Development and application of a modified EWMA control chart for early detection of process shifts in skewed distributions

Abstract

In this study, we introduce a novel control chart based on the Modified Exponentially Weighted Moving Average (MEWMA) statistic to swiftly detect shifts in processes, particularly for quality parameters following an exponential distribution. Our method is applied to monitor changes in the rate parameter within the Breast Cancer METABRIC database. Unlike traditional approaches that rely on the mean, which can be misleading for skewed distributions, our focus is on the maximum likelihood estimate of the rate parameter. This approach provides a more precise and responsive measure for detecting process shifts. We evaluated the performance of our proposed control chart using Average Run Length (ARL) across different shift levels, comparing it with Patel’s EWMA and traditional EWMA charts. The results demonstrate that the MEWMA statistic, with its lower variance, outperforms conventional charts in early shift detection. A practical example using the Breast Cancer METABRIC data underscores its effectiveness in identifying process shifts more accurately.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.