نوع مقاله : علمی - پژوهشی
نویسندگان
1 عضو هیئت علمی گروه عمران (راه و ترابری) دانشکده فنی، دانشگاه گیلان
2 دانشجوی کارشناسی ارشد- دانشگاه گیلان
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Researchers have utilized different forms of fixed and variable segmentation in accident frequency models to conduct safety studies on rural roads. The conducted research suggests that dynamic segmentation yields superior results in the realm of rural road safety, particularly when analyzing factors that influence the frequency of high-crash-potential road segments. Generally, the objective of analyzing the safety of rural road sections is to facilitate the development of effective and efficient strategies to enhance safety by identifying pertinent variables associated with accidents for users of these road types. Hence, this research aims to present a spatial approach based on GIS (Geographic Information System) to dynamically segment rural two-lane two-way roads and determine the factors influencing accident frequency within these segments, with an emphasis on road geometry and weather-related variables. To achieve this, the Poisson model with random parameters was applied to accident data from the years 2010-2015, pertaining to the dynamically segmented portion of the Lushan-Qazvin two-lane two-way road. The findings of the study reveal that, among the selected factors in the examined segments, the presence of a separate median, accident location, roadway conditions, clear weather, driver's haste and unnecessary acceleration, and collisions with fixed objects have the greatest impact on accident frequency, with respective marginal effects of 8.304, 4.763, 3.825, 1.909, and 1.205.
کلیدواژهها [English]