A meta-model based simulation optimization using hybrid simulation-analytical modeling to increase the productivity in automotive industry

Simulation modeling is one of the most useful techniques to analyze and evaluate the dynamic behavior of the complex manufacturing systems. Combining the mathematical power of an analytical method and the modeling capability of simulation with optimization approach called hybrid simulation-analytical modeling has been presented rarely. In this study a production control model is developed for a paint shop department in an automotive company in Turkey. As a real case study, the optimum operating setting of a paint shop production line of automotive company is determined using hybrid simulation optimization approach. In the optimization stage of the study Design of Experiment (DoE) is used to identify critical variables of the system by fitting a polynomial to the experimental data in a multiple linear regression analysis. The meta-model is validated and shown that it provides good approximations to simulation results. Findings from hybrid simulation-analytical optimization approach give invaluable knowledge to the company for the re-designing and control of current manufacturing system to increase its productivity.