Physics-informed offline reinforcement learning eliminates catastrophic fuel waste in maritime routing

TL;DR

PIER is a physics-informed offline reinforcement learning framework that significantly reduces catastrophic fuel waste and CO2 emissions in maritime routing by learning safety-aware, fuel-efficient policies from historical data without online simulation.

Contribution

The paper introduces PIER, a novel offline RL framework that eliminates catastrophic fuel waste in maritime routing using physics-calibrated environments and a safety shield, outperforming traditional methods.

Findings

Reduces mean CO2 emissions by 10% compared to great-circle routing.

Cuts extreme fuel waste episodes from 4.8% to 0.5%, a 9-fold reduction.

Lower per-voyage fuel variance by 3.5x, maintaining performance under forecast uncertainty.

Abstract

International shipping produces approximately 3% of global greenhouse gas emissions, yet voyage routing remains dominated by heuristic methods. We present PIER (Physics-Informed, Energy-efficient, Risk-aware routing), an offline reinforcement learning framework that learns fuel-efficient, safety-aware routing policies from physics-calibrated environments grounded in historical vessel tracking data and ocean reanalysis products, requiring no online simulator. Validated on one full year (2023) of AIS data across seven Gulf of Mexico routes (840 episodes per method), PIER reduces mean CO2 emissions by 10% relative to great-circle routing. However, PIER's primary contribution is eliminating catastrophic fuel waste: great-circle routing incurs extreme fuel consumption (>1.5x median) in 4.8% of voyages; PIER reduces this to 0.5%, a 9-fold reduction. Per-voyage fuel variance is 3.5x lower (p<0.001), with bootstrap 95% CI for mean savings [2.9%, 15.7%]. Partial validation against observed AIS vessel behavior confirms consistency with the fastest real transits while exhibiting 23.1x lower variance. Crucially, PIER is forecast-independent: unlike A* path optimization whose wave protection degrades 4.5x under realistic forecast uncertainty, PIER maintains constant performance using only local observations. The framework combines physics-informed state construction, demonstration-augmented offline data, and a decoupled post-hoc safety shield, an architecture that transfers to wildfire evacuation, aircraft trajectory optimization, and autonomous navigation in unmapped terrain.