To adapt changing market conditions, firms must make quick decisions and response them as fast as possible. Simulation is a powerful tool to analyze the effects of changes in an industrial or service system on a virtual environment and usage of simulation models have become widespread with the developments in computers. Simulation isn’t adequate to optimize the system parameters and additional methods are needed to integrate with simulation for optimization. In this study, a multi-objective optimization of a production system is considered. In this system, management aims to decide the optimal combination of workers in considered workstations. To cope with the problem a Data Envelopment Analysis (DEA) based metamodel is obtained and this metamodel is used as the objective function of the mathematical model with relevant constraints. In metamodeling stage two level factorial design is used.