Are you interested in applying data analytics and machine learning to real-world transport challenges? We are seeking motivated students to join an exciting research project on Transport Data-Driven Analytics for Big Telematics Data.
Using four years of longitudinal telematics data from heavy vehicles across Australia and New Zealand, this project will apply supervised and unsupervised machine learning techniques to: (i) identify trip ends and segment heavy vehicle trips; (ii) infer the type of stops and categorize activity stops in a tour; and (iii) provide insights into freight transport tour trip patterns and major activity points.
Why Join?
- Work with large-scale transport datasets and cutting-edge machine learning techniques.
- Gain valuable experience in transport data analytics, telematics, and freight transport research.
- Contribute to research with real-world industry and policy implications.
School
Civil and Environmental Engineering
Research Area
Transport engineering | Data science | Machine learning
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
This research will be undertaken jointly at the Research Centre for Integrated Transport Innovations (rCITI) and our telematic industry partner, Euclidic Systems at their CBD Office. rCITI is housed in the School of Civil and Environmental Engineering at UNSW Sydney. rCITI was established in 2011 as a strategic initiative to consolidate and expand the diverse landscape of transport research across the university. rCITI has continued to make remarkable strides since its inception. rCITI aims to reshape the field of multi-modal transport engineering and planning by introducing new innovative techniques and technologies, which enhance society, by integrating across methodological disciplines and contextual considerations.
1. Automated Trip Segmentation & Classification
- Develop models to accurately identify trip ends and segment heavy vehicle trips based on telematics data.
2. Stop Classification & Activity Inference
- Infer the type of stops (e.g., rest breaks, refueling, loading/unloading) and activity stops in a tour to better understand freight movement patterns.
3. Freight Transport Tour Insights
- Identify major freight corridors and activity hubs across Australia and New Zealand.
- Characterize common freight tour patterns and trip chaining behavior of heavy vehicles.
4. Machine Learning Framework for Transport Analytics
- Develop a reproducible machine learning framework that can be applied to other transport datasets for similar analytical purposes.
5. Policy & Industry Insights
- Provide insights that can inform freight transport policies, infrastructure planning, and logistics optimization.
- Support industry stakeholder in understanding fleet behavior and optimizing logistics operations.
6. Academic Contributions
- Potential to publish findings in transport and data science journals/conferences.
- Develop tools and methods that can be used for future transport modeling and freight analytics research.
