Abstract
This dissertation focuses on the macroscopic modelling of highway traffic flows. The overarchingaim is to understand the spatiotemporal dynamics of traffic flows in multi-class
multi-lane models. On the one hand, the research starts by developing a single-lane model
for heterogeneous traffic, that we call mixture theory, to explore the relationship between traffic
speeds and traffic compositions. On the other hand, the influence of lane changes on the
longitudinal behaviour of traffic flows is also explored, and this dissertation proposes novel
contribution-based ideas to model vehicles’ discretionary cross-lane behaviour. Traffic instabilities
driven by lane-changing effects are then revealed. Combining the experience in these two
respects, a multi-class multi-lane model is established, in which the single-class lane-changing
model is extended to the multi-class case and the mixture theory is used to compute traffic speeds.
This dissertation understands the model dynamics by exploring travelling wave solutions in it,
and a range of new and intriguing solution structures is discovered.
| Date of Award | 30 Sept 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | R E Wilson (Supervisor) & Anders F Johansson (Supervisor) |