An alternating variable method with varying replications for simulation response optimization

• Authors: Shih-Pin Chen
• Venue: Computers & Mathematics with Applications
• Year: 2004

Simulation response optimization is an important problem often encountered in behaviorinvestigation of systems that are so complicated that the performance can only be evaluated by using simulation. This paper modifies the alternating variable method used in deterministic optimization to suit the stochastic environment in simulation response optimization. The main idea underlying the proposed method is to conduct several replications at each trial point to obtain reli able estimate of the theoretical response. In particular, the number of replications is not fixed but is set to a variable automatically adjusted on the basis of the distance between the two successive trial points. To avoid misjudging the real different between two points due to the stochastic nature, a t-test instead of a simple comparison of the mean responses is performed. Empirical results from a stochastic Watson function with nine variables, a queueing problem, and an inventory problem indicate that this method is able to find the optimal solutions in a statistical sense, and the varying replications has demonstrated to be able to alleviate the computational burden in the whole optimization procedure. Moreover, this method is robust with respect to the parameter used in determining the varying replications conducted at each trial point.