Urban fleet simulator
Abstract
Urban mobility has changed drastically in recent years with the emergence of new modes of transportation and technological advances. That is why this paper presents a tool called SimFleet that allows simulating and evaluating new urban mobility models. The main objective of SimFleet is to allow users to develop and test new mobility management models or policies in the urban environment, as well as to analyze possible new coordination strategies and vehicle fleet regulation mechanisms that can improve efficiency in the distribution of people or goods in urban environments. SimFleet is a simulation tool that provides a convenient environment in which to develop and test complex coordination and negotiation scenarios in the context of a city containing one or more fleets of transport vehicles. The tool can be used to simulate any type of fleet in which a group of vehicles transport items (goods or people) from one location to another within the city. Examples of fleets would be, among others, courier companies, cab services, freight transport (in trucks), bicycle rental services, etc.
In the literature, there are different approaches that attempt to simulate, or propose strategies for the urban delivery problem. The best known simulation tools are SUMO and MatSim. SUMO can be considered as a simulation platform for traffic flows, including vehicles, any public transport and also pedestrians. It includes different tools and add-ons that offer functionalities such as route location or V2X communications. On the other hand, MatSim is an agent-based simulation framework for implementing large-scale agent-based transportation simulations. MatSim allows the design of agents consisting of a set of activities representing different transportation demands. These simulation tools are usually oriented towards general purpose applications and their specific use for designing and evaluating strategies for the urban delivery problem can be very complex. In contrast to these works, in SIMFLEET we propose a specific support tool to analyze and evaluate different models and strategies in the field of urban mobility and, specifically, in urban sharing scenarios. The developed tool is flexible enough to support various types of deliveries, transport companies, logistics and many other mobility requirements. Moreover, it allows the configuration of different scenarios reflecting new delivery modes such as collaborative delivery or car sharing solutions.
The tool has been built as a multi-agent system running on top of the SPADE platform, where the different actors in the simulation are modeled as agents that can interact with each other using SPADE’s communication facilities. The programming is done in Python and can run on any operating system. Furthermore, the tool has been designed to hide most of the complexity of developing a multi-agent application by providing the user with different interaction interfaces and a coherent internal architecture through which the tool can be easily adapted to the needs of a particular simulation scenario.
Technical specifications
Type of technology
SOFTWARE