How AI Optimizes Information Disclosure in Mobility‑On‑Demand Logistics

Recently, the International Conference on Knowledge Discovery and Data Mining (KDD) announced its accepted papers, selecting 138 out of 705 submissions (acceptance rate below 20%).

Among the accepted works, a paper from Huolala titled “Improving the Information Disclosure in Mobility‑on‑Demand Systems” stood out for its data‑driven approach to the intelligent order‑dispatch problem in internet logistics.

The current Huolala dispatch model uses a global broadcast: all orders within a time/distance window are pushed to all drivers, who then manually select orders. This leads to many drivers competing for the same high‑price orders, causing inefficient driver choices and reduced overall platform efficiency.

The proposed framework models the information‑disclosure process and solves an optimization problem that selects the most suitable driver for each order, thereby improving platform efficiency, user experience, and driver satisfaction.

The method is not limited to Huolala; it can be applied to any resource‑constrained recommendation system. As Huolala’s CTO Zhang Hao stated, “Technology changes logistics,” and the company believes that AI can bring revolutionary efficiency gains.

Huolala has been investing heavily in technical talent, hiring experts with backgrounds at Uber, Microsoft, and academic institutions, and plans to spend 1.5 billion RMB in 2021 to recruit over 1,000 engineers in big‑data, algorithm, and intelligent scheduling fields.