Assessing the Profit Impact of ARIMA and Neural Network Demand Forecasts in Retail Inventory Replenishment

Abstract

This study explores the integration of demand forecasting and inventory replenishment strategies to enhance retail profitability. Accurate sales forecasting is essential for efficient inventory replenishment decisions. Both traditional ARIMA and modern neural network models are utilized to predict future sales. These forecasts input into an integer programming model that strategically manages the inventory of stores across multiple retail routes. The optimization model considers transportation, sales loss, supply costs, and inventory dynamics to maximize retail profit with daily replenishment decisions. This approach enables us to assess the impact of forecasting accuracy on profitability over a multi-period planning horizon. The study is distinctive in its dual assessment: it evaluates both the accuracy of forecasting methods and their direct impact on profitability through systematic inventory decisions. Neural network architectures exhibit a 6% lower mean squared error compared to ARIMA models. For longer horizon predictions, the performance gap grows larger; for example, there is a 60% difference in predictions 15 days ahead. Predictions reflect 1.6% higher profits on average when neural network predictions and more efficient longer planning horizons of the optimization model are preferred. Planning 30 days ahead, optimizing with neural network predictions elicits 2.3% higher profits compared to those attainable based on ARIMA predictions. Our findings illustrate how different forecasting methods can affect firm profitability by shaping inventory replenishment strategies. By merging mathematical optimization with time series forecasting, this research provides a comprehensive evaluation of how advanced predictive technologies can enhance retail inventory practices and improve profitability. © 2024 Elsevier B.V., All rights reserved.