Frequency Modelling and Risk Evaluation of Road Crashes in Sylhet Region of Bangladesh
| Autor: | Md. Jahid Hasan, Md. Aminul Islam, Md. Ruhul Amin, Md. Bashirul Haque |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Driving factors
education.field_of_study General Neuroscience Applied Mathematics Population Negative binomial distribution Aerospace Engineering Policy analysis Computer Science Applications symbols.namesake Geography Overdispersion Control and Systems Engineering SAFER Environmental health Automotive Engineering symbols Poisson regression education Software Predictive modelling Information Systems |
| Zdroj: | International Journal of Intelligent Transportation Systems Research. 20:90-102 |
| ISSN: | 1868-8659 1348-8503 |
| DOI: | 10.1007/s13177-021-00275-0 |
| Popis: | Road traffic accident is a concerning issue for many low to middle-income countries (LMICs). An appropriate accident prevention strategy is thus necessary for ensuring a safer road that requires scientific investigation of accident risk and accident prediction. Therefore, this study aims for accident hazard risk analysis of Sylhet district of Bangladesh to identify the vulnerable zones of accident occurrence. Highly vulnerable areas are found geographically connected. To explore the driving factors of accident occurrence, accident frequency prediction models are developed. The negative binomial regression model is found to outperform over the Poisson regression model indicating the overdispersion of the data. The estimated model confirms a significant level of association of accident frequency with zone level sociodemographic, roadway and traffic characteristics such as population, number of commercial activities, length of the road, width of the road and average annual daily traffic. The findings of this study can be useful for relevant policy analysis to minimize the road crash severity and frequency in the cities of LMICs that share similar characteristics of the study area Sylhet. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink