Modeling the Demand for Electric Vehicles in Canadian Corporate and Government Fleets

This dissertation investigates factors affecting the acquisition of Electric Vehicles (EVs) in the Canadian fleet market. Data from a random sample of over 1,000 fleet operating entities (FOEs) that owned and operated light fleets (i.e., cars, pickup trucks and utility vehicles) in Canadian cities were collected via an online survey titled Canadian Fleet Acquisition Survey (CFAS) in December 2016. The CFAS gathered information about the general characteristics of the surveyed FOEs, their existing fleet characteristics, future acquisition plans and EV fleet prospects. A stated preference (SP) section was introduced in the CFAS to identify the circumstances that will lead to higher adoption rates of EVs for fleet usage. The SP responses were based on six choice scenarios, each featuring four powertrains (Internal-Combustion Engine Vehicles, Hybrid Electric Vehicles, Plug-in Hybrid Electric Vehicles and Battery Electric Vehicles). The CFAS also included attitudinal statements to understand the issues that support or deter EV acquisition in fleets. Chapters 4, 5 and 6 of the dissertation are dedicated to employing various modeling approaches including Exploratory Factor Analysis (EFA), Analytical Hierarchy Process (AHP), and advanced discrete choice models such as Latent Class (LC) and Ordered Logit (OL) models to investigate the feasibility of EVs in fleets from various perspectives. This includes investigating EV adoption with respect to entity type (i.e., corporate vs. government), fleet type (car fleets vs. pickup truck fleets vs. utility vehicles fleets), industry type (transportation and warehousing vs. retail trade) as well as the temporal dimension for fleet electrification (i.e., short run vs. long run). The estimated EFA models identify latent constructs of behavior on various aspects and attitudes relating to EV adoption and provides evidence of attitudinal heterogeneity in the corporate and government FOEs. The AHP approach validates the logical consistency of the attitudinal responses obtained from the sampled FOEs. The four latent classes of FOEs identified in the estimated LC choice model provide novel results regarding the factors that affect acquisition of EVs in fleets. The willingness-to-pay estimates from the LC model reflect the taste variation among the four latent classes for improvements in certain attributes of EVs. The results from the OL modelling exercise successfully explain the behavior governing the acquisition timeframe for battery electric vehicles in the sampled FOEs and highlight the heterogeneity in the factors affecting the acquisition timeframe. Finally, evidence-based policy guidelines are proposed to help stakeholders make informed decisions regarding the acquisition of EVs in fleets. Key guidelines include investment in public charging infrastructure, incentivizing on-site charging infrastructure, engaging FOEs with climate action plan, and harvesting positive attitudes towards fleet electrification through various campaigns. The research work described in this dissertation is the first of its kind to collect and analyze revealed and state preference data on the acquisition of EVs in Canadian fleets including the timeframes under which these vehicles will likely be acquired. The work is seminal as it fills an important gap in the current knowledge about the motivations and preferences towards fleet electrification in Canada