Dynamic Pricing in Shared Mobility on Demand Service

We consider a profit maximization problem in an urban mobility on-demand service, of which the operator owns a fleet, provides both exclusive and shared trip services, and dynamically determines prices of offers. With knowledge of the traveler preference and the distribution of future trip requests, the operator wants to find the pricing strategy that optimizes the total operating profit of multiple trips during a specific period, namely, a day in this paper. This problem is first formulated and analyzed within the dynamic programming framework, where a general approach combining parametric rollout policy and stochastic optimization is proposed. A discrete-choice-based price optimization model is then used for the request level optimal decision problem and leads to a practical and computationally tractable algorithm for the problem. Our algorithm is evaluated with a simulated experiment in the urban traffic network in Langfang, China, and it is shown to generate considerably higher profit than naive strategies. Further analysis shows that this method also leads to higher congestion level and lower service capacity in the urban traffic system, which highlights a need for policy interventions that balance the private profit making and the system level optimality.

PDF Abstract