Last modified: 2018-07-13
Abstract
This paper aims at identifying the best explicative models of the road adherence, lighting conditions and weather factors’ influence on road accidents occurred on various road features. The road accidents that caused victims between 01.01.2014 – 21.12.2016 in Romania have been analyzed and grouped on a monthly basis. The total number of road traffic accidents that caused injuries, accidents that occurred in alignments, in curves and intersections have been analyzed as dependent variables.
Twenty four different features of lighting and weather conditions and also of road adherence have been taken into consideration as explanatory variables. Multiple linear regressions have been applied and their global significances have been tested by using Fisher test. The individual meanings of the explanatory variables coefficients have also been tested, in order to choose the best explained model.
The wet roads and those covered by snow, cloudy sky and rains are the most present in these explanatory models. The slippery roadways, no street lighting and the rains explain the accidents produced in curves.
The cloudy sky, the functional and broken street lighting and the normal weather conditions explain the road accidents in a proportion of 98,90 %.