a large area, the follow-up of 
comsumption of these services and the evaluation of their needs is 
one of the main function of the program.
5. It provides unusual facilities for helping the manager to estimate his 
future needs (inventory extrapolation) and take account of his financial 
constraints.
	<Developed by> Programming: Paul-Marie BOULANGER from ADRASS and 
Alain
WODON from AEDES.
Analysis:Paul-Marie BOULANGER et Daniel VANDENBERG (AEDES)
	<Organization> A.D.R.A.S.S. (Association Pour le Developpement de 
la Recherche
Appliquee en Sciences Sociales) and A.E.D.E.S. (Association
Europeenne pour le Developpement et la Sante) or (European
Association for Health and Development).
	<Country> Belgium and Tchad (Africa).
	<Lines> 10500
	<Predicates> 480
	<Platform> 80386 25MHz with 4MB of RAM and 120MB HD.
MS-DOS 3.3
	<Type of Use> In-house
	<Current Status> Prototype
	<No. of Users> 1 (Tchadian Central Medical Store)
	<Information Date> 02/18/1992
</Program>

<Program>  PHILO THE LOGICIAN.
	<Number> 463
	<Domain> Introductory logic classes in which students are required to 
translate English sentences into truth-functional formulas.
	<Purpose> Exercises for translating English sentences into truth-
functional logic.
	<Description> PHILO is a standalone program which interacts with 
students as they try to give the logical analysis of English examples 
of truth-functional sentences.  The program reads back an English 
rendition of the student's attempt and checks whether it is logically 
equivalent to the instructor's answer; it does this
even if the student's answer is wildly different syntactically.  If the 
student's answer is not logically equivalent, a case is given in English 
where the student's answer differs from that of the instructor.  A record 
of errors is kept for the instructor to examine; a separate program, DUMP, 
prints these out in an informative way.  Exercise examples reside in a 
separate ASCII file which the instructor can alter using an editor or 
wordprocessor. 
	<Developed by> R. G. Wengert.
	<Organization> University of Illinois and IBM.
	<Country> U.S.A.
	<Prolog system> Arity Prolog compiler and interpreter.
	<Other languages>  None.
	<Lines> 779 lines.
	<Predicates>  126 predicates.
	<Platform> PCs and clones, from 8088s on up.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> January, 1986.
	<No. of Users> Over 500 students on this campus alone.
	<Information Date> November 6, 1992.
</Program>

<Program> PHOEBUS
	<Number> 144
	<Domain> Automation and Electronics : verification of conceiving 
hardware control systems.
	<Purpose> A Tool for Analysis of Hierarchial Finite State machines
	<Description> 
	<Developed by>  F. Martinolle, C. Rominger, C. Guibout
	<Organization> INSAT/DGE/GERII
	<Country> France
	<Prolog system>  MacProlog"Prolog_by_BIM, Release 4.0
	<Other languages> None.
	<Lines> 1200
	<Predicates> 160
	<Platform> Sun4 Sparcstation SLC / Unix_Berkeley
	<Type of Use> In-house
	<Current Status> Prototype
	<No. of Users> 2
	<Information Date> 09/89
	<References>  * F. Martinolle, "Analyse de la commandabilite et 
de l'obs
ervabilite de systemes modelises hierarchiquement. Etude et 
realisation de l'ou til informatique PHOEBUS", Ph. D. of the INSAT, 
Toulouse, France, Nov. 1990 (in french) 
* F. Martinolle, J.C. Geffroy, B. Soulas, "Functional Extraction of 
Hierarchical Sequential Systems", IFIP, Int. Workshop on Applied 
Formal Methods for Correct VLSI Design, Leuven, Belgium, pp. 216-225, 
Nov. 1989
* F. Martinolle, G. Motet, J.C. Geffroy, "Multi-level simulation and 
analysis of hierarchical models", SCS European Simulation 
Multiconference, Erlangen, Germ
any, June 1990
* F. Martinolle, J.C. Geffroy, B. Soulas, "Testability Analysis of 
Hierarchical Finite State Machines", IEEE European Design Automation 
Conference, Amsterdam,
 Feb. 1991
* F. Martinolle, J.C. Geffroy, K. El Maadani, "Test-Aided Design 
Evaluation Tools for Complex Systems", International Conference on 
Concurrent Engineering & E
lectronic Design Automation, Bournemouth, UK, March 1991
* F. Martinolle, J.C. Geffroy, B. Soulas, "PHOEBUS: A Tool for 
Hierachical Testability and Redundancy Analysis", poster session, 
Proc. of IEEE European Test Conference, Munich, Germany, Apr. 1991				     
* C. Rominger, J.C. Geffroy, B. Soulas, C. Guibout, "PHOEBUS: A Tool for 
Analysis of Hierarchical Finite State Machines", poster session, 1st 
Conference on the Practical Application of Prolog, London, UK, Apr. 1992 
</Program>

<Program> Phoebus
	<Number> 258
	<Domain> Systems testing
	<Purpose> analysis of hierarchical finite state machines
	<Description> This can be applied to complex sequential systems such as 
control parts for functional extraction and formal verification of a 
realization with the specifications and testability evaluation.
	<Developed by> C. Rominger, J.C.Geffroy, B. Soulas, C. Guibout
	<Organization> INSAT-GERII, EDF-DER
	<Country> France
	<Prolog system> Prolog by BIM
	<Platform> SUN1+Sparcstation with 20MB
	<References> Pheobus: A tool for analysis of hierarchical finite state 
machines. Abstracts of poster sessions, Int. Conference on Practical 
Application of Prolog, London 1992.
</Program>

<Program> Phosulk
	<Number> 99
	<Domain> Conservation Agriculture
	<Purpose> Land Conservation Advisor
	<Description> To facilitate farmers in high rainfall areas to make the 
correct fertliser deciion - avoid water catchment pollution and the devise 
most economical fertiliser strategy
	<Developed by> Michael Brown
	<Organization> Entretec Pty & Dept Agric, West Australia
	<Country> Australia
	<Prolog system> PDC Prolog v 3.21
	<Other languages> C about 10%
	<Lines> 4000
	<Predicates> 450
	<Platform> DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> June 1991
	<No. of Users> 6
	<Information Date> March 1992
</Program>

<Program> PHRASE
	<Number> 497
	<Domain> Projects involving safety, where human factor contribute to 
safe system operation.
	<Purpose> Human reliability Analysis.
	<Description> Currently, the only way to perform a human 
reliability analysis is to use standard paper-based procedures which 
lead the analyst through his calculations. PHRASE encapsulates 
several human reliability methods and data, and leads the analyst 
through the calculations using a hypertext interface
(based on the Hypertext Toolbox provided by PDC). Each time the user makes 
a choice by selecting a word on the screen, PHRASE performs calculations 
before the hypertext jump is made. The user is presented with "the answer" 
when he reaches the end of the hypertext "cook-book".
	<Developed by> Steve Lympany, Barry Whittingam.
	<Organization> Electrowatt Engineering.
	<Country> England.
	<Prolog system> PDC Prolog.
	<Other languages>  None.
	<Lines> 800
	<Predicates> 200
	<Platform> MS-DOS on IBM PC.
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Date deployed> Not yet released.
</Program>

<Program> PI
	<Number> 1
	<Domain> Theorem proving
	<Purpose> Interactive development and editing of formal proofs.
	<Description> PI is a system for the interactive development of proofs 
in the theory of partial inductive definitions. This mathematical theory 
can, in this context, be regarded as a general logic. Proofs are displayed 
graphically in a window. The user gives commands to the system using a 
mouse.
	<Developed by> Lars-Henrik Eriksson
	<Organization> Swedish Institute of Computer Science (SICS)
	<Country> Sweden
	<Prolog system> SICSTUS
	<Other languages> None
	<Lines> 3500, excl comment lines
	<Predicates> 290
	<Platform> Sun-3/4, UNIX w/ Xwindows
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> August, 1990
	<No. of Users> 2
	<Information Date> October 18, 1991
</Program>

<Program> PIAF
	<Number> 445
	<Domain> Electronic Design.
	<Purpose> Full custom integrated circuit floorplanning.
	<Description> PIAF is a combined knowledge-based/algorithmic 
full-custom integrated circuit floorplanning system. Knowledge is 
represented using transformation rules and guides the invocation of 
sophisticated graph processing algorithms and interprest their 
results.
Knowledge representations used in PIAF include: static and dynamic 
frames, predicate logic, transformation rules and procedural 
attachements. PIAF features a "context" representation that is 
backtrackable in a controlled fashion enabling controlled 
"forgetting" and permitting the avoidance of expensive re-
computations. The underlying structure of PIAF was actually separated 
at a later stage to produce a stand alone Knwoledge Based system 
shell called BREL. BREL was used for the rapid prototyping of 
Electronic
CAD tools.
	<Developed by> Marwan Anwar Jabri.
	<Organization> Systems Engineering and Design Automation 
Laboratory
Sydney University Electrical Engineering.
	<Country> Australia.
	<Prolog system> Quintus Prolog (first version) later on Sicstus Prolog.
	<Other languages> Pascal and C to implement the algorithmic library 
of
graph processing algorithms.
	<Lines> 16,000
	<Predicates> 5,000
	<Platform> Sparc station (Unix).
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 1989
	<No. of Users> 3
	<Information Date> 1992
	<References> M. Jabri, ``An Artificial Intelligence Approach to 
Integrated
Circuit Floorplanning'',  Springer-Verlag, Lecture Notes in Engineering, 
Vol 1.66, 1991.
</Program>

<Program> Pictorial Janus
	<Number> 16
	<Domain> Program Visualization
	<Purpose> Provide a visual programming environment for
concurrent logic (and more generally constraint) programming
	<Description> Pictorial Janus is a visual programming environment being 
built to support concurrent constraint (or logic) programming.  A visual 
syntax for Janus; a concurrent constraint programming language; has been 
designed.  Currently; two modules of Pictorial Janus are operational: a 
parser and an animator.  Visual programs can be created using nearly any 
illustration or CAD tool capable of producing a Postscript description of 
the drawing.  That description is executed by a Postscript interpreter; and 
a trace of the strokes and text drawn is saved to a file.  The Pictorial 
Janus parser then interprets the trace and produces input to the Pictorial 
Janus Animator.  The animator generates a stream of animation frames and 
associated sounds.  The colors; shape; fonts; line weights; used in the 
original drawing are preserved so that the animation displays these 
elements in the same graphical terms as they were conceived and created.
	<Developed by> Ken Kahn with help from Vijay Saraswat; Volker
Haarslev; and Susan Kropf
	<Organization> Xerox PARC
	<Country> USA
	<Prolog system> Strand88
	<Other languages> C (5%); Postscript (10%); cshell (5%)
	<Lines> 11000
	<Predicates> 500
	<Platform> SparcStations (stand-alone and networks of them)
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 3/91
	<No. of Users> 3
	<Information Date> 12/91
	<References> Kenneth M. Kahn and Vijay A. Saraswat 
"Complete Visualizations of Concurrent Programs and their Executions"
October1990
Proceedings of the IEEE Visual Language Workshop
Kenneth M. Kahn
Concurrent Constraint Programs to Parse and Animate Pictures of 
Concurrent Constraint Programs
1991
(submitted for publication)
Kenneth M. Kahn
Towards Visual Concurrent Constraint Programming)
1991
(submitted for publication)
Kenneth M. Kahn
Animation of Horn Clause Proofs
1991
(submitted for publication)
</Program>

<Program> PigE
	<Number> 43
	<Domain>  farm management.
	<Purpose> An expert system for optimising the growth of pigs raised 
in
pig pens.
	<Description> PigE is an intelligent back-end to a 
mathematicalmodelling package called AUSPIG. The model:
** simulates the growth and reproduction of pigs;
** identifies factors that limit optimal performance of the pig;
** identifies management strategies that maximise enterprise profit.
The expert system:
** presents an abstracted description of the output of the model in a
form that a non-mathematician can understand;
** suggests dietary; housing; genotype; or resource input changes 
that can improve the profitability of the herd.
Verification studies have demonstrated that the expert system can
significantly out-perform human experts interpreting the output of the 
model (performance measured in dollars per square metre per day).  In a 
usual case; the improvement is of the order of 10%.  If spread over the 300 
kilotonne $500 million dollar (per annum) Australian pig herds;this 
improvement would be in the order of $50 million.
	<Developed by> Dr. John Black Joel Fleming Tim Menzies Murray Dean
	<Organization> Black & Fleming:  Commonwealth Scientific 
Industrial
Research Organisation (CSIRO); Division of Animal Production; 
Prospect;
Sydney.  Menzies & Dean:  Stowe Computing Australia
	<Country> Australia
	<Prolog system> Arity Prolog
	<Other languages> Pascal; MKS Toolkit; some shell script.
	<Lines> Is this a serious measure of a program?  The code listing is an 
inch thick.  Many comments. Lots of files.
	<Predicates> 211
	<Platform> DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> Mid-1988 onwards
	<No. of Users> Farm managers in America; Holland; Belgium; France; 
Spain and
Australia.  Pig diet manufacturers in Europe and USA
	<Information Date> 23-2-92
	<References> "An Expert System for Raising Pigs"; in Proceedings of 
the First International Conference on the Practical Application of Prolog; 
April 2-3; 1992.
</Program>

<Program> PISSARRA
	<Number> 384
	<Domain> Maintenance and repair centres train tracks.
	<Purpose> Improve the management of trains maintenance and repair centre.
	<Description> Expert system for assisting the management of the 
maintenance and repair centre's train tracks for Ferrocarrils de la 
Generalitat de Catalunya. System at final installation phase.
	<Developed by>  Jose A. Marques and programmers team.
	<Organization> AIA
	<Country> Spain
	<Prolog system> PDC Prolog 3.31
	<Lines> 20000
	<Predicates> 625
	<Platform> Developed PC 386 / MS DOS
Delivery PC 386 / MS DOS Novell Netware DOS
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 07/1992
	<Information Date> 07/1992
</Program>

<Program> PLANE
	<Number> 145
	<Domain> Production planning.
	<Purpose> An evolutive planning system for scheduling Aircraft 
Production.
	<Description> PLANE is an evolutive planning system written using CHIP 
for the scheduling of aircraft production over a long term horizon.  The 
system has been developed as a production control system and as a 
management planning tool.  The user can test scenarios such as adding an 
order, change due dates for production  and immediately realise the 
production considerations.
	<Developed by> DASSAULT/COSYTEC
	<Organization> CLP Software and Services.
	<Country> France
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 1000
	<Platform> SUN/HP/ IBM/DIGITAL workstations
PC  386/486		    UNIX   DOS
	<Date deployed> 12/91
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Int. Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> PLANE:  PLANIFICATION EVOLUTION
	<Number> 322
	<Domain> Scheduling.
	<Purpose> Gives Schedule and load of the long term production of 
aircraft.
	<Description> The aim of the PLANE system is to pace the assembly 
lines, under a given set of constraints, so as to minimize combination of 
the stock costs and of the production rate changes.  The system takes into 
account most of the planning expert's usual work (his choice of heuristics, 
his planning and factory parameters, his planning method).  The average 
scale of the problem is to plan 250 planes over five years, in a factory 
using 8 assembly lines.  These numbers are taken from the data of the 
Mirage 2000 being used to test the system.  The PLANE system uses the 
constraints in the finite domains (integer numbers), one of the three 
constraints domains handled by the CHIP language.
	<Developed by> Jacques Bellone.
	<Organization> DASSAULT AVIATION
	<Country> France.
	<Prolog system> CHIP of COSYTEC
	<Lines> Kennel:  2,500, You Interface:  4,500
	<Predicates> Kennel:  100 - N:of constraints:  47
	<Platform> SUN 4 UNIX
		     X11
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 03/89
	<No. of Users> 2 for the evaluation
	<Information Date> 09/03/92
	<References> Constraint Bazed Tools for Planning and Scheduling,  
C. PRADELLES and A. CHAMARD, Proc. on A.I. and K.B.S for space 
(Workohop) ESTEC (91).
PLANE:  an Evolutive Planning System for aircraft production.  The 
practical application of prolog..1st International Conference (92).
</Program>

<Program> Planning (more than) one-level buildings
	<Number> 274
	<Domain> Architecture, Planning
	<Purpose> Supports the planning process of a building with more than one 
level.
	<Description> The system provides support in the stages of planning 
with more than one level. First it generates the variants of ground-plans 
of all flats according to the special requirements of customers. The 
customer is given the possibility to classify the variants, to choose the 
most advantageous ones, and to exclude the less favourable ones. The 
program assembles from chosen variants the plan of the whole building 
satisfying requirements for the horizontal and vertical arrangements, the 
given measurements, and the conditions depending on the building site.
	<Developed by> Z. Markusz 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> Z. Markusz (1980)
The application of PROLOG for many storied panel house design.
Informatcio-Elektronika XV. No 5., 1980, pp.256-263.
Z. Markusz (1980)
An application of PROLOG in designing many storied dwelling houses.
Reprints of Logic Programming Workshop, Debrecen, 1980. pp.249-260.
</Program>

<Program> Planning of panel buildings
	<Number> 286
	<Domain> Computer aided design
	<Purpose> Plans the architectural planning of panel buildings
	<Description> The system generates the floor plan variants of flats 
with given size, number of rooms or half rooms, using the panel elements 
stored in its database. The system designs all the possible variants of the 
floor plans of a flat, given an initial set of data.
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> Z. Markusz (1977)
How to design variants of flats using PROLOG.
proc. of IFIP'77, Toronto, pp.885-890.
</Program>

<Program> PLASMA
	<Number> 312
	<Domain> Re-Engineering.
	<Purpose> PLASMA is a reverse CASE tool to maintain large COBOL programs.
	<Description> PLASMA Targets the market of existing large scale 
COBOL programs which have been in productive use for years but have 
to be maintained anyway, for example to accomodate legal changes or 
new services.
Maintenance cost eats up more than 60% of the software cost over the 
lifecycle of large scale commercial software.
PLASMA reduces this maintenance cost by
1.  Analysing one or more COBOL programs and their interrelation.
2.  Visualizing the analysis result.
3.  Supporting software maintenance activities.
 
Application Areas:
1.  Software maintenance support.
2.  Job operation optimization support.
3.  Condition analysis during construction of data dictionaries.
	<Developed by> Computer Applications Corporation Tokyo, Japan.
	<Organization> Computer Applications Corporation Tokyo, Japan.
	<Country> Japan.
	<Prolog system> IF/PROLOG.
	<Other languages> IF/PROLOG, OSF/MOTIF.
	<Platform> all UNIX platforms.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 12/91.
	<Information Date> 03/ 92.
</Program>

<Program> PLASTIC
	<Number> 377
	<Domain> Train control centre.
	<Purpose> Intelligent operator control in a train dispatching.
	<Description> Expert system train coordination control (logical, 
physical 
trains, shifts, scheduling etc.). Project developed for Terrocarrils de la 
Generalitat de Catalunya and installed at its Dispatching Centre.
	<Developed by> Antoni Trias.
	<Organization> AIA
	<Country> Spain.
	<Prolog system>  PDC Prolog 3.31
	<Other languages> void
	<Lines> 14000
	<Predicates> 1200
	<Platform> Developed PC 386 / MS DOS
Delivered PC 386 386 / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1991
	<Information Date> 07/1992
</Program>

<Program> PLASYNT
	<Number> 64
	<Domain> Combinational logic synthesis
	<Purpose> Multiple-level combinational logic synthesis
	<Description> PLASYNT starts from a description of a 
combinational circuit, specified in Berkeley PLA format.  Internally, 
the descriptions are stored and manipulated in a directed acyclic 
graphs representation known as BDD.
The program includes algorithms for  finding optimal variable orderings in 
the BDDs, for assigning don't care values, for factoring Boolean functions 
to decrease the area of the circuits, and for mapping descriptions to 
specific technologies.  PLASYNT currently maps descriptions to two 
different technologies: (i) networks of NAND-gates and inverters, and (ii) 
gates from Plessey's gate library.
	<Developed by> Mats Carlsson
	<Organization> Swedish Institute of Computer Science
	<Country> Sweden
	<Prolog system> SICStus Prolog 2.1
	<Lines> 1045
	<Predicates> 120
	<Platform> UNIX
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 11/91
	<No. of Users> 3
	<Information Date> 24/3/92
	<References> Proc. The Practical Application of Prolog Conf., London 
1992.
</Program>

<Program> PLEX
	<Number> 455
	<Domain> Software Development.
	<Purpose> LEX for Prolog: A lexical analyzer generator for Prolog.
	<Description> PLEX creates a tokenizer from a lexical 
specification very
much like that for the UNIX(TM) LEX program. The major differences are that 
the right-hand (action) parts of the lexical rules are Prolog goals, the 
source-code output by PLEX is Prolog, and because of the dynamic nature of 
the Prolog environment, lexical analyzers can be built ``on-the-fly'' and 
used immediately.  That is, there is no need for intermediate compilation 
steps as when using LEX.  Together with MULTI, this program provides a 
complete environment for building parsers and generators from a single 
very-high-level specification.
	<Developed by> Suresh Rajgopal and Peter Reintjes.
	<Organization> DAZIX/Intergraph Corporation.
	<Country> USA
	<Prolog system> Quintus Prolog
	<Other languages> none.
	<Lines> 646
	<Predicates> 131
	<Platform>  Quintus Platforms.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 6/92
	<No. of Users> 3
	<Information Date> 11/92
</Program>

<Program> PPLC 2.0 -  flexible  process  control  by  Prolog  
represented
higher order Petri-nets .
	<Number> 485
	<Domain> Process Control.
	<Purpose>  Process  control  of  single  units  or  small  chemical   
and
biotechnological plants by a structure-based  model,  integrating the 
batch, sequential, feedforward and feedback control actions.
	<Description>  PPLC integrates the pseudo-parallel executing 
abilities of  the
Petri-like  nets  with  the   possibilities   provided   by   the
prototyping and unification characteristics of  the  PROLOG.  The 
continuous and/or batch control algorithms  are  described  by  a 
structural model consisting of active and passive  elements.  The 
passive elements are reserved places for  the  various  kinds  of 
control  information,  while  the   active   elements   are   the 
transitions determining the essential skeleton of the information 
processing. The architecture of the structural models is  similar to 
the higher order Petri-nets, however,  instead  of  tokens  or 
tuples, optional PROLOG terms can be associated with the  passive
elements and these terms are manipulated as information. The main executing 
rule of the kernel  algorithm  describes  that  if  all input signs and 
auxiliary conditions for a given  transition  are fulfilled, then the 
active element after an optional  time  delay fires, i.e. the prescribed 
output  signs  are  rewritten  in  the respective places. The structural 
model of the control is defined by a dynamic database or by  dynamic  
partitions  of  the  PROLOG.  The software  is  built  up  from  four  
shells  of  the user-friendly interface,  the  directly  editable  
database,  the actual includes and the kernel definitions, respectively.
	<Developed by> Bela Csukas, Laszlo Jambor, Rozalia Lakner .
	<Organization> University  of  Veszprem,  Supported  by  the  Hungarian  
Basic Research Grant N OTKA-2550 
	<Country>  Hungary.
	<Prolog system> TURBOPROLOG 2.0, PDC PROLOG 3.31, CSPROLOG 3.25; 
(ALL-
MULTILOGIC Ltd., Budapest).
	<Other languages>  Can also be implemented in object oriented languages.
	<Lines> 1000
	<Predicates> 150
	<Platform> PC/AT 386; transputer and work station versions in  preparation 
.
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 6/1990
	<No. of Users> 10
	<Information Date> 12/1993
	<References> In preparation.
</Program>

