Using Morris' randomized OAT design as a factor screening method for developing simulation metamodels

Simulation metamodels have been used for optimization, prediction, sensitivity analysis and understanding of complex, real-world systems. Since most simulation models contain a large number of input parameters, it is of great interest to determine the most important ones to include in a metamodel given a particular modeling context, i.e. given a particular set of questions which are to be addressed by the metamodel. This paper employs Morris' randomized one-at-a-time (OAT) design as a factor screening method prior to developing a number of simulation metamodels. The approach is illustrated with reference to a stochastic combat simulation, called SIMBAT.