2025. Enhancing Supply Lead Time Prediction with a Novel Hybrid Machine Learning Approach: Insights from a Retail Case Study
Invited abstract in session MA-28: Supply Chain and Logistics Management, stream Decision Support Systems.
Monday, 8:30-10:00Room: Maurice Keyworth 1.03
Authors (first author is the speaker)
| 1. | Kezban Mutlu
|
| Industrial Engineering, Marmara University |
Abstract
In supply chain management (SCM), supply lead time (SLT) is a crucial factor influencing inventory management and overall supply chain performance. Conventional methods often struggle to capture the complex, multivariate nature of SLT prediction, limiting their effectiveness in dynamic environments. This study proposes a novel hybrid machine learning approach, integrating supervised and unsupervised techniques to enhance predictive accuracy. Principal Component Analysis (PCA) is employed for dimensionality reduction and feature extraction, while the Gradient Boosted Trees (GBT) algorithm serves as the primary regression model alongside various other machine learning regression techniques. Using historical data from a retail company, the hybrid model achieved a superior predictive accuracy of 93% and a mean absolute error of 1.46 days, outperforming models without PCA transformation. These results demonstrate the efficacy of combining PCA and GBT for improving SLT estimation, providing valuable insights for inventory management and supply chain optimization. The study underscores the potential of advanced machine learning techniques in addressing real-world supply chain challenges and advancing data-driven decision-making processes.
Keywords
- Machine Learning
- Decision Support Systems
- Supply Chain Management
Status: accepted
Back to the list of papers