Traffic conflict modelling and analysis for Singapore
Traffic volume is widely used as the basis for a measure of exposure in traffic accident analysis. For example, accident rates on a road section are often expressed in number of accidents per million vehicle miles or kilometres of travel. Studies show that the correlation between traffic volume and...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78435 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Traffic volume is widely used as the basis for a measure of exposure in traffic accident analysis. For example, accident rates on a road section are often expressed in number of accidents per million vehicle miles or kilometres of travel. Studies show that the correlation between traffic volume and the accident rates on a section of road follow a generally proportional trend – meaning as traffic volume increases, it is expected that the number of traffic accidents will increase likewise. This study will be focusing on traffic junctions, as based on previous studies, it has been observed that a high proportion of crashes and serious injuries on urban roads occur at the junctions. It will focus on using microsimulation as a tool to realistically model conflicts at the traffic junctions and correlate the simulated conflicts with actual collisions. Using the traffic microsimulator VISSIM and actual traffic volume data, identified high-incident junctions in Singapore will be simulated for 3 separate time periods of the day – morning peak, evening peak, and midday. The simulation results are then run through the Surrogate Safety Assessment Model using predetermined parameters, yielding a detailed breakdown on the conflicts occurring during the simulation. Using QGIS, these conflicts are plotted onto the road network of the junction being simulated, and compare the conflicts and collision records and visualise the results of the pilot experiment. It was observed that during the simulation of the evening peak, where the traffic volume data is especially high, the number of conflicts increase disproportionately to the increase in volume. This leads to an inflation of conflict count in the simulation. The simulations done of the morning and afternoon peak yielded a simulated conflict count that was more closely related to actual collision count, and the simulations of the evening peak yielded a conflict count that is less closely related to the actual collision count |
|---|