<Program> Practitioner Awards KBS
	<Number> 191
	<Domain> Pensions administration
	<Purpose> 
	<Description> The system calculates retirement benefits for 
medical and dental practitioners employed by National Health Service. 
Conventionally this is a long and tedious task because the procedures 
used for assessment are defined by a large volume of ever changing 
legislation.
  Sets of practitioners records can be selected and downloaded from a 
mainframe to a 'master PC', and then distributed to other PCs over a LAN. A 
Prolog program converts incoming records to a frame-based representation 
prior to distribution. Each of the slave PCs contains a local copy of the 
Knowledge-based system which is used to maintain the content and accuracy 
of the mainframe data, generate graphical displays of the member's service 
history, calculate benefits and print out the results and appropriate 
documentation.
	<Developed by> Gerry Piper
	<Organization> Telecomputing PLC, IBiS Ltd.
	<Country> UK
	<Prolog system> LPA Prolog, flex
	<Platform> PC
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> June 1990
	<No. of Users> 11
	<References> KBS system calculates pension awards to GPs. Expert 
Systems Applications, Jan 1991. p21-22/
</Program>

<Program> Pre-Printed Blankrupter
	<Number> 395
	<Domain> Legal forms processing and decision making.
	<Purpose> Completely processes all the Federal Bankruptcy Forms for 
chapter 7, 11, 12 & 13 bankruptcy filings.
	<Description> Blankrupter processes and fills in Julius 
Blumberg's Pre-Printed Bankruptcy sets on dot matrix, daisy-wheel and 
laser printers complete documents for chapter 7, 11, 12 & 13 for the 
U.S. Bankruptcy Court. Sold nationwide through Julius Blumbergs Inc. 
distribution channel, this program is the newest upgrade to the 
Blankrupter program introduced in 1988. This latest update includes 
the ability to process Chapters' 11, 12 and 13 the creation of 
dynamic forms (ie. the amount of entry determines the number of
forms produced) and a new file achieving system designed for law firms. 
Price: $ 200.00.
	<Developed by> Norths Winds and Julius Blumberg Inc.
	<Organization> Norths Winds and Julius Blumberg Inc.
	<Country> USA.
	<Prolog system> PDC Prolog (100%)
	<Other languages> none.
	<Lines> 15.000
	<Predicates> 500
	<Platform> MSDOS 2.0, 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed>  10/1/88
	<Information Date> 9/1/92
	<References> Catalog.
</Program>

<Program> PRECIS
	<Number> 190
	<Domain> Systems analysis and design
	<Purpose> Analysis and CASE tool
	<Description> Precis offers the ability to record a business 
structure using three basic concepts: the business decision, the 
consequence and the business rule rlating a decision to its 
consequences. This is done using chart and flow diagrams and 
incorporates a code generator facility to automatically generate 
application code. Precis also provides facilities for documenting the 
system.
  Under development is a system of formalised English for specifying 
computer systems.
	<Developed by> John Warren, Basil Barnett
	<Organization> Precision Design
	<Country> UK
	<Prolog system> LPA Prolog Professional
	<Platform> IBM PS/2
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> Preditor
	<Number> 197
	<Domain> Electronics Design
	<Purpose> A VLSI design editor
	<Description> A design editor for VHDL design which works in 
association with the AUNT translator.
This replaced a similar editor written in 50,000 lines of C which was very 
difficult to maintain.
	<Developed by> Peter Reintjes
	<Organization> Microelectronics Center of North Carolina
	<Country> USA
	<Prolog system> Quintus
	<Lines> 5,000
	<Platform> Unix, VAX workstation
	<References> P.B. Reintjes: A VLSI Design Environment in Prolog. 
ICLP5, 1988, pp70-81
</Program>

<Program> PRESS 
	<Number> 77
	<Domain> Programming Language Interpreters
	<Purpose> (PRolog Execution of Standard Semantics) 
(1) To check and validate the informal semantics of draft standard 
Prolog, and
(2) To investigate techniques of writing high quality maintainable
Prolog programs.
	<Description> The informal semantics of Prolog in the draft 
standard defines the control constructs (i.e. conjunction/2, 
disjunction/2, true/0, fail/0, call/1, cut/0, if-then/2, if-then-
else/3, catch/2, throw/1) using a stack model. The details are 
complex, and this interpreter has been written in order to check the 
description is correct.
The subsidiary aims of the interpreter have been met by providing
(1) checks that internal predicates are called with valid arguments, 
and
(2) a set of test cases whose results can be easily compared
with model results. This checks that changes to one part of the interpreter 
have not broken other parts.
	<Developed by> Roger Scowen
	<Organization> National Physical Laboratory
	<Country> UK
	<Prolog system> SICStus 2.1 #3
	<Other languages> None
	<Lines> 1000 lines, approximately
100 -- history of PRESS,
250 -- internal checks,
200 -- subsidiary predicates,
100 -- test cases,
350 -- interpreter.
	<Predicates> 38
	<Platform> Sun 3/50, Sun OS Release 4.1
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> September 1991
	<No. of Users> 1
	<Information Date> March 1992
	<References> ISO/IEC JTC1 SC22 WG17 N92: Draft for Standard 
Prolog, March 1992.
National Physical Laboratory, Teddington, Middlesex.
</Program>

<Program> Print-A-Form
	<Number> 397
	<Domain> From Processing.
	<Purpose>  Automates the processing of a unique Deluxe Check Printer's 
Inc. Form.  For example an invoice or check.
	<Description> The Print-A-Form product line consists of 55 individual 
programs that process Deluxe Check Printer, Inc. standard business forms. 
The features include, database system, complete math, ASCII output, Journal 
report, macros, and much more. Individual form programs include invoices, 
purchase orders, checks, packing slips, statements, etc. Combined programs 
include Tax forms, constructions forms and medical forms.
	<Developed by> Norths Winds & Deluxe Check Printer Inc.
	<Organization> North Winds.
	<Country> USA.
	<Prolog system> PDC Prolog (100%)
	<Other languages> none.
	<Lines> 9.000
	<Predicates> 400
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/1987
	<Information Date> 9/15/92
	<References> Deluxe Catalogues 1988-92.
</Program>

<Program> Pro*CONSUL
	<Number> 242
	<Domain> Communications design
	<Purpose> to help the designer and implementer of communications 
protocols 
	<Description> A central repository stores all the information necessary 
for the design, testing and simulation of the protocols. This is surrounded 
by a number of components such as that finding test sequences for a given 
protocol.
	<Developed by> Peter Ecsedi-Toth
	<Country> Hungary
	<Prolog system> MProlog and FAIR
	<References> K. Tarnay, S Dibuz-W‡gner, Peter Ecsedi-T—th, P. A. 
W‡gner: A Knowledge-based approach to communication protocol endingeering. 
Proc Seminar on Future Trends in Information Engineering, Univ. Salzburg, 
Sep 1990, pp115-127
</Program>

<Program> Pro-Sylvia
	<Number> 146
	<Domain> Agriculture
	<Purpose> Optimization of cutting wood panels with minimum wastage
	<Description> The aim is to define a cutting diagram for a set of 
rectangular panels, each in considerable quantities, in a particular 
type of panel (e.g. white fibreboard, 267 by 207cm). The program 
makes it possible to calculate a cutting diagram for lists of 
approximately 30 different parts in 4 minutes and produces results 
for a dozen different types of panels with an average wastage rate of 
3-5%
	<Developed by> Michel Rueher, Bruno Legeard
	<Country> France
	<Prolog system> Prolog III
	<Information Date> 6/11/91
</Program>

<Program> PROCAPS
	<Number> 227
	<Domain> Electronic scheduling
	<Purpose> support and control the planning and scheduling of production 
lines of electronic components
	<Description> The Process Oriented Capacity Production Scheduler 
focuses on the optimisation of the use of available resources and on 
the reduction of the throughput time of the produced items. The 
system incorporates an explicit model of teh factory organization and 
includes a planning and scheduling functionality as they were 
previously performed manually. 
It is intended to integrate the system in the production line, which will 
require additional interfaces to corporate databases and control functions 
already available.
	<Organization> BIM, Philips of Eindhoven
	<Country> Belgium, Netherlands
	<Prolog system> Prolog by BIM
</Program>

<Program> PROCUT
	<Number> 481
	<Domain> Building and woodworking industry.
	<Purpose> Optimize layout plan for cutting of rectangular panels.
	<Description> Menu driven, single user program intended for use 
by the 
building and woodworking industry as part of quantitative analysis 
and works planning. The program provides a graphics display and 
printout of the final cutting schedule and materials list.
The database includes facility to store and retrieve all previous 
jobs and standard materials lists.
Program has three levels of optimization and allows for grain and no 
grain materials.
All layouts can be cut on a standard table or circular saw.
Performance benchmark:
Platform: AT/286 16 Meg CPU no EMS
Number of items: 1000
Best case: 20 seconds
Worst case: 17 minutes.
	<Developed by> Tinus Wille.
	<Organization> Tinus Wille and Compuchem cc (sponsor).
	<Country> South-Africa.
	<Prolog system> PDC Prolog.
	<Other languages>  Assembler routines for graphics hard copy
RTLink Plus for the overlay version.
	<Lines> 1670 Prolog source.
	<Predicates> 106 defined predicates.
	<Platform> (ISA) IBM PC/AT and compatible
640 K base RAM minimum
2 Meg EMS optional but recommended
Graphics support: Hercules (Mono) EGA and VGA
Printing Epson 9 pin
DOS 3.2 or later
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed>  Beta release 2/12/1991.
	<Information Date> 8/92
</Program>

<Program> ProdAction
	<Number> 345
	<Domain> Finite Capacity Scheduling in Discrete Manufacturing 
Industries.
	<Purpose> Finite Capacity Scheduling in Discrete Manufacturing 
Industries.
	<Description> The prodAction system allows the user to:  a)  Define his 
factory or discrete manufactoring environment (resources: machines, labour, 
tooling).  b)  Define the Work Order process flows.  c)  Define the 
Prioritizing Policies.
	<Developed by> Ignacio, Adrian Kromer, Alejandro Scheuer.
	<Organization> Invap S.E.
	<Country> Argentina.
	<Prolog system> PDC Prolog 3.20 for DOS, now upgrading to PDC Prolog 3.30 
for Windows.
	<Other languages> C (15%), Fox (20%), Quattro(5%).
	<Lines> 1.5 MB of Source code.
	<Predicates> approx. 800
	<Platform> 80386/486 IBM compatible, DOS 5.0 for development and delivery.
	<Type of Use> Licensed
	<Date deployed> 06/91
	<Information Date> 07/92
	<References> Sociedad Argentina de Investigacion Operativa - 
Sept. 1990 ADR:  Asignacion Dinamica de Recursos
Ignacio Grossi, Alejandro Scheuer, Marcelo Silvera.
</Program>

<Program> PRODATA
	<Number> 306
	<Domain> RDBMS
	<Purpose> Interface to commercial RDBMS
	<Description> PRODATA is a tightly coupled transparent interface 
between PROLOG SYSTEMS and RDBMS, which include ORACLE, INFORMIX 
and 
SYBASE.
	<Developed by> Rob Lucus
	<Organization> Keylink Computers Ltd
	<Country> England
	<Prolog system> LPA/QUINTUS/POPLOG
	<Other languages>    C
   20%
	<Lines> 1500 per I/F
	<Predicates> 90 per I/F
	<Platform> Mac/Sun/PC/HP/Sequent/Vax (Unix/vms)
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> Jan. 1988
	<No. of Users> 60
	<Information Date> 24/07/92
	<References> PROLOG-relational database INTERFACES.  R.J. Lucus.  
Information and Software Technology.  Dec. 1991 Vol. 33 No. 10.
</Program>

<Program> Prof'Expert
	<Number> 373
	<Domain>  Computer based training.
	<Purpose> The purpose of the system is to provide computer training in 
basic skills.
	<Description> The system is a hypertext which permits the learners to 
inform themselves about basic topics like French grammar, mathematics, etc. 
It is planned that the links between the information units adapt themselves 
auto-matically to the learner level. For some topics exercises of drill and 
practice are available. For these the teacher or the learner can select 
options like the presentation or the error admittance. They dispose of 
several help systems: How to obtain a solution or part of the answer etc. 
The learner and the teacher can examine and analyse the way that the 
problems were faced and solved. From a psychological point of view, the 
purpose of the system is to develop basic skills (automatisms) but with the 
possibility for the learner to examine his procedure and to develop 
metacognitiv capability. From an informatic point of view, the system is 
constitued of three generic expert systems. The first is the domain expert 
system (to perform calculus ....) the second is the didactical one. Its 
task is to establish the level of difficulty, classify the errors, etc. The 
third is the psychologist who should establish the learners profile.
	<Developed by> Jean-Pierre Baer, Marie-Louise Carrera, Anne 
Marechal,
Luc-Olivier Pochon, Christophe von Siebenthal.
	<Organization>  Centre de Formation Professionelle du Littoral 
Neuchatelois
(CPLN),
	<Country> Switzerland.
	<Prolog system> PDC Prolog.
	<Other languages>  IBM compatible, DOS, Windows planned.
	<Lines> 15000
	<Predicates> 300 relevants
	<Platform> IBM compatible, DOS, Windows planned.
	<Type of Use> Licensed
	<Current Status> Prototype
	<Date deployed> Oct. 1990
	<Information Date> August 1, 1992.
	<References> Pochon, L-O (1991) Le projet Prof'Expert. 
Neuchatel, CPLN, feb. 1 1991.
Vallet F. (1992) G_n_rateur de messages pour un syst_me d'EAO. Travail de 
diplome. Neuchatel, CPLN July 1, 1992.
</Program>

<Program> PROGART
	<Number> 280
	<Domain> Software applications
	<Purpose> To analyse programming style and effectiveness of syntactically 
correct PL/I and COBOL programs
	<Description> The system evaluates the programming style and 
effectiveness of syntactically correct PL/I and COBOL programs. Evaluation 
is done across the following dimensions: structure, style, effectiveness 
and complexity. In case where quality errors are discovered in a program 
the system suggests correct alternatives. During the structural analysis 
the system reveals and prints (in the form of a hierarchy diagrams) the 
logical structure of the program being analysed, and records the structural 
corrections that are to be performed.
	<Developed by> E. Halmay, P. Gero 
	<Organization> SZAMOK
	<Country> Hungary
	<Prolog system> MPROLOG
	<Other languages> PL/I
	<Information Date> 1984
	<References> E. Halmay, P. Gero (1981)
The PROGART system
SZAMOK report, 1981.
E. Halmay, P. Gero (1981)
PROGART: a computerised assistant for the programming instructor.
proc. of the 3rd Conference on Computer Education, 1981, Lausane.
</Program>

<Program> PROLAB
	<Number> 147
	<Domain> Medical
	<Purpose> PROLOG-based assistant for biochemical data interpretation
	<Description> Submitted to PAC. Paper 33
	<Developed by> Peter Hammond et al
	<Country> UK
	<Prolog system> MacProlog
	<Platform> Mac
	<Information Date> 31/10/91
</Program>

<Program> PROLEX
	<Number> 148
	<Domain> Expert system
	<Purpose> A Prolog Based Flexible Expert Sytem
	<Description> Submitted to PAC. Paper 221
	<Developed by> Julius Csonto
	<Country> Czechoslovakia
	<Prolog system> Arity Prolog
	<Platform> PC
	<Information Date> 18/10/91
</Program>

<Program> PROLOG VERIFIER
	<Number> 288
	<Domain> Program verification
	<Purpose> A system to verify Prolog programs
	<Description> The system aims at proving partial correctness of PROLOG 
programs. A subsystem, consisting of a program for formula transformation 
and a program for general theorem proving can be used in itself, for 
interactive theorem proving. The interactive formula transformation system 
performs natural deduction on the basis of either built-in or interactively 
generated transformational (inference) schemes. The general theorem prover 
is based on the resolution principle.
	<Developed by> K.Balogh
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> K. Balogh (1978)
An interactive  verifier for PROLOG programs.
Mathematical logic in Computer Science.
B. Domolki, T. Gergely (Eds)
Coll. Math. Soc. J.Bolyai, Vol. 26. Budapest, 1981, pp.111-142.
K.Balogh (1979)
On a logical method serving the proof of the semantic features of 
programs.
PhD thesis, Eotvos Lorand University, Budapest, 1979.
</Program>

<Program> Prolog-D-Linda
	<Number> 462
	<Domain>  Parallel Prolog Programming.
	<Purpose> To provide a Parallel Prolog Programming environment.
	<Description> Prolog-D-Linda (Prolog-Distributed-Linda) is an 
embedding of the Linda paradigm into SICStus Prolog. (Linda is a 
coordination language that is injected into existing programming 
languages to produce new parallel programming languages. Linda 
permits cooperation between 
parallel processes by controlling access to a shared tuple space.)  
Prolog-D-Linda supports a distributed tuple space, unification and 
Prolog style deduction in the tuple space, and a control hierarchy 
that provides remote I/O facilities for client processes.
The Prolog-D-Linda embedding of Linda in Prolog is very natural : the 
pattern matching and database features of Prolog have been used 
directly in the embedding; garbage collection and hashing in the 
tuple space are provided free by the Prolog implementation; the 
implementation of formals in tuples is direct; the way the tuple 
space is partitioned is specified by a 
Prolog program. The user controlled partitioning of the tuple space 
makes it possible to tune the use of the tuple space so that 
bottlenecks are avoided. The capability of deduction in the tuple 
space provides direct solutions to problems that were previously 
difficult or impossible. The remote I/O facility 
allows all client processes to be interactive.
	<Developed by> Geoff Sutcliffe, James Pinakis.
	<Organization> The University of Western Australia.
	<Country> Australia.
	<Prolog system> SICStus Prolog.
	<Other languages>  C - 40%.
	<Lines> 642
	<Predicates> 60
	<Platform>  SUN Sparc/SUN OS / Any/UNIX.
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 11/91
	<No. of Users>  C - 40%
	<Information Date> 2/11/92
	<References> Sutcliffe G., and Pinakis J. (To appear), Prolog-D-
Linda: An 
Embedding of Linda in SICStus Prolog, In Kacsuk P., Wise M. (Ed.), 
Proceedings of the Joint Workshop on Distributed and Parallel 
Implementation of Logic Programs (Washington, DC, 1992).
</Program>

<Program> PROM
	<Number> 29
	<Domain> Institute of Theoretical Computer Science
	<Purpose> flexible "make" tool
	<Description> PROM is a flexible "make" tool designed for the special 
purposes of large; portable software systems. It allows the combination of 
powerful default-rulesets with individual treatment of files requiring 
this. Rulesets may be developed for any language or tool operating on 
files.
	<Developed by> T. Kielmann
	<Organization> Technical University Darmstadt
	<Country> Germany
	<Prolog system> C-Prolog; V1.5
	<Lines> 500
	<Predicates> 80
	<Platform> HP 9000; S400(and other 32-bit UNIX machines with running C-
Prolog)
	<Type of Use> In-house
	<Current Status> Prototype
	<No. of Users> 5
	<Information Date> 1991
</Program>

<Program> PROM
	<Number> 277
	<Domain> Systems Programming/ Computer architecture
	<Purpose> Mapping of INTEL 3000 Microprograms
	<Description> PROM performs  the task for address assignment for 
the INTEL 3000 microprocessor.
The PROM database can be thought of as a matrix of INTEL instructions with 
each instruction in the matrix indicating its successor(s). There is only a 
limited set of matrix positions for placing a successor (e.g. in the same 
column as the predecessor) that depends on the type of instruction. PROM's 
input consists of a set of partially loaded store and a microprogram to be 
allocated in the given store. The system gives either a possible mapping 
plan as output or indicates the impossibility of mapping the given 
microprogram. 
	<Developed by> K. Balogh, Z. Farks, I. Futo, P. Garami, I. Herenyi, E, 
Santane-Toth, J. Szeredi, P. Szeredi, J.Visnyovszky 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> K. Balogh, Z. Farks, I. Futo, P. Garami, I. 
Herenyi, E, Santane-Toth, J. Szeredi, P. Szeredi, J.Visnyovszky 
(1978)
The application of Prolog to the design of software and hardware 
objects. 
vol. I -IV
NIM IGUSZI and SzKI reports. SOFTTECH D21-D25, D27,SZAMKI, 1978.
</Program>

<Program> Prometheus
	<Number> 409
	<Domain> Expert Systems Toolkit.
	<Purpose> To provide a flexible tool for developing KBS.
	<Description> Prometheus is a general purpose expert system 
toolkit.
It provides a rich set of facilities for representing
knowledge, including backward and forward chaining rules,
frames and a procedural control language.  It supports
automatic truth maintenance and also provides object
oriented style message passing.  These underlying facilities are used to 
support the tools that make up the development environment, but its open 
architecture allows for the user to call out of or into the toolkit from 
Prolog.  This means the system can be easily extended or tuned by the user, 
or embedded in other Prolog applications.
	<Developed by>  Steven Hedges, Mick Buckley, Paul Taylor, Tony Dodd.
	<Organization> Expert Systems Ltd.
	<Country> UK.
	<Prolog system> Prolog-2 from ESL.
	<Other languages> none.
	<Lines> 12889
	<Predicates> 955
	<Platform> PC/DOS and SUN/SunView.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> Dec 1988
	<No. of Users> 100 approx
	<Information Date> 25 March 1992
	<References> "Prometheus - an advanced logic based toolkit."		    
R. Broughton, J. Paine and M. Yazdani.
Intellect 1991.  ISBN 1-871516-13-7
</Program>

<Program> PROMOTEX
	<Number> 173
	<Domain> Automobile diagnosis
	<Purpose> Breakdown diagnosis for siz-cylinder Mercedes engines equipped 
with Bosch ignition
	<Description> 
	<Developed by> Reinhardt Skuppin
	<Country> Germany
	<Prolog system> Prolog III
</Program>

<Program> Propagator
	<Number> 149
	<Domain> TOOL
	<Purpose> A Prolog-Based Causal Reasoning System
	<Description> Submitted to PAC. Paper 211
	<Developed by> Ming Xi Tang
	<Country> UK
	<Prolog system> Poplog
	<Information Date> 1/10/91
</Program>

<Program> PROS 800 / PROVISA
	<Number> 469
	<Domain> Industrial uses, mainly in areas of thermo procedure 
techniques.
	<Purpose>  Process visualisation and registration, analysis of process 
data.
	<Description> 
	<Developed by> Peter Mueller Mannhardt.
	<Organization>  Hennig Software Engineering GmbH.
	<Country> Germany.
	<Prolog system> PDC Prolog.
	<Other languages> Stony Book Pascal, Turbo Assembler.
	<Lines> 20.000 LOC
	<Predicates>  500
	<Platform> DOS , IBM compatible.
	<Current Status> Released
	<Date deployed>  1989
	<Information Date> 07/92
</Program>

<Program> PROSPEX
	<Number> 179
	<Domain> 
	<Purpose> process scheduling and planning system for the food, chemcial 
and petrochemical industries.
	<Description> Configuration is a three stage process.
  A model of the production is built using the supplied library of 
resources including capability, connectivity, routing, manning 
requirements and shift patterns; the processes and operations 
performed such as distillation, filtration and blending; and the 
materials - raw, products, recipes.
  Orders are specified by their attributes such as quantity, earliest 
start time, due-by date and priority
  Scheduling rules and preferences are developed and entered.
  Once configured, PROSPEX can generate schedules either manually or 
by taking one order at a time, allowing the user to optimise the 
system interactively or automatically using the prioritised list ot 
orders together with the scheduling rules.
 Additionally PROSPEX can simulate and display the execution of 
alternative plans and schedules thus providing a powerful "What if?" 
facility.
May not be Prolog!
	<Organization> SIRA
	<Country> UK
</Program>

