The CANASTA experience: key management and technical decisions in a hybrid expert system project

A case study is presented of a successful intelligent system. CANASTA (the Crash Analysis Troubleshooting Assistant) is a knowledge-based system designed to assist support engineers in isolating the underlying causes of operating system crashes, whether due to hardware faults or system software bugs. CANASTA's success is largely due to a combination of project management actions and the innovative technical design and development of the system. The lessons learned from developing a large hybrid system such as CANASTA are presented.<
>