Assessment of Motorcycle Safety in Wyoming: Fatal and Severe Crashes, Contributing Factors and Potential Countermeasures [Addendum]
-
2023-06-01
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report Addendum (December 2022 – June 2023)
-
Contracting Officer:
-
Corporate Publisher:
-
Abstract:Motorcycle riders and passengers are much more likely to be killed or severely injured in a crash, and on average about 15 percent of all traffic fatalities include motorcyclists. Motorcyclists and their passengers are particularly vulnerable on the roads, which accounts for their higher percentage of fatal and severe crash accidents. To take appropriate safety measures, it is important to investigate the factors that affect crash injury severity. This part of the study explores the use of two machine learning methods: Random Forest and Support Vector Machine classifiers. The Random Forest model showed better performance, therefore it was used to assess the significance of each type of contributory variable on the crash, person, and vehicle level datasets. The most important factors identified from the crash-level analysis were driver action, vehicle maneuver, type of collision, junction relation and helmet use. The driver injury area, driver action, and presence of alcohol in particular were observed to influence crash injury predictions at the person-level, whereas the vehicle maneuver, most damaged area of the vehicle and the vertical grade of the roadway were found to be relevant to motorcycle injury severity predictions at the vehicle-level. Collectively, this study revealed how motorcycle crash injury severity can be predicted from the identified factors in each data dataset. This finding will assist WYDOT and other transportation agencies in providing proactive solutions for motorcycle crashes.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: