报告题目：Values of Personalization in O2O On-Demand Delivery with Crowd-Sourced Drivers
报 告 人：代宏砚
报告地点：腾讯会议（会议ID：741 291 383）
In O2O (Online-to-Offline) on-demand services, customers place orders online and the O2O platform delivers products from the stores to the customers within one hour. The platform usually hires crowd-sourced drivers as a cost-effective option due to their flexibility. However, the delivery speed and delivery capacity of the crowd-sourced drivers vary a lot. This service inconsistency brings challenges in precisely matching the delivery supply and customer demand, which may significantly decrease the delivery efficiency. This paper aims to address the challenges by proposing a personalized dispatch model, which integrates both the order and drivers characteristics during the order assignment and routing decision making process. To achieve this, two machine learning-based models are proposed to forecast the individual drivers delivery speed in real time and customize the drivers delivery capacity dynamically, in order to develop a portrait of each driver’s behavior. Next, a personalized O2O order assignment and routing model is proposed with integration of the two portrait models. Furthermore, we validate our model with a real dataset of one mainstream O2O platform in China with 1230624 orders. Through comparison with actual routing decisions by the drivers, we show that the proposed personalized model can reduce the average delay by 21.60%. We also run a comprehensive simulation to show the improvement in terms of on-time rate and delay time, brought by each personalization characteristics, viz, the personalized delivery speed and the customized delivery capacity. The theoretical and numerical results can shed light on the delivery management of the O2O on-demand services.
腾讯会议 会议ID：741 291 383