FairRide: A Cooperative-Game Approach to Fair Surge Pricing in Ridesharing

Abstract

Dynamic pricing is the core mechanism that allows rideshare platforms to balance demand and supply. While today’s surge strategies, like Uber’s surge multiplier model, achieve market efficiency, they often raise fairness concerns, disproportionately burdening riders in low-income areas with little or no access to public transit and creating inconsistent earning opportunities for drivers. We introduce FairRide, a cooperative game–theoretic framework that prices trips via Owen value to promote multi-sided fairness for both riders and drivers. We further propose two variants: FairRide+, which captures cross-zone demand interdependencies; and FairRide-Decay, which tempers volatility through temporal smoothing. Using a synthetic dataset of 10 zones (urban, suburban, rural), three vehicle categories, and 3,000 time steps, we compare our models against Uber-style surge and an additive-surge benchmark. FairRide-Decay reduces the incidence of extreme surges to below 8% while preserving rider equity and improving driver opportunity balance; all improvements are statistically significant (p < 0.001). These findings demonstrate that fairness-aware dynamic pricing is feasible at platform scale and establish a foundation for hybrid policies that jointly optimize efficiency, fairness, and driver incentives in real-world ridesharing systems.