Prediction inner-urban Accidents with regression models Case Study: Kermanshah Province

Accidents are the most important factors in today's world, and accidents within the city may lead to further damages and traffics , due to the importance of this issue ,further research should be focused on accidents. In this thesis, the statistics taken from traffic patterns in Tehran have been used to predict the traffic accidents in the city of Kermanshah with simple linear regression models as well as multiple linear regression correlation between variables in the model. The aim of this study was to predict the total number of crashes, accidents leading to property or bodily damages and or death , according to the statistics of the variables in the Kermanshah Province. Variables in this study or variables that are imported in the regression model are : the study of total number of accidents, accidents with damages, maim and death, the number of accidents on sunny, cloudy or rainy weathers, and other vehicles responsible for accidents within the city (Kermanshah Province) during the four years from 1383 to 1386 on a monthly basis. In this survey softwares like Excel or Spss 16 has been used for calculations. The results show that the average of total number of accidents in the city of Kermanshah is 478.6 per month and the average of 466.53 damages and 67.37 bodily damages and 2.63 death accidents per month. The correlation between the variables shows , the total number of accidents with number of damage accidents equal to 0.941 and With the number of Death accidents is 0. 603 and With the number of accidents on sunny days equal to 0. 660 , that is a strong positive correlation so the increase in each of them will lead to the increase of others.