Likelihood Ratio Test-Based Chart For Monitoring The Process Variability

Now, control charts have been widely used in process monitoring. If the process isout of control, it can timely remind us to take measures to prevent or eliminate the causesof drift. Some factors in manufacturing such as faulty raw material, unskilled/carelessoperators, and loosening of machine settings may lead to a change in process dispersionwithout necessarily influencing the level of the process mean. In this paper, we pro-pose a new chart which integrates the exponentially weighted moving average(EWMA)procedure with the generalized likelihood ratio(GLR) test statistics for monitoring theprocess variance. The new chart can be easily designed and constructed. Due to thegood properties of the GLR test and EWMA procedure, computation results show thatit provides quite a satisfactory performance, including the detection of the decrease invariability and the individual observation at the sampling point which are very importantin many practical applications but may not be well handled by the existing approachesin the literature. The optimal parameters that can be used as a design aid in selectingspecific parameter values based on the average run length(ARL) are described. Theapplication of our proposed method is illustrated by a real data example from chemicalprocess control.