Beyond Accuracy: Evaluating Forecasting Models by Multi-Echelon Inventory Cost
It addresses the need for data-driven decision-support tools in modern supply chains by integrating forecasting and inventory optimization, though it is incremental as it applies existing models to a new evaluation framework.
This study tackled the problem of evaluating forecasting models based on their impact on inventory costs rather than just accuracy, using the M5 Walmart dataset, and found that Temporal CNN and LSTM models significantly reduced inventory costs and improved fill rates compared to statistical baselines.
This study develops a digitalized forecasting-inventory optimization pipeline integrating traditional forecasting models, machine learning regressors, and deep sequence models within a unified inventory simulation framework. Using the M5 Walmart dataset, we evaluate seven forecasting approaches and assess their operational impact under single- and two-echelon newsvendor systems. Results indicate that Temporal CNN and LSTM models significantly reduce inventory costs and improve fill rates compared to statistical baselines. Sensitivity and multi-echelon analyses demonstrate robustness and scalability, offering a data-driven decision-support tool for modern supply chains.