Please use this identifier to cite or link to this item:
Title: Modelling traffic accidents using duration analysis techniques: a case study of Abu Dhabi
Authors: Al Kaabi, Abdulla Mohammed Saeed Khalaf.
Issue Date: 2013
Publisher: Newcastle University
Abstract: One of the main aims of Traffic Incident Management (TIM) is to reduce the duration of the disruption to traffic caused by an accident. Several approaches have been applied in the past in order to analyse and predict this. Incident duration can be broken down into four time intervals: reporting, response, clearance and recovery. Accurate models of each interval allow traffic controllers to deploy resources efficiently, thereby minimising an accident’s effect on traffic flow and congestion. This may, in turn, lead to a reduction in other adverse impacts of traffic accidents such as air pollution, fuel consumption and secondary crashes. A new approach to this problem, based on the accidents’ characteristics, was developed using a fully parametric hazard based modelling technique to predict accident durations. The road network around Abu Dhabi, capital of the UAE, was used as a case study. Data was obtained from the UAE Federal Traffic Statistics System (FTSS) and the Abu Dhabi Serious Collision Investigation Section (ASCIS). These data included the start and end of each time interval, the total accident duration, temporal, geographical, environmental and other accident characteristics. To analyse the total duration, the analysis was conducted using three time intervals. Accordingly, fully parametric Accelerated Failure Time (AFT) models were created for the purpose of reporting time, response time, and clearance time (all urban roads) and response time (rural freeways), depending on the data available. Analysis showed that the time intervals had different distributions. In addition, there was no similarity in the variable that affected each interval. The results also revealed that weaknesses exist in the current practices of TIM in Abu Dhabi. The results of the analysis were used to create decision trees to aid traffic controllers with decisions regarding traffic diversion and disseminating traffic information to travellers.
Description: PhD Thesis
Appears in Collections:School of Civil Engineering and Geosciences

Files in This Item:
File Description SizeFormat 
Al Kaabi, A.M.S.K. 13.pdfThesis2.86 MBAdobe PDFView/Open
dspacelicence.pdfLicence43.82 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.