2594. Discovering Import Container Dwell Time Determinants in Multipurpose Maritime Terminals: A Multi-port Analysis
Invited abstract in session TD-32: Container Stacking and Yard Planning-2, stream Maritime and Port Logistics.
Tuesday, 14:30-16:00Room: Maurice Keyworth 1.09
Authors (first author is the speaker)
| 1. | Sebastian Muñoz-Herrera
|
| School of Engineering, Universidad del Desarrollo | |
| 2. | Rosa G. González-Ramírez
|
| Universidad de Los Andes, Chile | |
| 3. | Karol Suchan
|
| Escuela de Informática y Telecomunicaciones, Universidad Diego Portales |
Abstract
This study examines Container Dwell Time (CDT) determinants across three Chilean multipurpose maritime terminals. While CDT research in container-specialized terminals is established, knowledge gaps exist regarding multipurpose terminal dynamics, hindering effective policy development and accurate estimation models. Based on three-year import container data, this research incorporates temporal, geographical, operational-logistical, and specialized cargo attributes. The methodology combines Cramer's V association analysis, Multidimensional Scaling, and Classification/Regression methods to identify key CDT determinants and generate good predictions. Results reveal significant differences in CDT determinants across terminals, showing that similar operational characteristics yield different CDT outcomes based on terminal-specific cargo profiles, operational rules, and regional economic markets. Findings indicate that accurate CDT estimation requires terminal-specific modeling rather than relying on generic prediction methods. Each terminal exhibits unique operational patterns influenced by local constraints, cargo composition, and market demands. The research contributes to terminal efficiency optimization literature by providing evidence-based insights for tailored CDT management strategies, thereby enhancing resource allocation, improving planning capabilities, and supporting more effective policy development for multipurpose maritime terminals.
Keywords
- Maritime applications
- Analytics and Data Science
- Machine Learning
Status: accepted
Back to the list of papers