On-demand ride-sharing with the walking incentive for meeting points

Abstract

The ride-sourcing market exhibits diverse travel demands, and travel platforms offer riders various service types. In such services, drivers usually need to travel sequentially to riders’ origins and destinations to provide door-to-door service, leading to longer travel times, especially when encountering road congestion or additional detours caused by one-way roads during pick-up or drop-off. Introducing meeting points into the mixed-demand pattern system can alleviate this problem. Assuming all riders exhibit the willingness to walk, the platform can implement incentive mechanisms to encourage walking to designated meeting points. Higher incentives can motivate riders to accept longer walking distances, thereby promoting sustainable mobility, enhancing system flexibility, and improving matching efficiency. However, offering high incentives may diminish system profits, necessitating a balanced optimization of incentive sizes. By screening the potential matching sets, two multi-objective optimization models with constant or optimized incentives are constructed with the objectives of maximizing the number of matched riders and total profit. Then, corresponding heuristic algorithms are designed to solve the model. Real-world-based results show that the introduction of constant walking incentives in the travel system increased the rider matching rate by an average of 4% and total profit by 3% across various travel scenarios. Conversely, matching under optimized incentives led to a 1.13% improvement in rider matching rate and a 1.27% increase in total profit compared to constant incentives. Introducing walking incentives provides higher system profits, superior rider service, and fewer vehicle dispatches and emissions. These findings can contribute to refining operational practices for on-demand ride-sourcing travel services.