Prediction of road accident severity using the ordered probit model



In 2012 almost 28,000 people died on European Union (EU) roads due to road traffic accidents. Portugal, regardless the remarkable progress in the last decades, is still well above the EU average number of road fatalities. As a result, the country established an ambitious target – “to reach 62 deaths per million inhabitants by 2015”. Practitioners involved in road safety management are encouraged to contribute to the fulfilment of that target. An effective road safety management requires a good insight in the factors that are believed to be related to road traffic accidents. Based on this framework, several research studies have been conducted over the years aiming at identifying factors that may influence both the frequency and the severity of road traffic accidents. However, additional research still pertinent for the Portuguese road network conditions. This study intends to contribute for a better understanding of the factors affecting the occurrence of accidents and, in particular, those that affect its severity at a national level. Firstly, a summarized state-of-the-art is presented aiming at identifying the more appropriate methodologies and the most significant explanatory variables affecting the road accident severity. Lastly, supported by accident data collected from an official accident statistics database, an ordered probit model is applied to examine the influence of a number of factors on the injury severity faced by motor-vehicle occupants involved in road accidents. The model estimation results suggest that some types of road accidents, namely the rollover-type, run-off-road, collisions against fixed objects and head-on collisions, appear to be the major contributors for the most severe injury level. Also, those who travel in a light-vehicle, at a two-way road and on dry road surface injuries tend to suffer more severe injuries than those who travel in a heavy vehicle, at a one-way road, and on a wet road surface. In contrast, the driver´s seat is clearly the safest seating position, urban areas seam to originate less serious accidents than rural areas, and women tend to be more likely to suffer serious or fatal injuries than men.


road accidents; accident severity models; ordered probit model


Transportation and Road Modelling

Related Project

TICE.MOBILIDADE - Sistemas de Mobilidade Centrado no Utilizador


Euro Working Group on Transportation 2014, April 2014

Cited by

Year 2016 : 2 citations

 Alkheder, Sharaf, Madhar Taamneh, and Salah Taamneh. "Severity Prediction of Traffic Accident Using an Artificial Neural Network." Journal of Forecasting (2016).

 Shibani, Abdussalam, and Most Shamim Ara Pervin. "Analysis of traffic accident severity on Great Britain roadways and junctions." International Journal of the Built Environment and Asset Management 2.1 (2016): 37-66.