Regime-Calibrated Fleet Repositioning with a Spatial Queue-Regret Decomposition

TL;DR

This paper introduces a regime-calibrated fleet repositioning method using a spatial queue-regret decomposition, improving wait times in ride-hailing scenarios through demand matching and stable queueing models.

Contribution

It develops a leakage-safe similarity gate, a spatial queue-regret decomposition, and evaluates these in a simulator, demonstrating improved wait times over baselines.

Findings

Spatial gate reduces mean wait to 82.3s from 85.3s and 85.8s in baselines.

Calibrated-demand gate outperforms hand-tuned similarity and distributional baselines.

Rebalancing controllers like scenario chance-MPC improve wait times compared to Wen-style rebalancing.

Abstract

Ride-hailing and autonomous mobility-on-demand operators reposition idle supply before future demand is fully observed. We study a retrieval-calibrated predict-then-optimize approach for this problem: historical demand regimes are matched to the current query block, combined into a calibrated demand prior, and passed to a fleet-balancing controller. The paper makes three contributions. First, we train a leakage-safe similarity gate whose objective penalizes demand error, pickup spatial mismatch, and queue shortage risk rather than retrieval rank alone. Second, we develop a spatial queue-regret decomposition for a stable queueing surrogate, linking demand-field error to wait through queueing sensitivity, allocator sensitivity, and Wasserstein pickup mismatch. Third, we evaluate learned retrieval and external-style rebalancing baselines in a common simulator. In the calibrated-demand gate experiment, across eight New York City scenarios and ten seeds, the spatial gate reduces mean wait to 82.3s, compared with 85.3s for hand-tuned similarity and 85.8s for a distributional-only baseline. In a separate replay-demand controller comparison, a scenario chance-MPC analog and a share-target transportation LP improve on Wen-style rebalancing (92.2s/92.2s vs. 100.1s), a reduced GPR chance-MPC comparator is intermediate at 94.4s, and an oracle MPC diagnostic is 91.3s.