Predicting Urban Commuters’ Transit Mode Preferences Using XGBoost Algorithms to Build Sustainable and Intelligent Mobility System in Urban Cities

Speedy urbanization swelling transit request, traffic bottleneck, and the environmental influence of private vehicle reliance have generated an immediate essential for intelligent and sustainable urban transit methods. Understanding commuters’ transit mode priorities is basic to framing applicable transportation rules, enhancing public transit facilities, and encouraging low-carbon travel attitudes. This research suggests an XGBoost (Extreme Gradient Boosting) based machine learning model for forecasting urban commuters’ desired in transportation modes using heterogeneous socioeconomic, demographic, trip-related, availability, and behavioral attributes. Unlike usual discrete choice methods which usually inflict predefined associations between explanatory variables and travel mode priorities, the planned method reveals the nonlinear interactions and complex decision limits inherent in commuter attitudes. The XGBoost approach is established and enhanced through organized hyperparameter tuning, cross endorsement, feature selection, and class imbalance supervision to develop foretelling strength and generalizability. Method execution is measured using accuracy, precision, recall, F1-score, receiver working characteristic area under the curve (ROC-AUC), and multiclass sorting metrics. Additionally, SHAP (SHapley Additive exPlanations) is combined to decode the contribution of individual parameters and discover heterogeneous things dominating commuters’ mode priorities, thereby addressing the interpretability restrictions usually related with machine learning methods. The recommended method is anticipated to specify the relative significance and nonlinear effects of variables such as travel time, travel cost, income, vehicle ownership, accessibility to public transportation, parking facility, service consistency, and environmental consciousness. The resulting predictive intelligence can assist transport organizers and strategy developers in building targeted requirement management rules, increasing public transportation magnetism, promoting sustainable mode shifts, and decreasing reliance on private automobiles. Finally, the research contributes to the growth of a data-driven Sustainable Intelligent Mobility System (SIMS) efficient of integrating attitudes prediction with proof based urban transportation preparation and easing more competent, equitable, strong, and environmentally sustainable mobility in quickly growing cities.

Authors

Publication Details

Journal
Dandao Xuebao/Journal of Ballistics
Published
2026-11-14
Primary Topic
Transportation Planning and Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Predicting Urban Commuters’ Transit Mode Preferences Using XGBoost Algorithms to Build Sustainable and Intelligent Mobility System in Urban Cities

M Rameshkumar J Sekaran
Dandao Xuebao/Journal of Ballistics
Transportation Planning and Optimization
article

Predicting Urban Commuters’ Transit Mode Preferences Using XGBoost Algorithms to Build Sustainable and Intelligent Mobility System in Urban Cities

M Rameshkumar J Sekaran
article en

Abstract

Speedy urbanization swelling transit request, traffic bottleneck, and the environmental influence of private vehicle reliance have generated an immediate essential for intelligent and sustainable urban transit methods. Understanding commuters’ transit mode priorities is basic to framing applicable transportation rules, enhancing public transit facilities, and encouraging low-carbon travel attitudes. This research suggests an XGBoost (Extreme Gradient Boosting) based machine learning model for forecasting urban commuters’ desired in transportation modes using heterogeneous socioeconomic, demographic, trip-related, availability, and behavioral attributes. Unlike usual discrete choice methods which usually inflict predefined associations between explanatory variables and travel mode priorities, the planned method reveals the nonlinear interactions and complex decision limits inherent in commuter attitudes. The XGBoost approach is established and enhanced through organized hyperparameter tuning, cross endorsement, feature selection, and class imbalance supervision to develop foretelling strength and generalizability. Method execution is measured using accuracy, precision, recall, F1-score, receiver working characteristic area under the curve (ROC-AUC), and multiclass sorting metrics. Additionally, SHAP (SHapley Additive exPlanations) is combined to decode the contribution of individual parameters and discover heterogeneous things dominating commuters’ mode priorities, thereby addressing the interpretability restrictions usually related with machine learning methods. The recommended method is anticipated to specify the relative significance and nonlinear effects of variables such as travel time, travel cost, income, vehicle ownership, accessibility to public transportation, parking facility, service consistency, and environmental consciousness. The resulting predictive intelligence can assist transport organizers and strategy developers in building targeted requirement management rules, increasing public transportation magnetism, promoting sustainable mode shifts, and decreasing reliance on private automobiles. Finally, the research contributes to the growth of a data-driven Sustainable Intelligent Mobility System (SIMS) efficient of integrating attitudes prediction with proof based urban transportation preparation and easing more competent, equitable, strong, and environmentally sustainable mobility in quickly growing cities.

Dandao Xuebao/Journal of Ballistics
Transportation Planning and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon