Application of Evolutionary Algorithms for Solving Multi-Objective Simulation Optimization Problems

Most optimization problems associated with real-world complex systems are too difficult to bemodeled analytically. The difficulty may come from the uncertainties involved in the input and output of the system, the complex (often nonlinear) relationships between the system decision variables and the system performances, and possible multiple, conflicting performance measures to be considered when choosing the best design. In light of this, discrete event simulation has become one of the most popular tools for the analysis and design of complex systems due to its flexibility, its ability to model systems unable to be modeled through analytical methods, and its ability to model the time dynamic behavior of systems [1]. However, simulation can only evaluate system performances for a given set of values of the system decision variables, i.e., it lacks the ability of searching for optimal values which would optimize one or several responses of the system. This explains the increasing popularity of research in integrating both simulation and optimization, known as simulation optimization: the process of finding the best values of decision variables for a system where the performance is evaluated based on the output of a simulation model of this system.