Design and Optimization of Road Networks for Automated Vehicles

Research output: ThesisDissertation (TU Delft)

6 Downloads (Pure)


Automated vehicles (AVs) are on the horizon, and they are expected to deliver traffic safety and efficiency benefits to transportation systems. There are different automation levels for AVs based on the functionalities of the automation systems and their operating design domain (i.e., on which conditions these functionalities can be realized). AVs with limited automation functions are already available on the market; however, fully automated vehicles with unlimited operational design domain (ODD) are not expected in the near future. Reaching a high market penetration rate of fully automated vehicles is a gradual process that can take several decades. Thus for a long time, a heterogeneous mix of traffic with AVs of different automation levels and regular vehicles on the roads will be inevitable. During this transition period with mixed traffic, relying on driving automation technology alone without infrastructure support might compromise the potential safety and efficiency gains of AVs. A proper infrastructure can support AVs’ functionalities, extend their ODD, and improve safety for all road users, while lack of proper infrastructure can negatively influence these factors. Besides, road infrastructure elements usually have a long lifetime and adjusting them can be costly. Hence, there is a strong need for research and planning to ensure that large infrastructure investments provide the highest societal benefits...
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Delft University of Technology
  • van Arem, B., Supervisor
  • van Nes, R., Supervisor
  • Snelder, M., Advisor
Thesis sponsors
Award date20 Jan 2021
Print ISBNs978-90-5584-272-8
Publication statusPublished - 2021


  • Network design problem
  • Automated vehicles
  • Optimizaion

Fingerprint Dive into the research topics of 'Design and Optimization of Road Networks for Automated Vehicles'. Together they form a unique fingerprint.

Cite this