Work

An Enhanced EWMA Model for Statistical Insights in Process Monitoring with Application in Brake Pad Failure and Carbon Fiber Strength

Year: 2025

Type: article

Abstract: Conventional control charts often assume normality, which may not hold for many engineering processes. In cases where processes follow an Inverse Maxwell $$\left( {{\text{IM}}} \right)$$ distribution, as seen in various industrial applications, it becomes crucial to employ suitable monitoring methods. To address this gap, this study introduces the hybrid exponentially weighted moving average ( $${\text{HEWMA}}_{{{\text{IM}}}}$$ ) chart for the $${\text{IM}}$$ distribution. Performance evaluation includes metrics like average run length, median run length, and standard deviation run length. Comparative analysis with existing IM distribution-based charts such as the Shewhart V chart ( $${\text{V}}_{{{\text{IM}}}}$$ ), exponentially weighted moving average ( $${\text{EWMA}}_{{{\text{IM}}}}$$ ), and extended EWMA ( $${\text{EEWMA}}_{{{\text{IM}}}}$$ ) charts reveal the $${\text{HEWMA}}_{{{\text{IM}}}}$$ chart’s superior efficiency. Real-world applications in brake pad production and carbon fiber strength testing validate its practicality and engineering applications. In conclusion, $${\text{HEWMA}}_{{{\text{IM}}}}$$ is a novel tool tailored to monitor $${\text{IM}}$$ processes efficiently, offering enhanced process monitoring for diverse industries.

Source: Journal of Statistical Theory and Applications

DOI: 10.1007/s44199-025-00110-5

Authors: Muhammad Waqas, Song Hua Xu, Syed Masroor Anwar, Zahid Rasheed, Gilbert Masengo