Dr. A.P. (Amir Pooyan) Afghari

Dr. A.P. (Amir Pooyan) Afghari

Profiel

Biografie

Amir is an assistant professor of probabilistic methods in transport safety at TU Delft. His research interest is at the intersection of transport engineeringsocial science and data science: understanding human behaviour and its impact on road safety using observational data, statistical and econometric methods and machine learning algorithms.
 

Prior to joining TU Delft, Amir worked at Queensland University of Technology, University of Queensland, McGill University, École Polytechnique Montréal, and Concordia University. He completed his PhD in road safety at the University of Queensland in Brisbane, Australia. 

Expertise

Amir is a methodologist. His expertise is in advanced statistical and econometric methods (e.g. structural equation models, simultaneous equation models, latent variable models, generalized linear models, discrete choice models, and Bayesian hierarchical models) applied to transport data. 
 

He has more than 14 years of experience in road safety research, and has published many peer-reviewed articles in prestigious journals in transport engineering and social sciences including:

  • Analytic Methods in Accident Research 
  • Journal of Choice Modelling
  • Accident Analysis and Prevention 
  • Transportation Research Part A: Policy and Practice
  • Transportation Research Part F: Traffic Psychology and Behaviour
  • Transportation Research Interdisciplinary Perspectives
  • Travel Behaviour and Society
  • Sustainable Cities and Society
  • Traffic Injury Prevention
  • Transportation Research Records

He has also contributed to a textbook on safe mobility, co-authored by the world’s leading academics in the field of road safety.

Some examples of Amir’s research are:

  • Behavioural interactions between pedestrians and cyclists
  • Social influence and speeding behaviour of drivers
  • Anticipation and road predictability
  • Driver sleepiness
  • Seatbelt use behaviour of vehicle occupants 
  • Mobile-phone distraction while driving
  • Risk-compensating behaviour of distracted drivers
  • Road users' receptivity towards autonomous vehicles
  • Missing data analysis in road safety
  • Bayesian inference in road safety
  • Analysis of crash data with excess zeros
  • Crash blackspot identification
  • Bicycle sharing systems
  • Automated data analytics for vulnerable road users.

Projects

IVORY: AI for vision zero in road safety - Horizon Marie Skłodowska-Curie Actions Doctoral Network [work package lead]. https://ivory-network.eu/ 

SOBER: Investigating the effects of social influence on driving behaviour in urban settings - TU Delft Safety and Security Institute [lead investigator]. https://www.tudelft.nl/2023/safety-and-security-institute/tu-delft-safety-security-institute-funds-8-seed-projects

PHOEBE: Predictive approaches for safer urban environments - European Union’s Horizon 2020 Research and Innovation Programme [work package lead]. https://www.phoebe-project.eu/

GAZETOAV: Investigating gaze behaviour of road users when interacting with automated vehicles in future mixed traffic using eye-tracking - TU Delft Safety and Security Institute [co-investigator]. https://www.tudelft.nl/tu-delft-safety-security-institute/research/seed-funding-projects

i-DREAMS: Smart driver and road environment assessment and monitoring system, a naturalistic driving study - European Union’s Horizon 2020 Research and Innovation Programme [work package co-lead].  https://idreamsproject.eu/wp/

Codes

You can use the following link to access Amir's GitHub repository, and the codes he has developed for advanced statistical and econometric models in his work: 

https://github.com/Econometrics-in-r 

These codes have been produced as part of his research during his academic career. Please cite the corresponding articles if you use them in any kind.

Supervision

“Explainable AI for road safety: benchmarking AI methods and data”, PhD, Safety and Security Science Section, Faculty of Technology, Polivy and Management, Delft University of Technology, Status: in progress.

“AI to mitigate driver distraction and drowsiness at different levels of automation”, PhD, Safety and Security Science Section, Faculty of Technology, Polivy and Management, Delft University of Technology, Status: in progress.

“An innovative integrated evidence-based decision making method for complex engineering structures and systems”, PhD, Safety and Security Science Section, Faculty of Technology, Polivy and Management, Delft University of Technology, Status: in progress.

“Roundabout safety: bicyclists’ perceptions”, Masters, Transport and Planning, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Status: graduated in 2024.

"New knee implant variants under the medical device regulation: A Bayesian approach to estimate their performance", Masters, Management of Technology, Faculty of Technology, Polivy and Management, Delft University of Technology, Status: graduated in 2024.

“Perceived risk of interaction with e-bikes”, Masters, Transport and Planning, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Status: graduated in 2023.

“Relationship between road geometric design consistency and vehicular crashes”, Ph.D., School of Civil Engineering, Queensland University of Technology, Status: graduated in 2023.

“A machine learning model for predicting individual charging profiles of electric vehicle users”, Masters, Complex Systems Engineering and Management (CoSEM), Faculty of Technology, Policy and Management, Delft University of Technology, Status: graduated in 2023.

“Safety effects of motorcycle paths at unsignalised intersections in Kampala, Uganda”, Masters, Transport and Planning, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Status: graduated in 2022.

“Dynamics of disruption and recovery in railway networks”, Masters, Transport, Infrastructure & Logistics, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Status: graduated in 2021.

“Developing a Risk Assessment Model for Road Segments in Queensland”, Bachelors, School of Civil Engineering, Queensland University of Technology, Status: graduated in 2019.

“Refining Australia’s road risk assessment model”, Bachelors, School of Civil Engineering, Queensland University of Technology, Status: graduated in 2019.

Media

MAR 2023 - Road safety and social pressure to speed: https://www.tudelft.nl/2023/safety-and-security-institute/road-safety-social-pressure-to-speed-a-chat-with-amir-pooyan-afghari

SEPT 2022 - Improving road safety with a novel driver intervention and feedback system: https://www.tudelft.nl/en/2022/tbm/improving-road-safety-with-a-novel-driver-intervention-and-feedback-system

JUNE 2022 - What can aviation, maritime and rail traffic learn from our experiences in the road sector and what can we learn from theirs? https://idreamsproject.eu/wp/wp-content/uploads/2022/06/EN-Newsletter5.pdf

MAY 2016 - New model to better predict motor vehicle crash blackspots: https://phys.org/pdf381482607.pdf

Lees meer

Publicaties