Development of robust X-bar charts with unequal sample sizes

TL;DR

This paper introduces a robust X-bar control chart method that effectively handles data contamination and unequal sample sizes, improving process monitoring accuracy in practical, non-ideal conditions.

Contribution

The paper presents a novel robust X-bar chart construction method that accounts for data contamination and unequal sample sizes, enhancing traditional control chart performance.

Findings

Traditional charts underperform with contaminated data

Proposed charts outperform traditional ones with outliers

Robust charts maintain accuracy with unequal sample sizes

Abstract

The traditional variable control charts, such as the X-bar chart, are widely used to monitor variation in a process. They have been shown to perform well for monitoring processes under the general assumptions that the observations are normally distributed without data contamination and that the sample sizes from the process are all equal. However, these two assumptions may not be met and satisfied in many practical applications and thus make them potentially limited for widespread application especially in production processes. In this paper, we alleviate this limitation by providing a novel method for constructing the robust X-bar control charts, which can simultaneously deal with both data contamination and unequal sample sizes. The proposed method for the process parameters is optimal in a sense of the best linear unbiased estimation. Numerical results from extensive Monte Carlo simulations and a real data analysis reveal that traditional control charts seriously underperform for monitoring process in the presence of data contamination and are extremely sensitive to even a single contaminated value, while the proposed robust control charts outperform in a manner that is comparable with the traditional ones, whereas they are far superior when the data are contaminated by outliers.