Response Surface Methodology in Optimization via Simulation

Simulation models are often applied to the optimization of complex problems that cannot be solved by analytical or mathematical methods. This paper seeks to analyze the application of Response Surface Methodology (RSM) as an organized and efficient way to pursue optimization in Discrete Event Simulation (DSS). It is categorized as a quantitative research using statistical tools and the experimentation research method following a normative empirical model. The results found are represented by means of the values of each input variable of the Design of Experiments (DOE) obtained at the optimal point. After that a comparison is made with the optimization results used here with those of a market software in which the effectiveness of the optimization model adopted in four study objects was verified. Furthermore the optimal point was reached with a reduction of about 80% in the number of experiments in addition to the sensitivity analyses generated by the MSR.