<Program> Protee
	<Number> 150
	<Domain> Biology research
	<Purpose> Protee : A Knowledge based system to assist biology researchers
	<Description> Submitted to PAC. Paper 382
	<Developed by> Le Dizes
	<Country> France
	<Prolog system> Prolog II + hyper card.
	<Platform> MacII FX
	<Information Date> 6/11/91
</Program>

<Program> Protein Structure
	<Number> 245
	<Domain> Genetics research
	<Purpose> represent and reason about protein structure topology
	<Description> A description of the relative position of protein 
structural deatures enables declarative representations of super-secondary 
structures to be used to search for structural motifs.
	<Developed by> Chris Rawlings, John Fox, W. Taylor, J. Nyakairuy, M. 
Sternberg
	<Organization> Imperial Cancer Research Fund
	<Country> UK
	<Prolog system> Decsystem-10 Prolog
	<Platform> DECsystem 2060
	<References> C.J. Rawlings, W.R. Taylor, J. Nyakairuy, J. Fox, M.J.E. 
Sternberg: Using Prolog to Represent and Reason about Protein Structures. 
ICLP3, 1986 pp536-543
</Program>

<Program> Protein Structure Prediction
	<Number> 101
	<Domain> Molecular biology
	<Purpose> 
	<Description> Joe Bryngelson and John Hopfield, two CalTech chemist, 
developed a 'C' program which analyses proteins and helps predict where the 
protein twists into a particular shape called an alpha helix. Sam Southard 
and Stephen Taylor, also of CaltTech, used Strand88 to implement a parallel 
version of this algorithm. This new program shows linear speed-up with up 
to 32 processor nodes.
	<Developed by> Caltech
	<Country> USA
	<Prolog system> Strand88
</Program>

<Program> ProTest
	<Number> 151
	<Domain> TOOL
	<Purpose> A Knowledge Analysis Tool for the Development of Expert Systems
	<Description> Submitted to PAC. Paper 305
	<Developed by> Dr I D Watson & Norman
	<Country> UK
	<Prolog system> Asymetrix's Toolbox
	<Platform> PC + Windows
	<Information Date> 31/10/91
</Program>

<Program> Prototyping in Prolog
	<Number> 434
	<Domain> Computer Aided Design of New Software Tools for Task  
Support 
	<Purpose> Communication aid for the purpose of system design .
	<Description> The system is used to describe simple tasks for
graphic presentation to end users in a development team. So
far the task set is very simple but we are working on
extending the types of tasks which can be described. The intention is to 
investigate whether it is possible to use such systems as design aids in 
the design process.
	<Developed by> Lars Oestreicher.
	<Organization> Uppsala University, NUTEC
	<Country> Sweden.
	<Prolog system>  LPA Mac Prolog/Prolog++ 4.0 (Object-Oriented 
version
of Prolog with graphics management)
	<Other languages> Under development is a graphical front-end which is 
supposed to replace graphics management under Prolog.
Currently, however, the system is completely contained
within Prolog.
	<Platform> Macintosh II fx (runs on any Macintosh II) 8 MB, RAM
	<Type of Use> 
	<Current Status> 
	<Date deployed> Early summer 1992.
	<No. of Users> 2-3
	<Information Date> 7 april 1992.
	<References> Oestreicher L. 1990, Logic Descriptions in Rapid
Prototyping of Applications, In: Diaper D., Gilmore D., Cockton G. & 
Schackel B., 1990, Human-Computer Interaction, Proceedings of the 
conference:  INTERACT'90, pp.  1029 - 1030,  North-Holland, Amsterdam						 
]
Oestreicher L., 1991, Executable Formal Descriptions
for Software design - Logic and Logic Programming as a
Design Tool, Ph.L. Thesis 11, Uppsala University, Uppsala.
		     
Oestreicher L., 1992, Relating Task Analysis to System
Development, Paper presented at the 11th
Interdisciplinary Workshop on"Informatics and Psychology", June 1992.  
</Program>

<Program>  PSI.
	<Number> 465
	<Domain> Information Management.
	<Purpose> PSI is an application for data management and retrieval,
intended for the non-commercial user.
	<Description>  PSI is a handy system for the management and 
retrieval of 
your personal data, be it addresses, CD collections, or bibliog
raphic references. It is intended for the non-commercial user. It may not 
be as full-fledged as some data-base systems, but has many features that 
you won't find in most commercial systems. Also, you may find it easier to 
set up and faster to use. PSI is useful for a broad range of data. Indexing 
with descriptors makes searching for the data you need fast, and the 
interface to other data-base formats (import and export) is quite powerful.
	<Developed by> Markus P.J. Fromherz.
	<Country> Switzerland, USA.
	<Prolog system> LPA MacProlog.
	<Lines> 5000 ?
	<Predicates> 500 ?
	<Platform>  Apple Macintosh.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/89
	<No. of Users> 40
	<Information Date> 7/92
</Program>

<Program> PULSE STREAM CLASSIFIER
	<Number> 299
	<Domain> Radar/ EW
	<Purpose> To assist in the classification of streams of radar pulses
	<Description> Reduces processing time for staggred pulse stream 
extraction and classification by an order of magnitude or more.  Implements 
a neural network of flexible size in 14 PROLOG predicates.
	<Developed by> Gregory Bourassa
	<Organization> Lockheed Canada Inc.
	<Country> Canada
	<Prolog system> PDC PROLOG  V3.21
	<Lines> ~200
	<Predicates> ~14
	<Platform> IBM 386
MS - DOS
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 01/91
	<No. of Users> 30
	<Information Date> 01/92
	<References> NATO AGARD CONFERENCE PROCEEDINGS 499
"Future ESM Systems and the Potential for Neural Processing", A. G. 
Self, G. Bourassa.
For distrubution contact:  DEFENCE RESEARCH INFORMATION CENTRE, 
KENTIGERN 
HOUSE,  65 Brown St. Glasgow G2 8EX.
</Program>

<Program> PX Telephone Attendant
	<Number> 83
	<Domain> Telephony Application
	<Purpose> Conversational Answering Machine
	<Description> Answers a users telephone and takes a message by 
repeatedly prompting the caller and recording the responses.
	<Developed by> Larry Brunet
	<Organization> Bell-Northern Research Ltd.
	<Country> Canada
	<Prolog system> BNR Prolog
	<Other languages> Pascal, C
	<Lines> 7500
	<Predicates> 500
	<Platform> Macintosh
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 12/90
	<No. of Users> 30
	<Information Date> 2/92
	<References> Brunet, L., PX Telephone Attendant, Proceedings of 
the AVIOS '90 Voice Input/Output Applications Conference
Brunet, L., A Telephone Attendant in Prolog Proceedings of the Practical 
Application of Prolog Conference April, 1992
</Program>

<Program> PYTHON
	<Number> 473
	<Domain>  Problem Solving Methodologies.
	<Purpose> Expert system for recognizing and performing squeeze plays, an 
advanced strategy in the game of bridge.
	<Description> PYTHON is an expert system for recognizing and 
performing squeeze plays, an advanced strategy in the game of bridge. 
It performs, in its limited domain, at a truly expert standard, 
comparable to players of national ranking. The development of PYTHON 
illustrates how a well-constructed logical solution to a simple case 
of a problem can be extended in several different directions, 
yielding new results and a powerful program. PYTHON's core recognizes 
when a simple squeeze exists according to well-established theory. 
The core was extended to handle more complicated squeezes, also 
described by theory, making PYTHON's performance truly expert.
The core was also extended to generate a plan for playing squeezes which 
has been successfully executed by a separate bridge playing program. 
Finally, methods were added for recognizing and executing squeezes not 
covered by existing theory by analogy with the other methods.
	<Developed by> Yossi Nygate & Prof. Leon Sterling.
	<Organization> Weizmann Institute of Science .
	<Country> Israel.
	<Prolog system>  Siscstus.
	<Other languages>  None.
	<Lines> 2500
	<Predicates> 300
	<Platform> VAX
	<Type of Use> Personal
	<Current Status> Evaluation
	<Date deployed> 12 months.
	<Information Date> 11/92.
	<References> Journal of Logic Programming, Vol 8, pp. 21-39, 1990.
</Program>

<Program> QD-Janus
	<Number> 431
	<Domain> programming language implementation.
	<Purpose> A translator for a sequential implementation of the 
language Janus.
	<Description> A compiler for a sequential implementation of the
concurrent constraint programming language Janus
[Saraswat, Kahn & Levy, NACLP-90]. Programs are
translated from Janus to Prolog.
	<Developed by> S. K. Debray.
	<Organization> The University of Arizona.
	<Country> USA
	<Prolog system> Sicstus Prolog v2.1
	<Other languages> none
	<Lines> 4500
	<Predicates> 430
	<Platform> Sparcstation-2
	<Type of Use> Free/Public Domain
	<Current Status> Prototype
	<Date deployed> 12/90
	<No. of Users> 6
	<Information Date> 6/91
	<References> S. K. Debray, "QD-Janus: A Prolog Implementation of
Janus", manuscript, Dept. of Computer Science,
The University of Arizona, Tucson, June 1991.
</Program>

<Program> Rapidforms Forms Template - Contractor
	<Number> 401
	<Domain> Form Processing.
	<Purpose> To process 9 contractor related forms.
	<Description> The Rapidforms Business Form Template program gives you a 
friendly easy way to fill in and process various pre-printed forms. The 
Contractor's system automates these forms: Proposed, Job Invoice, Letter of 
Transmittal. Job Estimate and Change Order. Information entered in one form 
passes into the others, keeping track of your jobs, while producing your 
forms. Features include, Forms seen on the screen, complete math 
capability, build job files and reports, macros, batch printing and our 
Intelligent Forms Management System.
	<Developed by> North Winds.
	<Organization> North Winds & Rapidforms, Inc.
	<Country> USA.
	<Prolog system> PDC Prolog
	<Other languages> None.
	<Lines> 10000
	<Predicates> 400
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> March 1991.
	<Information Date> 9/1/92
	<References>  Catalogs.
</Program>

<Program> Rapidforms Forms Template - Tax Template
	<Number> 402
	<Domain> Form Processing.
	<Purpose> To process W-2 and 1099 Tax Forms.
	<Description> The Rapidforms Tax Forms Template program processes 
various pre-printed tax forms. Features include: automatic calculation of 
W-3 and 1096 forms; multiple companies and employee files; company default 
information entered only once; forms can be printed one at a time or by 
company; company and employee information is saved for future use.
	<Developed by> North Winds.
	<Organization>  North Winds & Rapidforms, Inc.
	<Country> USA.
	<Prolog system> PDC Prolog
	<Other languages> None.
	<Lines> 10000
	<Predicates> 400
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> September 1991.
	<Information Date> 9/1/92
	<References> Catalogs.
</Program>

<Program> $RD
	<Number> 105
	<Domain> Software Develpment
	<Purpose> The Prolog Rapid Development Toolkit
	<Description> Submitted to PAC. Paper 273
	<Developed by> Harry Brown
	<Country> UK
	<Prolog system> LPA
	<Information Date> 30/10/91
</Program>

<Program> Real Time ABLE compiler
	<Number> 50
	<Domain> Multi-agent systems domains 
	<Purpose> To compile RTA programs
	<Description> ABLE (Agent Behaviour LanguagE) is a parallel production 
rule language for developing multi-agent systems by explicitly representing 
the behaviour of the agents and  of the world in which they are situated.  
Real Time ABLE (RTA) is a high-performance subset of ABLE which executes at 
up to 20;000 events per second and is targeted; inter alia; to the Philips 
CD-I player.  The RTA compiler contains an ABLE interpreter which  expands 
the RTA source program into a set of variable-  free rules which are then 
compiled into a representation  of an asynchronous digital logic circuit.  
This circuit  is emulated in the application by a separate C-based  module 
called the RTA kernel.
	<Developed by> Peter Wavish
	<Organization> Philips Research Laboratories
	<Country> UK
	<Prolog system> Quintus
	<Other languages> none
	<Lines> 4000
	<Predicates> 200
	<Platform> Sun
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> Jan 91
	<No. of Users> 5
	<Information Date> 8-3-92
	<References> P.R. Wavish & D.M. Connah; `Representing Multi-
agent Worlds in ABLE'; Philips Research Laboratories Technical  Note 
No. 2964; October 1990.
P.R. Wavish; `Real Time ABLE'; Philips Research Laboratories Redhill 
Review of 1990; 1991.
Michael Graham and Peter Wavish; `Simulating and Implementing Agents and 
Multiple Agent Systems';  Proceedings of the 1991 European Simulation 
Multi-Conference; Copenhagen; June 1991.
</Program>

<Program> REDO Window Manager
	<Number> 152
	<Domain> TOOL
	<Purpose> An Intelligent Software Maintenance Management System
	<Description> Submitted to PAC. Paper 223
	<Developed by> P. Breuer & Papapanagiotakis
	<Country> UK
	<Prolog system> "Quintus, Turbo"
</Program>

<Program> REKPERT PLUS (Rectification Knowledge based Expert System.
	<Number> 305
	<Domain> Chemical Engineering.
	<Purpose> REKPERT PLUS consists of several distributed cooperating expert 
systems to seperate multicomponant mixtures.
	<Description> For the implementation of an expert system for 
multicomponent rectification the technology of cooperating 
distributed problem solving is an adequate concept.
From an abstract point of view we differentiate between a manager 
expert system, several expert systems for special domains and service 
systems.  These systems (agents) communicate directly or via 
blackboard.
If a multicomponent mixture is to be seperated in pure components 
rectification could be an adequate operation.  The cooperating expert 
systems of REKPERT PLUS can be consulted to give an advice to 
separate the mixture economically.  The manager expert system of 
REKPERT PLUS decides which system is suitable to solve current 
subtasks.  The following special expert systems can be alternatively 
selected.
- REKPERT for usual columns and sidestreams.
_ TEAGPERT for special cases of the rectification like azeotropic or 
extractive distillation.
- SOLPERT choses suitable solvents.
- HEATPERT proposes heat integration possibilities.
-service systems with calculation methods and their applicability.
Alternative sequences, proposed by the system are ranked by the 
global knowledge of the manager system.
On the other hand in chemical engineering not only complex tasks arise, 
there are also isolated problems, e.g. if a company only needs a new 
solvent for their liquid withdrawals in a process.  So every expert system 
has to be able to operate in a process as a stand alone system and to 
cooperate with the other ones.
	<Developed by> A team of scientists
	<Organization> Department of Chemical Engineering at the University of 
Dortmund.
	<Country> Germany.
	<Prolog system> IF/PROLOG
	<Other languages> ORACLE (SQL), C, FORTRAN
X-Windows
	<Lines> 5MB
	<Platform> 386(Ix), Mac (aux), Apollo (Domain OS), LAN
	<Date deployed> 1984
	<No. of Users> several industrial requests
	<Information Date> 29. 04. 92
	<References> Erdmann, H.H.; Engelmann, H.D.; Burgard, W. und 
Simmrock, K.H.:  The Use of Expert Systems in Chemical Engineering, 
Chem.  Eng.  Process. ,  23  (1988  S. 125-133.
Engelmann, H. -D.; Erdmann, H.-H.; Funder, R.  ;  Simmrock, K. H.  :  
The Solving of Comples Process Synthesis Problems Using Distributed 
Expert Systems,  Comp.  chem.  Engng.  13  (1989)  Nr.  4/5,  S.459-
465.
Simmrock, D.H. ; Fried,  A. ; Funder, R. ;  Schuttenhelm, W.  :   
Cooperating Expert Systems in Process Synthesis, Computer Application 
in the Chemical Industry, Erlangen 23.-26.  April 1989, Dechema-
Monographie Vol.  116  S. 135-144, VCH-Verlagsgesellschaft  1989.
Engelmann, H.D. ,  Funder, R. , Simmrock, K. H.  :  Aspekte bei der 
Realisierung kooperierender Expertensysteme zur ProzeBgenerierung,  
Chem. -Ing. -Techn.  62  91990)  Nr. 5, S.395-397.
Bieker, T. ;  Funder, R.  Schuttenhelm,  W;  Simmrock, K.H. 
 :  Tools zur Entwicklung kooperierender Expertensystems fur 
computerunterstutzte ProzeBsynthese, Proceedings, IF/PROLOG User Day, 
7. 6. 1991.
</Program>

<Program> REPROBATE
	<Number> 323
	<Domain> Validation.
	<Purpose> Software valedation, re-streching and metrics for FORTRAN.
	<Description> REPROBATE (the Redo Esprit Project Reporter of Bugs and 
Typical Errors) is an infeltrated analysis, valedation and restreching tool 
set based around the concept of a control repository (PROLOG).  It passes 
FJN 77, 66, and much of Microsoft, vax 8 Sun FURTRANS into an intermediate 
representation, then ????
	<Developed by> C.P.W. Hornsby, M.J. Metcalfe.
	<Organization> Lloyd's Register, U.K.  (Esprit Redo Project).
	<Country> U. K.
	<Lines> 18,000
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 06/92
	<No. of Users> 2 
	<Information Date> 18/05/92
	<References> Hornsby C.P.W., "Software valedation for Marine Systems".  
Proceedings of the 2nd Intl conference on Machine Communications and 
Control, 21-23 November, 1990.
</Program>

<Program> RIGEN
	<Number> 79
	<Domain> Financial Modelling
	<Purpose> Regenerate user inputs to a Financial Model
	<Description> Rigen is a simple utility added to the existing 
MOSACRI service which the central Association provides to 
participating savings banks. MOSACRI contains a complete high-level 
Multiplan model of all of a bank's operations.
Each bank models its particular strategy to achieve its acceptable profit 
level. ACRI collates and consolidates the results (which arrive as ASCII 
files) on a mainframe system and offers comment and criticism to each bank, 
seen in the global picture. Rigen will now allow ACRI, centrally and 
without changing existing procedures, to regenerate the original Multiplan 
input worksheets and thus simulate alternative strategies to support their 
written comments. Rigen provides a friendly and secure front-end which 
ensures that the Multiplan worksheets it runs will not go wrong and produce 
unexpected, unintelligible, and out-of-context messages.
	<Developed by> Brian Read
	<Country> Italy
	<Prolog system> PDC Prolog 3.2 with Toolbox
	<Other languages> Microsoft Multiplan 3.0, approx. 50%
	<Lines> 550
	<Predicates> 40
	<Platform> IBM PS/2 DOS 3.3 development and delivery
	<Type of Use> Free/Public Domain
	<Current Status> Dead
	<Date deployed> 2/1992
	<No. of Users> 3
	<Information Date> 2/1992
</Program>

<Program> ROADS 
	<Number> 51
	<Domain> Communication networks; private exchanges
	<Purpose> The term ROADS stands for Rule Oriented Application 
Design System. ROADS is a tool for rapid prototyping and simulation 
of applications based upon finite state machine principles.
	<Description> ROADS supports the following main features:
- Implementation of finite state machines by using SDL-like design 
principles combined with the advantages of applying a rule-oriented 
approach.
- Prototyping of user interfaces by using the available BIM-Prolog 
functions to support the XView Windowing System and the possibility 
to connect any type of device; such as PCs using MS-Windows or 
workstations using X-Windows
  or telephonic devices. In the latter case the implementation of the 
prototyped user interface can be reused within the final product 
without any constraint.
- A window-oriented user interface for interacting with the tool in 
order to develop an application.
- Support for generating documentation.
For ROADS the behavior of an application is modelled by a set of 
Finite States Machines (FSM) similar to the principles of the well 
known Specification and Description Language (SDL).
	<Developed by> R.Schloeglhofer; M.Puchhammer
	<Organization> Alcatel Austria AG
	<Country> Austria
	<Prolog system> Prolog by BIM; release 3.1
	<Other languages> Some interface routines have been written in C
	<Lines> 4000
	<Predicates> 200
	<Platform> SUN4; XView
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> Nov 90
	<No. of Users> Application development engineers within Alcatel Austria 
AG
	<Information Date> June 90
</Program>

<Program> RoadWeather Pro
	<Number> 346
	<Domain> Decision support system estimating weather-related effects on 
highway maintenance operations, as well as on airports, transportation, 
recreational activities, agribusiness etc.
	<Purpose> User-interactive 24-hour weather prediction system for snow and 
ice control on highways.
	<Description> RoadWeather Pro consists of three parts:  1)  a numerical 
weather prediction system which forecasts from specified initial 
(radiosonde and surface observations) and boundary conditions (detailed 
terrain effects), using a system of hydrodynamic and thermodynamic 
differential equations.  This part of the system is written in Fortran; 2)  
A graphical user interface (GUI) written in Prolog under Windows 3.1 which 
is fully interactive and displays predicted weather phenomena as symbolic 
objects; 3)  an Expert Weather Advisor, written in Prolog and Windows 3.1, 
which permits mouse point-and-click manipulation of weather "objects", 
thereby allowing forecast upgrades based upon recent observational data 
received from sensors or human observers.
	<Developed by> Elmar R. Reiter and Luiz Teixeira
	<Organization> WELS Research Corporation.
	<Country> USA
	<Prolog system> PDC Prolog 3.30 for Windows 3.1.
	<Other languages> WATCOM FORTRAN used for "front-end" numerical prediction 
model, ca. 25,000 lines of code.
	<Lines> ca.  10,000 lines of Prolog code.
	<Predicates> ca.  1000
	<Platform> Minimum:  386/25 MHz with math coprocessor, 6 Mbyte RAM, 10 
Mbyte hard disk space available; Recommended:  486/50 MHz DX, 8 Mbyte RAM, 
10 Mbyte available hard disk space.
	<Date deployed> 05/92
	<Information Date> 07/92
ÇIF*** The text replacing this argument is too long 
***È<References> E. R. Reiter, 1991:  "Hybrid modeling in 
meteorological applications.  Part 1:  Concepts and approaches.  
"Meteorology and Atmospheric Physics, Vol. 46, pp 77-99.
E.R. Reiter, David K. Doyle and Luiz Teixeira, 1992:  Intelligent and 
localized weather prediction.  WELS Research Corporation, Final Report to 
Strategic Highway Research Program, National Research Council, 140 pp.
</Program>

<Program> RPROLOG - Re-evaluable Prolog .
	<Number> 487
	<Domain> Synthesis of Evolutionary Learning Algorithms.
	<Purpose> RPROLOG is a metainterpreter for the programming of the 
uncertain  possibilistic information feedback accompanying
evolutionary synthesis from genetic building blocks.
	<Description> The Re-evaluable PROLOG (RPROLOG) is  a  new  
approach  for the formalization of the possibilistic uncertain 
information feedback of the computer-aided synthesis. Using this 
metainterpreter,  the PROLOG s ability  for  the  determination  of  
the  all  possible solutions in principle remains, however instead of 
the systematic search, the successive variants tend to  satisfy  the  
previously
declared evaluating points of view. In the RPROLOG there are four 
kinds of predicates as follows: evaluated predicates,
evaluating predicates, re-evaluable predicates, and ordinary  
predicates.
The evaluated  clauses  may  have  multiple  solutions  that  are 
evaluated by one  or  more  aspects,  described  by  the  
relatedevaluating  clauses.  The re-evaluable  clauses   represent   
the significant part of the predicates playing role in the  proof  of 
the respective evaluated  clause.  The re-evaluable  clauses  are 
automatically supplemented by the various characteristics of  the 
value distribution functions of the solutions, deduced from them.
The connected ensemble of an evaluated predicate with the related 
evaluating  and re-evaluable  predicates  is  declared   by   the so-
called cybernetic modules. The evolutionary steps of  the  run are as 
follows:
-tHaving  selected  a  solution  for  an  evaluated  clause  the
  metainterpreter  evaluates  it  with  the  related   evaluating
  clauses.
-tNext  the  value  distribution  functions  of  the  prescribed
  re-evaluable clauses are modified by the new values.
-tFinally the re-valuable clauses  are  rearranged  according  to
  certain strategies.
	<Developed by> Bela Csukas, Rozalia Lakner, Laszlo Horvath.
	<Organization> University  of  Veszprem,  supported  by  the  Hungarian  
Basic Research Grant N OTKA-2550.
	<Country> Hungary.
	<Prolog system>  TURBOPROLOG 2.0, PDC PROLOG 3.31
	<Other languages> TURBOPASCAL and FORTRAN versions existed.
	<Lines> 2000
	<Predicates> 250
	<Platform>  PC/AT 386
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 12/1989
	<No. of Users> 20
	<Information Date> 12/1993
	<References> Csukas B., Lakner R.: Evolutionary Structures and 
Multicriteria Learning Languages -  the  RPROLOG  Concept,  Annales  Univ.  
Sci Budapest., Sect. Comp. 12 (1991) pp. 49-57
</Program>

