Improved network performance prediction through data-driven analytics and simulation
This project aims to improve the ability of road authorities to predict network performance in the short term using data-driven analytics and to estimate the impact of automated vehicles (AVs) in longer-term predictions.
The project has two subprojects:
• develop mathematical and data-driven empirical models for short-term traffic prediction (15-30 minutes timeframe). The prediction will be done on link level as well as area level. It aims at utilising emerging traffic datasets to improve network operations.
• simulate the Perth freeway traffic with various AV driving behaviour models and different penetration rates.
The main objectives are to:
- develop mathematical and empirical models to predict short-term traffic flow characteristics for both individual road links and broader sub-regions. At this stage, the models will work from offline historical data. However, with additional development, in the future, they can form the core of a prediction engine to inform Network Operations in real-time once all the live data feeds are available.
- investigate the probability of applying MFDs or area-based empirical models to guide network operations at a sub-regional level. It forms the first step of a paradigm shift from the conventional route-focused operations to a network-wide gating strategy.
- develop simulation models to identify possible impacts to the operation and planning of Perth’s freeways due to the introduction of AVs.