Adaptive Statistical Process Control by Managing Uncertainty in Process Monitoring

Statistical Process Control (SPC) is a widely used methodology for monitoring and improving process stability and quality. However, traditional SPC techniques rely on crisp control limits, which may be insufficient when dealing with uncertainty, variability, or imprecise data in real- world environments. This study introduces a fuzzy logic-based framework to enhance SPC by incorporating flexible and adaptive control mechanisms. In the proposed approach, process parameters such as mean, standard deviation, defect rate, and cycle time are transformed into fuzzy linguistic variables. A fuzzy inference system (FIS) is then designed to evaluate process conditions using expert-defined rules, providing an interpretable and continuous assessment of process stability. Unlike traditional control charts, which classify a process as "in-control" or "out-of-control", the fuzzy SPC approach allows intermediate states, such as "marginal" or "at risk," thereby enabling proactive intervention before severe deviations occur. The results demonstrate that fuzzy SPC provides greater robustness in handling uncertain data and offers a more realistic and actionable decision support system for quality management.