Artificial intelligence (AI) is playing an increasingly larger role in
production and manufacturing engineering. Much of this growth is the
result of special-purpose computer controlled machines that are
dominating modem manufacturing operations, such as computer numerically
controlled machines and robots, and production activities, such as
materials handling and process planning. Since a great deal of
production and manufacturing engineering knowledge can be put in the
form of rules, expert systems have emerged as a promising practical tool
of AI in solving manufacturing and production engineering problems. The
expert systems allow knowledge to be used for constructing human-machine
systems that have specialized methods and techniques for solving
problems in a particular application area. Over the years, many expert
systems have been developed for applications in manufacturing and
production engineering. Most of these expert systems, however, have been
of little use to practitioners at large. The primary reason for this
limited utility is that in most cases the developers do not divulge the
knowledge base and inference mechanism that form the backbone of an
expert system. Without the knowledge base, users can only derive a very
limited benefit from an expert system and, for all practical purposes, a
technical publication describing the expert system for the reader merely
becomes a publicity brochure. The reader must either develop his own
knowledge base or purchase the system from the developer, often at a
substantial cost.