Ensemble Learning for Demand Forecast of After-Market Spare Parts to Empower Data-Driven Value Chain and an Empirical Study

The bottlenecks of supply chains are shifting as individual companies are adopting smart production and reducing their inventories. Since the after-market orders mainly depend on repair and maintenance, addressing demand fluctuation is critical to reduce the risks of oversupply and shortage for sustainable supply chain. Intelligent data-driven technologies should be developed to promote value integration and value co-creation among supply chain partners for digital transformation. The shortening product life cycle and the reducing lot sizes of diverse products have increased the challenges of demand forecast and supply chain resilience. This study aims to develop stacking ensemble model to classify demand patterns and build individual forecasting models to improve forecasting performance. This study develops an alarm system to monitor the reliability of the model and a systematic adapt mechanism to enhance decision quality in real settings. An empirical study is conducted in a leading automotive after-market component manufacturer for validation. The results have shown the forecast errors and the total cost can be effectively reduced by the developed solution