Using Gis And Statistical Models For Traffic Accidents Analysis

This dissertation, "Using GIS and Statistical Models for Traffic Accidents Analysis: a Case Study of the Tuen Mun Town Centre" by C P, Eric, Yau, 丘之鵬, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of dissertation entitled Using GIS and Statistical Models for Traffic Accidents Analysis - A Case Study of the Tuen Mun Town Centre Submitted by Eric C. P. Yau For the degree of Master of Arts in Transport Policy and Planning at The University of Hong Kong in December 2006 Using the data from the Traffic Accident Data System (TRADS) of the Transport Department of Hong Kong, this study focused on finding the factors which had significant relationship with the occurrence of traffic accident. The study area is at the Tuen Mun town centre and the targeted dataset were acquired from the period of January to December of 2004. This study has used the Geographic Information System (GIS) and statistical modelling system to analyse the dataset. The independent variables being explored in the study included AADT, road gradient, speed limit, curvature and adjacency to the intersection of road. The dataset was being testified its severity of injury to the relationships of each independent variable. Statistical models like Chi-square test, Poisson regression model and the negative binomial model were employed to analyse the dataset. Modelled result have shown that speed limit was statistically significant to the occurrence of fatal and seriously injured accidents, while for those slightly injured accidents none of the variables were found to have direct influence to traffic accident. This study has laid its footprint on integrating GIS with statistical model for analysis of traffic accident. The findings and the designed prototype of this study can be useful for transport professionals to improve their planning as well as preventive measures. DOI: 10.5353/th_b3763911 Subjects: Traffic accidents - Statistical methods Geographic information systems Traffic accidents - China - Hong Kong Traffic accidents - Tuen Mun