Predictive Machine Learning to Increase the Throughput of Container Yards
Fuente:
arXiv
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914048217448448 |
|---|---|
| author | Cooper, Austin Ford |
| author_facet | Cooper, Austin Ford |
| contents | This study seeks to improve the throughput rates for shipping container terminals. In the United States, shipping ports link the domestic economy to global markets and are vital to sustain supply chain flow and economic stability. Maritime shipping accounts for nearly half of the U.S.'s annual international trade, two thirds of which are represented by container shipping. Previous studies highlighted the capability of automation in enhancing container processing; however, unlike in European and East Asian ports, full automation is limited in U.S. ports due to legal protections for human labor. Consequently, there is a need for alternative methods that deliver automation level efficiencies while maintaining the terms of cooperative agreements. This paper proposes an Intelligent Planning System (IPS) that applies the concept of Pareto Optimization to container yards through a mixed integer linear programming (MILP) based recursive appointment system. The results show an improvement from baseline for both daily terminal throughput volumes and processing times. The generated IPS can be employed to provide recommendations for container positioning and truck pickup appointments to optimize container yard layout and flow resulting in reduced realtime congestion and predictively mitigated future congestion. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16207 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Predictive Machine Learning to Increase the Throughput of Container Yards Cooper, Austin Ford Computational Engineering, Finance, and Science Optimization and Control 65K05 (primary) 90C11, 90C05 (secondary) This study seeks to improve the throughput rates for shipping container terminals. In the United States, shipping ports link the domestic economy to global markets and are vital to sustain supply chain flow and economic stability. Maritime shipping accounts for nearly half of the U.S.'s annual international trade, two thirds of which are represented by container shipping. Previous studies highlighted the capability of automation in enhancing container processing; however, unlike in European and East Asian ports, full automation is limited in U.S. ports due to legal protections for human labor. Consequently, there is a need for alternative methods that deliver automation level efficiencies while maintaining the terms of cooperative agreements. This paper proposes an Intelligent Planning System (IPS) that applies the concept of Pareto Optimization to container yards through a mixed integer linear programming (MILP) based recursive appointment system. The results show an improvement from baseline for both daily terminal throughput volumes and processing times. The generated IPS can be employed to provide recommendations for container positioning and truck pickup appointments to optimize container yard layout and flow resulting in reduced realtime congestion and predictively mitigated future congestion. |
| title | Predictive Machine Learning to Increase the Throughput of Container Yards |
| topic | Computational Engineering, Finance, and Science Optimization and Control 65K05 (primary) 90C11, 90C05 (secondary) |
| url | https://arxiv.org/abs/2509.16207 |