A generalized framework for characterizing the quality degradation and predicting failures among critical components within several computer numerically controled machine tools will be investigated. "Failure" in this context refers to degradation of the critical components within a machine tool and/or the quality of the finished product. This framework will integrate sensory devices with real-time knowledge bases and simulation models. The implemented framework will provide an integrated environment for a maintenance supervisor to understand the quality degradation process for various critical components within a machine tool and obtain early warnings about their failure time intervals. Although the framework is general, the implementation will be restricted to diagnosing quality degradation in grinding machines. The research will focus on building a tested knowledge base for characterizing the quality degradation process within grinding machines, and reducing the response time in providing decisions using knowledge bases and simulation. Technology transfer of the results will be achieved through interaction with industrial sponsors.
