EXHOLD: Experience-Aware Real-Time Hold Control for Large-Scale Ride-Hailing Matching at DiDi
2026-07-10 • Machine Learning
Machine Learning
AI summaryⓘ
The authors present EXHOLD, a system designed to improve ride-hailing by smartly deciding when to delay matching drivers with passengers. Instead of using simple rules, EXHOLD splits the task into two steps: first, it rates each driver-passenger pair based on overall experience goals, then it sets how long to wait before confirming matches to avoid unnecessary delays. Tested in Brazil with DiDi, EXHOLD showed better trip completion, higher driver earnings, fewer cancellations, and smoother matching. The authors found that both steps are important and the system adapts well to changing conditions. EXHOLD is currently in use in production.
ride-hailinghold controldriver-passenger matchingexperience-aware decision makingmulti-objective optimizationrandomized A/B testingconstrained optimizationmarketplace efficiencycancellation ratespatiotemporal heterogeneity
Authors
Xu Liu, Kai Wan, Zihao Lu
Abstract
In large-scale ride-hailing, hold control is a critical mechanism for improving passenger-driver experience. By selectively delaying certain driver-order pairs, the system waits for better opportunities, reduces cancellations, and mitigates wasted driver effort. However, existing industrial hold strategies often rely on heuristic thresholding over multiple predictive models, which can be brittle under non-stationary traffic and hard to optimize for multi-objective experience signals. We propose EXHOLD, a deployable two-stage framework decoupling experience-aware pair assessment from hold-time execution. In Stage I, we learn a decision model assigning each driver-order pair to discrete, interpretable experience tiers by optimizing a unified objective that aggregates satisfaction signals across the matching funnel. In Stage II, we solve for a monotone hold-time schedule via constrained optimization over empirical quantiles. This explicitly enforces service guardrails bounding the unnecessary holding of promising matches while maximizing overall experience improvement. We evaluate EXHOLD through randomized A/B experiments in DiDi's production system in Brazil. Results show consistent gains in marketplace efficiency and experience: EXHOLD increases trip completion and driver income, significantly reduces passenger cancellations, and improves funnel efficiency. Ablations and behavioral analyses confirm both stages are essential and that the policy makes calibrated decisions under spatiotemporal heterogeneity. EXHOLD is currently deployed, serving production traffic in Brazil.