A state transition model for rule-based expert systems
This dissertation describes a State Transition Model (STM) for building real-time, continuous-operation, transaction-processing expert systems for the maintenance of complex physical systems. An STM enhances the expressiveness of rule-based systems by making control knowledge explicit. The author has augmented the traditional rule-based production system architecture with a state transition model to develop a framework for building hybrid expert systems. The STM-based framework provides a unified way of representing and using rule-based empirical associations and model-based reasoning with procedural and control knowledge. An STM also simplifies the task of refining a knowledge base because the changes to the knowledge base can be identified by mapping them into the states and state transitions of an STM. The STM is well suited to implementing parallelism in expert systems. The STM has been successfully used in developing an expert system for telephone network maintenance.
- Research Organization:
- Rutgers--the State Univ., New Brunswick, NJ (USA)
- OSTI ID:
- 5815826
- Country of Publication:
- United States
- Language:
- English
Similar Records
Experiments with rule writer, a tool for building expert systems
Considerations concerning the construction of an expert system for control system design