Generic model for resource allocation in transportation. Application to urban parking management. (October 2016)

• Record Type: Journal Article
• Authors: Zargayouna, Mahdi; Balbo, Flavien; Ndiaye, Khadim
• Is Part Of: Transportation research. Volume 71(2016)
• Page Start: 538
• Page End: 554
• Publication Date: 2016-10
• Subjects: Resource allocation -- Urban parking -- VANET -- Multiagent systems -- Simulation; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2016.09.002
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: A model for resource allocation in transportation focused on space and time. A model for an example problem: urban parking management. A community-based solution to this problem composed of interacting agents in a VANET. Two cooperation models based on the willingness of drivers to share their intentions. Abstract: In this paper, we define the online localized resource allocation problem, especially relevant for modeling transportation applications. The problem modeling takes into account simultaneously the geographical location of consumers and resources together with their online nondeterministic appearance. We use urban parking management as an illustration of this problem. In fact, urban parking management is an online localized resource allocation problem, where the question is how to find an efficient allocation of parking spots to drivers, while they all have dynamic geographical positions and appear nondeterministically. We define this problem and propose a multiagent system to solve it. The objective of the system is to decrease, for private vehicles drivers, the parking spots search time. The drivers are organized in communities and share information about spots availability. We have defined two cooperative models and compared them: a fully cooperative model, where agents share all the available information, and a "coopetitive" model, where drivers do not share information about the spot that they have chosen. Results show the superiority of the first model.