Real-time prediction of crash risk on freeways under fog conditions
|نوع نگارش مقاله||
scopus – master journals – JCR
۴٫۲۷۶ در سال ۲۰۲۰
۲۶ در سال ۲۰۲۱
۰٫۹۰۱ در سال ۲۰۲۰
|شاخص Quartile (چارک)||
Q1 در سال ۲۰۲۰
خرید محصول توسط کلیه کارت های شتاب امکان پذیر است و بلافاصله پس از خرید، لینک دانلود محصول در اختیار شما قرار خواهد گرفت و هر گونه فروش در سایت های دیگر قابل پیگیری خواهد بود.
فهرست مطالب مقاله:
The study aimed to develop a real-time crash risk prediction model on freeways under fog conditions. The data used include traffic surveillance data, fog-related crashes data, road geometry data, and visibility data of Interstate-5 (I-5), Interstate-10 (I-10), Interstate-15 (I-15) and Interstate-405 (I-405) in California, United States. The random forests method was applied for variable selection to identify and rank the most important variables. And then, the Bayesian logistic regression model was employed to develop the real-time crash risk prediction model. The model estimation results show that the explanatory variables contributing to crash risk are different in different time slices before crashes. There are common features: (a) visibility is negatively-correlated with the real-time crash risk under fog condition, and (b) average and standard deviation of vehicle count at upstream station are positively-correlated with crash risk. The model of time slice 3 (interval between 10 to
15 min prior to a crash time) performs best with the lowest false alarm rate and the highest overall accuracy and the largest area under the receiver operating characteristic (ROC) curve. And it can identify over 72% of fog-related crashes with the pre-specified threshold of 0.2.
|بخشی از متن مقاله:|
Adverse weather conditions are known as an unsafe driving environment and can dramatically increase crash rates. The visibility reduction due to fog is one of the major concerns of safety management. Although the percentage of fog related crashes is small, there are still about ۳۰۰–۴۰۰ fog related fatal crashes happening every year in the United States (Hamilton et al., ۲۰۱۴). Crashes are often more severe in foggy weather, involving a large number of vehicles (Wang et al., 2017).
About the cause of fog-related crashes, Chengcheng et al. (2013) found that crashes under low visibility are highly related to the speed difference between the upstream and downstream; Peng et al. (2017)indicated that crash risk would increase significantly as the visibility and mean of headway decrease, while it would increase significantly as the mean speed and volume increase; Wu et al. (2018)suggested that the ‘‘Crash Risk Increase Indicator” worked well in evaluating the increase of crash risk under fog condition, and the crash risk was prone to increase at ramp vicinities in fog conditions.
For aspect of active traffic safety management, some researchers proposed real-time crash risk prediction models to unveil the crash precursors based on the advanced traffic surveillance system. Madanat and Liu (1995)developed a prototype system for real-time incident likelihood prediction based on the two-binary logit. Oh et al. (2005) used the Bayesian classifier to identify traffic flow that may lead to traffic crashes. Lee et al. (2002, 2003) investigated the potential for crashes, and refined the log-linear model. Abdel-Aty and Pande (2004) employed the matched case–control logistic regression modeling technique to predict freeway crashes based on loop detector data. And machine learning algorithms for crash risk prediction are also introduced by Abdel-Aty and Pande. Different real-time crash risk prediction models are developed using the prob- abilistic neural network (Abdel-Aty et al., ۲۰۰۴), multi-layer perceptron neural network (Pande and Abdel-Aty,
2006a,2006b), normalized radial basis function neural network (Pande and Abdel-Aty, 2006a,2006b), and support vector machine (Xu et al., 2017; Yu and Abdel-Aty, 2013), and so on. There is a new trend that combines multiple methods to pre- dict crash risk. For example, random forests method, classification and regression trees (CART) (Hossain and Muromachi,
2011) or random effect logit model (Xu et al., 2016) was applied to identify important factors, and then Bayesian network
(Sun and Sun, ۲۰۱۵; Hossain and Muromachi, 2012) was used to develop the real-time crash risk prediction model.
However, there were few studies that investigated the relationship between real-time traffic parameters and crash occur- rences under fog conditions. Abdel-Aty et al. (2012) developed crash risk prediction models for low visibility, Chengcheng et al. (2013) developed separate crash risk prediction models for different weather conditions, and Wu et al. (2018) propose a new algorithm to evaluate the rear-end collision risk under fog conditions considering reduced visibility. However, in those models, low visibility is a categorical variable rather than a continuous variable.
The primary objective of this study is to identify significant variables leading to fog-related crashes on freeways and develop a crash likelihood prediction model under fog conditions. Fig. 1 shows the flow chart of real-time prediction of crash risk under fog conditions. Random forests method was applied to identify and rank the most important variables via Mean Decrease Accuracy (MDA). And then the real-time crash risk prediction model on freeways under fog conditions was devel- oped using the Bayesian logistic regression model. Historical crash data, traffic flow data, and weather data are used to per- form variable selection, model building, and model estimation. Real-time data is detected by loop detectors and climatological stations on the freeway and it is input into the established model to obtain real-time traffic accident risk pre- diction. The research results will promote a better understanding of the relationship between traffic flow characteristics and crash risk under fog conditions. It will help transportation managers to develop better crash prevention measures under fog conditions.
لطفا پیش از ارسال نظر، خلاصه قوانین زیر را مطالعه کنید: فارسی بنویسید و از کیبورد فارسی استفاده کنید. بهتر است از فضای خالی (Space) بیشازحدِ معمول، شکلک یا ایموجی استفاده نکنید و از کشیدن حروف یا کلمات با صفحهکلید بپرهیزید. نظرات خود را براساس تجربه و استفادهی عملی و با دقت به نکات فنی ارسال کنید؛ بدون تعصب به محصول خاص، مزایا و معایب را بازگو کنید و بهتر است از ارسال نظرات چندکلمهای خودداری کنید.