Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimising Dynamic Traffic Distribution for Urban Networks with Answer Set Programming

Matteo Cardellini, Carmine Dodaro, Marco Maratea, Mauro Vallati

TL;DR

This work tackles dynamic traffic distribution in urban networks by presenting a centralized framework that uses Answer Set Programming (ASP) to compute optimal routes within a four-phase pipeline: Network Analyser, Domain Independent Search, ASP Optimiser, and Mobility Simulator. The Domain Independent Search generates diverse, acyclic candidate routes per vehicle using a time-expanded model with a discretisation step of $5s$, clustering routes by similarity and selecting the top $5$ from each group. The ASP Optimiser encodes route, timing, and capacity constraints as ASP facts and rules, guided by weak constraints to minimize congestion and travel time, and is validated on real-world networks in Bologna and Milton Keynes with up to $600$ vehicles, achieving significant KPI improvements in larger topologies. The results indicate that the ASP-based routing approach scales well in realistic urban settings and can support future integration into centralized urban traffic management systems.

Abstract

Answer Set Programming (ASP) has demonstrated its potential as an effective tool for concisely representing and reasoning about real-world problems. In this paper, we present an application in which ASP has been successfully used in the context of dynamic traffic distribution for urban networks, within a more general framework devised for solving such a real-world problem. In particular, ASP has been employed for the computation of the "optimal" routes for all the vehicles in the network. We also provide an empirical analysis of the performance of the whole framework, and of its part in which ASP is employed, on two European urban areas, which shows the viability of the framework and the contribution ASP can give.

Optimising Dynamic Traffic Distribution for Urban Networks with Answer Set Programming

TL;DR

This work tackles dynamic traffic distribution in urban networks by presenting a centralized framework that uses Answer Set Programming (ASP) to compute optimal routes within a four-phase pipeline: Network Analyser, Domain Independent Search, ASP Optimiser, and Mobility Simulator. The Domain Independent Search generates diverse, acyclic candidate routes per vehicle using a time-expanded model with a discretisation step of , clustering routes by similarity and selecting the top from each group. The ASP Optimiser encodes route, timing, and capacity constraints as ASP facts and rules, guided by weak constraints to minimize congestion and travel time, and is validated on real-world networks in Bologna and Milton Keynes with up to vehicles, achieving significant KPI improvements in larger topologies. The results indicate that the ASP-based routing approach scales well in realistic urban settings and can support future integration into centralized urban traffic management systems.

Abstract

Answer Set Programming (ASP) has demonstrated its potential as an effective tool for concisely representing and reasoning about real-world problems. In this paper, we present an application in which ASP has been successfully used in the context of dynamic traffic distribution for urban networks, within a more general framework devised for solving such a real-world problem. In particular, ASP has been employed for the computation of the "optimal" routes for all the vehicles in the network. We also provide an empirical analysis of the performance of the whole framework, and of its part in which ASP is employed, on two European urban areas, which shows the viability of the framework and the contribution ASP can give.
Paper Structure (18 sections, 3 equations, 4 figures, 1 table)

This paper contains 18 sections, 3 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background on ASP
  3. Syntax.
  4. Semantics.
  5. Syntactic shortcuts.
  6. Case Study and Proposed Solution Framework
  7. Problem Description
  8. The Solution Framework
  9. Network Analyser.
  10. Domain Independent Search.
  11. ASP for the Optimisation Phase
  12. Data Model.
  13. ASP Encoding.
  14. Experiments, Monitoring, and Execution
  15. Empirical Settings and Results
  16. ...and 3 more sections

Figures (4)

  • Figure 1: The solution framework.
  • Figure 2: asp Encoding used during optimisation.
  • Figure 3: The considered sumo model of the central Milton Keynes (left) and Bologna (right) urban areas. Please note that the maps are not in scale, so can not be directly compared.
  • Figure 4: (Top) Boxplot of the solving time of Clingo in correlation with the number of vehicles inside the networks. (Bottom) Histogram representing the number of instances (i.e., each time a new vehicle enters the network) w.r.t. the number of vehicles inside the map. In the two charts, the x-axis has been clustered in bins of $50$.