Also found in: Wikipedia.


A programming language for rule-based production systems. A rule consists of pre-condition(s) and a resulting action.

The system checks its working memory to see if there are rules whose pre-conditions are satisfied, if so, the action in one selected satisfied rule is executed.

There is a public domain implementation of an OPS5 interpreter written by Charles L. Forgy <forgy@cs.cmu.edu> in 1977. It was first implemented in Lisp and later in BLISS. It was also ported to Common Lisp by George Wood and Jim Kowalski.

CLIPS is a language for writing expert systems, with some of the capabilities of OPS5.

See also C5, OPS83, OPS4, OPS5+, OPS83.

Inference Engine Tech, Cambridge MA.

An OPS5 interpreter in Common LISP.

A version by Mark Kantrowitz. <mkant+@cs.cmu.edu>.

["Programming Expert Systems in OPS5", L. Brownston et al, A-W 1985].

["An OPS5 Primer", Sherman et al, comes with OPS5 for DOS].

["Rule-Based Programming in the Unix System", G.T. Vesonder, AT&T Tech J 67(1), 1988].
This article is provided by FOLDOC - Free Online Dictionary of Computing (foldoc.org)
References in periodicals archive ?
Jess's forward-chaining reasoning system was modeled after production systems such as OPS5 (Brownston et al.
This is identical to the interpretation of conditions in the CA rules of, for example, Ariel [Hanson 1992], RPL [Delcambre and Etheredge 1989], and set-oriented adaptations of OPS5 [Gordin and Pasik 1991].
These types of systems are usually implemented in a logic programming language or a production system that directly supports logical inference, such as Prolog or OPS5. Henderson [1984], Kung [1984], and Hummel [1989] are examples in this category.
The user module is written in PASCAL, a control module in REXX, a command language, and the evaluation module in LISP and OPS5. PEP has been implemented on the VM under the CMS operating system.
Shell prices range from $0 for public domain packages (e.g., OPS5) to under $100 for small PC-based packages (e.g., OPS-2000) to $110,000 packages (e.g., IBM's AD/Cycle TIRS) that run on expensive mainframes.
Most of the system is written in Jess, a Java-embedded descendant of OPS5 and Clips, Properly understood and used, rule-based programming and AI-style "second-order programming" based on constructs like eval, apply, and the Jess build are invaluable for writing and integrating complex algorithms in a concise and timely manner.
For example, the fact that R1 (23) performs a linear sequence of subtasks is not explicitly encoded; the system designer "encrypted," so to speak, this control in the pattern-matching of OPS5, the production-rule system in which RI is implemented.
These facilities all differ from the Starburst approach in that their rule language and semantics is based substantially on OPS5 [14]; hence, their rules respond to operations on a single row.
These additional actions may activate other rules, and a forward chaining control flow results, as was popularized in OPS5 [10].