RTESS : Real Time Expert System Shell

The improving performance of inference engine in expert system has become an important research in recent years. As it is not realistic to search through all production rule during each cycle using an exhaustive search. Expert systems with a large set of rules can be slow, and can not be suitable for real-time application. In this paper, new algorithm for forward chaining and backward chaining in inference engine was proposed. This algorithm accommodates balanced binary searched tree and binary tree sort that have good performance in large database. Moreover, this new inference engine is supported by certainty factor as well. Displaying image and other supporting materials as the answer is facilitated.