Simulation optimization of dual resource constrained CONWIP controlled lines using response surface methodology

This thesis deals with the use of flexible workers in CONWIP controlled serial manufacturing lines having parallel service facilities. The main objective of this thesis is to propose a worker flexibility policy for this type of system under saturated demand. The primary issue in developing this worker flexibility policy is to decide on the length of periodic controls for evaluating the current system status. To determine the optimum level of this decision variable, we developed a two-stage procedure involving simulation optimization using Response Surface Methodology. In implementing this two-stage procedure, we used a cost-based performance measure. Hence, another objective of this thesis is to develop a unit penalty cost function. To evaluate the performance of the proposed policy, we compared it with other alternative policies widely used in Dual Resource Constrained systems. The simulation models which serve to evaluate the proposed and alternative policies have been developed in ARENA V2.2. Keywords: Dual Resource Constrained Systems (DRQ, CONWIP, Simulation Optimization, Response Surface Methodology (RSM).