Forward-chaining rule-based programs, being data-driven, can function in changing environments in which backward-chaining rule-based programs would have problems. But, degugging forward-chaining programs can be tedious; to debug a forward-chaining rule-based program, certain ‘historical’ information about the program run is needed. Programmers should be able to directly request such information, instead of having to rerun the program one step at a time or search a trace of run details. As a first step in designing an explanation system for answering such questions, this paper discusses how a forward-chaining program run’s ‘historical’ details can be stored in its Rete inference network, used to match rule conditions to working memory. This can be done without seriously affecting the network’s run-time performance. We call this generalization of the Rete network a historical Rete network. Various algorithms for maintaining this network are discussed, along with how it can be used during debugging, and a debugging tool, MIRO, that incorporates these techniques is also discussed.
Real-time Machining Vibration Data Driven Milling Process Parameters Adaptive Optimization