Forecasting Inland Waterway Container on Barge Volume: A Machine Learning Approach Using Economic Features
This study presents a machine learning approach to predict Container-on-Barge (COB) volume in Inland Waterway Transportation (IWT) systems, focusing exclusively on using economic features as predictors. Five machine learning models were trained using European economic features to forecast COB volume...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Applied Artificial Intelligence |
| Online Access: | https://www.tandfonline.com/doi/10.1080/08839514.2025.2550462 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study presents a machine learning approach to predict Container-on-Barge (COB) volume in Inland Waterway Transportation (IWT) systems, focusing exclusively on using economic features as predictors. Five machine learning models were trained using European economic features to forecast COB volume, while historical COB volume was used solely for validation and hyperparameter tuning. Among these models, the convolutional neural network combined with long short-term memory (CNN-LSTM) exhibited superior performance, achieving a mean absolute percentage error (MAPE) of 1.08% when forecasting eight consecutive quarters of COB volume in Europe. The results demonstrate the feasibility of accurately forecasting COB volume using economic features. This research develops an alternative method to forecast COB volume and provides a foundation for developing transfer learning models to predict COB volume in other emerging markets where historical COB volume data is limited. The findings are expected to assist in strategic planning and infrastructure investment for efficient and sustainable COB IWT systems. |
|---|---|
| ISSN: | 0883-9514 1087-6545 |