Resumen de Multivariate Exponentially Weighted Moving-Average Chart for Monitoring Poisson Observations
Nan Chen, Zhonghua Li, Yanjing Ou
In many practical situations, multiple variables often need to be monitored simultaneously to ensure that the process is in control. This article presents a feasible multivariate monitoring procedure based on the general multivariate exponentially weighted moving average (MEWMA) to monitor the multivariate count data. The multivariate count data is modeled using Poisson log-normal distribution to characterize their interrelations. The authors systematically investigate the effects of different charting parameters and propose an optimization procedure to identify the optimal charting parameters. In particular, a design table is provided as a simple tool for quality engineers to design the optimal MEWMA chart. To further improve the efficiency, the authors integrate the variable sampling intervals (VSI) in the monitoring scheme. Simulation studies and an example are used to elicit the application of the proposed scheme. The results are encouraging and demonstrate effectiveness of the proposed methods well.
