Predict-then-Optimize for Seaport Power-Logistics Scheduling: Generalization across Varying Tasks Stream

TL;DR

This paper introduces a continual learning framework for seaport power-logistics scheduling that improves decision quality and generalizes across varying tasks by using Fisher information regularization and a differentiable surrogate.

Contribution

It proposes a novel decision-focused continual learning method with Fisher information regularization for better generalization in dynamic seaport scheduling tasks.

Findings

Outperforms existing methods in decision accuracy

Demonstrates strong generalization across task streams

Reduces computational costs during training

Abstract

Power-logistics scheduling in modern seaports typically follow a predict-then-optimize pipeline. To enhance the decision quality of forecasts, 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 only a specific task structure, and thus generalize poorly to evolving tasks induced by varying seaport vessel arrivals. We address this gap with a decision-focused continual learning framework that adapts online to a stream of scheduling tasks. Specifically, we introduce Fisher information based regularization to enhance cross-task generalization by preserving parameters critical to prior tasks. A differentiable convex surrogate is also developed to stabilize gradient backpropagation. The proposed approach enables learning a decision-aligned forecasting model across a varying tasks stream with a sustainable long-term computational burden. Experiments calibrated to the Jurong Port demonstrate superior decision performance and generalization over existing methods with reduced computational cost.