Balancing Profit and Traveller Acceptance in Ride-Pooling Personalised Fares

Michal Bujak,Rafal Kucharski

2024-11-05

Abstract:In a ride-pooling system, travellers experience discomfort associated with a detour and a longer travel time, which is compensated with a sharing discount. Most studies assume travellers receive either a flat discount or, in rare cases, a proportional to the inconvenience. We show the system benefits from individually tailored fares. We argue that fares that optimise an expected profit of an operator also improve system-wide performance if they include travellers' acceptance. Our pricing method is set in a heterogeneous population, where travellers have varying levels of value-of-time and willingness-to-share, unknown to the operator. A high fare discourages clients from the service, while a low fare reduces the profit margin. Notably, a shared ride is only realised if accepted by all co-travellers (decision is driven by the latent behavioural factors). Our method reveals intriguing properties of the shareability topology. Not only identifies rides efficient for the system and supports them with reduced fares (to increase their realisation probability), but also identifies travellers unattractive for the system (e.g. due to incompatibility with other travellers) and effectively shifts them to private rides via high fares. Unlike in previous methods, such approach naturally balances the travellers satisfaction and the profit maximisation. With an experiment set in NYC, we show that this leads to significant improvements over the flat discount baseline: the mileage (proxy for environmental externalities) is reduced by 4.5% and the operator generates more profit per mile (over 20% improvement). We argue that ride pooling systems with fares that maximise profitability are more sustainable and efficient if they include travellers' satisfaction. Keywords: ride-pooling, personalised pricing, individual discounts

Computer Science and Game Theory

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to balance the profits of operators and the acceptance of passengers in the ridesharing system. Specifically, the paper explores optimizing the economic and social benefits of ridesharing services through personalized pricing strategies. Traditional research usually assumes that passengers will receive a fixed discount or a discount proportional to the degree of inconvenience, while this paper proposes a personalized pricing method based on individual passenger characteristics (such as time value and willingness to share). This method not only considers the profit - maximization goal of operators but also the acceptance of passengers for the service, thereby achieving an overall improvement in the system's performance.

### Main contributions of the paper
1. **Personalized pricing model**: The paper proposes a personalized pricing model in a probabilistic framework, which can set personalized discounts according to the behavioral characteristics of different passengers (such as time value and willingness to share) to improve passenger acceptance and operator profits.
2. **System performance optimization**: Through personalized pricing, the paper shows how to increase the operator's profit per mile while reducing the total mileage (environmental benefit), thereby achieving an overall improvement in system performance.
3. **Consideration of behavioral heterogeneity**: The paper takes into account the heterogeneity of passenger behavior, that is, different passengers have different sensitivities to time and space, which makes the pricing strategy more flexible and effective.
4. **Experimental verification**: Through experiments on the data set in New York City, the paper verifies the effectiveness of the personalized pricing method. The results show that the total mileage is reduced by 4.5% and the operator's profit per mile is increased by more than 20%.

### Method overview
1. **Data collection and sharing graph construction**: First, collect trip requests and construct a sharing graph containing all potential shared trips. In this stage, the ExMAS algorithm is used to efficiently generate the sharing graph.
2. **Acceptance probability function**: For each feasible shared trip, calculate the acceptance probability of passengers. The acceptance probability depends on the personal characteristics of passengers (such as time value and willingness to share) and the discount offered.
3. **Discount space discretization**: To make the calculation feasible, discretize the discount space into a finite set for numerical approximation.
4. **Expected profitability calculation**: Combine the acceptance probability of passengers and revenue to calculate the expected profitability of each shared trip. The expected profitability is a comprehensive indicator that reflects the relationship between the operator's revenue and cost.
5. **Optimal matching**: Through integer linear programming techniques, select an optimal subset of shared trips so that each passenger is offered a shared trip proposal while maximizing the average expected profitability of the system.

### Conclusion
Through the personalized pricing strategy, the paper successfully balances the profits of operators and the acceptance of passengers in the ridesharing system and achieves an overall improvement in the system's performance. This method not only improves economic efficiency but also helps to reduce traffic congestion and environmental pollution and has important practical application value.