时间：2013年4月19日（星期五）下午15:00
地点：旗山校区理工楼实验楼109
主讲：澳门大学 舒连杰教授
主办：数学与计算机科学学院
专家简介：Dr. Lianjie Shu is currently an Associate Professor in Faculty of Business Administration at University of Macau. He received his Bachelor degree in Mechanical Engineering and Automation from Xi'an Jiao Tong University, and his Ph.D. in Industrial Engineering and Engineering Management from the Hong Kong University of Science and Technology (HKUST). He currently serves an Associate Editor on Journal of Statistical Computation and Simulation and a Senior Editor on Journal of the Chinese Institute of Industrial Engineers. His recent research interests include statistical quality control, healthcare surveillance, and statistical computing. His publications appear on a wide variety of journals such as Statistics in Medicine, Naval Research Logistics, IIE Transactions, Journal of Quality Technology, etc.
报告摘要：The statistical control chart is conventionally designed to minimize the detection performance at a particular shift while maintaining a fixed in-control performance. This optimal design is often carried out based on Monte Carlo simulations, which require a large number of simulation replicates in order to achieve high accuracy. This talk investigates the optimal design of Markovian-type control charts. In particular, a novel gradient approach is developed for this purpose when a control chart possesses the Markovian properties. This efficiency of this approach is illustrated based on two examples: the design of exponentially weighted moving average (EWMA) control charts, and the design of cumulative sum (CUSUM) control charts under random shifts.