The emergence of ride-hailing services enriches people's trip modes, which not only change traditional taxi operation market but has profoundly affected people’s travel behavior. In order to seek the sustainable development of ride-hailing services from the perspective of passengers’ mode choice behavior, a Multinomial Logit (MNL) model is built up and calibrated to describe passengers’ ride-hailing choice behavior through ride-hailing order data. From the results, it is found that the attribute of pick-up and drop-off position, time, travel distance, travel time and cost have strong impact on passengers’ choice behavior in different types of ride-hailing service. For example: passengers have different preference in pick-up and drop-off position which have different attributes. And passengers prefer to use the express at morning peak and evening peak, and during the night, passengers prefer to choose hitch. As travel distances increase, passengers prefer to choose hitch. The research results can be applied to relevant departments and various platforms to make ride-hailing better provide reference for social services in the future.

CICTP