Harvesting Efficient On-Demand Order Pooling from Skilled Couriers: Enhancing Graph Representation Learning for Refining Real-time Many-to-One Assignments

TL;DR

This work tackles real-time many-to-one assignment in on-demand food delivery by introducing the SC Delivery Network (SCDN), which leverages skilled courier trajectories to reveal latent order-pooling opportunities. It integrates an enhanced graph representation learning framework, EATNE, built on an attributed multiplex heterogeneous network (AMHEN) and regional negative sampling with a margin-ranking objective to produce robust FU embeddings. The system uses these embeddings to prune the huge MOA search space, guiding constructive heuristics and an imitation-learning-based MAKER (ILIMA) approach, while identifying scale-effect hotspots (SEHs) for focused deployment. Online experiments on Meituan Waimai show meaningful gains: a 5.3% improvement in MD scores, 51% cut in pickup time, 21% cut in delivery time, and a 45–55% boost in courier efficiency during peak hours, validating both the method and its real-world impact.

Abstract

The recent past has witnessed a notable surge in on-demand food delivery (OFD) services, offering delivery fulfillment within dozens of minutes after an order is placed. In OFD, pooling multiple orders for simultaneous delivery in real-time order assignment is a pivotal efficiency source, which may in turn extend delivery time. Constructing high-quality order pooling to harmonize platform efficiency with the experiences of consumers and couriers, is crucial to OFD platforms. However, the complexity and real-time nature of order assignment, making extensive calculations impractical, significantly limit the potential for order consolidation. Moreover, offline environment is frequently riddled with unknown factors, posing challenges for the platform's perceptibility and pooling decisions. Nevertheless, delivery behaviors of skilled couriers (SCs) who know the environment well, can improve system awareness and effectively inform decisions. Hence a SC delivery network (SCDN) is constructed, based on an enhanced attributed heterogeneous network embedding approach tailored for OFD. It aims to extract features from rich temporal and spatial information, and uncover the latent potential for order combinations embedded within SC trajectories. Accordingly, the vast search space of order assignment can be effectively pruned through scalable similarity calculations of low-dimensional vectors, making comprehensive and high-quality pooling outcomes more easily identified in real time. SCDN has now been deployed in Meituan dispatch system. Online tests reveal that with SCDN, the pooling quality and extent have been greatly improved. And our system can boost couriers'efficiency by 45-55% during noon peak hours, while upholding the timely delivery commitment.

Figures (16)

• Figure 1: A courier's concurrent execution and route sequence of four orders.
• Figure 2: Order pooling examples.
• Figure 3: Calculation volume and search space for modeling and solving MOA problems in each dispatch cycle.
• Figure 4: Illustration of AMHEN Construction, including 2 sessions. Session A contains 3 orders for FUs DE, FB and FC. The pick-up FU sequence is DE->FC->FB. And the delivery FU sequence is DE->FB->FC. Session B follows the same process.
• Figure 5: Illustration of the GRL Model.