Seventy percent of the traffic crash fatalities of Iran happen on rural roads, and a significant proportion of the rural roads network of this country is constituted of the main two-lane, two-way roads. The purpose of this study is to identify the most important factors which affect injury severity of drivers involved in traffic crashes on these roads, so that by eliminating or controlling such factors an overall safety improvement can be accomplished. Using the Classification and Regression Tree (CART), one of the powerful data mining tools, the crash data pertaining to the last three years (2006-2008) were analyzed. The variable selection procedure was carried out on the basis of Variable Importance Measure (VIM) which is one of the CART method outputs. The results revealed that not using the seat belt, improper overtaking and speeding are the most important factors associated with injury severity. KEYWORDS: injury severity; traffic safety; data mining; Classification and Regression Trees (CART); Variable Importance Measure (VIM)

Objective: To identify the significant factors that influence head-on conflicts resulting from dangerous overtaking maneuvers on 2-lane rural roads in Iran. Methods: A traffic conflict technique was applied to 12 two-lane rural roads in order to investigate the potential situations for accidents to occur and thus to identify the geometric and traffic factors affecting traffic conflicts. Traffic data were collected via the inductive loop detectors installed on these roads, and geometric characteristics were obtained through field observations. Two groups of data were then analyzed independently by Pearson's chi-square test to evaluate their relationship to traffic conflicts. The independent variables were percentage of time spent following (PTSF), percentage of heavy vehicles, directional distribution of traffic (DDT), mean speed, speed standard deviation, section type, road width, longitudinal slope, holiday or workday, and lighting condition. Results: It was indicated that increasing the PTSF, decreasing the percentage of heavy vehicles, increasing the mean speed (up to 75 km/h), increasing DDT in the range of 0 to 60 percent, and decreasing the standard deviation of speed significantly increased the occurrence of traffic conflicts. It was also revealed that traffic conflicts occur more frequently on curve sections and on workdays. The variables road width, slope, and lighting condition were found to have a minor effect on conflict occurrence. Conclusion: To reduce the number of head-on conflicts on the aforementioned roads, some remedial measures are suggested, such as not constructing long “No Passing” zones and constructing passing lanes where necessary; keeping road width at the standard value; constructing roads with horizontal curves and a high radius and using appropriate road markings and overtaking-forbidden signs where it is impossible to modify the radius; providing enough light and installing caution signs/devices on the roads; and intensifying police control and supervision on workdays, especially in peak hours.

Iran is a country with one of the highest rates of traffic crash fatality and injury, and seventy percent of these fatalities happen on rural roads. The objective of this study is to identify the significant factors influencing injury severity among drivers involved in crashes on two kinds of major rural roads in Iran: two-lane, two-way roads and freeways. According to the dataset, 213569 drivers were involved in rural road crashes in Iran, over the 3 years from 2006 to 2008. The Classification And Regression Tree method (CART) was applied for 13 independent variables, and one target variable of injury severity with 3 classes of no-injury, injury and fatality. Some of the independent variables were cause of crash, collision type, weather conditions, road surface conditions, driver's age and gender and seat belt usage. The CART model was trained by 70% of these data, and tested with the rest. It was indicated that seat belt use is the most important safety factor for two-lane, two-way rural roads, but on freeways, the importance of this variable is less. Cause of crash, also turned out to be the next most important variable. The results showed that for two-lane, two-way rural roads, "improper overtaking" and "speeding", and for rural freeways, "inattention to traffic ahead", "vehicle defect", and "movement of pedestrians, livestock and unauthorized vehicles on freeways" are the most serious causes of increasing injury severity. The analysis results revealed seat belt use, cause of crash and collision type as the most important variables influencing the injury severity of traffic crashes. To deal with these problems, intensifying police enforcement by means of mobile patrol vehicles, constructing overtaking lanes where necessary, and prohibiting the crossing of pedestrians and livestock and the driving of unauthorized vehicles on freeways are necessary. Moreover, creating a rumble strip on the two edges of roads, and paying attention to the design consistency of roads can be a helpful factor in order to prevent events such as "overturning" and improve the overall safety of freeways.