A fuzzy logic application in SPC evaluation and control

Control chart pattern identification is an important aspect of statistical process control (SPC). In the conventional use of SPC, the user may just need to know whether or not the process is out of control. What actions should be taken to adjust the process is uncertain. This is a major drawback of control charts and, as a consequence, the user cannot employ SPC technique to automatically control the process. This paper provides a new fuzzy-SPC evaluation and control method. Fuzzy logic is applied to control charts to create a fuzzy inference system using the fuzzy logic toolbox in MATLAB. Different membership functions were chosen and their fuzzy control bases were generated as outputs of a fuzzy inference system based on zone rules. They are executed in a simulation system written in Visual C++. Results of the adjusted and controlled SPC charts are displayed on the screen. Analysis of the effect of changing the membership function is presented.