EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3522. RORO Stowage Planning with Reinforcement Learning

Invited abstract in session WA-62: Machine Learning and Optimization in Ports I, stream OR in Port Operations.

Wednesday, 8:30-10:00
Room: S12 (building: 101)

Authors (first author is the speaker)

1. Alastair Main
Management, Technical University of Denmark
2. Filipe Rodrigues
DTU Management, Technical University of Denmark
3. Dario Pacino
Department of Management, Technology and Economics, Technical University of Denmark

Abstract

Reduction of Greenhouse Gas (GHG) emissions in the shipping industry has been of great focus over the past years. Within Roll-on/Roll-off (RORO) shipping, GHG emissions can be reduced by decreasing the turnaround time at port calls.
Decreasing the turnaround time for RORO vessels results in more time being available for sailing between port calls. As more time is allotted for sailing, the vessel can significantly reduce its speed and, thereby, fuel consumption.
If RORO vessels have several port calls, the cargo is handled following an approximate First-in-last-out queue. Therefore, in some cases, cargo to be discharged at the port can become stuck behind cargo not to be unloaded. This results in a necessary re-handling of cargo, increasing time spent at the port.
Optimizing the loading sequence, positioning of cargo, and cargo routing within the vessel reduces the number of cargo that needs to be re-handled.
Prior research has utilized mathematical models and heuristics to solve similar problems. The authors present preliminary results for using a reinforcement learning framework to optimize the RORO stowage planning problem. Leveraging reinforcement learning allows for the development of an algorithm that can provide executable actions at the port based on current information in real time.

Keywords

Status: accepted

Back to the list of papers