Traffic congestion modelling (in collaboration with BMW)

• Main Author: Khurana Dhriti
• Other Authors: Justin Dauwels
• Format: Theses and Dissertations
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/68679
• Institution: Nanyang Technological University

With rapid urbanization and increasing urge for economic productivity there has been a high growth of migration into urban areas, consequently increasing the problems of traffic congestion in cities like Singapore. In context of the highly complex urban transportation system, it is estimated that 50% of the traffic congestion and time delay takes place due to factors other than the peak hours. These factors include increased vehicle volume, road incidents, work zones and bad weather. Hence there is an urgent need for an advanced traffic predictive models that can guide the commuters to take an appropriate alternative route in order to avoid the congestion. This thesis contributes to this problem by incorporating spatiotemporal data sets such as road incidents, weather information and commuters’ mobility patterns (morning rush hours, day/night time, etc.) to design methods required to build a traffic congestion model. Building an accurate traffic prediction model involves analyzing large sets of historical traffic data. The raw data sets of traffic volume, rainfall intensity and road incidents are extracted and analyzed. The aim of this work is to help avoid traffic jams and accidents by designing methods that can build the urban traffic prediction model using MATLAB. Consequently, traffic jams can be controlled and eliminated. We can thereby, save time and fuel by reducing the total congestion. The decreased emission from vehicles and lower transportation costs benefits the national economy as a whole.