Accident Severity Analysis On the North-South Expressway Using Binomial Logistic Regression
Keywords:
Accident severity models, pavement conditions, binomial logistic regressionAbstract
Accidents on Malaysian expressways must be monitored on a regular basis since severe accidents occur on expressways due to higher posted speeds than on other roadways. This study aims to analyse accident severities based on pavement conditions using binomial logistic regression. Data on accident severities and pavement conditions were gathered from the Malaysian Highway Authority (MHA) and the The Royal Malaysian Police (PDRM) respectively. Three binomial logistic regression models were constructed based on four accident severity categories of death, serious injury, minor injury and damage. The accident severity was grouped into different classifications in which Model 1, Model 2 and Model 3 were developed. To assess the model's predictive capabilities, predicted accident severity levels were distinguished with actual accident severity levels. Based on the results, Model 1 and Model 2 except Model 3 have significant pavement conditions and are viable for accident severity predictions whereby both models exhibited high prediction accuracy, has good fit and good in differentiating between two classifications. The model's classifier is better at classifying accident severity classes with more samples than categorizing accident severity classes with fewer data. The odds ratio of both Model 1 and Model 2 revealed that International Roughness Index (IRI) has greater influence in predicting accident severity compared to rut depth (RD) and mean texture depth (MTD) particularly on predicting death. This study suggested that a comparable study be conducted on other road classifications.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2023 International Journal of Integrated Engineering
![Creative Commons License](http://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Open access licenses
Open Access is by licensing the content with a Creative Commons (CC) license.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.