Optimal Model for Supply Chain System Controlled by Kanban under JIT Philosophy by Integration of Computer Simulation and Genetic Algorithm

In this study, multi-stage supply chain system (SCS) controlled by kanban system is evaluated. In the kanban system, decision making is based on determination of batch size for each kanban. This paper attempts to simulate supply chain system regarding the costs under just-in-time (JIT) production philosophy. Since the adopted model is of backward type, the desired output is given in order to find the parameters and/or the structure of the model producing the output. This backward problem is non-analytic and often seems to be even more complex than the forward one. The paper applies Genetic Algorithm (GA) to optimize this simulation model. A simple real-coded GA is presented and used to change the simulation model parameters. With each new set of parameters, a simulation run is performed. From the statistics gathered by running the simulation, a goal function is constructed to measure the quality of these parameters. GA successfully provides a parameter set to demonstrate its capability to solve such difficult backward problems even in the area of complex simulation model optimization specially when there is no prior knowledge of simulation model behavior. Significance: Since supply chain management has drawn much attention in industrial and academic fields, various techniques are developed to model, analyze and solve complex decision making problems in supply chains. Computer-based simulation with its own strength on evaluating variations and interdependencies in a complex system is one of those promising techniques. This paper addresses the successful application of GA-simulation, which is a random search technique, to simulation model optimization and design, though the stochastic behavior of SCS. Hence, we have used kanban to achieve SCS control goals under JIT philosophy. Moreover, an optimization algorithm is coupled to the model that directly changes control parameters of the model to increase its performance