CYCLED: city-wide exposure modelling to advance bicycling

2021–2023

This project aims to develop a world-leading platform for city-wide modelling of cycling exposure. This project will provide unparalleled insights into cycling exposure by combining multiple cycling data sources through the use of advanced spatial statistical and machine learning techniques. The expected outcomes of this project are a novel inventory of cycling infrastructure, a cycling route choice modelling system and robust predictions of cycling volumes on individual streets. This project will deliver a step change in cycling that will lead to increased cycling participation, enhanced safety, and improved infrastructure planning, thereby resulting in substantial gains in population and environmental health.

This project is led by Associate Professor Ben Beck and is funded by an Australian Research Council (ARC) Discovery Project (DP210102089).

Read more about the CYCLED Study here.

Relevant publications

  • Pearson L, Bhowmick D, Winters M, Saberi M, Nelson T, Pettit C, Nice K, Dai D, Gupta M, Beck B. Gender differences in bicycle infrastructure use and preferences: A disconnect between ideals and reality. International Journal of Sustainable Transportation. 2025 Dec 22:1-2. doi.org/10.2139/ssrn.5108271
  • Lilasathapornkit T, Bhowmick D, Beck B, Wu H, Pettit C, Nice K, Seneviratne S, Gupta M, Vu HL, Nelson T, Saberi M. Cycling route choice preferences: A taste heterogeneity and exogenous segmentation analysis based on age, gender, Geller typology, and e-bike use. Transportation Research Part A: Policy and Practice. 2025 Nov 1;201:104679. doi.org/10.1016/j.tra.2025.104679
  • Gupta, M., Bhowmick, D., Saberi, M., Pan, S., & Beck, B. (2025). Evaluating the effects of data sparsity on the link-level bicycling volume estimation: A graph convolutional neural network approach. Journal of Cycling and Micromobility Research, 100086. doi.org/10.1016/j.jcmr.2025.100086
  • Gupta, M., Bhowmick, D., Newbury, R., Saberi, M., Pan, S., & Beck, B. (2025). INSPIRE-GNN: Intelligent Sensor Placement to Improve Sparse Bicycling Network Prediction via Reinforcement Learning Boosted Graph Neural Networks. arXiv preprint arXiv:2508.00141. doi.org/10.48550/arXiv.2508.00141
  • Gupta, M., Bhowmick, D., & Beck, B. (2025). BikeVAE-GNN: A Variational Autoencoder-Augmented Hybrid Graph Neural Network for Sparse Bicycle Volume Estimation. arXiv preprint arXiv:2507.19517. doi.org/10.48550/arXiv.2507.19517