Optimum Mean Location in a Poor‐Capability Process

Abstract Many processes may be stable but not capable of meeting the specification limits; consequently, a higher number of nonconforming items than the expected will be observed and monetary losses are expected. A natural solution is to reduce process variability, but in many cases, it may be expensive and take time to get meaningful results. Here we propose a procedure that minimizes losses by adjusting the process mean in an economically ideal way. We concentrate on obtaining an optimum mean value for a two‐sided specified normal process, employing a quadratic cost function. Because an analytical solution for the optimum mean is difficulty to obtain, we developed a program that allows the user to find the optimum mean value easily.