Scientific Research and Essays

  • Abbreviation: Sci. Res. Essays
  • Language: English
  • ISSN: 1992-2248
  • DOI: 10.5897/SRE
  • Start Year: 2006
  • Published Articles: 2768

Full Length Research Paper

Accident severity analysis on national highways in Bangladesh using ordered probit model

Md Tanvir Siddique
  • Md Tanvir Siddique
  • School of Engineering and the Built Environment, Transport Research Institute, Edinburgh Napier University, Edinburgh, EH10 5DT, Scotland, UK.
  • Google Scholar

  •  Received: 09 August 2018
  •  Accepted: 29 August 2018
  •  Published: 15 September 2018


National Highways of Bangladesh, where majority of fatal accident takes place, are considered as the accident-prone location. Statistics revealed that despite some major initiatives undertaken to improve road safety, death rate at National Highways of Bangladesh is quite high compared to other developed as well as developing nations. To understand the relationship between injury severity and accident factors, investigation concentrate on different influential factors are responsible for injury severity. Micro Computer Accident Analysis Packages (MAAP5), data of Bangladesh Police from 2004-2015, is used in this investigation to identify the probability of fatal and grievous injury. Preliminary analysis revealed that almost 80% accident recorded as fatal accident and majority of the accidents occurred in fair-weather condition, at straight portion of the road, in non-rush hour and in broad daylight. Conversely, three level of injury severity (dependent variables) was considered in the econometric analysis namely, fatal, grievous and simple injury. Ordered Probit Model examined the dependent variables based on several explanatory variables such as time factors, road geometry, environmental conditions, vehicle categories, driver attitudes, passenger and pedestrian behavior. Considering different level of independent variables, it is found that under the heading of "collision type", "hit Pedestrian" is the statistically significant variable that creates severe injury. Considering vehicle characteristics, non-motorized vehicle, bus and motorcycle was found to be the most vulnerable group of road users in Bangladesh and the probability of fatal injury is much higher for vehicle with multiple defects. This investigation also pointed on some remedial measures that will help the decision maker to set up long as well as short term strategies. For instance, installation of physical barrier at the accident-prone location will minimize the pedestrian fatalities and isolation of non-motorized vehicle and motorcyclist from main carriageway, which would dramatically reduce injury severity on highways. Finally, this investigation also provides some specific guideline for future steps on injury severity analysis in Highways of Bangladesh.

Key words: National Highway, Fatal Accident, Ordered Probit Model, Variables