Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
A Computing Framework for Modeling the Functional Relationship Between Electric Vehicle Uptake and Demographic Characteristics at a Granular Level Throughout the U.S
Abstract
Though Electric Vehicles (EVs) have seen substantial growth in the last decade, this growth has been far from uniform geographically. In this dissertation, a computing framework model is developed to analyze the functional relationship between the many quantifiable demographic factors that characterize an area at the ZIP code level and the EV uptake in those regions. This research fills an important knowledge gap of EV uptake, both by analyzing it at a finer geographic granularity, and by discovering the demographic factors that correlate with this growth. Heteroskedasticity and multi-collinearity exist for the independent and dependent variables, because of which new composite variables are generated using different constraints and hypothesis testing. A total of 82 demographic features are selected or engineered from census data, and evaluated as independent variables in regression models for the prediction of EV uptake at the ZIP code level. Using data from 11 states across the U.S., several established machine learning algorithms are compared, and recursive feature elimination is applied to reduce the input features and maintain prediction accuracy. The XGBoost regression model is determined to be the best-performing model, with a 20-feature subset having an adjusted R2 of 0.88. Shapley values are used to quantify the sensitivity of EV uptake to individual features in the regression model. Housing units where residents earn more than $75k and spent more than 30% on housing is the most impactful feature. In general, features describing population with income greater than $75k have a positive impact on EV uptake, whereas population having income less than $75k, population having zero vehicles, and population commuting less than 60 minutes have a negative impact on EV uptake. This framework provides localized information on the functional relationship of demographic features to EV uptake in any ZIP code for many potential applications. The prediction model is applied in Nebraska as a test-case study, where the EV uptake is unknown at the ZIP code level.
Subject Area
Architectural engineering|Automotive engineering|Energy
Recommended Citation
Shom, Subhaditya, "A Computing Framework for Modeling the Functional Relationship Between Electric Vehicle Uptake and Demographic Characteristics at a Granular Level Throughout the U.S" (2023). ETD collection for University of Nebraska-Lincoln. AAI30487615.
https://digitalcommons.unl.edu/dissertations/AAI30487615