Prediction of walking occurrence and frequency among the residents of city community housing projects using machine learning techniques

Closed

Rosilawati Zainol, Tin Tin Su, Hazreen Abdul Majid, Cheong Peng Au Yong, Mahdi Aghaabbasi, Nithiah Thangiah, Dan He, Fadly Usman

2026 Journal of Transport and Health Vol. 49 Article Cited by 0 Quartile

Abstract

Introduction Walking is one of the most affordable and accessible forms of daily physical activity, yet it remains an unsafe and under-supported mode of transport in many low-income urban neighborhoods. Residents of public housing communities often face inadequate pedestrian infrastructure, traffic safety risks, and limited access to essential services, all of which discourage walking and exacerbate health inequalities. Understanding the complex interplay of these factors is critical for improving mobility and public health outcomes among vulnerable urban populations. Methods This study applies a Bayesian Network model to predict the occurrence and frequency of walking among 680 residents across five low-income public housing projects (PPRs) in Kuala Lumpur, Malaysia. Using cross-sectional data on individual demographics, travel behavior, built environment perceptions, and health status, the model identifies probabilistic relationships between key factors influencing pedestrian activity. Results Results indicate that gender, employment status, and perceived walkability significantly affect the likelihood and frequency of walking trips. Notably, the perceived quality of pedestrian infrastructure and accessibility to public facilities also emerged as influential predictors. Conclusions The results highlight the urgent need to design safer and more inclusive walking environments in low-income urban areas. The study demonstrates the practical utility of Bayesian modeling in capturing interdependent factors influencing walking behavior and provides actionable insights for planners and policymakers seeking to promote equitable, health-oriented, and pedestrian-friendly urban mobility. © 2026 Elsevier Ltd.

Affiliations

Department of Urban and Regional Planning, Centre for Sustainable Urban Planning and Real Estate (SUPRE), Faculty of Built Environment, Universiti Malaya, Malaysia; Jeffrey Cheah School of Medicine and Health Sciences, Victorian Heart Institute (VHI), Monash University, Malaysia; Department of Social and Preventive Medicine, Faculty of Medicine, Universiti Malaya, Malaysia; Department of Building Surveying, Faculty of Built Environment, Universiti Malaya, Malaysia; Urban Digital Twin Lab, School of Modeling, Simulation, and Training, University of Central Florida, 3100 Technology Parkway, Orlando, 32826, FL, United States; Centre for Civilisational Dialogue, Universiti Malaya, Malaysia; Centre for Transportation Research (CTR), Faculty of Engineering, Universiti Malaya, Malaysia; Department of Urban and Regional Planning, Universitas Brawijaya, Indonesia