Multi-modal Matching Problem of Shared Mobility (2109.09456v1)

Abstract: Rideshare is one way to share mobility in transportation without increasing traffic demand, where travel mobility and usage of vehicle capacity can be improved. However, current literature on rideshare has allowed only one-modal trips and may be limited in the matching efficiency, especially when there is a large gap between the supply and demand of mobility. Therefore, the objectives of this paper are first to develop a multi-modal matching framework of shared mobility with public transportation to maximize the performance of a rideshare system, and second to evaluate the effect of the public transportation and of the schedule flexibility on the matching efficiency. To fulfill the first objective, a multi-modal matching framework is developed to allow rideshare with both private and public vehicles with detailed design of detour, using Genetic Algorithm. Also for the second objective, the effects of public transportation and schedule flexibility are evaluated with a simplified network of Sioux Falls. The results show that public transportation helps the match rate slightly at a low supply of private vehicle, but this must be evaluated for practical implementation as different cities may bring different results. Also, a larger schedule flexibility helps greatly in increasing match rate even at a lower supply level. As well, the planning subject of time schedule is benefited more with larger schedule flexibility, in this paper the drivers, on the matching efficiency. Moreover, a rideshare system with private vehicles outperforms a public transportation system, possibly due to the rigid route of public transportation that takes no detour burden. This confirms the need for a flexible design of sharing mobility, as can be fulfilled with the multi-modal matching framework developed in this research.