Decision-Focused Continual Learning for Seaport Power-Logistics Scheduling: Generalization across Varying Tasks
arXiv:2511.07938v3 Announce Type: replace Abstract: Power-logistics scheduling in modern seaports typically follows a predict-then-optimize pipeline. To enhance the decision quality of predictions, decision-focused learning has been proposed, which aligns the training of forecasting models with downstream decision outcomes. However, this end-to-end design inherently restricts the value of forecasting models to a specific task structure and therefore generalizes poorly to evolving tasks induced by varying vessel arrivals. We address this gap with a decision-focused continual learning framework
This research addresses the current limitations of decision-focused learning models, which struggle with dynamic environments, a critical challenge for real-world AI deployment.
Improving AI's ability to adapt to varying operational conditions will lead to more robust, autonomous, and scalable AI systems, directly impacting efficiency and decision-making in complex logistical operations.
The proposed decision-focused continual learning framework allows AI to generalize more effectively across diverse scenarios, enhancing its practical utility beyond static task structures.
- · Logistics companies
- · Seaport operators
- · AI/ML developers
- · Supply chain management
- · Inefficient scheduling methods
- · Static AI models
Increased automation and efficiency in port operations and supply chain logistics due to more adaptive AI models.
Reduced operational costs and improved throughput in global shipping, potentially alleviating supply chain bottlenecks.
Enhanced resilience and responsiveness of critical infrastructure to unforeseen disruptions, leveraging advanced AI for adaptive management.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG