A fuzzy rule-based algorithm to improve the performance of SPC in quality systems

Fuzzy logic has rarely been applied to statistical process control (SPC) in quality control systems of industrial applications. Instead, control charts have extensively been utilised for monitoring the production processes. The measurement process of the control parameter involves usually random disturbance that arises from the environment or the measuring instrument itself. The control parameter may exceed the predetermined control limits of the plant due to this disturbance and generate false alarms. Consequently, it will be very difficult for the traditional charts to distinguish between real faults and false alarms. In the paper a fuzzy control algorithm has been developed. Its rules are trained to cope with false alarms and efficiently detect real faults in the SPC of the industrial processes. The design and implementation of the developed control algorithm are presented and tested