Dynamic traffic assignment modelling and evaluation: a case study for the Sydney region

Congestion is detrimental to economic productivity and quality of life in a city. With limited revenue to deliver infrastructure improvements, it is essential that transportation planning tools are employed to inform decisions by quantitatively ranking projects according to their value for money. However, existing methods have limitations in realistically representing transport networks from the assumption of steady state conditions over the analysis period. Consequently, there is an emerging focus on dynamic traffic assignment to address the shortfalls of traditional procedures and provide a more realistic basis for evaluating competing proposals. This study stems from developing the Greater Sydney model, which is the first large scale, simulation-based dynamic traffic assignment model in Australia. In contrast to existing methods, dynamic traffic assignment models capture time-varying system conditions. These models are well established in literature and employ a mathematical, simulation-based framework to predict route choice. However, the computational burden of extensive networks and model complexity constrain their widespread application. Analysing the Greater Sydney region is an intriguing task due to its corridor-based road network with minimal grid structure and pronounced peaks in daily traffic flows. The core contribution of this investigation is to form a case study on the practicalities of using regional dynamic traffic assignment models for scenario analysis. Data informing this assessment includes parameters from existing models of Sydney and government plans for capacity improvements by the year 2031. Comparing various situations enables the effectiveness of network changes to be evaluated, and provides insight into the influence of travel demand assumptions. The outcomes highlight the importance of employing dynamic traffic assignment models to rigorously assess alternatives and realistically represent scenarios as traffic flows are inherently time dependent in reality. With further development, this dynamic traffic assignment model would become an invaluable tool for informing transportation planning policy.


  • English

Media Info

  • Pagination: 17p

Subject/Index Terms

Filing Info

  • Accession Number: 01676255
  • Record Type: Publication
  • Source Agency: ARRB
  • Files: ITRD, ATRI
  • Created Date: Jul 26 2018 10:46AM