Page 15 - tmp_Neat

the Australian Bureau of Statistics. This study also is using multinomial logit model in

explaining and predicting heavy-vehicle crash severity.

Study three examines the factors contributing to injury severity in angle crashes involving
heavy vehicles. The data used in the study include all police-reported collisions in Victoria,

Australia, from 2006 to 2016 and information on traffic volumes and road features from
AURIN. This study also compares the binary logit, skewed logistic (Scobit) and random

parameters logit (with uniform and normal distributions) models.

1.5 Research Contributions

This research provides evidence-based recommendations to improve the safety of all road

users in general and heavy vehicle drivers in particular on Australian roads. It is hoped that it
will save lives and prevent injuries on Australian roads.

This study contributes to advancing knowledge in the field for the following reasons:

 To date no research has been conducted to understand the characteristics of single
heavy-vehicle collisions at intersections and mid-blocks.

 To date no study has investigated the effects of neighbourhood socioeconomic
characteristics of both the crash location and road users' residence, on traffic crashes

involving heavy vehicles.

 No previous research has examined the factors contributing to angle crashes in collisions

involving heavy vehicles.

 This research compares several advanced methods to model crash severity, including
binary logistics, skewed logistics and mixed logit models, to provide road safety

professionals with more information on the relative strengths and weaknesses of these

statistical models.

4