<Program> RTSC - Real Time Supervisory Control by structure-based  PROLOG 
program.
	<Number> 489
	<Domain> Real Time Supervisory Control and Database Management.
	<Purpose> Supervisory control of  a  flexible  multiproduct  batch  
plant
including supervision  of  a  PLC  and  a  six-channel periphery interface, 
as well as receipt interpreter  software  balance  and database 
manipulation.
	<Description> In  the  structure-based  supervisory  control  and  
real-time
database management the continuous and  discrete  activities  are 
represented by a unified  structural  model.  In  this  structure 
there are 9 types of passive elements including  reserved  places for 
the various kinds  of  Prolog  terms  referring  to  keyboard inputs, 
periphery  inputs,  periphery  outputs,  internal  hidden dynamic  
information,  internal  monitored  dynamic  information, problems 
activating  various  tasks,  database  inputs,  database outputs, 
printer outputs and screen outputs. There are 2 types of the active 
elements describing  the  rules  and  the  dynamically
called program modules of the individual problem  solving  tasks.
This architecture of the control model supports  the  in-parallel
execution that can be realized either by a concurrent operational 
system (e.g. CDOS) or by the  processes  of  a  CSPROLOG  program 
running in a multiputer machine.
     As an  example  in  a  flexible  multiproduct  plant  powder
mixtures for welding electrodes are produced  according  to  cca. 100 
receipts from 1-20 components stored in 28 silos.  There  are three  
feeding  lines  (weigh  hoppers  and  rope  crabs),   four automatic 
weighers, two manual balances, conveyor,  mixer  and  a drum filling 
station. The local control is performed by  a  FESTO PLC  (FPC-405),  
and  the  communication  is   supported   by   a six-channel 
periphery interface, both of them  connected  to  the
SICOMP PC via the serial ports. The  PROLOG  supervisory  program 
decomposes  the  actual  tasks,  coordinates  and  controls   the feeding, 
weighing, mixing  and  filling  processes,  updates  the software balance, 
accomplishes logging and performs the  database handling  functions  
(including  task   generation,   interactive database   manipulations,   
statistical   data   processing   and archivation).
	<Developed by> Bela Csukas, Karoly Varga, Laszlo  Jambor,  Lajos  Nagy,  
Tibor Chovan, Gabor Wittinger.
	<Organization> University  of  Veszprem,  supported  by  the  Hungarian  
basic Research Grant N OTKA-2550.
	<Country> Hungary.
	<Prolog system> TURBOPROLOG 2.0; PDC PROLOG 3.31
	<Other languages> Certain tasks of low level periphery communication and 
database management can be advantageously implemented in C.
	<Lines> 5000
	<Predicates> 600
	<Platform> PC/AT 286 and 386 optionally under CDOS .
	<Type of Use> Sold Product
	<Current Status> 
	<Date deployed> 9/1991
	<No. of Users> 1
	<Information Date> 12/1993
	<References> In preparation.
</Program>

<Program> SALI
	<Number> 153
	<Domain> Power distribution
	<Purpose> a Substation ALarms Interpreter
	<Description> Submitted to PAC. Paper 364
	<Developed by> Vlado Milosevic
	<Country> Switzerland
	<Prolog system> AAIS Prolog
	<Platform> Macintosh II
	<Information Date> 6/11/91
</Program>

<Program> SAME
	<Number> 187
	<Domain> Systems design
	<Purpose> Structured Analysis Modelleling Environment
	<Description> SAME provides a set of tools implementing Structured 
Analysis methodology, based primarily on data flow diagrams.This includes 
the ability to 'run' dataflow diagrams.
	<Developed by> Tom Docker
	<Country> New Zealand
	<Prolog system> LPA MACProlog
	<Platform> Macintosh
	<References> T.W.G. Docker: SAME - A structured analysis tool and its 
implementation in Prolog. ICLP5, 1988, pp82-95
</Program>

<Program> SAMPLE: The Simplest AI Language Interpreter
	<Number> 80
	<Domain> Knowledge Representation
	<Purpose> Development and implementation of a new technology for 
creation of
expert systems in any field of symbol-numeric data processing by 
problem
analysem
analysts rather than professional programmers.
The technology main feature: the user composes a program from fielded
data examples (generalizing and building up some structure on them).
	<Description> The system contains menu service, editor, means for 
control over SAMPLE Interpreter options, tutorial and 20 demo 
programs. The SAMPLE language short description. Syntax is 
intentionally made trivial (only 4 key words). The only semantic 
concept: matching text abstractions (which may contain variables 
instead of abstracted fragments). Clauses and program performance are 
analogous to the Prolog ones. Link with Prolog is used for machine-
oriented objects and effects. The main feature of the language is the 
most possible similarity  of constructions for writing algorithms to 
the data being processed. SAMPLE may be considered as simplification 
of Prolog with special conventions and additional powerful 
possibilities for list matching, which are related to Refal, Snobol 
or Planner languages.
Uses: Linguistics: Check spelling (inflected languages as Finnish); 
Research Music: Generation and analysis of specific dodecaphonic series and 
tunes Mathematics: symbolic differentiation; logic formulas transformation; 
Education: demonstration of Logic Programming and other AI concepts in 
their simplest form. Aprobated with more than 50 program examples (with 2-
300 clauses SAMPLE source code sizes)
	<Developed by> A.H. Gleibman, A.G. Sokolsky
	<Organization> Institute of Theoretical Astronomy of the Russian Academy 
of Sciences
	<Country> Russia
	<Prolog system> Arity Prolog
	<Other languages> No
	<Lines> 2500
	<Predicates> 90
	<Platform> IBM PC/XT/AT and compatible
	<Current Status> Released
	<No. of Users> Universities, Colleges (education process);
CS and AI labs (expert systems development);
Enterprises (creation of expert systems for technologies)
Conservatory (musicologic analysis)
	<Information Date> February 1982
ÇIF*** The text replacing this argument is too long 
***È<References> 1. Gleibman A.H. SAMPLE: New Programming Technology 
And AI
Language (subtitle: Data processing examples abstraction immediately
becomes a useful program if text matching is in focus). To be 
appeared.
2. Gleibman A.H. and Kirsanov N.O. Using SAMPLE for Algebra and
Linguistics. To be appeared. (Both papers are available in TeX 
format.
Contact the Authors)
3. Gleibman A.H. Synthesis of Text Processing Programs by Examples:
the SAMPLE Language. Preprint of the Institute of Theoretical
Astronomy of the USSR Academy of Sciences No. 15, St. Petersburg, 
1991
(in Russian). English translation is available in the ITA.
4. SAMPLE. The Simplest AI Language Interpreter. User's manual. ITA
of the Russian Academy of Sciences, 1992
</Program>

<Program> SANOCOR
	<Number> 73
	<Domain> Preventive Medicine
	<Purpose> Evaluate personal risk of coronary disease
	<Description> Expert system. User enters personal data - program 
evaluates risk of coronary heart disease. Features a meta-system that can 
explain its advices and reasonings. Context sensitive help.
	<Developed by> Audun Toennesen
	<Organization> STATOIL - The Norwegian National Oil Company
	<Country> Norway
	<Prolog system> PDC ESTA
	<Other languages> void
	<Lines> void
	<Predicates> void
	<Platform> IBM PC/ DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> May 1987
	<No. of Users> 50
	<Information Date> March 1992
	<References> Internal company publications, seminars on 
Artificial
Intelligence.
</Program>

<Program> SAR
	<Number> 388
	<Domain> Electrical dispatching.
	<Purpose>  Restoration of an electrical power network after a general 
black-out.
	<Description> Expert system for restoration electrical distribution 
networks after a general black-out. Under development for the Spanish 
electrical companies ENHER, FECSA, HEC and REE.
	<Developed by> Antoni Trias and programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.3 for Windows, Delphia Prolog on RISC 
station.
	<Other languages>  Oracle, STORIA, XMOTIF, C.
	<Lines> unknown.
	<Predicates> unknown.
	<Platform> Developed PC 386 / MS DOS and RISC station
Delivery		  RISC station (SUN and IBM)
	<Type of Use> Sold Product
	<Current Status> Prototype
	<Information Date> 07/1992
</Program>

<Program> SAX
	<Number> 439
	<Domain> Natural Language Processing.
	<Purpose> A parser generator for DCGs.
	<Description> A parser generator for a Definite Clause Grammar 
which
implements a concurrent bottom-up Chart parser.  The system provides other 
facilities such as a simple English morphologica analyzer, a tree printer, 
and an interface from a Japanese morphological analyzer.
	<Developed by> Yuji Matsumoto.
	<Organization> Deptartment of Electrical Engineering, Kyoto University.
	<Country> Japan.
	<Prolog system> SICStus Prolog
	<Other languages> No
	<Lines> 1500
	<Predicates> 150
	<Platform> SUN SPARCstation.
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> October 1991
	<No. of Users> 5
	<Information Date> 22 April 1992
	<References> Yuji Matsumoto and Ryoichi Sugimura, A Parsing System 
based on Logic Programming, IJCAI-87, Aug. 1987.
</Program>

<Program> SCE
	<Number> 154
	<Domain> TOOL
	<Purpose> A Tool for defining Routing Information for the freephone 
Service in AXE.
	<Description> Submitted to PAC. Paper 121
	<Developed by> Hans Nilsson
	<Country> Sweden
	<Prolog system> Prolog 2
	<Platform> IBM PC family
	<Information Date> 30/10/91
</Program>

<Program> Scheduling and Planning
	<Number> 97
	<Domain> Car Manfacturing
	<Purpose> 
	<Description> Scheduling and Planning - Manufacturing Sector  A 
production line at a car manufacturing plant can accept a very large number 
of different model configurations. This flexibility means that the process 
of 'line balancing'. optimising the 'build' process of each car and 
supplying the line with skilled workers and materials is extremely complex. 
The basic rules of line balancing are already well known to skilled 
industrial engineers at the plant and had already been expressed by the 
customer's staff using Quintec-Prolog on a PC. Quintec Systems Ltd. was 
commissioned to provide and intuitive graphical system that allowed 
engineers to specify and manipulate a 'build precedence' network. A layered 
technique was employed so that the system could handle very detailed 
information without obscuring the network layout. The system was contructed 
using Quintec-Objects on a PC under Microsoft Windows and is currently 
undergoing trials at the plant. The manufacturer hopes to employ the system 
at their other plants in future.
	<Developed by> Quintec Systems Ltd
	<Country> UK
	<Prolog system> Quintec
	<Platform> PC with Windows
	<Type of Use> In-house
	<Current Status> Evaluation
</Program>

<Program> Schikorre
	<Number> 155
	<Domain> Power distribution
	<Purpose> Schikorre: A Knowledge Based System for the Correlation of 
Stratigraphic Sequences in Prolog
	<Description> Submitted to PAC. Paper 249
	<Developed by> Dr Lutz Plumer & Baumbach
	<Country> Germany
	<Prolog system> LPA Prolog
	<Platform> PC
	<Information Date> 31/10/91
</Program>

<Program> SCORE
	<Number> 22
	<Domain> Airline Fleet Management
	<Purpose> Decision support system for traffic planning and coordination 
in large airports.
	<Description> SCORE is an airport management system which provides 
planning for arrival and departures. In major airports with a large number 
of carrier movements; the task of efficient traffic coordination becomes a 
very complex when historical precedence; IATA regulations and local 
procedures have to be observed. SLOT automates this task.
	<Developed by> Dan Ole Johansen
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Other languages> none
	<Lines> >10.000  <100.000
	<Platform> DOS; OS/2
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1/12 1988
	<No. of Users> 15
	<Information Date> 01/92
</Program>

<Program> Scrivenerª - A Document Assembly Legal Expert System
	<Number> 89
	<Domain> Legal drafting
	<Purpose> an expert system based document assembly program
	<Description> To create a document for a user which is compatible 
with Word Perfect¨, Scrivener asks for necessary information to 
select and combine standardized sentences, paragraphs or other blocks 
of text, inserting the relevent names and other variable information 
into the selected text to create a customized document. It therefore 
reduces the time needed to prepare common forms of document and the 
possibility of errors in choosing among numerous standardized 
provisions. 
Scrivener offers the following features: ¥ The user is only asked necessary 
and relevant questions when assembling a document and can change any or all 
answers at any time and then see the document reassembled with the revised 
answers. ¥ Scrivener can display the model in outline form, with the 
conditions for including (or excluding) each clause, and provides pull-down 
menus and other development tools for quick and easy model editing.. ¥ 
Scrivener uses natural language names for variables and clauses with no 
artificial numbering or labelling and supports a variety of data types, 
including text, numbers, dates and multiple choice and a variety of data 
functions including arithmetic, string manipulation, date arithmetic and 
financial functions.  Dianoetic also offers a library of legal documents in 
model form, such as incorporation documents, leases, contracts, wills and 
trusts, for attorneys who wish to purchase and modify prewritten models.
	<Developed by> Dianoetic Development Company, Pennsylvania, U.S
	<Country> USA
	<Prolog system> Arity Prolog
	<Other languages> 30 lines of C
	<Lines> 7,500 lines of Prolog 
	<Platform> IBM PC 
	<Type of Use> Sold Product
	<Current Status> Released
</Program>

<Program> SD-Scicon Analyst System
	<Number> 91
	<Domain> Computer Aided Software Engineering
	<Purpose> provides a complete and automated support system for specifying 
project requirements
	<Description>  Analyst provides the software developer with a powerful 
tool for capturing a thorough specification of the project at different 
levels of abstraction the CORE methodology. It takes full advantage of the 
well-known user interface on the Apple Macintosh on which the system runs, 
and provides comprehensive graphics facilities using the full functionality 
of the Macintosh Toolbox. The high-level graphics programming features and 
window-style bit-mapped graphics combine to give Analyst the complete 'look 
and feel' of a standard Macintosh application. Other activities also 
related to the production of a requirements specification, such as 
documentation and diagrams, can be performed on the same Macintosh system. 
It is generally recognized that Prolog's strength is in the expression of 
logical relations and rules. SD therefore used Prolog for Analyst's rule 
checking subsystem. It is used to express the methodology rules and to 
store and retrieve application information. Pascal is used for handling 
user-interaction and interactive graphics in line with SD's normal policy. 
Development started in 1983. 
	<Developed by> SD-Scicon
	<Country> UK
	<Prolog system> LPA MacProlog
	<Other languages> Pascal and Assembler
	<Platform> Macintosh
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1983
	<Information Date> 1991
</Program>

<Program> SEAP (Systeme expert d'alerte precoce).
	<Number> 74
	<Domain> Famine Prediction
	<Purpose> Help the team in charge of the Malian "Early Warning 
System"
(SAP:Systeme d'Alerte Precoce), to predict and assess the
nutritional status of the Malian population.
	<Description> Early Warning Systems aims at preventing the 
occurence of famines in sahelian coutries, such as experienced in 
1973-1974 and 1984. With the technical support of AEDES, the malian 
EWS collect all through the year agricultural and economical 
indicators about harvest, prices and availability of crops on 
markets; cattle and so on, and also more behavioral indicators 
related to migrations, changing of food habits...about some 200 
administratives areas of the North Mali.
SEAP is an information system acting first as a database manager for the 
stockage and restitution of all these indicators and secondly as a kind of 
expert system for their interpretation. By the confrontation of these 
conjonctural indicators with structural data about each of the 
administrative sectors, and with production rules, it produces early in the 
year a prognostic about the future nutritional status of the population and 
at the end of the year an end-state diagnostic. The outputs of the system 
help the technical team in formulating recommendations about food 
distribution and in orienting the collect of data.
	<Developed by> Programming: Paul-Marie BOULANGER from ADRASS and 
Jean-Michel
GUILLAUME from AEDES.
Analysis:Francoise ANDRE et Jean-Michel GUILLAUME (AEDES),Paul-
Marie BOULANGER (ADRASS) + all the technical staff of the
"SAP-MALI" (Systeme d'Alerte Precoce).
	<Organization> A.E.D.E.S.: Association Europeenne pour le 
Developpement et la
Sante (European Association for Health and Development).
	<Country> Belgium and Mali (Africa).
	<Lines> 3580
	<Predicates> 330
	<Platform> 80286 with 1MB of RAM and 42 MB of HD.
MS-DOS 3.3
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 02-1990
	<No. of Users> 1
	<Information Date> 02/18/1992
	<References> F.ANDRE,P-M BOULANGER & J-M GUILLAUME:"Le systeme 
d'alerte precoce au Mali:l'approche expert", AEDES, Bruxelles, 1991.
</Program>

<Program> SECReTS
	<Number> 189
	<Domain> Financial advice
	<Purpose> supporting senior decision makers within bankers organisations
	<Description> An expert system which provides statistical and 
analytic information for credit extension or rationing, and the 
control of risks. It uses information supplied to the Italian central 
Risk Centre about loans and loads on clients.
Installed at The Banco Populare di Sondrio and all its branch offices.
	<Organization> Formula plu
	<Country> Italy
	<Prolog system> LPA MacProlog
	<Platform> Macintosh
	<Type of Use> 
	<Current Status> Released
</Program>

<Program> See Mail, Version 1.3
	<Number> 411
	<Domain> Communications and Networking, group work support tools.
	<Purpose> Comprehensive e-mail system for DOS based LAN, with MCI Mail 
for external e-mail.
	<Description> This is a comprehensive e-mail system including a 
powerful range of tools & utilities to manage a complete e-mail 
system, not just basic message passing. The internal mail system 
supports groups, password security & mail attachments. The MCI mail 
access includes both on-line
access & full unattended background mail transfers. Extended support 
is provided for mail storage, including filing cabinet with drawers & 
folders, mail compression, & archiving. The system also has a 
configurable built-in PIM database system for addressing & other 
uses. Mail management tools including key word searching across 
folders, bulk move, delete, & archive tools. A multi-font/page 
formatting print tool rounds out the package. Supports mouse, 43/50 
line modes,& setting of
UI defaults. Recognises MCI, Compuserve, GE mail, configurable to read old 
mail from other systems for seamless upgrade.
	<Developed by> David RR Webber.
	<Organization>  Gnosis Inc.
	<Country> USA.
	<Prolog system> PDC 3.21.
	<Other languages> Assembler.
	<Lines> 30,000
	<Predicates> 6,000
	<Platform> PC compatibles, PS/2.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> March 1991
	<No. of Users> 100+
	<Information Date> March 1992
	<References> None.
</Program>

<Program> SEED
	<Number> 156
	<Domain> Software Development
	<Purpose> Using Prolog to Develop Tools Supporting Formal Methods in 
Software Development
	<Description> Submitted to PAC. Paper 276
	<Developed by> Betty Cheng & Gannod
	<Country> USA
	<Information Date> 6/11/91
</Program>

<Program> SEILBAHN
	<Number> 347
	<Domain> Technical and financial calculation of airial tramways.
	<Purpose> void.
	<Description> The first part of the system calculates a complete 
ropeway system with all the technical data and parameters based on the 
specific country and construction laws (15 countries, 22 types and so far).  
The second part calculates all prices and weights for a complete system 
with all the details.  Using the companies catalogue the system leads the 
user through all possible options.  The output is a correct offer, based on 
the selected technical data.  The user can pick a language and can ask for 
different output language.
	<Developed by> H. Geschwindner.
	<Organization> Sisyphos/Doppelmayer.
	<Country> Germany/Austria.
	<Prolog system> Turbo Prolog/PDC Prolog 2.0/ PDC Prolog 3.30.
	<Other languages> None.
	<Lines> in Prolog:  With comments 13000 lines.
	<Predicates> void.
	<Platform> PC's 80286/386/486
MS-DOS 640 KB RAM
Diskspace 20 MB
EGA/VGA
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/89
	<No. of Users> void.
	<Information Date> 05/08/92
</Program>

<Program> SEL Compiler
	<Number> 157
	<Domain> PAR
	<Purpose>  Compiler For Subset Equational Language & Abstract Analyzers
	<Description> Submitted to PAC. Paper 347
	<Developed by> Giancarlo Succi
	<Country> Italy
	<Platform> RISC Sun 4 architecture
	<Information Date> 6/11/91
</Program>

<Program> SERPES
	<Number> 11
	<Domain> Water Engineering Expert Systems
	<Purpose> Perform Sewerage Rehabilitation Planning
	<Description> SERPES performs three of the major phases from the 
standard SewerageRehabilitation Planning Procedures.  This includes 
interfacing to andinterpreting the output of a large commercial Fortran 
simulation program;WASSP. 
	<Developed by> Charles Hornsby; Mike Metcalfe
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog; Quintec Prolog
	<Other languages>  Quintus ProWindows; Fortran
	<Lines> 40000
	<Predicates> 1500
	<Platform> Sun 4 / SunOS4
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1987
	<No. of Users> n/a
	<Information Date> Oct 1991
	<References>  Ahmad; Holmes-Higgin; Hornsby & Langdon; 1988; "Expert 
Systems for Planning andControlling Complex Physical Networks"; Knowledge 
Based Systems Journal 1 (3)pp 153-165.
</Program>

<Program> Sesam-Chimio Hypertext system
	<Number> 207
	<Domain> Medical information
	<Purpose> Browsing of patient medical records
	<Description> Sesam-Chimio is itself a prototype knowledge-based system 
to monitor the progress of patients undergoing chemotherapy. The hypertext 
system uses a library of hypertext functions BIM-ProbeText, which is 
integrated within BIM_Probe, an obect-oriented system implemented in 
Prolog.
	<Developed by> I. de Zegher, Philippe Jassem
	<Organization> BIM
	<Country> Belgium
	<Prolog system> Prolog by BIM, with BIM_Probe
	<Platform> SUN
	<Type of Use> In-house
	<Current Status> Evaluation
	<No. of Users> 3
	<References> I. de Zeger, P. Jassem. Coupling hypertext to an object-
oriented environment. Artificial Intelligence in Medecine 2, (1990) pp205-
216.
</Program>

<Program>  SEXSY (Studiefinancierings EXpert Systeem // Scholarship 
Expert System)
	<Number> 371
	<Domain> Expert system, legislation on scholarships.
	<Purpose>  SEXSY is a tool to compute the composition and amount of 
governmental scholarships.
	<Description>  SEXSY is a data driven rule-based expert system 
with hypertext facilities. It's an event driven system with a text-
mode windowing user inter-face. It supplies the decentralised 
"Steunpunten" (information burea) of the
Ministry of Education and Science with a tool to answer complex 
questions of students about the amount and composition (gift, loan) 
of their governmental
scholarship they may expect, given their personal situation. SEXSY 
consists of about 100 rules which apply on about 1500 facts. The user 
interfaces consists of multiple "forms" in resizeable and moveable 
windows. In these
forms the user may specify the situation of the student. SEXSY 
applies the scholarship legislation to these facts and computes the 
scholarship the student may expect. The system will be extended with 
a hyperbase containing the
legislation, further regulations, and jurisprudence on scholarships.
	<Developed by> Dick Schrauwen, Miguel Carrasquer.
	<Organization> Joint-venture of Niceware and Hogeschool Rotterdam en 
Omstreken.
	<Country> The Netherlands.
	<Prolog system> PDC Prolog 3.30.
	<Other languages> Borland C++ 3.0, PharLap DOS-Extender, PDC 
Hypertext 
Toolbox and regular PDC Prolog Toolbox.
	<Lines> 5000 (+/- 30% excluding toolboxes)
	<Predicates> 478 (excluding toolboxes and libraries)
	<Platform> MS-DOS, 286-extended DOS, OS/2 (textmode).
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> August 1992
	<Information Date> 12th of August 1992
	<References> 12th of August 1992.
</Program>

<Program> Simapro
	<Number> 391
	<Domain> Environmental analysis of products.
	<Purpose>  Calculating environmental impact of production/use/waste 
handling of products.
	<Description> Description Experimental program that reflects Life 
Cycle
Analysis method for products. The program delivers an overview of 
environmental impacts (emisions and use resources) caused by 
production, use and waste handling of products. Program has four 
parts: A database module, a product definition module, a calculation 
module and a presentation module. In the database module process 
information (use resources, use intermediate products,
emissions to air/water/soil and product of process) in stored and 
maintained. The calculation module takes as input a product design, 
retrieves additional information from the database, and delivers raw 
information with in the 
presentation module is displayed (tables, graphs).
	<Developed by> R. Heule/L. van Oers/J. Guinae/P. Mulder.
	<Organization> Centre for environmental Science of the Leiden 
University.
	<Country> Holland.
	<Prolog system> PDC Prolog 3.2
	<Other languages> None.
	<Lines> Approx. 3000
	<Predicates> Approx. 250
	<Platform> MS-DOS 5.0
	<Type of Use> Sold Product
	<Current Status> 
	<Date deployed> April 1991.
	<Information Date>  April 1991.
	<References> Simapro manual (1991).
</Program>

<Program> SIMPOS
	<Number> 254
	<Domain> Operating systems
	<Purpose> operating system for the personal sequential inference machine
	<Description> SIMPOS has been developed under the following 
principles:
  To provide a good programming environment for logic programming
  To provide a good man-machine interface through a multi-window 
system
  to provide the necessary computer network facilities
  to provide a unique language system based on a logic programming 
language, ESP
  to develop an object-oriented paradigm which would realize a 
simple, consistent and flexible system.
It consists of a programming system (PS) and operating system (OS). It 
includes a Kernel, Supervisor, I/O Media system, programming system and 
Manipulator.
	<Organization> ICOT
	<Country> Japan
	<Prolog system> ESP
	<Lines> 145,000
	<Predicates> 15,000
	<Platform> PSI
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 1985?
	<Information Date> 1986
	<References> K. Fuchi, K. Furukawa: The Role of Logic Programming in 
the Fifth Generation Computer Project. ICLP3, 1986. pp1-24.
</Program>

<Program> SimSched
	<Number> 413
	<Domain> Design and  control of Flexible Manufacturing Systems.
	<Purpose> Creating manufacturing schedule and analyzing  the work of 
Flexible Manufacturing Systems using simulation.
	<Description> This CS-PROLOG based simulation system creates 
manufacturing schedule for given manufacturing tasks. 
Each resource of the manufacturing cell  and each
workpiece is represented by individual, parallel,
communicating processes. Animation Package is  connected to system. 
	<Developed by> CIM Research Laboratory.
	<Organization> Computer and Automation Institute, Hungarian 
Academy 
of Sciences.
	<Country> Hungary.
	<Prolog system> CS-PROLOG (Communicating Sequential PROLOG,  developed by 
Multilogic Ltd., Hungary).
	<Other languages> Microsoft C and Graphic Software  Systems*   Computer 
Graphic Interface (GSS*CGI) for the connected  Animation Package.
	<Lines> 800
	<Predicates> 200
	<Platform> IBM PC AT, OS/2;  MicroVAX, VMS
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> July 1991 
	<Information Date> Marc 1992
	<References> Publication: George Kovacs, Istvan Mezgar and Sandor 
Kopacsi: Concurrent design of automated manufacturing systems using 
knowledge processing technology; IMS'91-Learning in IMS, Computer in 
Industry 17 (1991) pp.257-267, Elsevier.
</Program>

<Program> SINTESIS
	<Number> 158
	<Domain> Software Development
	<Purpose> Towards an Automated Software Production Environment
	<Description> Submitted to PAC. Paper 290
	<Developed by> I Ramos et al
	<Country> Spain
	<Prolog system> BIM Prolog
	<Platform> Sun 3/XX Workstations
	<Information Date> 31/10/91
</Program>

<Program> SIPS
	<Number> 159
	<Domain> CIM, QA, Education, Artificial Intelligence.
	<Purpose> Inductive Analysis:  A machine learning algorithm as an 
alternative statistical technique.
	<Description> The theoretical bases of this program are Quinlan's 
ID3-Induction Algorithn and Cendrowska's PRISM-algorithm.  ID3 
produces decision trees, PRISM produces modular rules.  Own research 
involved several aspects, such as dealing with real-typed variables, 
missing data, noisy data, contradictory data, finding a way to merge 
ID3 and PRISM, making the program output understandable for technical 
oriented persons and so on.
Development started as an exercise to learn how to write a program in 
Prolog.  Its further growth was driven by the R&D division of a steel 
cord multi-national , who collected huge amounts of data on the work 
floor as a first step towards CIM and -what is now called:  QA- and 
by R&D work at the depertment of Building Sciences at a University 
where one of the questions was how to distinguish experts from 
novices, and to find out what does expertise really mean, and how 
efficient the transfer of knowledge is in this domain.
In both domains mentioned above, the program yielded results of great 
practical importance.  Some relations found in the data set collected on 
the woek floor had such potential that when other statistical techniques 
could not confirm those relations, the company resorted to laboratory 
experiments that did indeed confirm the findings by SIPS.
	<Developed by> Guido Cosemans
	<Organization> V.I.T.O. (The Flemish Institute for Technology and 
Research)
	<Country> Belgium
	<Prolog system> Turbo Prolog (now PDC Prolog).
	<Other languages> Assembler (100 lines).
	<Lines> 18,000
	<Predicates> 1500
	<Platform> development:  386-PC, MS-DOS 3, PDC Prolog
delivery:  PC-AT, 640 KB RAM and hard disk
SUN (in PC-DOS envirinment)
	<Date deployed> 06/92
	<No. of Users> Void.
	<Information Date> 09/07/92
ÇIF*** The text replacing this argument is too long 
***È<References> 1)  Cosemans g., Dijkstra J., Hajek J., Post W. 
Schaefer W, Vavra F. and Melichar J., Methoden voor klasse-
identificatie empirisch vergeleken, Internal Report, University of 
Eindhoven, Dept. of Building Sciences, 1991.
2)  Cosemans g.  and Samyn J. Inductive Analysis of Data Sets, Proceedings 
of the SEAS Annual Meeting, Amsterdam, September 25-28.
</Program>

<Program>  SIRIUS
	<Number> 379
	<Domain> Train dispatching and control.
	<Purpose>  Decrease the operator resolution time of failures on trains on 
the track lines by means of dialoging expert system.
	<Description> Expert system for train failures treatment for the series 
UT 111 cars. Project developed for the Ferrocarrils de la Generalitat de 
Catalunya.
	<Developed by> Xavier Fustero and programmers team.
	<Organization> AIA
	<Country> Spain.
	<Prolog system> PDC Prolog 3.31
	<Other languages> void
	<Lines> 4790
	<Predicates> 354
	<Platform> Developed PC 386/ MS DOS
Delivery PC / MS DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1992
	<Information Date> 07/1992
</Program>

<Program> SKIL
	<Number> 160
	<Domain> CASE
	<Purpose> Synthesizing Knowledge in Intuitionistic Logic
	<Description> Submitted to PAC. Paper 387
	<Developed by> D Galmiche, Conraud, & Herman
	<Country> France
	<Prolog system> Quintus
	<Information Date> 6/11/91
</Program>

<Program> Skilaki
	<Number> 62
	<Domain> Expert systems
	<Purpose> Logic based expert system shell
	<Description> An expert system shell similar to `apes'. It 
provides
a query the user facilty, conditional answers a form of
constructive negation, and an execution trace. It uses a
compilation aproach rather than a meta interpreter.
	<Developed by> Marek Sergot and Yannis Cosmadopoulos
	<Organization> The developers
	<Country> UK
	<Prolog system> LPA MacProlog
	<Lines> 1200
	<Predicates> 150
	<Platform> Apple Mac
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> Month/year of inital practical use of program
	<No. of Users> 10
	<Information Date> 27/3/92
ÇIF*** The text replacing this argument is too long 
***È<References> Peter Hammond, APES: A User Manual",
    Report 82/9, Department of Computing, Imperial College", 1982
Yannis Cosmadopoulos and Richard W. Southwick, Using Meta-Level 
Information for Expert System Control: A Blending Transformer 
Approach,
In Nigel Shadbolt (Ed) "Research and Development in Expert Systems 
VI, Cambridge University Press, pp54-65,1989
Yannis Cosmadopoulos and Marek Sergot and Richard W. Southwick. Data-
driven Transformation of Meta-interpreters: A Sketch. In Michael M. 
Richter, Harold Boley (Eds) Proceedings of the International Workshop 
on Processing Declarative Knowledge. Springer Verlag, 1991
</Program>

<Program> SKOL EXPERT
	<Number> 410
	<Domain> Education.
	<Purpose> Introduce pupils to expert systems / logic / automated 
reasoning.
	<Description> SKOL ("school") EXPERT is a C&M-derived inference 
engine with elements of cumulative uncertainty evaluation, running in 
a semi-graphic shell featuring decision-tree display, rule-base 
verification/consistency
check/editing, file mgr, etc.  Two different rule-syntaces are available 
for compatibility with an existing Swedish Prolog, and the inference engine 
will optionally forward-chain.
	<Developed by> Per Gregers Bilse, Goeran Nydahl.
	<Organization> Prolog Development Center, Swedish Ministry of 
Education,
respectively.
	<Country> Denmark, Sweden.
	<Prolog system> PDC Prolog.
	<Other languages> 5% C, by volume.
	<Lines> 4500
	<Predicates> 400
	<Platform> DOS
	<Type of Use> 
	<Current Status> 
	<Date deployed> Autumn 1988
	<No. of Users> Not reported.
	<Information Date> 25th 1992
	<References> "SKOL EXPERT User's Guide" by Goeran Nydahl, 
Swedish
Ministry of Education.
</Program>

<Program> SMES
	<Number> 314
	<Domain> Monitoring.
	<Purpose> 
Monitoring a tunnel-drill-machine.
	<Description> The system monitors the drilling process in 
realtime.  In case of malfunction (pipe-clogs, influxes, .....) 
diagnosis is started.
The system has two parts:
Part 1.  measures up to 96 analog and 128 digital sensor-readings and 
visualises them on a PC screen.
Part 2.  analyses abnormal values, finds relations between 
malfunction and offers a fix-procedure.
The usefulness of this system was proven at building sites with difficult 
geological environments and with high water pressures.  In these cases a 
very precise and sensitive control of the drilling process is necessary.
	<Developed by> Dr. Oskar Bartenstein.
	<Organization> Interface Computer Japan Ltd.
	<Country> Japan.
	<Prolog system> IF/PROLOG
	<Other languages> GKS
	<Platform> HP 9300
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 08/87
	<No. of Users> 1.
	<Information Date> 14/02/92
</Program>

<Program> SMI
	<Number> 161
	<Domain> OODB
	<Purpose> A Meta-Integration System for a Heterogeneous Object-Oriented 
Database Environment - Implementation in Prolog
	<Description> Submitted to PAC. Paper 361
	<Developed by> Qutaishat, Fiddian, & Gray
	<Country> UK
	<Prolog system>  POPLOG
	<Platform> Sun
	<Information Date> 6/11/91
</Program>

<Program> SMS
	<Number> 271
	<Domain> Mathematics
	<Purpose> To simplify mathematical structures
	<Description> The objective the system is to simplify expressions of a 
very broad class of mathematical structures (including groups, rings, 
fields, Boolean lattices, etc). Expressions are traversed bottom up and 
represented by binary trees. In case of an associative chain of operators 
the program performs sorting according to the appropriate ordering aspects.
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> V. Kofalusi (1979)
On Simplification in mathematical structures.
SOFTECH D42, SZAMKI, 1979, pp.12-86.
</Program>

<Program> Socrate
	<Number> 498
	<Domain> Computer based training.
	<Purpose>  Is a CAI system for developing lessons by teachers
don*t know anything about computers.
	<Description> The program works for two purposes: Building 
lessons by the teacher and learn this lessons by a student.
Teacher:		  - Can write question in multiple choice, programming 
instruction, true/false type.
 -		 Have to insert the right answer to questions.
 -		 Have to insert a number positive & negative for the 
wrong / right answers.
 -		 Have many utilities like a electronic register.
Student:		  -		 Have to answer to question inserts by 
teachers.
 -		 The system give an evaluation (positive/negative) to each 
question & to the entire exercise
 -		 Furthermore give statistic & graphical supports 
(= histogramme) to describe the students behaviour.
	<Developed by> Dr. Luciano Pes .
	<Country> Italy.
	<Prolog system> PDC Prolog (Turbo Prolog 2.0 From Borland, under DOS with 
8086 platform using BGI & Microsoft mouse.)
	<Lines> 1.600
	<Predicates> 103
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> September *92
	<Information Date> 10/12/92
</Program>

<Program> Software development information system
	<Number> 284
	<Domain> Software development
	<Purpose> An information system for the "Answer" software development 
system
	<Description> The system is used as a submodule of the ANSWER 
software development environment for CDL programs. The main function 
of the system include:
- checking for the correct connections between CDL program modules
- tracing the effects of a modification of a CDL object
- searching a CDL module for a given task. 
	<Developed by> P. Ban, J. Kohegyi, Gy. Suhai, A. Veszepremi, L. Zsako
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> P. Ban, J. Kohegyi, Gy. Suhai, A. Veszepremi, L. 
Zsako (1979)
An information system for Answer
ELTE report.
SOFTTECH D20, SZAMKI, 1979.
</Program>

<Program> SOLITAIRE FOUNDATION PLANNING
	<Number> 291
	<Domain> 
	<Purpose> To automate the planning of Solitaire Foundation
	<Description> The problem solved by the system  is the selection of 
bodies, usable under columns, from a given set of system components. An 
arbitrary system is loaded (the load list) and a distance list can be given 
in the goal statement and a result consists of the identifiers of the 
foundation bodies to be used at the loading forces.
	<Developed by> D. Holnapy
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> D. Holnapy (1979)
Mathematical foundations of the automatised technical planning.
ETI report, 1979.
</Program>

<Program> Sonar performance
	<Number> 174
	<Domain> Sonar design
	<Purpose> Optimization of sonar performance
	<Description> This enables the designer to forecast the performance of 
a device and dimension it according to the specification. These operations 
are governed by a certain number of equations that take into account the 
propagation conditions of the relevant evironment, the state of the sea, 
the range, emission frequencies, etc.
	<Developed by> Thomson SIntra ASM, Toulon
	<Country> France
	<Prolog system> Prolog III
</Program>

<Program> SP-Motif
	<Number> 428
	<Domain>  Graphical User Interface.
	<Purpose> Prolog Interface to OSF/Motif.
	<Description> It provides an Object Oriented Interface to the 
OSF/Motif 
Toolkit and so, eases the task of implementing Graphical 
User Interface based on OSF/Motif.  Each Motif pseudo class is mapped 
onto a true SP-Prolog class where Motif functions are seen  as 
methods sent to instances of these classes, some Xt  functions are 
seen as predicates.  A SP-Motif programm can be interpreted so the 
development
time is reduced. The final programm can be compiled.
	<Developed by> Ph. Coq ,J. Dalle Rive.
	<Organization> Bull S.A.
	<Country> France.
	<Prolog system> SP-Prolog Compiler (Bull S.A)
(SP-Prolog Compiler has been extended to incorporate
smalltalk-like object oriented mechanisms)
	<Other languages> C
	<Lines> 6000
	<Predicates> 600
	<Platform> DPX5000,DPX2000,DPX/2 100,200,300 MIPS, AIX RS/6000
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 02/91
	<No. of Users> 15
	<Information Date> 03/92
</Program>

<Program> SPADE
	<Number> 492
	<Domain> High integrity software.
	<Purpose> To provide formal validation and verification of high-
integrity
software.
	<Description> The SPADE interactive Proof Checker was developed 
to provide formal validation and verification of high-integrity 
software. It can be applied to software written in a variety of 
programming languages: Pascal, 68000 assembler, a provable Ada subset 
("SPARK"). The Proof Checker polices proof
attempts, preventing the user from making incorrect deductions or 
manipulations, and it assists the user in searching for a proof by 
applying a large library of replacement and inference rules using 
pattern matching
techniques. Systematic validation by SPADE during a program 
implementation also significantly reduces the burden of subsequent 
testing and maintenance.
POPLOG was selected for development of the Proof Checker because of its 
conformity to Prolog standards, its rigour in handling integers of 
arbitrary length, and its availability on a number of platforms.
	<Developed by> Ian O'Neill
	<Organization> Program Validation Limited.
	<Country> U.K.
	<Prolog system>  POPLOG Prolog.
	<Other languages> No.
	<Lines> 17,000
	<Predicates> 700
	<Platform> VAX   VMS   SUN    MicroVAX
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 10/87 (practical use)
	<No. of Users> 20
	<Information Date>  19.1.93
ÇIF*** The text replacing this argument is too long 
***È<References> Logic Programming Tools and Techniques for 
Imperative Program Verification
Ian O"Neill Ph.D. Thesis, University of Southampton Department of 
Electronics and Computer Science, April 1987
A Proof Tool for Program Proving
Ian M. O'Neill
Prodeedings of BCS FACS Term Rewriting Workshop, Bristol, September 
1988.
The Formal Verification of Safety-Critical Assembly Code
I.M. O'Neill, D.L. Clutterbuck, P.F. Farrow, P.G. Summers and W.G. 
Dolman
Proceedings of SAFECOMP'88 Fulda, F.R.G., Pergamon Press Nov 1988.
SPADE - un outil pour la preuve de la correction des programmes
Ian M. O'Neill
in La Lettre de la Surete de Fonctionnement, no. 4 Nov 1988.
Tool Support for Software Proof
Ian O'Neill and Denton L. Clutterbuck
Proceedings of IEE Colloquium on the Application of Computer Aided 
Software Engineering Tools, Feb 1989.
Machine-Assisted Program Proof
Ian M. O'Neill
Proceedings of SafetyNet '89: Industrial Experience of Formal 
Methods, November 1989.
Validation and Verification of Low-Level Language Code
Ian M. O'Neill
Proceedings of Colloquium on High Integrity Systems, organised by BCS 
Specialist Group on Safety-Related Systems, Warwick, April 1990.
</Program>

<Program> Speech synthesis
	<Number> 235
	<Domain> Speech synthesis
	<Purpose> high quality speech synthesis system
	<Description> This uses a non-procedural declarative approach to 
knowledge representation in speech synthesis. The Unification Grammar 
formalism is adopted using directed acyclic graphs as the data structure 
for the representation of phonological words and phrases. The accoustic 
interpretation of such structures is determined by simple constraint-
satisfaction techniques rather than a laboured set of destructive 
transformation rules. This approach avoids many of the practical problems 
encountered in conventional text-to-speech systems, yielding high quality 
speech.
	<Organization> York University
	<Country> UK
	<Prolog system> Poplog
	<Platform> MicroVax 3400
</Program>

<Program> Spiral Menu
	<Number> 513
	<Domain> Sprial menu is a graphics user interface, hence a general 
purpose routine. Most efficient use will be in applications where 
items to select may be arranged in alphabetical or numerical order, 
or some other logical order like menus, directories, customer decks, 
geographic indexes, lexical
collections, catalogues, etc.
	<Purpose> Thanks to mouse selection, Sprial Menu allows random access to 
up to 53 items, in graphics mode.
	<Description> Spiral Menu becomes of use in applications 
requiring fast 
selection from a large number of items.
Up to 53 items may be displayed on one single screen.
Random, rather than sequential access is provided for, thanks to 
mouse point and click.
Discrimination and progress analysis between items is enhanced by a 
spiral background, and each box around one name receives a 
distinctive musical identifier.
When mouse arrow enters one box, said box starts blinking and own 
tune goes on, for easy visual and auditive tracking.
Potential enhancements include:
-- multiple selection, color changes, etc.
-- crunching an even larger number of entries, maybe a few hundreds, 
on one single screen.
-- organizing sequences on other compact patterns, maybe dynamic ones.
	<Developed by> Pierre Chirouze.
	<Organization> None.
	<Country> France.
	<Prolog system> PDC Prolog 3.31a
	<Other languages>  None.
	<Lines> 1000
	<Predicates> 18
	<Platform> 386, VGA monitor, DOS 4.01 and above.
	<Type of Use> Personal
	<Current Status> Evaluation
	<Date deployed> 4/92
	<No. of Users> 10
	<Information Date> 9 feb 1993
</Program>

<Program> SRI Business Analysis
	<Number> 199
	<Domain> Systems analysis
	<Purpose> 
	<Description> Prolog was used to build an interface to an 
existing C program which provided an analytic hierarchy model. An 
optimization program was then developed in Prolog to allow the user 
to modify information in the ranking algorithm.
  The analysis begins with a reasoning process which deduces a 
recommended positioning for the new product. From that information a 
market function is build. The optimization program then runs with 
control of the rule-based deductive reasoning system. If at any time 
the systems analysis does not seem to fit with the reality of the 
scenario, changes can be made in the program definition.
  The system reduces typical project time from 6 months to two weeks.
	<Organization> SRI International
	<Country> USA
	<Prolog system> Quintus
	<Other languages> C
	<Platform> Unix
</Program>

<Program> SSPA
	<Number> 163
	<Domain> PLAN
	<Purpose> A KBS for Writing Safe Sequences of Operations on a High 
Voltage Electricity Network
	<Description> Submitted to PAC. Paper 201
	<Developed by> Brint et al
	<Country> UK
	<Prolog system> Quintec Prolog
	<Platform> "Sun 3, to PCs running windows"
	<Information Date> 31/10/91
</Program>

<Program> SSTREN
	<Number> 356
	<Domain> Computer Aided Training for mentally impaired pupils.
	<Purpose> Software/Hardware training package for severely mentally 
inpaired.
	<Description> Activation of the switchpad on the supplied special 
keyboard gives the pupil auditive and visual feedback on different 
subjects.  Several modules are planned.
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> PDC Prolog 3.3 + PDC Toolbox.
	<Other languages> None.
	<Lines> 600
	<Predicates> 20
	<Platform> MS-DOS.
	<Type of Use> Sold Product
	<Current Status> Evaluation
	<Date deployed> 08/92
	<Information Date> 12/08/92
</Program>

<Program> Staff
	<Number> 20
	<Domain> Staff Scheduling
	<Purpose> Creation and maintenance of work schedules for large companies 
and institutions.
	<Description> Staff is a system for decentralized creation and 
maintenance of work schedules for larger companies and institutions. The 
system works on a PC with optional communication with a central computer 
system for transfer of payroll data. Staff automatically generates and 
optimizes staff schedules based on skill requirements within departments; 
holiday plans and shift preferences. The Staff planning procedure uses a 
rule based expert system which specifies the rules for work schedule 
generation; based on governing laws; local agreements and planning 
heuristics.
	<Developed by> Jens Kjaerum; Lars Malmgren-Hansen
	<Organization> Prolog Development Center
	<Country> Denmark
	<Prolog system> PDC Prolog
	<Lines> >10.000   <100.000
	<Platform> IBM PC with DOS version 3.30 or OS/2 version 1.1 or later, with 
640K RAM (OS/2 version 4M RAM)
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 02/89
	<No. of Users> 200
	<Information Date> 01/92
</Program>

<Program> STEUFF
	<Number> 480
	<Domain> Nuclear technique still in use ! in preliminary use at a 
nuclear station.
	<Purpose> Producing in-data for a computer that surveys and controls the 
stearing mechanism in a KWU hot-water reactor.
	<Description> XPS with a nice graphical user interface, mouse, 
OOP, print
options of end forms.
	<Developed by> Siemens KWU0143, Joachim Lorek.
	<Organization>  Siemens KWU internally.
	<Country> Germany.
	<Prolog system> PDC Prolog 3.0 (formerly Turbo Prolog 2.0)
	<Other languages> C for a printer routine.
	<Lines> approx. 16.000
	<Predicates> approx. 1500
	<Platform> DOS 5, PC 286...486
	<Current Status> Prototype
	<Date deployed> 10/91
	<Information Date> 10/92
	<References>  Expert Systems and Computer Simulation in Energy 
Engineering, March 92, 8520 Erlangen
</Program>

<Program> STRAND88 Programming System
	<Number> 26
	<Domain> practical (symbolic) parallel processing
	<Purpose> a portable programmng system for parallel computers
	<Description> The STRAND88 system is extended to permit multiple users 
to share a running Strand system. The very fine granularity of a CLP 
system; coupled with the stream based nature of user interactions through 
an X window interface; makes practical very flexible information systems 
which offer users considerable scope for sharing data. A demonstration of 
concurrent user updates to a shared spreadsheet and a shared PERT chart 
editor with inter-operability between the tools proves the point. 
	<Developed by> RGT Barnes MA Gittins WB Pickles et al
	<Organization> The STRAND Group
	<Country> UK
	<Prolog system> STRAND88
	<Other languages> interface to X-windows
	<Lines> 10000
	<Predicates> ~800
	<Platform> networks of workstations
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> September 1991
	<No. of Users> n/a
	<Information Date> October 1991
</Program>

<Program> Studygram
	<Number> 164
	<Domain> NL
	<Purpose> Prolog for Natural Language Morphology
	<Description> Submitted to PAC. Paper 208
	<Developed by> Peter Greenfield et al
	<Country> UK
	<Information Date> 31/10/91
</Program>

<Program> SUMO_SAMA
	<Number> 477
	<Domain> Process Engineering for manufacture of circuit packs using 
Surface Mount technology.
	<Purpose> SUrface MOunt Slot And Machine Allocator.
	<Description> For a family of circuit boards specified by the 
user,
SUMO_SAMA allocates all board components to feeder slots
on either FUJI CP3 or FUJI IP2 SMT component placement
machines using either a time-balance heuristic or a 1st choice 
component preference heuristic selected by the user
For components allocated to the FUJI IP2 machine,
SUMO_SAMA also attempts to allocate feeder slots based
on component quantities to efficiently use the 4 placement
nozzles available.  Feeder slot dedications to components
can also be specified by the user to further constrain a solution.
	<Developed by> Russell G. Bulsis.
	<Organization>  AT&T - Denver Works, Denver, Colorado.
	<Country> U.S.A.
	<Prolog system>  Quintus Prolog Release 3.1.1
	<Other languages> Quintus ProWindows 1.2 - SunView Version.
	<Lines> 5000
	<Predicates> 500
	<Platform> Sun SPARCstation 1+ running SunOS 4.1.1
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 9/91
	<No. of Users> 6
	<Information Date> 11/25/92
</Program>

<Program> SXL
	<Number> 165
	<Domain> Software Development
	<Purpose> Modeling and Prototyping with Structured Logic
	<Description> Submitted to PAC. Paper 138
	<Developed by> Stanley Lee
	<Country> USA
	<Prolog system> "C Prolog, Quintus Prolog and Arity Prolog."
	<Platform> SXL 2.0 + 3.0 Unix systems and IBM compatible personal
	<Information Date> 6/11/91
</Program>

<Program> Syclops II
	<Number> 443
	<Domain> Software development.
	<Purpose> Developer's workbench for database applications.
	<Description> SyclopsII is a developer's workbench to build GUI 
interface and DB access methods.
	<Developed by> ISAC, Inc.
	<Organization> ISAC, Inc.
	<Country> Japan.
	<Prolog system> K-Prolog.
	<Other languages> C (90%).
	<Lines> 4000 lines.
	<Predicates> 600 predicates.
	<Platform> Sparc,HP-9000,SONY NEWS.
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> June 1, 1991.
	<No. of Users> unknown persons (at 60 sites).
	<Information Date> April 17, 1992
	<References> N/A
</Program>

<Program> SYLAE
	<Number> 231
	<Domain> expert systems interface
	<Purpose> natural language generation interface for expert systems etc.
	<Description> IBM uses a standard semantic representation in its expert 
systems interface. SYLAE translates this into a set of natural language 
sentences, using pronouns, varied word order, etc. This considerably 
increases the portability of other modules across language and domain 
barriers and is considerably more flexible than the normal 'frozen 
sentence' construction.
	<Organization> IBM Paris Scientific Center
	<Country> France
	<Prolog system> IBM Prolog
</Program>

<Program> SYLLOG
	<Number> 48
	<Domain> Deductive database; expert system;  natural language query 
	<Purpose> Strongly declarative deductive database system; with end-user 
level knowledge authoring and use. Direct execution of near-English 
specifications.
	<Description> Strongly declarative use of end-user specifications via 
an inference method called backchain iteration.  Includes automatic 
generation of SQL database queries.
	<Developed by> Adrian Walker; Daphne Tzoar; Peter Sheridan. Gustaf 
Neumann. Andrew Taylor.
	<Organization> IBM Research Division
	<Country> USA
	<Prolog system> IBM Prolog Program Product
	<Other languages> SQL; REXX
	<Lines> 10,000 lines of Prolog
	<Predicates> 4,000
	<Platform> VM; on an IBM 3090 computer
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> Jan 1987
	<No. of Users> 20
	<Information Date> March 1992
ÇIF*** The text replacing this argument is too long 
***È<References> Theory basis is "Towards a Theory of Declarative 
Knowledge" by Apt; Blair and Walker; in: Foundations of Deductive 
Databases and Logic Programming; J. Minker (Ed.); Morgan Kaufman 
1988.
 A. Walker; M. McCord; J. Sowa and W. Wilson. "Knowledge Systems and 
Prolog: Developing Expert; Database; and Natural Language Systems"; second 
edition; Addison-Wesley; 1990. SYLLOG user notes; available from Adrian 
Walker.
</Program>

<Program> Symbolic Differentiatio and Integrator (DS)
	<Number> 499
	<Domain> Instruction and practical use of Calculus.
	<Purpose> This program symbolically differentiates all common 
mathematical functions and solves most simple integrals ( including 
integration by parts, but not substitution ).
	<Description>  This program symbolically differentiates and 
integrates 
mathematical expressions. It is complete and comes with an easy-to-
use Borland-style user interface with pull-down menus and help 
facility.  Expressions are entered in normal mathematical notation 
(e.g. 5x ^ 3 + tanh (x ^ 2)), are		parsed and then the result is 
displayedin normal notation. The program performs integration and 
differentiation at first-year college level. It automatically uses 
logarithimic differentiation
and integration by parts, but   cannot do variable substitution. It 
has a powerful algebraic simplifier and also numerically evaluates 
expressions. User-defined functions and their		  derivatives and 
integrals can be defined, loaded and saved. Program sessions can be 
loaded and saved. The program 
comes with an on-line user manual in English and Swedish. A 45-page thesis 
describing the  inner-workings and algorithms inside the program is 
available only in Swedish.
	<Developed by> Martin Szummer.
	<Organization> Hj. Lundbohmsskolan, Kiruna (school).
	<Country> Sweden.
	<Prolog system> PDC Prolog v. 3.20
Turbo Prolog Toolbox v. 1
	<Other languages> None.
	<Lines> 3000
	<Predicates> 300
	<Platform>  DOS v. 3.3
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> May 1990
	<Information Date> January 1993
	<References> "Symbolisk derivering och integrering i Prolog - 
ett specialarbete
" Martin Szummer.
		  (Unpublished thesis, in Swedish.)
</Program>

<Program> System 9
	<Number> 166
	<Domain> Hospital information
	<Purpose> System 9: An Application Environment for Hospital Information 
Systems
	<Description> Submitted to PAC. Paper 119
	<Developed by> Dr. Bart Van den Bosch
	<Country> Belgium
	<Prolog system> Prolog by BIM.
	<Platform> "Macworkstation, Apple terminal Services"
	<Information Date> 6/11/91
</Program>

<Program> SysView
	<Number> 427
	<Domain> System Administration.
	<Purpose> Distributed System Administration.
	<Description>  - SysView is a set of applications which allows 
centralized
administration of a cluster of UNIX machines on a local
network. The architecture of Sysview is based on the ISO model for 
distributed systems management: An Agent runs on each machine, and 
implements the low level system dependent functionnalities, 
applications implements the user interface  (using OSF/Motif) and the 
high level functionnalities. These applications allows the management 
of  system resources like printers, devices, file systems,
users and so on. 
- SysView works with a Router (OEX) which manage the
information flow between applications and several agents
(commands, responses, events). Applications and Agents
communicate with CMIP protocol. OEX is an SP-Prolog application
- Managed resources are seen like objects, each object is
an instance of a class. Classes are described by ASN.1
templates. Templates are compiled in a schema. Applications
can access template information via OEX. The schema compiler is also a SP-
Prolog application.
	<Developed by> Ph. Donz's team.
	<Organization> Bull S.A.
	<Country> France.
	<Prolog system>  SP-Prolog Compiler (Bull S.A)
(SP-Prolog Compiler has been extended to incorporate
smalltalk-like object oriented mechanisms)
	<Other languages> C
	<Lines> 30000 lines
	<Predicates> 3000
	<Platform> DPX/2, AIX RS/6000
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 01/92
	<Information Date> 03/92
</Program>

<Program> Tangram
	<Number> 176
	<Domain> Teaching programming
	<Purpose> Teaching system for programming
	<Description> in PrologIA Applications notes (French)
	<Developed by> A. Lucci
	<Country> France
	<Prolog system> Prolog II, v.2.2 to 2.4
	<Platform> Macintosh Plus, SE or II
	<References> A. Lucci. Le systéme Tangram; un environment de 
production de logiciels d'enseignement de l'algorithmique et de la 
programmation. Conference dans le cadre des options E.A.O. des D.E.S.S., 
Grenoble, 7 Juin 1988.
</Program>

<Program> TAPPS
	<Number> 442
	<Domain> Production management system.
	<Purpose> Parts selection for another type of modular house.
	<Description> For each customer's housing plan,
1) enter the plan using GUI operation,
2) solve the constraints between the parts,
3) select the adequate parts to satisfy the plan.
	<Developed by> ISAC, Inc.
	<Organization> Sekisui Chemical Corporation.
	<Country> Japan.
	<Prolog system> K-Prolog
	<Other languages> As of HAPPS.
	<Lines> 50,000 lines (including parts database).
	<Predicates> 1,500 methods.
	<Platform> SONY NEWS; 4.2/4.3BSD (development / delivery).
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> February 21,1992.
	<No. of Users> 6 persons (at 2 sites, including 1 plants).
	<Information Date> April 15, 1992.
	<References> N/A.
</Program>

<Program> TARIFICA DATABASE
	<Number> 56
	<Domain> Telecommunications Research
	<Purpose> An Intelligent Costing system for International Leased-line 
circuits
	<Description> An Intelligent Costing system for International Leased-
line circuits running under Microsoft windows on an IBM PC-compatible.
	<Developed by> David Whitgift (of Logica plc) and David Wightman
	<Organization> Intelidata Ltd
	<Country> UK
	<Prolog system> Prolog-2 for Windows 3
	<Other languages> NONE
	<Platform> 80386 IBM-PC Compatible with 4Mb of RAM. Also Windows 3.0 and 
3.1
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 2/91
	<No. of Users> 15
	<Information Date> 30/1/92
</Program>

<Program> TEMA
	<Number> 357
	<Domain> Installation program.
	<Purpose> Installation of auxilaries to an existing program package.
	<Description> Utility to install some modules I sell to an already 
installed program on the customers machine.  It checks to see if it can 
find the new modules, and the old program, and asks the customer where they 
are if it can't.  It transfers the modules, and shows the README files.
	<Developed by> Sigve B_
	<Organization> SigveSaker
	<Country> Norway.
	<Prolog system> PDC Prolog 3.21
	<Other languages> None.
	<Lines> 220
	<Predicates> 8
	<Platform> MS-DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 03/91
	<Information Date> 12/08/92
</Program>

<Program> Term Refiner/Browser
	<Number> 13
	<Domain> Relational Databases
	<Purpose> Generic Relational Database Frontend
	<Description> The Term Refiner and Term Browser provide a high-level 
interface to theORACLE relational database system.  Built as part of the 
ESPRIT TranslatorsWorkbench (TWB) project; they are fully customisable to 
other databaseapplications.  Both hypertext-like navigation and intelligent 
datamodification are provided in a window and dialog based environment.
	<Developed by> Paul Holmes-Higgin
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog
	<Other languages> Quintus ProWindows;  Quintus ProDB
	<Lines> 4000
	<Predicates> 550
	<Platform> Sun 4 / SunOS4
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1991
	<No. of Users> n/a
	<Information Date> Oct 1991
	<References> Ahmad; Fulford; Griffin & Holmes-Higgin; 1990; "The 
Translator's Workbench";In (Ed.) C.Picken; 'Translating and the Computer 
II'; pp. 394-403. ASLIB; London.
</Program>

<Program> Termbase.
	<Number> 400
	<Domain> Natural Language Processing, Computer Aided Translation.
	<Purpose> To provide a multilingual terminology management system for 
translators.
	<Description> Termbase/DOS (and Termbase/Windows) is a 
networkable 
multilingual terminology management system for translators. It is intended 
to be used in conjunction with any word processor. Currently supporting 
German, English, French, Italian and Spanish, it incorporates morpho-
syntactic featuers of all these languages to provide a terminology database 
that is ideally suited for translators' needs. Used by itself, it is 
dictionary making  tool.
	<Developed by> Mr. V. Srinivasan.
	<Organization> None.
	<Country> Germany.
	<Prolog system> Initially Turbo Prolog then PDC Prolog 2...finally PDC 
Prolog 3.3
	<Other languages> A little LISP (approx. 3%)
	<Lines> Approx. 2000
	<Predicates> 312 (DOS) add a percentage for Windows.
	<Platform> IBM PC/AT or compatibles, DOS 2.11 or greater, Windows 3.0 & 
3.1
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed>  1.7.87
	<Information Date> 28.9.92
	<References> ASLIB Proceedings London 1989, Univ of Saabr_cken Report 
on CAT systems 1990 (?), Forthcoming PH.D. dissertation, Univ. of Mainz, 
Faculty of Applied Linguistics in 6728 Germersheim.
</Program>

<Program> TES (TErm rewriting System)
	<Number> 368
	<Domain> Term rewriting, completion.
	<Purpose> Programming environment for term rewriting systems  (term 
rewriting laboratory).
	<Description> TES is an experimental environment for computing 
with
term rewriting systems. It offers
 - term reduction
 - critical pair computation
 - checking of local confluence
 - proving termination of term rewriting systems
 - checking of sort decreasingness of term rewriting systems  (order-
sorted version)
 - completion of term rewriting systems.
	<Developed by> Roland Dietrich (original version),
Wolfgang Wagner (KA-Prolog version),						Christoph Brzoska 
(reduction ordering extensions) and  Andreas Werner (order-sorted version).
	<Organization> University of Karlsruhe, partly supported by the Deutsche 
Forschungsgemeinschaft (SFB 314, project S2).
	<Country> Germany.
	<Prolog system> C-Prolog (original version of the program),  KA-
Prolog, 
LOPSTER  (LOgic Programming with SubTypes and Extended
 Resolution, order-sorted version of the program).
	<Lines> KA-Prolog version: circa 5500
LOPSTER version: circa 6500
	<Predicates> KA-Prolog version: circa 360
LOPSTER version: circa 440
	<Platform> UNIX, SUN workstations.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> 1985 (original version), 
1989
(KA-Prolog version),
1991
(order-sorted version).
	<No. of Users> 3
	<Information Date>  August 12th, 1992.
	<References> R. Dietrich: Eine Programmierumgebung fuer 
Termersetzungssysteme, Arbeitspapiere der GMD, Nr. 130,
1985.
</Program>

<Program> Test and modelling of self reproducing biochemical processes
	<Number> 283
	<Domain> Biochemistry
	<Purpose> To test and model self reproducing biochemical processes
	<Description> The system permits the analysis and modelling of any 
biochemical self-reproducing cycle. With the replacement of the build-in 
database any cycle can be examined. The input of the system are the formal 
reaction equations; in the course of processing the nutriments, end 
products and attractors are selected.
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
</Program>

<Program> Test Tool
	<Number> 422
	<Domain> Natural language processing.
	<Purpose> A tool for testing Prolog programs.
	<Description> This test tool allows execution of a module, a 
combination of modules, or an entire program using predefined test 
batches. It is used to test and verify the (intermediate)
results of a natural language processing program.  The test tool also 
stores intermediate and final results/structures and produces reports 
comparing test results with old results/structures,  indicating where they 
differ.
	<Developed by> NLP Department, IBM Nordic Laboratories.
	<Organization> IBM.
	<Country> Sweden.
	<Prolog system> IBM Prolog
	<Other languages> None.
	<Lines> Around 2000 (comments not counted).
	<Predicates> Not counted.
	<Platform> VM/XA for IBM Mainframe.
	<Type of Use> In-house
	<Date deployed> February 1992
	<No. of Users> 10
	<Information Date> March 1992
</Program>

<Program> TESTING COMPUTER ARCHITECTURES
	<Number> 292
	<Domain> Computer architecture design
	<Purpose> To generate programs for testing computer architectures
	<Description> The system is an experimental tool for the purpose 
of generating  programs. In designing and evaluating a computer 
architecture supporting high level programming, good use can be made 
of small-sized runnable programs written in the language, that 
exhibit certain statistical features absorbed from the analysis of 
users' programming style. Example of such statistical features are 
the rate of occurrence  of instruction data types. The system's
input consists of the syntax rules of the high-level language and the 
statistical features that we want the generated program to have.
	<Developed by> V. Kiss, G. Simor
	<Organization> SzKI
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
	<References> V. Kiss, G. Simor(1978).
A preliminary specification of an architecture design environment and 
the analysis of the programming tools applicable in the environment.
SzKI report, 1978.
</Program>

<Program> TEX4
	<Number> 348
	<Domain> CASE/Document Research.
	<Purpose> Specification Research Tool.
	<Description> Tex4 is a Cross-reference/search engine that 
provides its user with a selected topical document search.  Results 
may be stored and reviewed as a hypertext chain.  Searches return a 
high semantic correlation without direct lexical (keyword) linkage.
Searches focused to return >90% of desired references in a document 
set will contain under 60% undesireable references.  Recursive 
application of proximity/weighting waveform analysis provides 
algorythm and focus/zoom is the interface paradym.
Used primarily in house for specification analysis, this tool is 
equally applicable to symbolic languages other than english.
Technology tracking and historical analysis are seen as areas of high 
potential applicability.
	<Developed by> M Cavanagh, D. Campbell.
	<Organization> NeoText SophtWear.
	<Country> Canada.
	<Prolog system> PDC Prolog.
	<Other languages> C
	<Lines> 2500
	<Predicates> 750
	<Platform> UNIX, MSDOS.
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 04/91
	<Information Date> 08/92
</Program>

<Program> The Configurer Graphics Service, CGS
	<Number> 415
	<Domain> Automatic Diagram Generation.
	<Purpose> To produce professional quality diagrams of ICL hardware
configurations for use by salesforce in proposals.
	<Description> The Configurer Graphics service is a value added 
service provided to the users of the ICL Configurer service - 
effectively the entire ICL international sales
force and a number of third party traders.  Configurer allows users 
to create complete and technically correct configurations of the 
entire range of ICL hardware. CGS uses the output from Configurer to 
produce a diagram of this
configuration which can then be printed at source and sent through 
the post to the requesting user, or sent as a POSTSCRIPT file to be 
printed locally.
The application is written in PROLOG (and a shell built on top) with 
calls being made to UNIX scripts. The diagram is built up using a 
declarative language called Maker Interchange Format, which is 
processed through Frame
Technology's FrameMaker package to produce POSTSCRIPT.
	<Developed by> Steve Sharman, Dick Higgs.
	<Organization> ICL Computers Limited.
	<Country> United Kingdom.
	<Prolog system> DECISIONPOWER PROLOG
	<Other languages> awk, C shell, Maker Interchange Format (Framemaker)
	<Lines> unknown, but not much below 8000.
	<Predicates> unknown, in the hundreds.
	<Platform> Sun 4, running SunOS 4.1.1
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 11/90
	<No. of Users> 14,000 !!
	<Information Date> 25/03/92
	<References> none.
</Program>

<Program> THE INSPECTION NETWORK
	<Number> 339
	<Domain> Remote access expert system, pay-per-use network.
	<Purpose> Inspection Analysis.
	<Description> General purpose inspection analysis/processing 
system.  User selects modules (summary reports, inspection 
categories) to create inspection format, system produces tailored 
checklist, user enters results, system links over phone with expert 
system, user prints finished report or sends over electronic delivery 
system to distant customer.
User charged for each access to expert system (except for resends).  
System manages access accounts - warns when balance getting low, 
denies access when balance <=0.  Host expert system is distributed 
parallel processing system with loosly linked multiple line 
processors and supervisor.  Each line processor supports multiple 
closely linked parallel processors for concurrent or sequential 
processing.  Each processor is a separate computer linked by serial 
or modem.  System can act as pay/per/use gateway to 3rd party 
software executed on demand on parallel processors with results 
passed through or merged with output of expert system.  Est. over 
90,000 inspections processed since 1987.
	<Developed by> Kort E. Patterson.
	<Organization> Overall Technology.
	<Country> USA.
	<Prolog system> PDC  Prolog 3.30
	<Other languages> C/10%, ASM/10%, Turbo Pascal/10% (aux programs).
	<Lines> unknown
	<Predicates> System composed of several programs with 1000+ preds 
each.
	<Platform> MS-DOS/DR-DOS
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 11/87
	<Information Date> 08/03/92
</Program>

