عنوان مقاله [English]
The main reasons of injury and death all over the world and especially in Iran are mainly run off road, roll over, and crushing with the fixed objects at the roadside. This is why the road safety engineers should pay special attention to roadside hazards, otherwise as a consequence the number of fatalities would increase consequently although exact data are not available regarding this issue in Iran, evidence shows that a large number of dangerous road accidents in Iran is due to vehicle run-off-the road. Nowadays introduction of a model which would have the capability of anticipating the number of run-off-roads would assist the engineers to evaluate the number of run-off in order to carry out a proper incident management In this research a model has been introduced that is capable of anticipating the number of roll-over due to run-off-roads in each direction of roads. The effect of each parameter has been thoroughly investigated on the roadside safety, and finally its effects on the number of accident occurred in Iranian Roads. In this research, two models for prediction of vehicle run – off - road were evaluated.For evaluating the differences between the obtained amounts by selected models and the real rates for the run off roads, the X test was used. This test was chosen due to the quantitative characteristics of accidents (which are usually numbered) and the real amount of run-off- road also due to the number of accidents in each direction of roads and the fitness of theoric models. The following model had the most realistic similarities with the predicted rates and therefore was chosen for utilization. This model is shown below:ADT/1000000)×exp(βst-(0.04×ADT/1000)+lnf + 0.12HC+0.05VG) ×E= (365 Where:E = expected number of road side encroachment per kilometer annually.ADT = average annual daily traffic (in number of vehicles) from 1000 to 12000.B st = state constant with a default value of 0.45 for those areas (or states) where rural two- lane roads data are available, it is recommended that B st be estimated as the natural log of the run-off-road accident rate for road segment with low ADT ( e.g., 2000) that are relatively straight (e.g., horizontal curvature < 3 degrees) and level (e.g., vertical grade < 3%) .
Lnf= 0, 0.20, and 0.44, respectively, for road segments with 3.6m 3.3m 3 m ft wide lane.HC = horizontal curvature) for 0 to 30 degree. VG = vertical grade (in percent) from 0 to 10 percent.Using the above model it would be possible to predict the total run off road accidents.
On the other hand, the main objective of this research is estimating the amount and percentage of rollover due to run off roads. Due to this fact it was tried to consider an index namely Roadside Hazard Index. By using this index the above objective may be achieved.Using this index which is always >1 it would be possible to estimate the rollover due to run off roads, as follows:The road side safety condition is depended to RH index.RH = the length of risk ratio to the length of axis: