A Distributed Approach for Inferring Production Systems

Abstract

Tools for building production systems encounter the problem of low performance, and many researchers are working on the improvements of performance of th~se to?ls. This paper proposes a distribu~ed al?pr<?ach for mfen:ng production systems. The resultmg ~~s~buted productH;m systems are expected to be built over dtstnbuted ~ystems ~1th broadcast capability, and production rules on differc:nt ~ttes work in a cooperative way with only a few_c~mmumc~u.ons between them. Working memory on a local stte IS made vlSlble to rules on remote sites. A tool for building distributed production systems, called DPS, has been imp~emented . D~S not only supports elegant constructs for ~xpressmg th~ ~~pactty of distributed inference but also provtdes the facthttes for building clusters of rules. Wi~ these facilities, ~PS allows users to make the inference engme focus on a particular set of rules. This paper also describes the knowledge representation and other features about DPS.