Maintenance anomalies in rule-based expert systems

Maintenance has been identified as a major difficulty in expert system implementations. Maintenance on XCON: a 17,500 rule system (in 1988), requires that half of the code be rewritten annually (V.E. Barker and D.E. O´Connor, 1989). Surprisingly, the problem of maintenance has only recently begun to receive attention in expert system research. The author introduces the concept of maintenance anomalies in expert system rule bases. He presents a heuristic that eliminates implicit anomalies and two computable algorithms that minimize explicit anomalies. In addition to lower maintenance demands, this approach favorably affects storage requirements, computational efficiency, and verification