Effects of Measurement Error on Process Signal Detection

Abstract:

Many data comes from a process with variables having non-normal distributions. The commonly used Shewhart variable control charts which depend heavily on the normality assumption should not be applied here. Further, measurement error of an instrument or gauge is an important factor in industry that influences the outcome of a process. Incapability of the instrument or gauge would cause the observation measure deviates from the true value and consequently lead to a false discrimination between good and bad service quality and manufacturing parts. In this study, we propose a new EWMA control chart to monitor the exponentially distributed service time between consecutive events with the measurement error instead of monitoring the number of events in a given time interval. We showed that, in the case of the measurement error following a normal distribution, the performance of the EWMA-M chart is significantly affected. We suggest letting for keeping the chart with the same detection ability; otherwise, the chart’s detection ability would be much poor than the charts with no measurement errors

• Periodical: Applied Mechanics and Materials (Volumes 263-266)
• Pages: 496-500
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.263-266.496
• Online since: December 2012
• Authors: Su Fen Yang, Chung Ming Yang, Ging Yun Cheng
• Keywords: Control Chart, Measurement Error, Signal Detection

References

[1] L.Y. Chan, M. Xie, T.N. Goh, Cumulative quantity control charts for monitoring production process, International Journal of Production Research. 38(2) (2000) 397-408.

DOI: 10.1080/002075400189482

[2] J.Y. Liu, M. Xie, P. Ranjan, A study of EWMA chart with transformed exponential data, Engineering Management Conference Proceedings IEEE International. 3 (2004) 1061-1065.

[3] J.Y. Liu, M. Xie, T.N. Goh, P.R. Sharma, A Comparative study of exponential time between events charts, Quality Technology of Quantitative Management. 3 (2006) 347-359.

DOI: 10.1080/16843703.2006.11673120

[4] J.Y. Liu, M. Xie, T.N. Goh, L.Y. Chan, A study of EWMA chart with transformed exponential data, International Journal of Production Research. 45(3) (2007) 743-763.

DOI: 10.1080/00207540600792598

[5] Y.J. Xie, K.L. Tsui, M. Xie, T.N. Goh, Monitoring time-between-events for health management, Prognostics and Health Management Conference. PHM'10 (Jan 2010 Macau), (2010) 1-8.

DOI: 10.1109/phm.2010.5413412

[6] T. Kanazuka, The Effects of Measurement Error on the Power of charts, J. Quality Technology. 18 (1986) 91-95.

[7] H.J. Mittag, Measurement Error Effect on Control Chart Performance, Annual Proceedings of the American Society for Quality Control. 49 (1995) 66-73.

[8] H.J. Mittag, D. Stemann, Gauge Imprecision Effect on the Performance of the Control Chart, Journal of Applied Statistics. 25 (1998) 307-317.

DOI: 10.1080/02664769823043

[9] K.W. Linna,W, H. Woodall, Effect of measurement error on Shewhart control charts, Journal of Quality Technology. 33 (2001) 213-222.

DOI: 10.1080/00224065.2001.11980068

[10] H. Shore, Determining measurement error requirements to satisfy statistical process control performance requirements, IIE Transactions. 36(9) (2004) 881-890.

DOI: 10.1080/07408170490473051

[11] P.E. Maravelakis, J. Panaretosand, S. Psarakis, EWMA chart and measurement error, Journal of Applied Statistics. 31(4) (2004) 445-455.

DOI: 10.1080/02664760410001681738

[12] S. Yang , H. Ho, M.A. Rahim, Economic Quality Control. 22(1) (2007) 1-14.

[13] S.A. Abbasi, On the performance of EWMA chart in presence of two component measurement error, Quality Engineering. 22(3) (2010) 199-213.

DOI: 10.1080/08982111003785649

[14] R.D. Luce, Response Time: Their role in inferring elementary mental organization, New York Oxford University Press, 1991.