Combining simulation with genetic algorithm for solving stochastic multi-product inventory optimization problem

All companies are challenged to match supply and demand, and the way the companytackles this challenge has a tremendous impact on its profitability. Due to the fact that markets are rapidlyevolving and becoming more complex, flexible, and information-intensive, notorious binging-andpurgingapproach is inappropriate. Scuh an approach, in which product is, firstly, overpurchased or overproducedin order to prepare for expected demand spikes and then discarded by sharp decline in price.Thus, in order to tailor inventory control to urgent industrial needs, the discrete-event simulation modelis proposed. The model is stochastic and operates with multiple products under constrained total inventorycapacity. Besides that, the model under consideration is distinguished by uncertain replenishmentlags and lost-sales. The paper contains both mathematical description and algorithmic implementation.Besides that, an optimization framework based on genetic algorithm is proposed for deriving an optimalcontrol policy. The proposed approach contributes to the field of industrial engineering by providing asimple and flexible way to compute nearly-optimal inventory control parameters.