Inventory Policy Design under Seasonal Demand and Partial Sales Loss Conditions

Abstract

Inventory management plays a pivotal role in the smooth functioning of supply chains, particularly in industries experiencing seasonal fluctuations in demand. Seasonal demand patterns introduce complexity into inventory decisions, as demand peaks and troughs affect both the timing and volume of stock replenishment. Simultaneously, the phenomenon of partial sales loss—where a portion of customer demand is lost due to stockouts but may be partially recovered later—adds another layer of intricacy to policy design. Traditional inventory models often assume either complete backordering or total loss of sales, thereby failing to capture the nuanced realities of modern marketplaces.

This paper addresses the dual challenge of designing inventory policies that are both responsive to seasonal demand patterns and robust against partial sales loss conditions. A mathematical model is developed to optimize ordering policies by incorporating key parameters such as holding cost, ordering cost, demand seasonality, and the probability of partial sales recovery. The model aims to minimize total system cost while maintaining service levels that reflect customer expectations during seasonal peaks.

In addition, the study integrates forecasting techniques such as Holt-Winters and ARIMA to anticipate seasonal demand, providing a more dynamic foundation for decision-making. A numerical case study demonstrates the effectiveness of the proposed policy in reducing lost sales and optimizing inventory turnover. The findings offer practical insights for industries like apparel, agriculture, and consumer electronics, where seasonality and lost sales coexist frequently.