Dynamic Pricing Method for One-Way Car Sharing Service to Meet Demand and to Maximize Profit Under Given Utility Function

Abstract

A car sharing service is a model of car rental where people rent cars for short periods of time. A round-trip car sharing service, in which a user must return the car to the same station where the user picked up the cars is generally used for shopping, leisure and occasional trips; on the other hand, one-way car sharing service such that a user can return a car at an arbitrary station is used for all purposes and more convenient for users than a round-trip car sharing service. However, due to imbalance of demand between users’ departures and destinations, when a user want to use a car at an arbitrary station, no cars may be returned yet till the time at the station and the user cannot always use a car. Some strategies have been proposed to solve this problem, one of which is a dynamic pricing. Charging higher price decreases demand, and charging lower price increase demand; thus, we can make demand match supply of available cars by using the strategy of dynamic pricing. This strategy is effective; however, too high price decreases demand too much, so that the total profit may decrease. On the other hand, too low price increase demand too much, so that the imbalance of demand cannot be sufficiently balanced. Therefore, a control method of the dynamic pricing to maximize profit with matching demand to supply is needed. In this paper, we formulate the optimization problem to determine price between any two station at each time slot. This problem is formulated as a network flow problem on a time and space network. We evaluate this control method by numerical experiments. The result shows that the proposed method achieves more profit than a fixed pricing.