Simulation optimization research and development

TL;DR: Six domains that are common to any automated simulation optimization tool are identified and are the cornerstones for a unified strategy for simulation optimization and should guide future research in the field and development of next generation simulation optimization tools.

Abstract: Simulation optimization is rapidly becoming a mainstream tool for simulation practitioners. Simulation optimization is the practice of linking an optimization method with a simulation model to determine appropriate settings of certain input parameters so as to maximize the performance of the simulated system. Requirements for an automated simulation optimization tool for practitioners were formulated in the early 1970s and the first widely used commercial product appeared in 1995. The paper identifies six domains that are common to any automated simulation optimization tool. The domains are Methods, Classification, Strategy and Tactics, Intelligence, Interfaces, and Problem Formulation. These domains are the cornerstones for a unified strategy for simulation optimization and should guide future research in the field and development of next generation simulation optimization tools. This paper describes the six domains, presents recent research, and discusses research issues for two-phased optimization techniques.