Understanding the Long-Term Evolution of Electric Taxi Networks
Guang Wang,Fan Zhang,Huijun Sun,Yang Wang,Desheng Zhang
DOI: https://doi.org/10.1145/3393671
IF: 5
2020-08-31
ACM Transactions on Intelligent Systems and Technology
Abstract:Due to the ever-growing concerns over air pollution and energy security, more and more cities have started to replace their conventional taxi fleets with electric ones. Even though environmentally friendly, the rapid promotion of electric taxis raises problems to both taxi drivers and governments, e.g., prolonged waiting/charging time, unbalanced utilization of charging infrastructures, and inadequate taxi supply due to the long charging time. In this article, we conduct the first longitudinal measurement study to understand the long-term evolution of mobility and charging patterns by utilizing 5-year data from one of the largest electric taxi networks in the world, i.e., the Shenzhen electric taxi network in China. In particular, (1) we first perform an electric taxi contextualization about their operation and charging activities; (2) then we design a generic charging event extraction algorithm based on GPS data and charging station data, and (3) based on the contextualization and extracted charging activities, we perform a comprehensive measurement study called ePat to explore the evolution of the electric taxi network from the mobility and charging perspectives. Our ePat is based on 4.8 TB taxi GPS data, 240 GB taxi transaction data, and metadata from 117 charging stations, during an evolution process from 427 electric taxis in 2013 to 13,178 in 2018. Moreover, ePat also explores the impacts of various contexts and benefits during the evolution process. Our ePat as a comprehensive measurement of the electric taxi network mobility and charging evolution has the potential to advance the understanding of the evolution patterns of electric taxi networks and pave the way for analyzing future shared autonomous vehicles.
computer science, information systems, artificial intelligence