Analysing Commuters’ Transit Mode Chosen Pattern in Urban Cities to Build Sustainable Intelligent Transit Mobility

Abstract

Brisk urbanization and growing transportation demand have deepened the essential for sustainable and intelligent transit systems. Understanding commuter travel behaviour is important for building active transportation plans and mobility solutions. This research examines the commuter mode-choice behaviour using advanced approach based on the Random Forest algorithm. Here, travel dataset containing of 544 observations and 44 descriptive variables was used to analyse the aspects dominating the chosen of among private motorized transport, public motorized transport, and non-motorized transport modes. The Random Forest algorithm have attained an overall prediction accuracy of 89.02%, demonstrating superior capability in identifying complex nonlinear relationships among socio-economic, demographic, and travel-related variables. Feature importance analysis expressed that specified mode priority, automobile requirement, travel cost, driving license ownership, and travel distance meaningfully influence the commuter decisions. The findings give valuable insights for urban organisers and policymakers in scheming sustainable mobility plans and intelligent transportation systems possible of plummeting the congestion, emissions, and dependence on private vehicles.