Abstract: The combination of simulation and optimization is one of the hottest topics in the scientific community today. This paper describes an algorithmic structure of combining simulation and optimization techniques both in theory and practice. The response surface methodology has been used to optimize the decision variables under the environment of simulation.The algorithm includes a 2k-P fractional factorial design of simulation experiment. The simulation replication can be performed at the design experiment points. By using an ordinary least-square regression analysis, the simulation responses and the associated decision variabls can be fit to be a first or second order response surface model. The improved solution of the system can then be obtained at the steepest gradient on the response surface . As this procedure continues, the optimal solution of the system is given.Based on the GASP-4 simulation language, a simulation-optimization software GASPRSM has been developed and implemented successfully on an IBM-PC computer. The decision variables of any simulation model which modeled by GASP-4 can be optimized and simulated by running this software.As an application of GASPRSM, we have established an aggregate production planning simulation-optimization model in which the basic ideas of the parametric producton planning are involved. The objective of this model is to minimize the production cost and generate an optimal production plan and inventory control strategy for an aircraft manufacturing factory. Running this model with GASPRSM shows that the optimal production plan results in significant economy in propduction.