On Developing Robust Monitoring Scheme for Location Parameter

ABSTRACT Statistical Process Control (SPC) charts are extensively utilized in quality monitoring to maintain process stability and identify shifts within production systems. Conventional location control charts, like the Shewhart chart, are designed based on the premise that process data adheres to a normal distribution. Nevertheless, in practical industrial environments, data frequently diverges from normality due to factors such as skewness, heavy tails, or the occurrence of outliers and extreme values. These violations reduce the efficiency and detection capability of conventional charts, leading to delayed identification of process shifts and increased production risks. To address these challenges, a robust monitoring scheme for the location parameter is developed that can perform reliably under both normal and contaminated environments. The efficiency and robustness of different location estimators are first evaluated, with particular focus on their performance in the presence of outliers and non‐normal data structures. Based on these insights, a new location control chart has been developed to address the limitations of conventional methods. The effectiveness of the proposed chart is evaluated through a comprehensive analysis of its run length characteristics across different distributional scenarios. Additionally, the application of the proposed chart is illustrated to showcase its practical use in process monitoring. The effectiveness of the proposed scheme is illustrated through a real‐life example, highlighting its usefulness for quality practitioners and industrial applications.