International Journal of Industrial Engineering & Production Management

 Statistical process control charts are generally designed assuming that when the process is in control the observations are independent and identically distributed (i.i.d.) over time. However, the assumption of independence is easily violated when a process inherently generates auto correlated observations. When traditional control charts are applied to such processes then the false alarm rate experienced would be higher than what is expected. Machining process, due to the tool wear out, usually generates auto correlated observations. If such phenomenon is not incorporated in the chart design then one should expect a pattern in the plotted observations that will eventually lead to false alarms from time to time. This paper discusses the application of logistic regression to model and eliminate patterns that appear on fraction nonconforming items chart because of tool wear. Numerical results indicate significant improvements.