Urban traffic network congestion region identification

Analysing the traffic data is a very important topic to improve traffic efficiency. It has no way to manage the flow of traffic if the congestion region is not identified correctly. Thus, identifying the congestion region and its level of congestion are essential for authorities to plan for smooth t...

Full description

Saved in:
Bibliographic Details
Main Author: Soe, Moe Zaw
Other Authors: Su Rong
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2021
Subjects:
Online Access:https://hdl.handle.net/10356/149602
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Analysing the traffic data is a very important topic to improve traffic efficiency. It has no way to manage the flow of traffic if the congestion region is not identified correctly. Thus, identifying the congestion region and its level of congestion are essential for authorities to plan for smooth traffic flow of the cities. Nowadays, there are several tools for studying the traffic data. Machine learning also plays an important part in analysing the traffic data. Researchers utilize it to improve traffic signal control at urban intersections around the world.[1]In this project, simulated traffic network for Woodlands area is first developed using VISSIM software. Subsequently, the extracted data from simulation is utilized for identifying the congestion level of each link. For this part, two different traffic data from peak hours and off-peak hours are being analysed and compared. In order to identify the congestion level, different methods and definitions will be studied to choose the most suitable one for this project. Lastly, machine learning technique will be used to cluster the links with similar congestion level.In clustering experiments, traffic data from two different simulation periods are studied. This project will provide the congestion level of each link and congested regions obtained from clustering links with high congestion level.