Overall Research Project Description Despite billions of dollars of federal investments that promote transportation electrification, rural communities are still largely left behind with very low EV penetration and few charging resources. EV usage in rural communities may be more beneficial for environmental, social, and economic sustainability due to the longer driving distances and high driving frequencies of rural communities for different personal and professional uses (e.g., for purchasing daily necessities, recreation, work, and transporting agricultural implements, etc). Therefore, adopting EVs in rural communities not only helps improve environmental benefits but also has the potential to promote an equitable transition to clean energy and mitigate the wealth disparity between rural communities and urban communities. In this project, we propose to develop a data-driven socially informed framework to help promote EV adoption for rural innovation towards environmental, social, and economic sustainability. In doing so, we aim to address multiple complicated technical and real-world challenges by performing the following tasks.
(i) Investigating drivers and barriers to EV adoption in rural communities. We plan to conduct comprehensive analyses of barriers and drivers from different stakeholders’ perspectives (e.g., rural residents, charging infrastructure operators, utility agencies, and governments). We will use mixed methods, including surveys, panels, and big data analytics.
(ii) Predicting charging demand in rural areas. Based on our accessed large-scale fine-grained human mobility data and socioeconomic data from both urban and rural areas, combined with EV charging data from urban areas, we plan to extract social activity patterns (e.g., visitation frequency and dwelling time) and charging patterns (e.g., charging duration and frequencies). We then plan to predict charging demand in rural areas based on rural mobility patterns, socioeconomic status, and charging patterns in urban areas.
(iii) Combining the predicted charging demand with other real-world conditions like population and POI distributions, we plan to design a decision-support tool to help deploy charging infrastructure by balancing different practical factors, such as (a) improving charging infrastructure accessibility to satisfy charging demand of EVs, (b) increasing charging resource utilization to improve the profitability of operators, and (c) reducing potential impacts on power grids, etc.