An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach

Decarbonization of the world is greatly contributed to by the recent technological advancements that have fostered the development of electric vehicles (EVs). The EVs relieve transportation dependence on natural fossil fuels as an energy source. More than 50 % of the petroleum products produced worl...

Full description

Saved in:
Bibliographic Details
Main Authors: Cuthbert Ruseruka, Judith Mwakalonge, Gurcan Comert, Saidi Siuhi, Debbie Indah, Sarah Kasomi, Tumlumbe Juliana Chengula
Format: Article
Language:English
Published: Elsevier 2024-09-01
Series:Transportation Research Interdisciplinary Perspectives
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2590198224001970
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850283060910620672
author Cuthbert Ruseruka
Judith Mwakalonge
Gurcan Comert
Saidi Siuhi
Debbie Indah
Sarah Kasomi
Tumlumbe Juliana Chengula
author_facet Cuthbert Ruseruka
Judith Mwakalonge
Gurcan Comert
Saidi Siuhi
Debbie Indah
Sarah Kasomi
Tumlumbe Juliana Chengula
author_sort Cuthbert Ruseruka
collection DOAJ
description Decarbonization of the world is greatly contributed to by the recent technological advancements that have fostered the development of electric vehicles (EVs). The EVs relieve transportation dependence on natural fossil fuels as an energy source. More than 50 % of the petroleum products produced worldwide are estimated to be used in the transportation sector, accounting for more than 90 % of all transportation energy sources. Consequently, studies estimate that the transportation sector produces about 22 % of global carbon dioxide emissions, posing significant environmental issues. Thus, using EVs, particularly on road transport, is expected to reduce environmental pollution. To accelerate EV development and deployment, governments worldwide invest in EV development through various initiatives to make them more affordable. This research aims to investigate the changing needs of EV users to establish factors to be considered in the selection of charging demands using machine learning, using an extreme gradient boosting model. The model reached high accuracy, with an R2-Score of 0.964 to 1.000 across all predicted needs. The model performance is greatly affected by age, median income, education, and car ownership. High values of people with high income, high education, and age between 35–54 years show a positive contribution to the model’s performance, contrary to those with 65+, low income, and low education attainment. The outcomes of this research document factors that influence EV charging needs; therefore, it provides a basis for decision-makers and all stakeholders to decide where to locate EV charging stations for usability, efficiency, sustainability, and social welfare.
format Article
id doaj-art-ae681ea9c0584f82ad9cd4897999f610
institution OA Journals
issn 2590-1982
language English
publishDate 2024-09-01
publisher Elsevier
record_format Article
series Transportation Research Interdisciplinary Perspectives
spelling doaj-art-ae681ea9c0584f82ad9cd4897999f6102025-08-20T01:47:51ZengElsevierTransportation Research Interdisciplinary Perspectives2590-19822024-09-012710121110.1016/j.trip.2024.101211An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approachCuthbert Ruseruka0Judith Mwakalonge1Gurcan Comert2Saidi Siuhi3Debbie Indah4Sarah Kasomi5Tumlumbe Juliana Chengula6Department of Engineering, South Carolina State University, 300 College Avenue, Orangeburg, SC 29117, USA; Corresponding author.Department of Engineering, South Carolina State University, 300 College Avenue, Orangeburg, SC 29117, USANorth Carolina A&T State University, Computational Data Science and Engineering Department, 1601 E. Market Street, 519 McNair Hall Greensboro, NC 27411, USADepartment of Engineering, South Carolina State University, 300 College Avenue, Orangeburg, SC 29117, USADepartment of Engineering, South Carolina State University, 300 College Avenue, Orangeburg, SC 29117, USAHDR Inc., 76 S Laura Street, Suite 1600, Jacksonville, FL 32202, USADepartment of Engineering, South Carolina State University, 300 College Avenue, Orangeburg, SC 29117, USADecarbonization of the world is greatly contributed to by the recent technological advancements that have fostered the development of electric vehicles (EVs). The EVs relieve transportation dependence on natural fossil fuels as an energy source. More than 50 % of the petroleum products produced worldwide are estimated to be used in the transportation sector, accounting for more than 90 % of all transportation energy sources. Consequently, studies estimate that the transportation sector produces about 22 % of global carbon dioxide emissions, posing significant environmental issues. Thus, using EVs, particularly on road transport, is expected to reduce environmental pollution. To accelerate EV development and deployment, governments worldwide invest in EV development through various initiatives to make them more affordable. This research aims to investigate the changing needs of EV users to establish factors to be considered in the selection of charging demands using machine learning, using an extreme gradient boosting model. The model reached high accuracy, with an R2-Score of 0.964 to 1.000 across all predicted needs. The model performance is greatly affected by age, median income, education, and car ownership. High values of people with high income, high education, and age between 35–54 years show a positive contribution to the model’s performance, contrary to those with 65+, low income, and low education attainment. The outcomes of this research document factors that influence EV charging needs; therefore, it provides a basis for decision-makers and all stakeholders to decide where to locate EV charging stations for usability, efficiency, sustainability, and social welfare.http://www.sciencedirect.com/science/article/pii/S2590198224001970Utility of EV Charging NetworksElectric VehiclesAlternative Fuel VehiclesSHapley Additive exPlanations
spellingShingle Cuthbert Ruseruka
Judith Mwakalonge
Gurcan Comert
Saidi Siuhi
Debbie Indah
Sarah Kasomi
Tumlumbe Juliana Chengula
An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
Transportation Research Interdisciplinary Perspectives
Utility of EV Charging Networks
Electric Vehicles
Alternative Fuel Vehicles
SHapley Additive exPlanations
title An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
title_full An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
title_fullStr An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
title_full_unstemmed An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
title_short An Investigation of factors Influencing electric vehicles charging Needs: Machine learning approach
title_sort investigation of factors influencing electric vehicles charging needs machine learning approach
topic Utility of EV Charging Networks
Electric Vehicles
Alternative Fuel Vehicles
SHapley Additive exPlanations
url http://www.sciencedirect.com/science/article/pii/S2590198224001970
work_keys_str_mv AT cuthbertruseruka aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT judithmwakalonge aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT gurcancomert aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT saidisiuhi aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT debbieindah aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT sarahkasomi aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT tumlumbejulianachengula aninvestigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT cuthbertruseruka investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT judithmwakalonge investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT gurcancomert investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT saidisiuhi investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT debbieindah investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT sarahkasomi investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach
AT tumlumbejulianachengula investigationoffactorsinfluencingelectricvehicleschargingneedsmachinelearningapproach