Researchers have developed EXHOLD, a novel two-stage framework for optimizing ride-hailing matching systems. This system aims to improve both passenger and driver experience by intelligently delaying certain driver-order pairings. The framework first assesses pairs based on aggregated satisfaction signals and then determines an optimal hold-time schedule to maximize overall experience while adhering to service guardrails. EXHOLD has been successfully deployed in DiDi's production system in Brazil, demonstrating significant improvements in trip completion, driver income, and a reduction in passenger cancellations. AI
IMPACT This framework could set a new standard for optimizing matching algorithms in large-scale service industries, improving efficiency and user satisfaction.
RANK_REASON The cluster describes a research paper detailing a new framework and its deployment in a production system.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
