Published
Evaluation of the Mean Control Chart Under a Bayesian Approach
Evaluación de la carta de control para la media de un proceso bajo un enfoque bayesiano
DOI:
https://doi.org/10.15446/rce.v45n1.93588Keywords:
Control charts, Bayesian Approach, ARL, Conjugate prior, Informative prior (en)Gráficas de control, Enfoque Bayesiano, ARL, Distribución a priori conjugada, Distribución a priori informativa (es)
Downloads
A previous study on the evaluation of control charts for the mean with a Bayesian approach, based on predictive limits, was performed in such a way that neither prior nor sample information was taken into account. This work was developed to make a more complete study to evaluate the influence of the combination of the prior distribution with the sample information. It is assumed that the quality characteristic to be controlled can be modeled by a Normal distribution and two cases are considered: known and unknown variance. A Bayesian conjugate model is established, therefore the prior distribution for the mean is Normal and, in the case where the variance is unknown, the prior distribution for the variance is defined as the Inverse-Gamma(ν, ν). The posterior predictive distribution, which is also Normal, is used to establish the control limits of the chart. Signal propability is used to measure the performance of the control chart in phase II, with the predictive limits calculated under different specifications of the prior distributions, and two different sizes of the calibration sample and the future sample. The simulation study evaluates three aspects: the effects of sample sizes, the distance of the prior mean to the mean of the calibration sample, and an indicator of how informative is the prior distribution of the population mean. In addition, in the case of unknown variance, we study what is the effect of changing values in the parameter ν. We found that the false alarm rate could be quite large if the prior distribution is very informative which in turn leads to an ARL (average run length) biased chart, that is, the maximum of the ARL is not given when the process is under control. Besides, we found
great influence of the prior distribution on the control chart power when the size of the calibration and future samples are small, particulary when the prior is very informative. Finally, regarding the effect of the parameter ν, we found that the smaller the value, which means having a less informative prior distribution, the lower the power of the control chart.
Un estudio previo sobre la evaluación de las gráficas de control para la media con un enfoque Bayesiano, basadas en límites predictivos, fue realizado de tal manera que no se tuvo en cuenta ni la información a priori ni la información muestral. En este trabajo hemos desarrollado un estudio más completo para evaluar la influencia de la combinación de la distribución a priori con la información muestral. Se asume que la característica de calidad a controlar puede modelarse mediante una distribución Normal y se consideran dos casos: varianza conocida y desconocida. Para la aproximación Bayesiana se establece un modelo conjugado, por lo tanto la distribución a priori para la media es Normal y, en el caso donde la varianza es desconocida, se define como distribución a priori para la varianza la Gamma-Inversa(ν, ν). La distribución predictiva posterior, que también es Normal, es utilizada para establecer los límites de control de la gráfica. Se utiliza la probabilidad de señal para medir el desempeño de la gráfica en la denominada phase II de control, con los límites predictivos calculados bajo diferentes especificaciones de las distribuciones a priori, del tamaño de la muestra de calibración y del tamaño de la muestra futura. El estudio de simulación evalúa tres aspectos: efectos del tamaño de muestra, de la distancia de la media a priori con relación a la media de la muestra de calibración, y un indicador de cuán informativa es la distribución a priori de la media poblacional. Adicionalmente, cuando la varianza es desconocida, se estudia el efecto de los valores del parámetro ν. Se encuentra que la tasa de falsas alarmas puede ser exageradamente grande si se especifica una a priori muy informativa, lo que a su vez puede conducir a una gráfica de control con una ARL (average run length) sesgada, es decir, que el máximo de la ARL no se dará cuando el proceso está en control. Además, cuando el tamaño de las muestras de calibración y de la muestra futura son pequeñas, hay gran influencia de la especificación de la a priori sobre la potencia de la gráfica de control, en especial cuando la a priori es muy informativa. Finalmente, en cuanto al efecto del parámetro ν, se encuentra que entre más pequeño es su valor, lo cual indica que la distribución a priori para la varianza es menos informativa,
menor es la potencia de la gráfica de control, en especial si los tamaños de
muestra son pequeños.