<Program> The Interpreter and Explainer, (TIE), version 3.5
	<Number> 514
	<Domain> Language transliteration. Education. Bulletin Board 
communication.
	<Purpose> TIE is both a transliterator and instantdictionary tool. 
It can be used to convert words from one language into another or used as a 
transcriber of terms, especially for research work.
	<Description> The Interpreter and Explainer enables users to use 
specialized dictionaries of languages or subject matter ot obtain 
explanations of terms. This can be either single terms or two term 
combinations. Textual information can be read 
into TIE or actually typed on the spot.
By placing the cursor at the beginning of a term then pressing the 
return button, a term match will be bought from the appropriate 
dictionary. Successive tapping of the return key or holding it down 
will result in a quick scan of each word or term in a sentence or 
phrase.
Information that is found will be displayed in one of two windows, 
depending upon the length of the explanation. Users can quickly 
define their own terms if they so wish. This enables existing 
dictionaries to be appened to or 
entirely now ones to be created. Language based dictionaries are 
being made that compise some 10,000 common words each, e.g. Italian, 
Spanish and German.
Although the languae conversions are quasi-literal, the context and 
format of conversions makes "reinterpretation" suprisingly good. All 
conversions can be recorded to a log file name.
The program has application for specialized fields of knowledge where 
new terms are constantly being created or is very useful for 
concentrating on (say) biology, medical, law terminology etc. Even 
school children would find it useful in reading sessions.
An advanced version of the program is being planned.
There are many good dictionary programs around but virtually no 
transliteratios of practical worth.
	<Developed by> Mr. Craig Walter
.
	<Organization> Software Writers Independent Marketing or SWIMware.
	<Country> Australia.
	<Prolog system> PDC Prolog (formerly Turbo Prolog version 2.0)
	<Other languages> No other languages for this version.
	<Lines> Including the "include" files, approximately 1,500.
	<Predicates> 55 in main module, plus 40 or so in "include" 
files. 
Over 20 database declarations.
	<Platform> IBM and MSDOS compatibles, CGA< EGA, VGA. Hercules displays 
suitable. Not mouse dependt for this version.
	<Type of Use> Licensed
	<Current Status> 
	<Date deployed> 22/5/92
	<References> Australian PC World magazine, September issue n the 
shareware section.
</Program>

<Program> The Logic Programming Tutor.
	<Number> 407
	<Domain> Prolog Teaching.
	<Purpose> Front-end for teaching Prolog to novices.
	<Description> Provides a simple front-end for introducing Prolog to 
novices, allowing it to be entered in an English-like infix notation called 
"Logic". System includes a Basic-like clause editor, switches for 
translating clauses and queries between Logic and Prolog, and commands for 
showing how variables are quantified and for tracing and explaining answers 
to queries. The Tutor comes with a set of scripts divided into screen-sized 
sections: these set exercises and suggest experiments, and are displayed as 
the student proceeds.
	<Developed by> Jocelyn Paine.
	<Organization> Oxford University.
	<Country> England.
	<Prolog system> Poplog.
	<Other languages> Pop-11.
	<Lines> 10000.
	<Predicates> 1000.
	<Platform> VAX/VMS.
	<Type of Use> Free/Public Domain
	<Current Status> Released
	<Date deployed> 2/1985 - to date.
	<No. of Users> 1 (organisation).
	<Information Date> 3/1992.
	<References> "The Logic Programming Tutor", Jocelyn Paine, Intellect 
Books, 1992.
</Program>

<Program> THE MANAGER'S EDGE
	<Number> 341
	<Domain> Leadership; Participatory Management; Group Dynamics.
	<Purpose> Portrays trade-off between efficiency and participation for 
autocratic and group leadership styles known to be effective in a 
situation.
	<Description> Does your organization have difficulty making 
decisions?  This expert system will help.  The explanation portion of 
the expert system displays a decision tree to asist the user in 
determining why a question is being asked or a solution has been 
identified.  It is based on over 35 years worth of literature and 
investigation regarding the situational approach to decision making 
and participatory management.  The knowledge acquisition portion 
allows answers with a degree of certainty.  The report portion 
includes the ability to manage filed reports.  One result of a dialog 
with the expert system is a bar chart displaying effective leadership 
styles for the situation.
Hypertext database and context sensitive help convey the theory behind the 
system and detailed explanations regarding the dialog and menus.  The 
software is useful in determining when meetings are inefficient or 
ineffective.  It is also useful in determining when autocratic leadership 
is risky and conversely when group decision making is inappropriate.
	<Developed by> Mighetto and Associates.
	<Organization> Tribase.
	<Country> USA.
	<Prolog system> PDC Prolog 3.20
	<Other languages> None.
	<Lines> 5000 not including standard modules.
	<Predicates> 1622 total; 222 local;43 global;1347 public.
	<Platform> IBM PC, DOS 3.0 and above, 512K, CGA, VGA, EGA, Hercules Mono.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 04/91 as prototype.
	<Information Date> 07/92
	<References> National Science Foundation Proposed Project June 17 
1992, Title "Individual/Group Decision Analysis Software Acceptance".
</Program>

<Program> THE MAX AUDIOMETRIC TESTING SYSTEM
	<Number> 307
	<Domain> Medical diagnosis.
	<Purpose> Expert System for Audiometric screening testing for industrial, 
medical and school markets.
	<Description> LOGIC PROGRAMMING, SQL and machine learning (via 
Abductive Polynomical nets) have been combined to produce an expert system 
for automatic hearing testing and classification.  The system is in daily 
use by major corporations such as GM, FORD, GULFOIL.  Distribution to 
medical clinics is through STARKEY HEARING AIDS.  The program is fully 
compiled and requires the MAX AUDIOMETRIC hardware connected to the serial 
port.  There is a charge test of $5.00.
	<Developed by> David Lypka, Errol Davis.
	<Organization> Sound Linked Data Inc. and Future Logic Inc.
	<Country> Canada
	<Prolog system> 95% ARITY PROLOG (DOS)
	<Other languages> 1% Assembler (8086), ARITY SQL
1% MSC 5.1
1% GSS Graphics Library
1% AIM  Machine Learning (by ABTECH)
	<Lines> 100,000+
	<Predicates> 1000+
	<Platform> MS DOS, 286/386, companion Audiometer
4 megs RAM, 30 meg disk, VGA
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 07/89
	<No. of Users> 100
	<Information Date> 13/08/92
	<References> Intelligent Systems Review (I think), 1991 (Don't 
remember exact name or date).
</Program>

<Program> The T-Prolog Interpreter
	<Number> 276
	<Domain> Programming languages
	<Purpose> An interpreter for the language T-Prolog
	<Description> T-Prolog is a Prolog based simulation language with a 
capability  for explicit and implicit time handling. The interpreter of the 
language is capable of running an arbitrary number of PROLOG goals in 
parallel. The process executing the goals can communicate through logical 
variables, through the database and by means of "demon mechanism". In the 
case of deadlock further paths are explored through backtracking.
	<Developed by> I. Futo, J.Szeredi, K. Szenes 
	<Country> Hungary
	<Prolog system> MPROLOG
	<Information Date> 1984
ÇIF*** The text replacing this argument is too long 
***È<References> I. Futo, J.Szeredi, K. Szenes (1980)
A modelling tool based on mathematical logic - T-Prolog.
Acta Cybernetica, Tom. 5., Fasc. 3., 1981. pp. 68-74.
I. Futo, J. Szeredi (1982)
A discrete simulation system based on artificial intelligence 
methods.
In Discrete Simulation and Related Fields
A. Javor (Ed)
North Holland, Amsterdam, 1982. pp. 135-150.
I. Futo, J. Szeredi (1982)
T-Prolog: a very high level simulation system.
User's manual, SzKI Report, 1982.
</Program>

<Program> Theorist-S
	<Number> 251
	<Domain> Theory
	<Purpose> a theory formation system
	<Description> definite clauses are augmented with the power to express 
potentially inconsistent scientific theories, by partitioning them into two 
categories: ordinary assertions and integrity constraints. This provides an 
interesting reformulation of rule-based diagnosis systems such as MYCIN.
	<Developed by> Randy Goebel, David Poole, Koichu Furukawa
	<Organization> University of Waterloo, ICOT
	<Country> Canada, Japan
	<Prolog system> Waterloo ?
	<Platform> IBM VM/CMS ?
	<Type of Use> In-house
	<Current Status> Prototype
ÇIF*** The text replacing this argument is too long 
***È<References> Randy Goebel, David Poole, Koichu Furukawa: Using 
definite clauses and integrity constraints as the basis for a stheory 
formation approach to diagnostic reasoning. ICLP3, 1986. pp211-222.
D.L.Poole, R.G. Goebel, R. Aleliunas: Theorist: a logical reasoning system 
for defaults and diagnosis. in N.J.Cercone, G. McCalla (eds) Knowledge 
Representation. Springer-Verlag, 1986.
</Program>

<Program> THREAD
	<Number> 495
	<Domain> 
Any large project where there is a lot of documentation (such as letters, 
safety reports etc.) where there are many common issues traversing the 
documents.
	<Purpose>  Intelligent document tracking database.
	<Description> The program was written specifically for Safety 
Case 
documentation, where authorities need to track live issues and problems via 
letters. Sometimes a letter may contain references to several issues, and 
several parties may have responses to one issue. THREAD allows the section 
of the letters and documents pertaining to an issue to be isolated, and the 
user may go forwards or backwards along the thread to check the history of 
that issue. The user may also jump from one issue "thread" to another 
within the currently open document, and also to search the whole database 
for keywords.
	<Developed by> Steve Lympany.
	<Organization> Electrowatt Engineering.
	<Country> England.
	<Prolog system> PDC Prolog.
	<Other languages> None.
	<Lines> 1000
	<Predicates> 100
	<Platform> MS-DOS on IBM PC.
	<Type of Use> Personal
	<Current Status> Prototype
	<Date deployed> Not released.
</Program>

<Program> Timetable Scheduler.
	<Number> 493
	<Domain> Planning, Scheduling.
	<Purpose> Producing timetables for a Polytechnic.
	<Description> A Prolog-based assignment algorithm was written to 
plan for the teaching timetables of the Electronic & Computer 
Engineering Department in Ngee Ann Polytechnic, Singapore. The 
timetables were for 1744 students in the first two years of study, 
taught by 137 lecturers in 63 teaching
locations. The students were separated into 15 classes which are sub-
divided into 94 groups, each of about 20 or less persons. Each of the 
94 groups of students will have a timetable with about 30 hours of 
lessons per week. Each lecturer has about 20 hours of teaching 
assignments besides
certain other administrative duties.
		  The program has to satisfy numerous constraints and the most 
difficult is due to the heavy utilizations of all the physical facilities. 
The large number of students' curriculum hours per week also complicates 
the task especially towards the end of the scheduling process.
	<Developed by> Ngee Ann Polytechnic.
	<Organization> Ngee Ann Polytechnic.
	<Country> Singapore.
	<Prolog system> Poplog Prolog.
	<Other languages> Nil.
	<Lines> 1500
	<Predicates> 88
	<Platform> Sun SPARCStation.
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> May 1992. 
	<No. of Users> 1
	<Information Date> Dec 1992
ÇIF*** The text replacing this argument is too long 
***È<References>  Integrating Genetic Algorithms with a Prolog 
Assignment Program as a Hybrid Solution for a Polytechnic Timetable 
Problem - Ling Si Eng 1991.
		  Logic Programming for solving the timetable problem, Journal of 
Ngee Ann Polytechnic, Ling Si Eng, Oct. 1992.
</Program>

<Program> TMDS Comm (Tandem Maintenance and Diagnostic System, 
Communications).
	<Number> 358
	<Domain> Computers System Maintenance and Diagnostics for communications 
controllers.
	<Purpose> Diagnosis of communications-related equipment problems.
	<Description> The TMDS software product is designed to help Customer 
Engineers service hardware problems both at the customer site and remotely 
from a National Service Centre.  TMDS Comm. is the communications-related 
portion of this system.  For support of a new communcations controller a 
new architecture was developed in which an embedded prolog system is the 
heart of the diagnostic.  The prolog portion of the product stores 
information about devices and their characteristics.  Upon a command from 
the user, the prolog program generates a sequence of low-level commands to 
hardware-specific servers and other software components which have the 
intended effect.  Any number of devices may be the target of a command.  
Use of prolog is transparent to the user.
	<Developed by> Prolog:  Rahdall McRee/Heather McKelvey
TAL  (Tandem Application Language):  Ian Earnest/ Paul Komosinski.
	<Organization> Tandem Computers Inc.
	<Country> U.S.A.
	<Prolog system> TL, Test Language interpreter, by Randall McRee.
	<Other languages> TAL (60%)    (TAL is comparable to "C").
	<Lines> 7250  (including comments), ~2000 just prolog.
	<Predicates> 568  (including builtins).
	<Platform> Tandem Computers Guardian 90 systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/91
	<No. of Users> 50
	<Information Date> 03/92
	<References> TMDS Reference Manual, Tandem part number 061249.
</Program>

<Program> TML Genealogy System
	<Number> 374
	<Domain> Genealogy.
	<Purpose> A genealogy expert system.
	<Description> Data format conversion in and out of GEDCOM format, 
context sensitive date conversions, and projections of "educated 
guesses" for missing information in genealogical research.  Also does 
some text and graphical report
generation for formats not presently support by COTS products.
	<Developed by> T. Michael Louden.
	<Country> U.S.A.
	<Prolog system> 3.3 PDC Prolog WINDOWS.
	<Other languages> Borland C++
	<Lines> 3,935 lines prolog source and 1.084 lines of C++ source
	<Predicates> 409 (12 linked to C++)
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> First version 1978, First prolog based version = June 
1988.
	<No. of Users> 5
	<Information Date> 17 Aug 1992.
	<References>  none.
</Program>

<Program> TOERNOOI ASSISTENT
	<Number> 349
	<Domain> Sports (tennis).  Administration, ranking. planning and 
scheduling.  Tournament optimalisation.
	<Purpose> To automate everything around tennis tournaments.  Optimal 
planning and scheduling of matches.  Easy and fast replanning in case of 
rain (!).
	<Description> Stand-alone menu- and table-driven program to handle 
tennis tournaments.  Administering the players.  Seeding and drawing of 
lots for the arrangement of the players.  Fast planning and scheduling of 
the matches according to proprietary optimalisation algorithms.  Easy 
replanning in case of rain.  Processing of the match results, generation of 
press-releases.  Financial support.  Lots more. 
	<Developed by> J.G. de Lint, private.
	<Organization> RAET automatisering as the distribuution and support 
channel.
	<Country> Netherlands.
	<Prolog system> PDC Prolog 3.3.
	<Other languages> Some Turbo C (<5%).  Use of PDC Prolog Toolbox.
	<Lines> 15000
	<Predicates> 730
	<Platform> PC with 640k and hard disk, PCDOS-MSDOS-0S/2.  Developed under 
OS/2 and ported (cross-compiled) to DOS.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 03/91
	<Information Date> 09/07/91
	<References> Some press releases.  Program has 'official' status with 
Royal Dutch Tennis Union (KNLTB).
</Program>

<Program> TopNet
	<Number> 223
	<Domain> Network design
	<Purpose> an environment to support network design tasks
	<Description> Network design can be seen as the definition of 
topologies with particular link capacities and the assignment of 
flows on the basis of information about node locations, traffic 
generated and received, and the cost of data communications (leased 
line, modems, X.25 PSDN tariffs, etc.) The design matches reliability 
and delay/throughput constraints with the aim of minimizing the 
global cost. 
The product provides a user-friendly environment (windowing, graphics, 
mouse interaction etc.) which integrates the various tools needed for 
hierarchical network design.
	<Organization> BIM
	<Country> Belgium
	<Prolog system> Prolog by BIM
	<Type of Use> In-house
	<Current Status> Released
</Program>

<Program>  Total Network Management - Event Correlation Package.
	<Number> 472
	<Domain> Telecommunications - network monitoring and 
troubleshooting.
	<Purpose> User defined correlation of anomolous  events in 
telecommunication networks.
	<Description> Large telecommunication networks require constant 
monitoring of their switches and facilites to detect and correct any 
problems due to hardware and software failures.
Often a small number of contemporaneous events result in a 
larger problem which consequently generate an additonal number of 
alarms messages. The user has to sort through this dynamically 
changing log to separate the consequences (to try and reduce their 
impact on the customers) and the causes (which need to be fixed).
TNM (Total Network Management), and it's international version - 
MFOS, are an AT&T developed product sold throughout the U.S and the 
world to track and resolve these problems.
Our feature package is a user defined module with a "natural
language" interface that allows the correlation of multiple events  into 
logical groups reflecting a particular problem.
	<Developed by> Yossi Nygate, Warren Britt.
	<Organization> AT&T.
	<Country> U.S.A
	<Prolog system> American Interface Computer Inc.
	<Other languages>  C++ and C.
	<Lines>  2500 of Prolog code.
	<Predicates> 300.
	<Platform> Tandem (an FT platform running  UNIX V.3).
	<Type of Use> Sold Product
	<Current Status> 
	<Date deployed> 1/18/93.
	<No. of Users> Initially deployed at 5 sites in NYNEX and New 
England
Telephone Regional Bell Operating Companies. It will then be 
included in all future releases of TNM - at least an additional 20 
sites -
in the U.S. and Europe.
	<Information Date> Current.
</Program>

<Program> Towa Mold Press Diagnostic & Maintenance Expert System (TMES)
	<Number> 466
	<Domain> electronic, mechanical, pneumatic & hydraulic diagnostics.
	<Purpose> To diagnose failures and prescribe repairs for a 60-ton mold 
press used in molding plastic semiconductor packages.
	<Description> A diagnostic and maintenance expert system for an 
automated semiconductor mold press which provides consistent
diagnostic expertise, auditability and maintenance of the 
domain expertese by the experts, and significant cost reductions throughout 
the diagnostic and maintenance process.  The system receives 90% of its 
input directly from the mold press interface (10% from the operator) and 
reduces mean time of diagnosis from over 90 minutes to under 2 minutes.
	<Developed by> M. Alan Newman (Lester Ticey & Jim Hodges, domain 
experts)
	<Organization> Motorola.
	<Country> U.S.A.
	<Prolog system> Arity V.5 compiler.
	<Other languages>  Interfaced to existing network interface in C.
	<Lines> 2800 (plus about 400 domain rules).
	<Predicates> 350.
	<Platform> Desqview 2.0/DOS 3.1, developed on a 386, delivered on a 
286.
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> March, 1990.
	<No. of Users> 60
	<Information Date> November 3, 1992.
	<References>  "Custom Shells and Automated Factories",  M. Alan 
Newman, AI Expert, Sept. 1990.
</Program>

<Program> TransLexis dictionary conversion programs
	<Number> 36
	<Domain> Natural Language Processing
	<Purpose> Conversion of dictionary entries from and into the 
TransLexis lexical
database
	<Description> TransLexis conversion programs translate from LMT and 
Collins dictionaries into DB format and from DB format to LMT format
	<Developed by> R. Wirth
	<Organization> IKBS
	<Country> Germany
	<Prolog system> VM/Prolog
	<Other languages> REXX
	<Lines> 6000
	<Predicates> 1000
	<Platform> VM
	<Type of Use> In-house
	<Current Status> No longer supported
	<Date deployed> 12/84
	<Information Date> 1/92
</Program>

<Program> TSM - Twin Structure Model of controlled objects .
	<Number> 486
	<Domain> Analysis and Synthesis of Controlled Objects.
	<Purpose> Dynamic simulation of controlled process units or technological 
systems by  the optionally in parallel  executable   coupled structural 
models of conservation and information processes.
	<Description> The Twin Structure Model (TSM) is a coupled pair of  
structural models representing the conservation and  information  
processes, respectively. In the conservational structure the active 
elements correspond to the elementary physical and chemical changes, 
while the state of the system is modeled by  the  extensive  
quantities stored in the passive elements. In the information 
structure  the active  elements  describe  the  various  information  
processing rules, whilst the passive elements are places for the 
signs.  The
connection between the conservation  and  information  structures can 
be interpreted as transformations determining the information sources 
and the intervention of the control.
     The Twin Structure Models can  adequately  be  described  as 
communicating sequential processes. In the PROLOG implementation, the  
structural  elements  are  described  by  dynamic  facts  or clauses, 
while the various tasks can be executed by  the  
generalmetainterpreting clauses of the kernel. The communication 
between the conservation and information  models  can  be  organized  
via special messages.
From the analysis of the Twin Structure Models the so-called
cybernetic characteristics of the semiring-like structure can  be
studied. The design of the control algorithms can be interpreted as 
synthesizing feedforward routes and  feedback  loops  for  the achievement  
of  the  prescribed,  stable  functioning of the conservation system in 
question.
	<Developed by> Bela Csukas, Karoly Varga, Laszlo Jambor .
	<Organization> University  of  Veszprem,  supported  by  the  Hungarian  
Basic Research Grant N OTKA-2550.
	<Country> Hungary.
	<Prolog system> TURBOPROLOG 2.0, PDC PROLOG 3.31, CSPROLOG 3.25; (ALL-
MULTILOGIC Ltd., Budapest). 
	<Other languages>  Can also be implemented in object oriented languages.
	<Lines> 3500
	<Predicates>  400 
	<Platform> PC/AT 386; transputer and work station versions in preparation 
.
	<Type of Use> In-house
	<Current Status> Prototype
	<Date deployed> 1/1992
	<No. of Users> 5
	<Information Date> 12/1993
	<References> In preparation.
</Program>

<Program> TWAICE
	<Number> 4
	<Domain> knowledgebased system
	<Purpose> Open development environment for knowledge based applications.
	<Description> TWAICE is an open expert system shell for the 
development and integration of knowledge based applications.  As a 
shell; TWAICE allows the application the comfortable use of 
predefined and broadly applicable standard forms for the inference 
process; exlanation; user interface and integration into the DP 
environment.  Further TWAICE is OPEN: through defined interfaces 
TWAICE may be adapted to special needs.  E.g. one can implement 
special problem solution strategies.
TWAICE is a hybrid tool because it integrates different forms of 
knowledge representation: rules; objects; frames and logic (Prolog).
TWAICE has an expandable user interface; efficient knowledge engineering 
tools and a high degree of connectivity to other software systems.
	<Developed by> AI Centre;  AP333
	<Organization> Siemens Nixdorf Informationssysteme AG
	<Country> Germany
	<Prolog system> Prolog from Siemens Nixdorf and
IF/Prolog from InterFace Computer GmbH
	<Other languages> C; 15 %
	<Lines> 70000 (without comments); 93000 (with comments)
(Prolog)
	<Predicates> 4.750 (14000 clauses)
	<Platform> SINIX Systems; UNIX Systems;
MS/DOS System (with DOS extender)
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 1985
	<No. of Users> 200
	<Information Date> October 1991
</Program>

<Program> UBL
	<Number> 260
	<Domain> 
	<Purpose> A problem solver with learning capabilities
	<Description> UBL learns solving methods by analyzing solutions. 
Trying to understand these solutions then allows generalizations to 
be drawn. A problem is presented as a set of n-ary relations between 
objects, consdered as hypotheses, and by a specific relation 
representing the goal to prove. UBL uses knowledge given by the user 
or learned previously expressed in the form if "conjunction of 
relations" then "conjunction of relations". 
It uses sophisticated unification using an equational theory, a 
compromise between forward checking and backward chaining, an 
analysis of rules and a learning method.
Its main testing area has been in Euclidean geometry.
	<Organization> IBM France?
	<Prolog system> IBM Prolog
</Program>

<Program>  UNZ & Co. Export Software
	<Number> 399
	<Domain> 
	<Purpose> Process the required documents for completing U.S. and Canada 
export forms.
	<Description> Unz & Co Export Software allows you to automate 
preparation of export documents - a process that is usually 
repetitive and time consuming. It is a friendly, easy way to fill in 
and process export forms using your micro computer. Information is 
typed into an electronic image of any of the eight export forms. Its 
ability to remember information you enter and to automatically enter 
the information as appropriate in each other form in the
series is its major advantage. Features include, context sensitive, Help, 
data-base management, journal reports, math and conversation calculation, 
macros, ASCII conversion, batch printing and much more.
	<Developed by> North Winds and UNZ & Co.
	<Organization> North Winds.
	<Country> USA.
	<Prolog system> PDC Prolog.
	<Other languages> None.
	<Lines> 10000
	<Predicates> 400
	<Platform> MSDOS 2.0 512K memory, most dot matrix, daisywheel & laser 
printers, hard or floppy drive systems.
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> Summer 1989.
	<Information Date> September 14, 1992.
	<References> Product Literature.
</Program>

