Dynamic Pricing, Route Planning, and Order Assignment in Dada's Instant Logistics Platform

Background Dada, a leading instant logistics information service platform in China, fulfills a wide range of orders—including JD.com grocery, food‑delivery, e‑commerce, and long‑distance same‑city deliveries—using only crowd‑sourced couriers. To handle the complexity of these services, the platform relies heavily on AI technologies.

Dynamic Pricing Balancing supply and demand across spatial, temporal, and order dimensions is critical. Dada visualizes supply‑demand heat maps for each region, accounts for weather, holidays, and peak hours, and evaluates order difficulty. By applying real‑time, per‑order dynamic pricing, the platform aligns courier incentives, user experience, and overall profitability, achieving a “price‑per‑order” strategy.

Supply‑Demand Imbalance Dimensions • Spatial imbalance – heat‑maps of each area. • Temporal imbalance – variations caused by weather, festivals, and rush hours. • Order imbalance – each order’s unique difficulty influences dispatch decisions.

Route Planning Couriers often handle multiple orders simultaneously, making route optimization a core challenge. The problem is modeled as a Traveling Salesman Problem (TSP), which is NP‑hard. Dada employs a genetic algorithm as a heuristic to generate near‑optimal routes within milliseconds, even for more than ten stops, while maintaining high accuracy.

With the base algorithm, the system can quickly assess whether two orders are on the same route and determine the suitability of a courier for a given set of orders.

Order Assignment Dada uses a hybrid of order‑grabbing and dispatch mechanisms. Assignment considers route compatibility, courier preferences, capacity, activity level, and fairness, framing the problem as a constrained multi‑objective optimization. This approach enables most orders to be allocated within one minute of being placed.

Conclusion The logistics domain presents novel algorithmic challenges with highly non‑standardized inputs and multiple optimization goals, far exceeding typical internet‑scale problems. Advances in AI and the rise of new retail are driving rapid transformation, and Dada‑JD Daojia aims to stay at the forefront of this innovation.

Author Bio Liao Ruiqi – Head of Dada’s Delivery Algorithm Team, with extensive experience in machine learning, logistics algorithms, and computational advertising. Joined Dada in early 2016, built the algorithm team from scratch, and now leads work on order assignment, dynamic pricing, and route planning.