Abstract
Urban traffic routing has to deal with individual mobility and collective wellness considering citizens, multi-modal transport, and fleet traffic with conflicting interests such as electric vehicles, local distribution, public transport, and private vehicles. Different interests, goals, and regulations, suggest the development of new multi-objective routing mechanisms which may improve traffic flow. In this work, Traffic Weighted Multi-Maps (TWM) is presented as a novel traffic routing mechanism based on the strategical generation and distribution of complementary cost maps for traffic fleets, oriented towards the application of differentiated traffic planning and control policies. TWM is built upon a centralized control architecture, where a Traffic Management Center generates and distributes customized cost maps of the road network. These maps are used individually to calculate routes. In this research, we present the TWM theoretical model and experimental results based on microscopic simulations over a real city traffic network under multiple scenarios, including traffic incidents management. Experimental evaluation takes into account driver's adherence to the system and considers a multi-objective analysis both for the global network parameters (congestion, travel time, and route length) and for the subjective driving experience. Experimental results deliver performance improvements from 20% to 50%. TWM is fully compatible with existing traffic routing systems and has promising future evolution applying new algorithms, policies and network profiles.
Highlights
Modelling and design of traffic management systems and services have still important challenges to address, such as matching multi-objective demand and resources in an optimal and automated way
In this paper we focus on the effects of multimaps on travel time and route length using randomly generated weights for traffic dispersion in real urban traffic network
In this paper we address static traffic routing with Traffic Weighted Multi-Maps (TWM), leaving the dynamic routing based on using different routing algorithms ( Dijstra) for future works
Summary
Modelling and design of traffic management systems and services have still important challenges to address, such as matching multi-objective demand and resources in an optimal and automated way. In one side Traffic Control Systems (TCS) measure and react over the traffic network (resources) to coordinate traffic demand by means of signaling systems, traffic information panels, and regulatory policies and restrictions [36]. We have vehicles represented by their traffic agents, that plan routes dynamically, and react in real time to traffic data input: traffic network status, signaling directives and congestion information [35]. The need of individual route generation and dynamic re-planning from a TCS has been addressed by many different approaches and commercial proposals [1], [2], [5], [30].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
