Improving inventory management in an automotive supply chain: a multi-objective optimization approach using a genetic algorithm

Abstract

Inventory management represents a cornerstone inherent to any supply chain, regardless of industry type. Nevertheless, uncertainty phenomena related to demand and supply can induce overstock or even inventory stock-outs occurrences which, in turn, jeopardize one of the major principles of supply chain management: deliver the right product at the right place, at the right time and to the right cost. This situation may also be aggravated in automotive supply chains, due to their complexity in terms of entities involved. This research paper explores a multi-objective optimization model and applies it to a real industrial company, to address an inventory management problem. Moreover, a genetic algorithm is used to determine solutions corresponding to the order size and to a safety factor system. The obtained results are compared to the current strategy adopted by the company. At this point, the advantages and the drawbacks of the model implementation are assessed. Based on a set of logistic performance indicators, it is showed that the adoption of a smaller order size is potentially beneficial to the overall levels of inventory and to the value of inventory on–hand, without compromising the service level. Assertively, the proposed model reveals to be an useful tool to practitioners involved in automotive electronic supply chains.