Leveraging explainable artificial intelligence and big trip data to understand factors influencing willingness to ridesharing

Carpool-style ridesharing, compared to traditional solo ride-hailing, can reduce traffic congestion, cut per-passenger carbon emissions, reduce parking infrastructure, and provide a more cost-effective way to travel. Despite these benefits, ridesharing only occupies a small percentage of the total ride-hailing trips in cities. This study integrates big trip data with machine learning and eXplainable AI (XAI) to understand the factors that influence willingness to take shared rides. We use the City of Chicago as a case study, and results show that users tend to adopt ridesharing for longer distance trips, and the cost of a trip remains the most important factor. We identify a strong diurnal pattern that people prefer to request shared trips during the morning and afternoon peak hours. We also find socio-economic disparities: users who requested trips from neighbourhoods with a high percentage of non-white, a low median household income, a low percentage of bachelor's degrees, and high vehicle ownership are more likely to share a ride. The findings and the XAI-based analytical framework presented in this study can help transportation network companies and local governments understand ridesharing behaviour and suggest new strategies and policies to promote the proportion of ridesharing for more sustainable and efficient city transportation.