Simulation-based optimization of parameters in (s,S) inventory control system

Inventories in manufacturing companies have a significant impact on the costs of the company. Companies can determine an inventory control policy appropriate to their constraints in order to increase their profitability and reduce their costs. With effective stock control, companies can continue their production without interruption and without holding excessive stock, reduce their costs, and regularly balance their stock for cost reports. In this study, a single-item inventory control model was stochastically modelled with the simulation method in a manufacturing company. Using a genetic algorithm, simulated annealing algorithm, and simulation optimisation tool integrated with the simulation program, the inventory control parameters that give the optimum total inventory cost were tried to be determined, and the results were compared. The parameters used in genetic algorithms and simulated annealing algorithms are essential for the effective operation of the algorithm. Therefore, a series of experiments were conducted with parameters to improve the performance of genetic algorithm and simulated annealing algorithms. In order to examine how inventory costs are affected under different conditions, the program was run for varying levels of order quantity, lead time and time between arrivals, and the total inventory cost values obtained were compared. Afterwards, comparative experiments were conducted to examine the effects of these factors.