Measurement Error in Monitoring Dependent Process Steps

In the application of quality control charts for monitoring multi-stage processes, most research has been based on the assumption of accurate measurements. However, measurement errors often exist in many manufacturing and service environments. To address this issue, the researchers considered a linear measurement error model with an AR(۱) process in the state space model. This model could be equivalently re-written as an ARMA(۱,۱) process, where the MA(۱) parameter is related to the variance of measurement errors. The researchers then designed a new control chart to discover the effect of measurement errors in this model. Simulation studies under different shift scenarios and situations were conducted to evaluate the capability of the proposed control chart using well-known performance criteria for detecting measurement errors. The results demonstrate that the remedial approach can effectively identify the measurement error model in multi-stage processes. This finding highlights the importance of accounting for measurement errors when monitoring complex, multi-stage processes in order to ensure accurate quality control.

Keywords:

Measurement errors , multi-stages processes , statistical process monitoring , Phase II , EWMA control chart