Inertia measurement of exponentially weighted moving average control chart based on autocorrelated attribute data

Abstract

Purpose: This paper aims to study the effect of inertia on the performance of the EWMA control chart for attribute characteristics and to investigate the effects of inertia properties on control charts with attribute characteristics.Design/methodology/approach: By calculating the inertia criterion, the effectiveness of the control chart can be measured in identifying changes and how one can reduce the inertial properties of the control charts. MATLAB software has been used to conduct simulation studies. The effect of inertia on the performance of an exponentially weighted moving average control chart and modified exponentially weighted moving average control chart for observations caused by correlated processes based on Poisson distribution has been investigated.Findings: After calculating and analyzing the results, it was determined that the modified exponential moving average control chart performs better than the exponential moving average control chart based on the Poisson distribution in the worst-case signal resistance. Also, by decreasing the parameter λ, the inertia value of the control charts increased.Research limitations/implications: Other control charts that follow different distributions can be used, or the effect of inertia on the performance of control charts in multivariable processes can be considered as a subject for further study.Practical implications: After analyzing the positive properties of inertia, it is possible to increase productivity in production processes. This research purposed new indicators to evaluate the performance of control charts in terms of inertial properties.Social implications: In the field of manufacturing industries, it is possible to increase productivity by examining the appropriate characteristics of inertia. On the other hand, regarding the use of control charts for monitoring chemical and petrochemical processes and evaluating the inertia index, it is possible to reduce the variability of the process in long term, which will have an impact on environmental issues.Originality/value: Measuring the inertia of the exponentially weighted moving average control chart when the observations are independent and follow the Poisson distribution, in addition to measuring the inertia of the exponentially weighted moving average control chart in the presence of correlated observations distinguishes this research from previous studies.

Keywords

• exponentially weighted moving average
• inertial properties
• statistical process control
• correlation
• poisson distribution