HRHao Ren

Hao Ren

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hao Ren has a total of 4 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    An effective matching algorithm with adaptive tie-breaking strategy for online food delivery problem

    by Jing-Fang Chen, Ling Wang, Shengyao Wang, Xing Wang, Hao Ren

    With the prosperity of e-commerce, ordering food online has become increasingly prevalent nowadays. Derived from the dispatching problem in Meituan, a real online food delivery (OFD) platform in China, this paper addresses an OFD problem (OFDP). To solve the OFDP efficiently, an effective matching algorithm with adaptive tie-breaking strategy (MAATS) is proposed by collaboratively fusing the optimization methods with machine learning (ML) techniques. First, to efficiently generate a partial solution with a certain quality, a best-matching heuristic is proposed. Second, to break the ties occurring in the best-matching heuristic and obtain a complete solution with high quality, multiple tie-breaking operators are designed. Third, to adapt to different scenarios, the tie-breaking operators are utilized in a dynamic way which is achieved by using ML methods including decision trees and a specially-designed deep neural network. Fourth, problem-specific features are extracted as decision information to assist the ML models to predict the best tie-breaking operator for use in the current scenario. Preliminary offline simulations are carried out on real historical data sets to validate the effectiveness of the proposed algorithm. Moreover, rigorous online A/B tests are conducted to evaluate the performance of MAATS in practical applications. The results of offline and online tests demonstrate both the effectiveness of MAATS to solve the OFDP and the application value to improve customer satisfaction and delivery efficiency on Meituan platform.