Prediction of the impacts of traffic incidents

• Main Author: Sim, Obenza Jun Kai
• Other Authors: Mao Kezhi
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/70982
• Institution: Nanyang Technological University

Traffic incidents have significant negative impacts on human lives, directly or indirectly. They can cause traffic congestion, economic loss, personal casualty etc. Thus, traffic management has grown in its importance. Good traffic management consists of various aspects, including traffic congestion control. Traffic congestion is a growing problem that everyone faces today. Using data sets from San Francisco, this project aims to understand the traffic behaviours better by analyzing several aspects of a traffic network. This report focuses on forecasting the severity of an incident occurrence. Resulting impacts of the incident, such as incident queue lengths, will also be investigated. Clustering techniques will be applied to improve prediction accuracy and regression methods will be used in the approach to predict incident duration. Discussion of the suitability and effectiveness of these methods will also be discussed in the conclusion.