A fairness-aware joint pricing and matching framework for dynamic ridesharing

Enabled by the prevailing mobile devices, novel mobility services, such as ridesharing, have a great potential to change the mobility pattern of metropolis inhabitants. In this study, we focus on the pricing and matching challenges faced by a mobility service platform when both ridesharing and non-shared mobility services are provided. A joint pricing and matching framework is proposed to efficiently dispatch vehicles and deliver explicit trip time and fare information in real-time. Besides, we define six principles to evaluate the fairness of pricing methods and develop a discount function considering the features of passengers’ shared rides. In simulation experiments where passengers can choose from different service types, we show that our method can significantly increase the system’s profit without violating the fairness principles among co-riders.

I. Introduction

Dynamic ridesharing, as a way to utilise one vehicle to deliver multiple passengers whose trips share spatial-temporal similarities, is expected to improve the car occupancy rate, thus serving more customers. Passengers can also benefit from ridesharing as they share the travel fare with co-riders, nevertheless sacrificing time and comfort. Therefore, ridesharing is commonly provided to users as a budget complement to the conventional ride-hailing solo-trip service. Compared to static ridesharing, where only empty vehicles are considered for matching, dynamic ridesharing is more flexible yet more challenging in terms of matching and pricing. In this paper, we target the mobility platform’s daily operation problem of how to dispatch vehicles and price passengers considering stochastic user behaviour in a dynamic ridesharing context.