|
Forecasting warehouse volume using machine learning: a case study in sporting goods logistics |
|---|---|
| รหัสดีโอไอ | |
| Title | Forecasting warehouse volume using machine learning: a case study in sporting goods logistics |
| Creator | Sripornchai Kolsrichai |
| Contributor | Warut Pannakkong, Advisor |
| Publisher | Thammasat University |
| Publication Year | 2568 |
| Keyword | Warehouse volume forecasting, Machine learning, Artificial neural network, Support vector machine, Decision tree, Sporting goods logistics |
| Abstract | Accurate warehouse volume forecasting plays a critical role in modern logistics, where fluctuations in inbound and outbound flow directly affect capacity planning, resource allocation, and operational efficiency. This study aims to develop machine learning–based forecasting models for a sporting goods distribution center, focusing on three key forecasting perspectives: weekly shipped quantity, weekly order generation, and daily seven-day-ahead order generation. These three perspectives represent different stages of warehouse activity, where order generation reflects early customer demand entering the system and shipped quantity represents the final processed volume.Using historical data from 2023 to 2025, three machine learning techniques—Artificial Neural Network (ANN), Support Vector Machine (SVM), and Decision Tree (DT)—were applied within a rolling forecasting framework to simulate real operational conditions. The models were evaluated on their ability to capture temporal patterns and adapt to changing demand behaviors.The findings indicate that weekly forecasting provides stable and practical insights into warehouse volume trends, with shipped quantity demonstrating more consistent behavior than order generation. Daily forecasting, however, exhibited high variability and was not suitable for the current dataset. Overall, the study highlights the potential of machine learning models to support warehouse volume prediction and provides a foundation for improved planning and decision-making in logistics operations. |
