A multi-agent architecture for fault diagnosis in manufacturing systems

Rapid and accurate diagnosis of faults in computer integrated manufacturing systems is essential to avoid costly downtime. Many artificial intelligence approaches to automated diagnosis use either symptom-based or functional reasoning. Symptom-based approaches are unable handle exceptions, while functional approaches are often computationally expensive. Current hybrid approaches which combine the two methods are too structured in their approach to switching between the reasoning methods and, thus fail to provide rapid response comparable to humans. This paper presents a robust, extensible approach to fault diagnosis that allo vs unstructured sit,itching between reasoning methods using multiple intelligent agents that examine the problem domain from a variety of perspectives.