Prioritising People: Advancing Active Transport through Engineering, Modelling, and Technology
Delve into the heart of active transport with a focus on creating people-first traffic and transport engineering solutions, from enhancing pedestrian trip generation models to identifying optimal cycling routes at intersections. Explore the scalability challenges of machine learning models for estimating walking and cycling volumes in large networks, and learn how these innovations are shaping the future of sustainable urban mobility.
Session Summary
- The Heart of Road Use - Building a people-first approach into traffic engineering | Professor Anna Timperio, Deakin University
- Spatial Transferability of Pedestrian Trip Generation Models | Fatemeh Nourmohammadi, UNSW
- Identifying preferred cycling options at intersections | Natalie Sham, Stantec
- Scalability challenges of machine learning models for estimating walking and cycling volumes in large networks | Dr Meead Saberi, UNSW
Presenters
The Heart of Road Use - Building a people-first approach into traffic engineering
Professor Anna Timperio
Deakin University (Vic)
Prof Anna Timperio is a Deputy Director of the Institute for Physical Activity and Nutrition (IPAN) and Deputy Head of the School of Exercise and Nutrition Sciences at Deakin University. Prof Timperio is a global leader in physical activity research with more than 24 years of experience. Their research focuses on the behavioural epidemiology of physical activity. A large focus of that work has been on understanding influences on physical activity, active travel and sedentary behaviours among children and young people, particularly the role of the neighbourhood environment. She has published >270 peer-reviewed papers and 10 book chapters, and has been recognised as a 'Highly Cited Researcher' (ranked in the top 1% of most cited researchers in their category) seven times since 2015. They have previously held Associate Editor roles for two journals.
Abstract Synopsis
This abstract highlights the importance of a people-first approach in traffic engineering to promote heart health by prioritizing equitable, accessible, and inclusive road design. It presents evidence for shifting road planning from car-centric designs to those that enhance transport choice, quality of life, and walkability.
The presentation will discuss topics including Australia’s car dependency, links to health, economic benefits of active and public transport, traffic calming, speed reduction, and streetscape improvements. Based on the latest research, the session explores how transport mode choice impacts physical activity, sedentary behavior, and cardiovascular health, guiding best practices for healthier, more livable urban environments.
Spatial Transferability of Pedestrian Trip Generation Models
Fatemeh Nourmohammadi
UNSW
Fatemeh Nourmohammadi was born in 1997 and is a second-year PhD student at the University of New South Wales (UNSW), specializing in pedestrian demand modeling. She completed her master's degree in transportation engineering in South Korea. Fatemeh combines her background in industrial engineering with expertise in data science, data engineering, AI, and statistical modeling to develop innovative solutions for urban mobility challenges. Her research focuses on creating advanced models to better understand pedestrian behavior, aiming to enhance walkability and optimize city infrastructure. With a strong foundation in data-driven methods, she bridges the gap between traditional transportation engineering and modern AI applications, contributing valuable insights for urban planners and policymakers.
Abstract Synopsis
The availability and consistency of pedestrian travel data vary across different locations, often requiring the transfer of estimated models in the absence of comprehensive local data. However, the extent to which pedestrian demand models are spatially transferable is not well understood.
This abstract examines the spatial transferability of pedestrian trip generation models using data from Sydney, Melbourne, and Brisbane. Aggregated models (e.g., at Local Government Area level) show reasonable transferability under certain conditions, while disaggregated models face more limitations. Bayesian regression models effectively balance performance and interpretability, making them valuable for estimating walking trip generation.
By combining various modeling approaches, the research provides insights into walking behavior at city-wide and localised scales, aiding urban planning in data-scarce areas. This is the first study to evaluate the transferability of walk trip generation models across different scales.
Co-Author
Professor Taha Hossein Rashidi
UNSW
Dr Taha Rashidi is a Professor in Transport Engineering at the School of Civil and Environmental Engineering at UNSW and a member of the Research Centre for Integrated Transport Innovation (rCITI).
Prof Rashidi's research and teaching demonstrate how effectively transport engineering can draw upon the strengths of a broad range of disciplines to inform smart-city solutions.
Prof Rashidi is currently leading research into the interconnectivity between travel behaviour and time use and the potential of new mobility technologies to influence this paradigm. Prof Rashidi is also examining the capacity of social media data to complement existing data resources as part of the development of an integrated multi-level modelling framework to demonstrate the relationships between land use and transport systems and the consequences this has for city planning and travel behaviour more broadly.
Identifying preferred cycling options at intersections
Natalie Sham
Stantec
Natalie thrives to achieve sustainable outcomes and believes leading positive changes in communities through her work. With over three years of experience in the industry, Natalie has proven track record of project management and delivery. Her extensive research and data analysis skills have been supporting a number of transport strategies with data-driven insights focused on human-centred design.
Abstract Synopsis
There is no consensus among traffic engineers and transport planners at a national level on how to determine the most appropriate cycling infrastructure at intersections. As a consequence, good treatment options are often value engineered out of intersection designs, leaving gaps within cycling networks.
With consideration to movement and place, this presentation focusses on evaluating infrastructure options at intersections and utilising a multi-criteria assessment method to inform a decision on the preferred option. In particular, the MCA framework would take into consideration elements such as accident exposure, land constraints, demand of cyclists, cycling route choice and connectivity to the broader network.
This presentation outlines the above approach, its potential usage through a number of case studies and the opportunities and limitations with respect to its practical implementation.
Scalability challenges of machine learning models for estimating walking and cycling volumes in large networks
Dr Meead Saberi
UNSW
Dr. Meead Saberi is an Associate Professor in the School of Civil and Environmental Engineering at the University of New South Wales (UNSW), Sydney, and co-founder of footpath.ai, a UNSW spinout that leverages GeoAI and computer vision to map pedestrian infrastructure. He leads the CityX research lab within the Research Centre for Integrated Transport Innovation (rCITI), which focuses on advanced transportation network modeling, pedestrian dynamics, and urban data analytics. Dr. Saberi's research spans diverse areas including traffic flow theory, large-scale transportation network simulation, complex networks, and urban data visualization. He earned his PhD in Transportation Systems Analysis and Planning from Northwestern University, USA, and holds a Master's degree in Transportation Engineering from Portland State University, USA. With a strong background in academic and applied research, his work aims to provide innovative, data-driven solutions to enhance urban mobility, pedestrian safety, and the sustainability of transport infrastructure.
Abstract Synopsis
The surge in active transport initiatives has emphasized the need for accurate estimates of walking and cycling volumes. Addressing scalability challenges in applying machine learning models to extensive urban networks, this study focuses on the New South Wales Six Cities Region. Models were developed to estimate walking volumes across 188,999 links and cycling volumes across 114,885 links.
Traditional methods relying on crowdsourced and mobile phone data, such as Strava and telecommunications-based movement patterns, are limited by data gaps and biases. To address these limitations, this study integrated diverse datasets, including population, land use, and official pedestrian and cycling counts, and utilised spatial and temporal cross-validation to enhance model accuracy.
The findings of this study will be outlined in this presentation and provide actionable insights for designing sustainable, data-driven transport systems that encourage active travel modes and support healthier, environmentally friendly cities.
Co-Author
Tanapon Lilasathapornkit
UNSW
Tanapon Lilasathapornkit is the Chief Technology Officer (CTO) and co-founder of footpath.ai, a UNSW spinout company focused on automating the mapping of pedestrian infrastructure using GeoAI and computer vision. He leads the technological innovations at footpath.ai, driving the development of cutting-edge tools to improve urban walkability and sustainability through advanced data analytics and AI-powered solutions.
