Carbon emission reduction potential of on-demand transit replacing fixed-route transit

This study examines the potential of on-demand transit to reduce carbon emissions compared to fixed-route transit in Shanghai, China, with both services using electric vehicles. We first analyze how carbon emission reductions vary across different time periods. The results show that on-demand transit is more effective in replacing fixed-route transit during evening and night periods to achieve carbon emission reduction. Next, we use CatBoost models to explore how route characteristics influence carbon emission reduction. For per capita carbon emission reductions, demand, route length, and route curvature are important factors. On the other hand, when considering the maximum demand for achieving carbon emission reductions (critical demand threshold), the distribution of passengers across the route (sectional load factor) plays more important roles than the physical characteristic of the route. Additionally, increasing the number of vehicles while reducing their capacity can accommodate more passengers and improve the potential for emission reduction.