The electrification of transport plays a critical role in reducing the footprint of greenhouse gas (GHG) emission. The total number of electric buses in service is forecast to more than triple, from 386,000 in 2017 to about 1.2 million in 2025, equal to about 47% of the worldwide city bus fleet. However, electrification of transit buses faces many issues, such as massive capital investment for acquiring vehicles, reduced bus capacities because of space occupied by carrying batteries, and potential disrupted operations due to the short range of electric buses, to name a few. It is challenging for a public transit authority to plan the process of electrifying its bus fleet and continue to operate its mixed fleet cost-effectively. To complicate it even more, uncertainties exist during the electrification process, such as the evolution of battery technologies, change of ridership, and operational variations.
The objective of this proposed study is to provide a decision support tool to public transit authorities for facilitating the process of electrifying their transit buses. Specifically, given the periodical budget and transit network and features, the tool will provide outcomes at different stages including (1) which routes the acquired electric buses should serve; (2) where to deploy charging facilities (both plug-in at stations and dynamic wireless charging facilities embedded in road pavement); and (3) what should be the right size of onboard battery for a specific route. We call this tool “Multi-stage Planning for Electrifying Transit Bus Systems with Multi-format Charging Facilities.”
The decision support tool will help transit authorities select the routes that the acquired electric buses will serve and determine locations of charging facilities and on-board battery size to guarantee cost-efficiency, service continuity, and environmental friendliness under the constraints of each financial cycle’s limited budget. Two optimization sub-problems will be modeled. For route selection, based on historical operational data (passenger boarding and alighting, travel time, bus seat capacity, operational frequency and time duration, etc.) and charging facility locations, binary integer programming will be developed to select the most cost-effective routes to be served by the E-buses. The general cost will be a weighted sum of several KPIs of the transit system, subject to transit authority preference. For the integrated optimization of multi-format charging facility location and on-board battery size, bi-objective mixed integer linear programming will be developed. One objective function is the lifecycle cost of the designed charging system, which includes the construction cost of DWCFs, battery cost, and energy consumption cost. The other objective is the environmental impacts caused by the electrification of buses.
The proposed tool kit will help transit authorities to improve their key performance indicators (KPIs), including (1) ridership productivity (bus passengers per revenue hour), (2) efficiency (gross cost per revenue mile), (3) quality of service (complaints per 100,000 passengers and mean distance between vehicle failures), (4) on-time performance (1 minute early to 5 minutes late at scheduled time points), and (5) finance (growth of fund balance and fiscal sustainability).