Professor Dai Hongyan from Our School Published an Academic Paper in the Authoritative Journal “Journal of Management Science in China”

Recently, Professor Dai Hongyan from our School collaborated with Dr. Tao Jiawei from Tsinghua University, Professor Jiang Hai from Tsinghua University, and Professor Zhou Weihua from Zhejiang University to write the article "Paradigm shift for big data-driven decision making: New paradigm for O2O on-demand logistics.” This article was published in the authoritative academic journal "Journal of Management Science in China.”

The Journal of Management Science in China was founded by the Management Science Department of the NSFC in 1998 and is a national comprehensive academic journal in the field of management science. The journal is identified by the Management Science Department of the NSFC as an A-level significant journal in management science. It has been chosen as a "High-Quality Chinese Journal of FMS Management Science" and has been continuously ranked as the "Most Internationally Influential Chinese Academic Journal" for ten years from 2012 to 2021. In 2021, it was among the first to be selected into the Think Tank Journal Group 1.0.

The paper, which is set in the context of O2O instant service, a novel business model, examines the O2O instant logistics scheduling model under the new decision paradigm driven by big data. The paper integrates logistics sector, other operational sectors, and external environment information across domains to form a comprehensive data set. It also relaxes the classical assumptions of the traditional decision paradigm, achieving a shift from uniform delivery times to personalized delivery times, and from demand following a prior distribution to considering the spatiotemporal attributes of future demand. The paper combines machine learning and operational optimization methods to realize the O2O instant logistics scheduling model under the new decision paradigm. At the prediction level, the paper builds an individualized crowdsourcing delivery time prediction model and a demand scenario prediction algorithm based on the spatiotemporal similarity of order sets. At the decision level, it simultaneously considers the point estimate of the individualized model and its uncertainty, as well as the spatiotemporal distribution of future order sets. It constructs a scheduling model that accounts for prediction uncertainty and designs a synchronous prediction and decision algorithm for solution. The paper, in cooperation with China's mainstream O2O supermarket platforms, verifies the feasibility and effectiveness of the O2O instant logistics scheduling model under the new decision paradigm through simulation and validation based on real data.

Professor Dai Hongyan has engaged in extensive and in-depth research in areas such as data-driven optimization decisions, data-driven operational optimization, new retail supply chain management, and the Internet of Things, achieving abundant research outcomes. He has achieved abundant research outcomes, published over 40 articles, including in top domestic and international journals like MS, EJOR, Journal of Management Science in China, and Journal of Industrial Engineering and Engineering Management. He has also directed multiple national and provincial/ministerial key research projects, such as the NSFC’s Major Research Plan Cultivation Project and the NSFC’s General Program Projects.