Sustainable Inventory Management Model for High-Volume Material with Limited Storage Space under Stochastic Demand and Supply

Inventory management and control has become an important management function, which is vital in ensuring the efficiency and profitability of a company’s operations. Hence, several research studies attempted to develop models to be used to minimise the quantities of excess inventory, in order to reduce their associated costs without compromising both operational efficiency and customers’ needs. The Economic Order Quantity (EOQ) model is one of the most used of these models; however, this model has a number of limiting assumptions, which led to the development of a number of extensions for this model to increase its applicability to the modern-day business environment. Therefore, in this research study, a sustainable inventory management model is developed based on the EOQ concept to optimise the ordering and storage of large-volume inventory, which deteriorates over time, with limited storage space, such as steel, under stochastic demand, supply and backorders. Two control systems were developed and tested in this research study in order to select the most robust system: an open-loop system, based on direct control through which five different time series for each stochastic variable were generated, before an attempt to optimise the average profit was conducted; and a closed-loop system, which uses a neural network, depicting the different business and economic conditions associated with the steel manufacturing industry, to generate the optimal control parameters for each week across the entire planning horizon. A sensitivity analysis proved that the closed-loop neural network control system was more accurate in depicting real-life business conditions, and more robust in optimising the inventory management process for a large-volume, deteriorating item. Moreover, due to its advantages over other techniques, a meta-heuristic Particle Swarm Optimisation (PSO) algorithm was used to solve this model. This model is implemented throughout the research in the case of a steel manufacturing factory under different operational and extreme economic scenarios. As a result of the case study, the developed model proved its robustness and accuracy in managing the inventory of such a unique industry