Optimization of Inventory Costs under Exponential Deterioration, Shortage, Inflation, and Power Dependent Demand using AI- Based Techniques

Abstract

This research introduces dynamic inventory model that combines exponential linear power dependent demand, in which the demand rate depends on the selling price. The model describes the reality of market behaviour in that items that decay with time are taken into account and demand changes with the decisions made on prices. In order to reflect the current economic environment, inflation is also factored in by use of present value discounting methods to allow a better evaluation of the inventory related expenses throughout the planning time frame. Particle Swarm Optimization (PSO) algorithms based on Artificial Intelligence (AI) are further used to optimize the model to find the cost minimizing inventory policy. Shortages are also allowed and are assumed to be fully backlogged, and therefore the model is applicable when delayed fulfilment is tolerated in industries. The entire numerical example is presented to show the structure of calculations and prove the efficiency of the methodology. Besides that, an analysis of the effect of the most important parameters on the total cost of the system is carried out through a sensitivity analysis; these parameters include the rate of deterioration and inflation, demand exponent, holding cost and shortage cost. The findings indicate that the joint impact of deterioration, inflation and price-sensitive demand play a significant role in optimal ordering decisions. The optimization with the use of AI provides more powerful and efficient solutions compared to the traditional mathematical methods. In general, this research may be useful to managers and practitioners who need to address perishable or deteriorating products in inflationary conditions so that they could implement cost-efficient and evidence-based decision-making procedures.