ECONOMIC-STATISTICAL DESIGN OF THE MULTIVARIATE SYNTHETIC T2 CHART USING LOSS FUNCTION

Authors

  • Wai Chung Yeong Universiti Sains Malaysia
  • Michael Boon Chong Khoo Universiti Sains Malaysia
  • Mohammad Shamsuzzaman University of Sharjah
  • Philippe Castagliola Universite de Nantes

DOI:

https://doi.org/10.23055/ijietap.2014.21.5.1481

Keywords:

economic-statistical design, factorial design, Hotelling’s T2 chart, MEWMA chart, multivariate loss function, synthetic chart

Abstract

This paper proposes the economic-statistical design of the synthetic T2 chart. In this design, the optimal chart parameters are obtained by minimizing the expected cost function, subject to constraints on the in-control and out-of-control average run lengths (ARL0 and ARL1). The quality cost is calculated by adopting a multivariate loss function. This paper investigates the effects of certain input parameters, shift sizes and multivariate loss coefficients toward the optimal cost, choice of chart parameters and optimal ARLs. This is done by using factorial design, which allows interaction effects to be identified. Besides that, comparisons are made between the significant parameters of the synthetic T2 chart with that of the Hotelling’s T2 and Multivariate Exponentially Weighted Moving Average (MEWMA) charts. Conditions where the synthetic T2 chart shows better economic-statistical performance than the Hotelling’s T2 and MEWMA charts are identified. The synthetic T2 chart compares favorably with the other two charts.

Published

2014-10-20

How to Cite

Yeong, W. C., Khoo, M. B. C., Shamsuzzaman, M., & Castagliola, P. (2014). ECONOMIC-STATISTICAL DESIGN OF THE MULTIVARIATE SYNTHETIC T2 CHART USING LOSS FUNCTION. International Journal of Industrial Engineering: Theory, Applications and Practice, 21(5). https://doi.org/10.23055/ijietap.2014.21.5.1481

Issue

Section

Statistical Analysis