`` Direct Search'' Solution of Numerical and Statistical Problems
Journal of the ACM (JACM)
Design and Analysis of Experiments
Design and Analysis of Experiments
Expert Systems with Applications: An International Journal
Economic design of EWMA control charts based on loss function
Mathematical and Computer Modelling: An International Journal
| Hi-index | 12.05 |
The multivariate exponentially weighted moving average, MEWMA, control chart is an effective statistical tool for detecting small shifts in process mean vectors. On the one hand, the economic design process of a MEWMA control chart involves determining the main parameters of the chart, namely, the sample size n, the sampling interval h, the smoothing constant r, and the control limit L such that a quality cost function is minimized. On the other hand, the statistically constrained economic design of MEWMA chart is to determine the chart parameters such that a cost function is minimized while the statistical performance of the chart is maintained at a desire level. In this paper, the statistically constrained economic design model of the MEWMA control chart is first extended by the Taguchi loss function to improve its effectiveness. Next, an ant-colony optimization algorithm is proposed to solve the model in which a Markov chain approach is developed to calculate the average run lengths, ARLs. Then, the main parameters of the employed ant colony algorithm are tuned by means of a response surface methodology, RSM, approach. Finally, a sensitivity analysis on the main parameters of both the cost function and the control chart is performed using experimental design approach. The results show that augmenting a proper statistical nonlinear constraint to the model will improve the effectiveness of the model without a significant increase in the cost.