COMPUTER-AIDED PLANNING AND DESIGN OF MANUFACTURING SIMULATION EXPERIMENTS

An simulation has become a major tool in studying discrete manufacturi ng systems, experimental design issues have started to receive much at tention. Although the importance of proper experimentation is often em phasized in the literature on discrete event simulation, it is neglect ed in most practical simulation studies. The main reason for this negl ect is that the design and analysis of a simulation experiment require expertise in experimental design methodology as well as familiarity w ith traditional statistical output analysis methods. This article pres ents an attempt to structure the process of planning and designing sim ulation experiments, and proposes a knowledge-based system, namely Des ign Of Experiments for Simulation (DOES), that assists the inexperienc ed analyst in this process. The proposed system takes the analyst thro ugh a planning process based on the simulation objectives.