Pedestrian safety models for urban environments with high roadside activities

Download Statistics

Downloads

Downloads per month over past year

Kraidi, Raid A. (2019). Pedestrian safety models for urban environments with high roadside activities. University of Birmingham. Ph.D.

Preview

Kraidi2019PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (3MB) | Preview

Abstract

Pedestrian safety is of paramount importance in road safety management. Yet to date there have been a limited number of models capable of capturing satisfactorily all aspects of pedestrian movement in road environments and subsequently suggesting measures to make roads safer for them. This study aimed at investigating such a model with emphasis on road side activities in urban environments. The work followed a methodology which analysed the problem at hand, suggested appropriate mathematical models, and tested the models.

Based on data gathered from appropriately chosen road links and other records, the study showed that the number of bus stoppings, parking activity and pedestrian crossing violations, the traffic speed variation, the number of intersecting side roads, main road traffic volume and intersecting traffic volume were the most significant risk factors related to pedestrian crash risk. In addition, the traffic operating speed was found to be the predominant factor determining the crash severity. Two models were produced to link these variables with: (a) the risk of a pedestrian crash using generalised regression models (GLM); and (b) with the crash severity using logistic regression. Model selection tests were conducted and it was found that the models can capture the pedestrian crash risk satisfactorily. In addition, these models were tested and validated using three evaluation measures including the goodness of fit assessment, a graphical exploration of the observed versus the predicted crash numbers and a field validation.

Furthermore, the study proposed the integration of the new models within the iRAP, an established road safety management system; it demonstrated how this may be achieved and examined their applicability if used in practice. The integration was systematically tested and ii validated and it was found that the new model provided better results compared to the original iRAP.

The developed model, albeit estimated using data from urban roads in Birmingham in the UK, it can be transferred and ultimately used in developing countries subject to appropriate adaptation to local conditions and associated data.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):