Optimizing Spare Parts Inventory Management Using Genetic Algorithm

Having spare components on hand is essential to keeping systems operating at their best. Many scholars have looked into problems with inventory management for spare parts over the last decades. Proving that (s,S) policies are optimal for specific dynamic inventory models with random periodic requests has been a major contribution to inventory theory. The widely used maintenance models make the assumption that there is an endless supply of replacement parts. Nonetheless, in certain real-life scenarios, the availability of spare components plays a significant role in choosing the best course of action for the entire system. The main goal from this research is to minimize management expenses by identifying the most efficient method for figuring out how much spare parts to order at each period. These expenses cover a range of charges related to inventory management. In order to reach all that, the study suggests a mathematical framework which relies on a Markov decision model, and then solve the optimality equation and derive optimal (s, S) policies through the application of the recurrence approach. As for the part of numeric, the paper is also presenting a new method that introduces genetic algorithms for the maximization of the service levels and the minimization of the expenses at the same time. After all, this work aims to establish strong inventory management plans for a variable periodic demand inventory model, which will guarantee effective resource allocation and operational excellence.