<Program> US Export Control Expert System
	<Number> 180
	<Domain> Shipping regulations
	<Purpose> To determine US license requirements for shipments
	<Description> This system was developed for Reuters to handle the 
increasingly onerous controls on the export and re-export of US 
technical items. It interfaces with two databases previously 
developed by the Corporate Export Control department containing lists 
of more than 10,000 items commonly used by Reuters and information 
about the export control status of countries in which Reuters 
operates.
The system requests information about proposed shipments and recommends the 
appropriate licensing, keeping a log of decisions for auditing purposes.
	<Organization> Quintec Systems
	<Country> UK
	<Other languages> Dbase III+
	<Type of Use> In-house
	<Current Status> Released
</Program>

<Program> Ushell
	<Number> 46
	<Domain> Programming environments
	<Purpose> To provide an environment for an introductory logic programming 
course for developing simple logic programs
	<Description> Ushell is a simple environment for developing and testing 
simple logic programs. The application is based on a Prolog meta-level 
interpreter with iterative deepening. This method overcomes some of the 
completeness problems of the Prolog interpreter by displaying solutions up 
to a preset limit in the search tree. The shell also provides user 
explanation of the deductions.
	<Developed by> L. Umit Yalcinalp
	<Organization> Case Western Reserve University
	<Country> USA
	<Prolog system> XPRO; Sicstus Prolog; Quintus Prolog
	<Lines> 550
	<Predicates> 60
	<Platform> Atari; Sun/Unix
	<Current Status> Prototype
	<Date deployed> 1988
	<No. of Users> 130
	<Information Date> 1992
	<References> L. Umit Yalcinalp; Ushell: An Environment for 
Introduction to Logic Programming Technical Report; Center of Automation 
and Intelligent Systems Research; Case Western Reserve University; 1992. 
</Program>

<Program> UTI
	<Number> 238
	<Domain> medical treatment
	<Purpose> expert system for treatment of Urinary Tract Infections
	<Description> UTI advises on antibiotic treatments of urinary 
tract infections. The advice is based on the evaluation of the 
patient's characeristics and microbiological data. The system suggest 
the most effective antibiotics with the minimal risk of toxicity and 
a relatively low price, according to bacteriological considerations 
and the patient's clinical status.
The system covers 21 pathogens, 43 antibiotics, 11 drugs and 13 
concomitant illnesses.
Clinical validation has been performed in in Péterfy S‡ndor Hospital and 
beta testing in Korvin Ott— Hospital and Szent-Gyšrgyi Albert Medical 
University .
	<Organization> IQSOFT and Medical division of Pétefy S‡ndor Hospital, 
Semmelweis Medical University
	<Country> Hungary
	<Prolog system> MProlog 2.3 and MProlog Dialog
	<Lines> 1250 clauses in 25 modules
	<Predicates> 450
	<Platform> PC 286 and 386 with MS-DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<References> Ch.13 Planning of antibiotic therapy. in T. Deutsh, E.R. 
Carlson, E. Ludwig: Computer assisted Clinical decision making (A control 
systems perspective), Plenum Press 1991 (2?)
</Program>

<Program> VALERE
	<Number> 350
	<Domain> Environment, Chemical analysis.
	<Purpose> Quality Assurance in an Environment Laboratory:  Water Quality.
	<Description> Implementation of the procedures described in 
ISO/DP 8466 including extensions towards non-linear calibration 
functions.  The kernel of the Draft International Standard ISO/DIS 
8466-1 (1988-05-5) and Draft Proposal ISO/DP 8466-4 (1989-04-01) are 
statistical techniques to be used for:  Calibration, Evaluation of 
Analytical Procedures, Estimation of Performance Characteristics, The 
Limit of Detection and the Limit of Determination of an Analytical 
Basis Method.
The program supports the daily work in the laboratory.
	<Developed by> Guideo Cosemans.
	<Organization> V.I.T.O.  (The Flemish Insitute for Technology and 
Research)
	<Country> Belgium.
	<Prolog system> PDC Prolog version 2.
	<Other languages> Assembler (100 lines).
	<Lines> 5000
	<Predicates> 370
	<Platform> Development:  386-PC + co-processor, MS-DOS 5, PDC Prolog 
delivery:  PC-XT with co-processor, 640 KB RAM and hard disk.
	<Type of Use> In-house
	<Date deployed> 04/90
	<Information Date> 09/07/92
</Program>

<Program> Vascular Surgery decision support
	<Number> 259
	<Domain> Medical decision support
	<Purpose> providing advice for clinical management decisions
	<Description> The system provides a simple and flexible way of 
reasoning about time in the context of illnesses. A detailed domain model 
for arterial disease is provided. This provides not only access to 
historical patient information but recommendations as to what actions 
should be carried out next at any time.
	<Developed by> Paul Soper, Geetha Abeysinghe, Charles Ranaboldo
	<Organization> University of Southampton, Royal South Hampshire Hospital
	<Type of Use> In-house
	<Current Status> Prototype
	<References> Paul Soper, Geetha Abeysinghe, Charles Ranaboldo:  
Knowledge based decision support for Vascular Surgery. Abstracts of poster 
sessions, Int. Conference on Practical Application of Prolog, London 1992.
</Program>

<Program> Vegetation Management Tools
	<Number> 5
	<Domain> Forest management
	<Purpose> To provide a framework within which forest managers can 
select
the best herbicide treatment for a particular set of weeds and crop
situation.
	<Description> The system is configured as a domain-specific shell and 
consists of two programs; one for information input; and one for use of 
information.  The input program allows experts to enter relevant 
information about weeds; herbicides; surfactants; application methods; and 
their interactions.  Input is both in structured form (stored in frames); 
and unstructured (stored as text in a database).  The other program asks 
managers to define their sites; and then allows them to explore options for 
management.  The program selects the most cost- effective treatments (given 
criteria nominated by the user); and provides warnings of environmental 
hazards.  Unstructured information can be accessed when relevant.
	<Developed by> Euan Mason; Don Geddes; Brian Richardson; Noel Davenhill
	<Organization> Forest plantation establishment
	<Country> New Zealand
	<Prolog system> PDC Prolog 3.2
	<Other languages> None
	<Lines> 7000 approx
	<Predicates> 300 approx
	<Platform> MS-DOS
	<Type of Use> In-house
	<Current Status> Released
	<Date deployed> 10/90
	<No. of Users> 20
	<Information Date> 24 October 1991
	<References> Mason; E.G.; D.J. Geddes; B. Richardson; & N.A. 
Davenhill;
1991; Application of Knowledge-based programming techniques to
cost-effective selection of herbicides in forestry
</Program>

<Program> VEX/ATL
	<Number> 67
	<Domain> Planning in distributed networks
	<Purpose> Production of routing tables for a distributed telecom network
	<Description> The task performed by VEX/ATL is to produce control 
data for the switches in a distributed telecom net. The control data 
produced is in the form of tables. The data tables control the 
priority path for routing a call through the network. The information 
used for  constructing the tables is the network's topology and the 
selection  algorithm used by the switches to chose an outgoing line 
from its table. Complexity arises due to the use of different switch 
types in the network which use different algorithms. Heuristics have 
been built up by a small group of experts working over many years. 
VEX/ATL has incorporated these heuristics into a system and has 
automated this previously manual task. A significant amount of code 
is used in implimenting a user-friendly interface for entering 
network information and presentation of produced data.
	<Developed by> Mark Plotnick, Infologics AB
	<Organization> FMV Sweden, (Swedish Defense Materials Authority)
	<Country> Sweden
	<Prolog system> Prolog-2 (8086 version) from Expert Systems Int'l
	<Other languages> C used minimally for external device access (plotter) & 
access to Operating System functions ( 200 lines)
	<Lines> 9000
	<Predicates> 1483
	<Platform> MS-DOS (Both development and delivery)
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> January 1991
	<No. of Users> 3
	<Information Date> November 1991
	<References> Expert Systems Support in Telecommunication, Proceedings 
of the IFIP TC 6 International Conference on Information Network and Data 
Dommunication, ed. D.Khahkar V. Iversen, Elseveir science publishers, 1988 
ISBN: 0 444 70503 1
</Program>

<Program> View Creation System
	<Number> 209
	<Domain> Database design
	<Purpose> Design of datbases used by large complex organizations
	<Description> The user interacts with the expert system by entering 
into a question answer dialog. Generally more than one user will be using 
the database and will enter their parameters. Once all needs are entered, 
the system sorts overlapping requirements and designs the database covering 
every requirement. Once completed the design is ready for individual 
assembly and use.
	<Developed by> Dr. Glen Cooper, Dr. Goldstein
	<Country> Canada
	<Prolog system> Arity Prolog
	<Platform> IBM PC
</Program>

<Program> VLSI Test pattern generator
	<Number> 247
	<Domain> Electronic design
	<Purpose> A hierarchical approach to fault injection as a means of 
diagnosing VLSI circuits
	<Description> The program allows the designer to specify, simulate and 
test circuits in a uniform manner without resorting to a different tool for 
each activity.
	<Developed by> Rajiv Gupta
	<Organization> SUNY at Stony Brook, Dept of Computer Science
	<Country> USA
	<References> Rajiv Gupta: Test-pattern generation for VLSI circuits in 
a Prolog environment. ICLP3, 1986. pp528-535.
</Program>

<Program> VOR (VHF omniderectional Range) Simulator
	<Number> 351
	<Domain> Real Time simulation.
	<Purpose> Simulation of multi-channel air navigation aid (ATC beacon).
	<Description> Program(s) provide four-channel simulation of VOR 
equipment interface to remote monitoring system.  Interface behaviour 
modeled and validated (formal closure shown).  Facilities for logging and 
analysis of all interface transactions provided.  Interactive control 
provided.  Simultaneous windowed display provides visual feedback on any of 
the simulated channels.
	<Developed by> Jai Hakim, Duncan Campbell.
	<Organization> Canadian Marconi Corporation.
	<Country> Canada.
	<Prolog system> PDC Prolog.
	<Other languages> C, Assembler.
	<Lines> 3000
	<Predicates> 700
	<Platform> DesqView 386
	<Current Status> Released
	<Date deployed> 04/90
	<Information Date> 08/92
</Program>

<Program> VORAN/ABS
	<Number> 241
	<Domain> Banking advice
	<Purpose> advisory system for small investors in bank
	<Description> 
	<Organization> Zentralsparkasse und Kommerzialbank, IQSOFT
	<Country> Hungary
	<Prolog system> MProlog
	<Platform> IBM MVS/TSo
	<Current Status> Prototype
	<References> Imre Kili‡n: An approach to integrate the spreadsheet 
paradigm and logic programming. Proc 2nd Conf. on AI, Budapest, Jan 1991.
</Program>

<Program> VSTA
	<Number> 167
	<Domain> Digital design
	<Purpose> Using Prolog as a Tool for Systolic Array Designs
	<Description> Submitted to PAC. Paper 419
	<Developed by> Fuyau Lin, Shih, Ling & Davis
	<Country> USA
	<Prolog system> SICSTUS
	<Platform> Sun 4
	<Information Date> 6/11/91
</Program>

<Program> WADNES
	<Number> 10
	<Domain> Water Distribution
	<Purpose> Assist Engineers in Water Distribution Emergencies
	<Description> WADNES provides support for a range of emergencies that 
can occur in awater distribution system.  It uses a model of the 
distribution networkto suggest various methods of resupplying water; 
informing appropriatebodies and managing remedial resources. WADNES creates 
input datafilesfor a large commercial Fortran simulation program; WATNET; 
runs it andinterprets its output files as part of its intelligent 
processing.  Developed using WIESSE.
	<Developed by> Paul Holmes-Higgin; Andrew Langdon
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog; Quintec Prolog
	<Other languages>  Quintus ProWindows; Fortran
	<Lines> 30000
	<Predicates> 1200
	<Platform> Sun 4 / SunOS4
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1987
	<No. of Users> n/a
	<Information Date> Oct 1991
	<References>  Ahmad; Holmes-Higgin; Hornsby & Langdon; 1988; "Expert 
Systems for Planning andControlling Complex Physical Networks"; Knowledge 
Based Systems Journal 1 (3)pp 153-165.
</Program>

<Program> WALLS
	<Number> 168
	<Domain> Engineering design
	<Purpose> A program to aid in the building of poured concrete structures
	<Description> Submitted to PAC. Paper 196
	<Developed by> Sharon Sickel Beach & Brian Beach
	<Country> USA
	<Prolog system> Turbo
	<Platform> PC 286
	<Information Date> 17/10/91
</Program>

<Program> Weeds In Sugar Cane
	<Number> 169
	<Domain> Agriculture
	<Purpose> An Example of Maintainable Program Design in Prolog : Weeds In 
Sugar Cane
	<Description> Submitted to PAC. Paper 198
	<Developed by> Coulston, Smith and Tilley
	<Country> Australia
	<Prolog system> LPA Prolog.
	<Platform> Any IBM compatible computer with graphics card.
	<Information Date> 6/11/91
</Program>

<Program> WFProlog (well-founded Prolog).
	<Number> 506
	<Domain> Default reasoning systems needing an underlying inference 
system (Prolog style interpreter) based on well-founded semantics.
	<Purpose> A Prolog meta-interpreter that computes well-founded semantics 
for normal logic programs using negation as failure.
	<Description> The interpreter is based upon a formal, top-down, tree 
specification for well-founded semantics (see reference).  The interpreter 
is, in fact, an executable specification of  the top-down tree semantics in 
the form of a Prolog program (written in Quintus Prolog).  The resulting 
procedure -- which we call SLD3NF -- computes 3-valued well-founded answers 
to queries.   The interpreter is quite general, programs and goals can have 
variables,  any built-in goals can be called, and programs can be 
conveniently loaded from source files in usual ways.
	<Developed by> John R. Fisher.
	<Organization> California State Polytechnic University.
	<Country> USA
	<Prolog system> Quintus Prolog.
	<Other languages> None (at present).
	<Lines> 75
	<Predicates> 14
	<Platform> Easily modifiable for any Prolog.
	<Type of Use> Free/Public Domain
	<Current Status> Evaluation
	<Date deployed> February 1992.
	<No. of Users> 6
	<Information Date> 9 February 1993.
	<References> Fisher, J.R., Tree semantics for logic programs 
with
negation as failure, Proceedings of the Third California
State University A.I. Symposium, June, 1992, pp. 158-66,
Computer Science Department TR#1993-03.
</Program>

<Program> WHALE
	<Number> 320
	<Domain> Logistic.
	<Purpose> Analyse warehouse placement.
	<Description> The Warehouse Assignment and Location Environment is a 
tool for analysing warehouse placement according to a specified customer 
base.  The objective of the tool is to minimize the overall costs, taking 
into accout the fixed and variable cost incurred in establishing and 
maintaining the warehouses.
	<Developed by> COSYTEC
	<Organization> CPL Software and Services.
	<Country> France.
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 1100
	<Platform> SUN/HP/IBM/DIGITAL workstations
PC  386/486		  UNIX   DOS
	<Type of Use> In-house
	<Current Status> Evaluation
	<Date deployed> 02/92
	<No. of Users> 2
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Int. Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> WIESSE
	<Number> 9
	<Domain> Water Engineering Expert Systems
	<Purpose> Development of Expert Systems for the Water Industry
	<Description> WIESSE provides an expert systems development 
environment customised to thespecial needs of the Water Industry.  
Developed as part of the WIESC AlveyProject; it has been used to 
develop two major expert systems: WADNES andSERPES.  WIESSE compiles 
its Knowledge Representation Language to Prolog.
	<Developed by> Paul Holmes-Higgin; Charles Hornsby
	<Organization> University of Surrey
	<Country> England
	<Prolog system> Quintus Prolog; Quintec Prolog
	<Other languages> Quintus ProWindows
	<Lines> 7542
	<Predicates> 1170
	
	<Platform> Sun 4 / SunOS4
	<Type of Use> Licensed
	<Current Status> Released
	<Date deployed> 1987
	<No. of Users> n/a
	<Information Date> Oct 1991
	<References> Hornsby; Holmes-Higgin & Ahmad; 1987; "The Water 
Industry Expert SystemSupport Environment".  In (Ed) D.S.Moralee; 
'Research and Developments inExpert Systems IV'; Cambridge University 
Press. 
</Program>

<Program> WSUFP
	<Number> 364
	<Domain> Language development, parallelism, and functional programming.
	<Purpose> To develop an FP interpreter in Prolog.
	<Description> An FP(Language developed by Backus, FP stands for 
functional programming) interpreter  developed in relation to a reduced 
message passing scheme.  Also for the study of  functional programming 
lnaguages.
	<Developed by> Chris Fickert.
	<Organization> WRight State University.
	<Country> U.S.A.
	<Prolog system> Quintos
	<Lines> 700
	<Predicates> 55
	<Platform> Ultrix(unix-based).
	<Date deployed> Fall 1988
	<No. of Users> unknown 
	<Information Date> July 1992
	<References> Sigplan Notices November 1992.
</Program>

<Program> XLSIOR
	<Number> 170
	<Domain> Insurance
	<Purpose> An Intelligent System for Reinsurance Recoveries Calculation
	<Description> Submitted to PAC. Paper 207
	<Developed by> Martin Wood, Corlett, & Hopson
	<Country> UK
	<Prolog system> SD-Prolog
	<Information Date> 6/11/91
</Program>

<Program> XPL (eXplanation for ProLog)
	<Number> 45
	<Domain> Expert System Shells
	<Purpose>  Explaining behaviour of an expert system written in Prolog
	<Description> XPL is an explanation shell built in Prolog by using 
meta-programming techniques. The heart of the shell is an integrated meta-
interpreter which handles both successes and failures in layers. The 
layered design allows handling of negation and failure; including cuts; and 
successes in a single interpreter. The shell provides why; how; whynot and 
bywhom type of explanations in one shell and allows dynamic user input to 
the knowledge base by query-the-user facilities . It also illustrates the 
interactions of cuts with the execution.
	<Developed by> L. Umit Yalcinalp
	<Organization> Case Western Reserve University
	<Country> USA
	<Prolog system> Sictus Prolog; Quintus Prolog
	<Other languages> C
	<Lines> 2000
	<Predicates> 240
	<Platform> Sun/UNIX
	<Type of Use> Free/Public Domain
	<Current Status> Prototype
	<Date deployed> Feb 88
	<No. of Users> 2;3
	<Information Date> Feb 92
ÇIF*** The text replacing this argument is too long 
***È<References> L. Sterling and L. U. Yalcinalp; Explaining Prolog-
Based Expert Systems Using a Layered Meta-Interpreter; in Proceedings 
of 11th International Joint Conference in Artificial Intelligence; 
Morgan--Kaufmann; pp 66-71;1989    
L. U. Yalcinalp and L. Sterling; An Integrated Interpreter for Explaining 
Prolog's Successes and Failures; in Meta Programming in Logic Programming; 
eds. Abramson and Rogers; pp. 191-203; 1989.
</Program>

<Program> XTEL
	<Number> 313
	<Domain> Techn. configuration.
	<Purpose> Configuration of Telephone Systems.
	<Description> Customers of the German Post System can state their 
wishes, concerning their own inhouse telephone system.  The system 
generates possible layouts and tests them against legal guidlines.
	<Developed by> Mr. Thuy, Mr. Schnupp.
	<Organization> Interface Computer GmbH.
	<Country> Germany.
	<Prolog system> IF/PROLOG
	<Other languages> Siemens Collage.
	<Platform> Siemans MX300
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 10/89
	<No. of Users> 150
	<Information Date> 14/02/92
	<References> Der Postexperte
KI, 1-89
Oldenbourg Verlag, Munich.
</Program>

<Program> XTRACT
	<Number> 451
	<Domain> Text analysis for information extraction.
	<Purpose> Extract and analyse costing information from capture files 
generated from communication dialogues with on-line databases.
	<Description> The application consists of two parts:
1. Information extraction from text files
2. Database reports on accumulated information .				  
Information extraction:
----------------------
Through the use of user definable text templates, the system will 
search a text file for matches. The templates consists of fixed text 
expressions and variable names. A template matches a block of text 
when all the fixed text components
can be matched in that block, the variables are then assigned to the 
text at their position eg.  Template:
		  Total cost at HR; MIN is COST dollars
Will match:
		  Total cost at 3:15 is 20.32 dollars
With variables bound as follows:
		  HR=3. MIN=15, COST=20.32
The user may also supply equations and look up tables to be used in 
calculating other variable values. Variables with reserved names may 
then be asserted into a database.		     
Database reports:
-----------------
The database may be inspected and reports generated in either text or 
comma deliminated file formats. These may be used for reporting, 
invoicing etc.  The system currently supports six international 
information hosts, another ten 
or so are to be added before product release.
	<Developed by> John Jones, Jim Smith, Jill Cousins.
	<Organization> Trade Ware Systems, First Contact.
	<Country> Australia, UK.
	<Prolog system> PDC Prolog version 3.3. Extensive use of the PDC Prolog 
Toolbox in the following areas: Parser generator, User interface tools, 
Screen definition / data entry tools.
	<Other languages> None.
	<Lines> 2500 Does not include the large amount of toolbox code used.
	<Predicates> 180 Does not include the large amount of toolbox code 
used.
	<Platform> IBM PC compatible, DOS version 3.3 or later.
	<Type of Use> Licensed
	<Current Status> Evaluation
	<Date deployed> October 1992
</Program>

<Program> XWIN
	<Number> 316
	<Domain> Graphical interface.
	<Purpose> UIMS tool for building graphical user interfaces.
	<Description> The XWIN editor is a graphical tool provided with the 
CHIP development system to build graphical user interfaces for CHIP 
applications using the XGIP graphical environment.  Unlike other graphical 
dialog generators, it not only allows users to define dialog fields with 
buttons, sliders etc, but also allows for a simple generation and set up of 
drawing areas used for diagrams, charts and more advanced graphical 
interfaces.
	<Developed by> COSYTEC.
	<Organization> CLP Software and Services.
	<Country> France.
	<Prolog system> CHIP
	<Other languages> C  X11
	<Lines> 1500
	<Platform> SUN/HP/IBM/DIGITAL workstations.
PC  386/486		    UNIX		DOS
	<Type of Use> Sold Product
	<Current Status> Released
	<Date deployed> 01/92
	<No. of Users> 25
	<Information Date> 03/92
	<References> M. Dincbas, P. Van Hentenryck, H. Simonis, A. 
Aggoun, T. Graf, F. Berthier, The Constraint Logic Programming 
Language CHIP.
In Proc. of The Int. Conf. on Fifth Generation Computer Systems, p693-702.  
Tokyo, Japan, November 1988.
</Program>

<Program> Ytoolkit
	<Number> 255
	<Domain> Graphics
	<Purpose> portable graphics package forProlog applications
	<Description> The system is composed of a lower level interface to a 
windowing system, a graphical interface toolkit and development tools to 
build and enhance user interfaces.
	<Developed by> José Paulo Leal, Luis Damas
	<Organization> Universidade de Porto, Centro da Inform‡tica
	<Country> Portugal
	<Prolog system> Many
	<References> José Paulo Leal, Luis Damas: Ytoolkit: a portable 
graphics package forProlog applications. Abstracts of poster sessions, Int. 
Conference on Practical Application of Prolog, London 1992.
</Program>

<Program> ZEXPERT
	<Number> 171
	<Domain> Finance
	<Purpose> a Prolog-based expert system shell
	<Description> Submitted to PAC. Paper 332
	<Developed by> Zsuzsa Farkas et al
	<Country> Austria
	<Prolog system> MProlog
	<Information Date> 6/11/91
</Program>