Coordination of Urban On-demand Transportation Services

**Coordination of Urban On-Demand Transportation Services**:

- Réf **ABG-129280**
- Sujet de Thèse
- 07/03/2025
- Contrat doctoral
- Université Gustave Eiffel - Site de Marne-la-Vallée
- Lieu de travail- Champs sur marne - Ile-de-France - France
- Intitulé du sujet- Coordination of Urban On-Demand Transportation Services
- Champs scientifiques- Sciences de l’ingénieur
- Mots clés- Transportation engineering, New mobility services, Ride-sharing, Optimization

**Description du sujet**:
**Context**

The rise of on-demand transportation services has significantly altered urban mobility patterns. While these services offer convenience to users, if not properly managed or coordinated, they may worsen urban congestion, increase the travel time, energy consumption in the end. For instance, in the absence of coordination, two people that go in the same direction within nearly the same time slot require two vehicles whereas they can share the same vehicle. This research aims to evaluate the different strategies of coordinating on-demand services provided by different companies, in terms of number vehicle-miles for given demands. It is worth noting that minimizing the number of vehicle-miles is in line with minimizing the congestion and energy consumption. Centralized optimization which uses all the resources (vehicles, drivers,) initially owned by different service providers (i.e., pools the resources they provide) to satisfy all the requests obviously is the best strategy in terms of the number of vehicle-miles. But it may hurt the interests of some and more crucially, it requires that the centralized platform have complete access to the information of the service providers and therefore creates reluctance of the latter. As a consequence, such a strategy may be inapplicable in practice.

This research investigates two categories of service provision: cooperative-direct, and cooperative-indirect, examining how these strategies create values to customers and society in terms of cost (vehicle-miles) and congestion, respectively. Note that centralized optimization strategy applies to monopoly case and cooperative-direct category. Monopoly corresponds for instance to situations where the service is provided by government as part of public service. In case of cooperative-direct, although the service providers are willing to share the information and resources, each of them has its own financial interests. As a result, it is necessary to develop proper mechanism to share the value created due to this cooperation.

The main parts of an on-demand service are the passengers, the ride provider, and the matching algorithm. The passenger seeks a ride to pick her/him up at the origin point and drop her/him off at the desired destination within a time interval. The ride provider has a fleet of vehicles (taxi, van, autonomous car, etc.) that is ready to serve the passengers’ requests. The matching algorithm orchestrates the pairing of passengers with available vehicles, seeking to optimize both travel time and vehicle-miles.

We classify this system into two categories, considering the interactions between different fleet providers and the performance of the matching algorithm. Fleet providers can work competitively or cooperatively. In the case of cooperative interaction, they share the fleet information so the passenger may get the ride easier, and the benefits of ride-sharing will be achievable more easily. In the competitive scenario, each provider tries to gain a higher profit even at the expenses of others. The matching algorithm communicates with the providers directly or indirectly. This classification can be expanded for other new mobility services as well.

Existing literature has predominantly explored the mechanics of ride-sourcing and ride-splitting while inadequately addressing their impact on urban congestion and energy consumption. This research aims to bridge this knowledge gap by investigating the influence of competitive and cooperative strategies, coupled with direct and indirect dispatching algorithms, on the performance of the whole service that can be measured in vehicle-miles that heavily impact traffic congestion and energy consumption. This study holds paramount importance for devising targeted optimization strategies that reduce the vehicle-miles and hence create values for passengers as well as the whole society through reduced congestion and energy consumption.

**Objectives**:

- On-demand service providers categorization: We aim to establish a clear taxonomy of operational models based on competitiveness and dispatching strategies.
- Optimization algorithms and value-sharing mechanism design: We aim to develop and propose optimization strategies for each category and, if necessary, mechanism of sharing the value created due to the cooperation among service providers so as to incentivize their willingness for cooperation.
- Comparative analysis and sensitivity analysis: We aim to analyze the impacts o