Date of Award

Spring 2017

Degree Name

Master of Science in Engineering

First Advisor

Dr. Lee J. Wells

Second Advisor

Dr. Steven E. Butt

Third Advisor

Dr. Azim Houshyar

Access Setting

Masters Thesis-Open Access

Abstract

Over the past half-century, Exponential Weighted Moving Average (EWMA) control charts have been widely used in Statistical Process Control to detect small and persistent process shifts. In theory, EWMA control limits monotonically increase over time to account for the continual growth of the EWMA statistic’s variance. These control limits are often assumed constant and are set to their respective asymptotic limits. While not theoretically correct, this simplifies the process of applying and analyzing the EWMA control chart while having little effect on the chart’s performance. One-Sided EWMA charts are often implemented when it is only desirable to detect shifts in a specific direction.

When using One-Sided EWMA charts, reflecting boundaries (resets) can be used to prevent the statistic from drifting too far from the chart’s control limit, which can delay shift detection. There have been several research efforts into the design and performance of One- Sided EWMA charts with reflecting boundaries; however, these efforts have maintained the constant control limit assumption. When implementing a reflecting boundary, the EWMA statistic’s variance constantly being reset to zero, which may significantly affect the constant control limit assumption’s validity. This study focused on understanding the impact into chart’s performance with and without the constant limit assumption. Such comprehension has led to interesting performance comparisons between One-sided time-varying limit EWMA scheme and the combined Shewhart-EWMA control chart.

Share

COinS