Big Data for Urban Sustainability: Integrating Personal Mobility Dynamics in Environmental Assessments
To alleviate fossil fuel use, reduce air emissions, and mitigate climate change, “new mobility” systems start to emerge with technologies such as electric vehicles, multi-modal transportation enabled by information and communications technology, and car/ride sharing. Current literature on the environmental implications of these emerging systems is often limited by using aggregated travel pattern data to characterize personal mobility dynamics, neglecting the individual heterogeneity. Individual travel patterns affect several key factors that determine potential environmental impacts, including charging behaviors, connection needs between different transportation modes, and car/ride sharing potentials. Therefore, to better understand these systems and inform decision making, travel patterns at the individual level need to be considered. Using vehicle trajectory data of over 10,000 taxis in Beijing, this research demonstrates the benefits of integrating individual travel patterns into environmental assessments through three case studies (vehicle electrification, charging station siting, and ride sharing) focusing on two emerging systems: electric vehicles and ride sharing. Results from the vehicle electrification study indicate that individual travel patterns can impact the environmental performance of fleet electrification. When battery cost exceeds $200/kWh, vehicles with greater battery range cannot continuously improve travel electrification and can reduce electrification rate. At the current battery cost of $400/kWh, targeting subsidies to vehicles with battery range around 90 miles can achieve higher electrification rate. The public charging station siting case demonstrates that individual travel patterns can better estimate charging demand and guide charging infrastructure development. Charging stations sited according to individual travel patterns can increase electrification rate by 59% to 88% compared to existing sites. Lastly, the ride sharing case shows that trip details extracted from vehicle trajectory data enable dynamic ride sharing modeling. Shared taxi rides in Beijing can reduce total travel distance and air emissions by 33% with 10-minute travel time deviation tolerance. Only minimal tolerance to travel time change (4 minutes) is needed from the riders to enable significant ride sharing (sharing 60% of the trips and saving 20% of travel distance). In summary, vehicle trajectory data can be integrated into environmental assessments to capture individual travel patterns and improve our understanding of the emerging transportation systems.
Charging station siting, Electric vehicles, Environmental assessments of emerging transportation systems, Individual travel patterns, Ride sharing, Vehicle trajectory data
Cai, Hua. (2015) “Big Data for Urban Sustainability: Integrating Personal Mobility Dynamics in Environmental Assessments.” Doctoral Dissertation, University of Michigan: Ann Arbor. 1-104.