Search

The awareness of pollution in urban areas is currently rising. Among the various possible solutions, the reduction of carbon dioxide emissions from the use of fossil-fuel-powered vehicles stands out. To this end, private vehicles and other services such as taxis could be replaced by more sustainable alternatives, including carsharing fleets. This type of vehicle, generally electric, is being introduced in more and more cities, offering a green mobility alternative. In this paper, we analyse the efficiency of the current taxi fleet in Valencia through multi-agent simulations. We then assess possible carsharing fleets with different configurations. Our results show the potential that a combination of the two types of fleets would have to meet the current demand and, at the same time, improve the sustainability of the city.

[EN] To ensure cities sustainability, we must deal with, among other challenges, traffic congestion, and its associated carbon emissions. We can approach such a problem from two perspectives: the transition to electric vehicles, which implies the need for charging station infrastructure, and the optimization of traffic flow. However, cities are complex systems, so it is helpful to test changes on them in controlled environments like the ones provided by simulators. In our work, we use SimFleet, an agent-based fleet simulator. Nevertheless, SimFleet does not provide tools for easily setting up big experiments, neither to simulate the realistic movement of its agents inside a city. Aiming to solve that, we enhanced SimFleet introducing two fully configurable generators that automatize the creation of experiments. First, the charging stations generator, which allocates a given amount of charging stations following a certain distribution, enabling to simulate how transports would charge and compare distributions. Second, the load generator, which populates the experiment with a given number of agents of a given type, introducing them dynamically in the simulation, and assigns them a movement that can be either random or based on real city data. The generators proved to be useful for comparing different distributions of charging stations as well as different agent behaviors over the same complex setup., This work was partially supported by MINECO/FEDER RTI2018-095390-B-C31 project of the Spanish government. Pasqual Martí and Jaume Jordán are funded by UPV PAID-06-18 project. Jaume Jordán is also funded by grant APOSTD/2018/010 of Generalitat Valenciana - Fondo Social Europeo.

With the number of people that live in cities increasing every year, the complexity of urban traffic increased as well, making it more necessary than ever to find solutions that are good for the citizens, energy-efficient, and environmentally friendly. One of the systems that are becoming more popular is carsharing, specifically free-floating carsharing: fleets of cars that are parked around a city that can be temporarily booked for private use within the borders of the city by the system users. In this work, we implement one of these systems over SimFleet, an agent-based fleet simulator. We present how the original SimFleet agents are adapted to our system and how they interact with each other, as well as the strategies they follow to address the urban traffic problem efficiently. Our implementation for the simulation of free-floating carsharing scenarios is crucial for companies or municipalities to make the necessary tests before deploying the systems in real life.