A Minimum Discrimination Information Control Chart for Stationary Autocorrelated Processes
Abstract
A procedure for simultaneous monitoring of both the process mean and the standard deviation with a single control chart based on the Kullback-Leibler (KL) statistic is proposed for stationary Gaussian autocorrelated processes. The residual based MDI control chart is developed and then the performance is investigated through Monte Carlo simulations in situations of sustained shifts using the average run length (ARL) as the preferred criterion for comparisons. The results show that the proposed chart performs exceptionally well for detecting shifts in process mean even when standard deviation changes.
DOI: http://dx.doi.org/10.4038/sljastats.v12i0.4973
Sri Lankan Journal of Applied Statistics Vol.12 2011 pp.161-174
© 2012 Pairoj Khawsithiwong, Nihal Yatawara, published by The Institute of Applied Statistics, Sri Lanka
This work is licensed under the Creative Commons License.