Temporal equilibrium for electrified ride-sourcing markets considering charging capacity and driving fatigue

Abstract

Electric vehicles (EVs) are now being introduced to the ride-sourcing market to catalyze the realization of sustainable transportation. Given their demand for charging, ride-sourcing EV drivers may have distinct work schedules from gasoline vehicle (GV) drivers, yielding significant impacts on the market supply when their penetration becomes high. Driving fatigue is another factor affecting ride-sourcing drivers’ working hours and work schedules, but has been seldom deliberated. This paper proposes an equilibrium modeling framework based on a time-expanded network to describe the work schedules of EV and GV drivers with the consideration of their driving fatigue and EVs’ charging opportunities subject to the limited charging infrastructure. Specifically, a novel cost formulation is proposed to capture the impact of driving fatigue on drivers’ choices of work schedules. To solve the equilibrium model, we develop a gap function-based method coupled with the column generation scheme in which a non-additive shortest path (NSP) problem appears as a subroutine. A customized multi-objective label correcting algorithm enhanced with several problem-tailored speed-up techniques is then designed to solve the NSP problem efficiently. Experimental results suggest that the speed-up techniques contribute significantly to the reduction of computational time. Numerical examples reveal that the temporal equilibrium of the electrified ride-sourcing market is moderated by the charging capacity, EV penetration as well as the competition among drivers; and driving fatigue may yield considerable impacts on drivers’ work schedules and the market performance.