Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This work studies the role of proactive and targeted ride-splitting incentives on the service level and total revenue of a ride-sourcing platform, which is built as a discrete event simulator that incorporates simulated taxi data within a congestible road network. While shared trips offer riders a discount to compensate for any additional travel or waiting time, the success in matching shared trips relies on riders’ attitudes towards the trade-off between travel time and cost. Therefore, formulated as a multinomial logit model, alternative-specific coefficients characterize the probability of a random draw among three options: solo ride, shared ride, or a public-transit-like service. A multi-objective analysis shows that for regions where empty vehicle depletion rate is high, by offering incentives with an additional discount for a shared trip, the platform can proactively rebalance vehicle supply in high-demand regions during peak-hours, and succeed with a small abandonment for trips with high waiting or detour times.
Michel Bierlaire, Evangelos Paschalidis
Nikolaos Geroliminis, Lynn Fayed