References
Ali, S. & Riaz, M. (2020), ‘On designing a new bayesian dispersion chart for process monitoring’, Arabian Journal for Science and Engineering 45, 2093–2111. DOI: https://doi.org/10.1007/s13369-019-04036-w
Bhat, S. V. & Gokhale, K. D. (2014), ‘Posterior Control Chart for Process Average under Conjugate Prior Distribution’, Economic Quality Control 29(1), 19–27. DOI: https://doi.org/10.1515/eqc-2014-0003
Chen, Y. (2016), ‘A New Type of Bayesian Nonparametric Control Charts for Individual Measurements’, Journal of Statistical Theory and Practice 10(1), 226–238. DOI: https://doi.org/10.1080/15598608.2015.1117405
Colosimo, B. M. & del Castillo, E. (2007), Bayesian process monitoring, control and optimization, Chapman & Hall/CRC, Boca Raton, FL. DOI: https://doi.org/10.1201/9781420010701
Gelman, A. (2006), ‘Prior distributions for variance parameters in hierarchical models: Comment on article by browne and draper’, Bayesian Analysis 1(3), 515–534. DOI: https://doi.org/10.1214/06-BA117A
Jensen, W. A., Jones-Farmer, L. A., Champ, C. W. & Woodall, W. H. (2006), ‘Effects of parameter estimation on control chart properties: A literature review’, Journal of Quality Technology 38(4), 349–364. DOI: https://doi.org/10.1080/00224065.2006.11918623
Menzefricke, U. (2002), ‘On the evaluation of control chart limits based on predictive distributions’, Communications in Statistics - Theory and Methods 31(8), 1423–1440. DOI: https://doi.org/10.1081/STA-120006077
Monfared, M. E. D. & Lak, F. (2013), ‘Bayesian estimation of the change point using ¯X control chart’, Communications in Statistics - Theory and Methods 42(8), 1572–1582. DOI: https://doi.org/10.1080/03610926.2011.594536
Nenes, G. (2013), ‘Optimisation of fully adaptive Bayesian charts for infinitehorizon processes’, International Journal of Systems Science 44(2), 289–305. DOI: https://doi.org/10.1080/00207721.2011.600475
Nikolaidis, Y. & Tagaras, G. (2017), ‘New indices for the evaluation of the statistical properties of Bayesian ¯x control charts for short runs’, European Journal of Operational Research 259(1), 280–292. DOI: https://doi.org/10.1016/j.ejor.2016.09.039
Psarakis, S., Vyniou, A. K. & Castagliola, P. (2014), ‘Some recent developments on the effects of parameter estimation on control charts’, Quality and Reliability Engineering International 30(8), 1113–1129. DOI: https://doi.org/10.1002/qre.1556
Saghir, A. (2015), ‘Phase-I Design Scheme for ¯x chart Based on Posterior Distribution’, Communications in Statistics - Theory and Methods 44(3), 644–655. DOI: https://doi.org/10.1080/03610926.2012.752846
Shewhart, W. A. (1931), Economic control of quality of manufactured product, Van Nostrand., (Reprinted by The American Society for Quality Control, 1980), New York.
Tagaras, G. & Nikolaidis, Y. (2002), ‘Comparing the effectiveness of various bayesian ¯X control charts’, Operations Research 50(5), 878–888. DOI: https://doi.org/10.1287/opre.50.5.878.361
Tsiamyrtzis, P. & Hawkins, D. M. (2005), ‘A bayesian scheme to detect changes in the mean of a short-run process’, Technometrics 47(4), 446–456. DOI: https://doi.org/10.1198/004017005000000346
Woodall, W. H. (2000), ‘Controversies and contradictions in statistical process control’, Journal of Quality Technology 32(4), 341–350. DOI: https://doi.org/10.1080/00224065.2000.11980013
Woodall, W. H. & Montgomery, D. C. (1999), ‘Research issues and ideas in statistical process control’, Journal of Quality Technology 31(4), 376–386. DOI: https://doi.org/10.1080/00224065.1999.11979944
Woodall, W. H. & Montgomery, D. C. (2014), ‘Some current directions in the theory and application of statistical process monitoring’, Journal of Quality Technology 46(1), 78–94. DOI: https://doi.org/10.1080/00224065.2014.11917955
Xiaosong, W., Rui, M., Xinyi, Z., Zhibin, J. & Xuening, C. (2015), ‘A conjugate Bayesian approach to control chart for multi-batch and low volume production’, International Journal of Production Research 53(7), 2179–2185. DOI: https://doi.org/10.1080/00207543.2014.975857
How to Cite
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
License

This work is licensed under a Creative Commons Attribution 4.0 International License.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).