I. A. Urrutia, N. C. Sanchez, D. Antonelli, L. Alonso and K. Larson, "Mode Choice Modeling for Sustainable Regional Commuting Using Machine Learning: A Case Study in Gipuzkoa, Spain," 2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), Edmonton, AB, Canada, 2024, pp. 2125-2130, doi: 10.1109/ITSC58415.2024.10919588.

Abstract

Addressing mobility is essential for achieving mid-century emissions reduction targets. The role of local governments is pivotal in shaping this future, yet they often lack adequate tools for informed decision-making regarding appropriate interventions within their unique contexts. To bridge this gap, we developed a machine learning (ML) model based on a random forest algorithm that predicts the impacts of different interventions on mobility mode choices and CO2 emissions. Moreover, we address the model's interpretability by offering a quantitative analysis of feature importance. Through a case study focused on commuting trips in Gipuzkoa, Spain, we explore a range of intervention scenarios, illustrating potential emissions reductions between 2% to 58%. Additionally, we integrated this model into a user-friendly web-based interface, which could support local governments in strategic mobility planning, thereby facilitating more informed and effective policy decisions.