This study develops new real-time freeway rear-end crash potential predictors using support vector machine(SVM) technique. The relationship between rear-end crash occurrences and traffic conditions were explored using...This study develops new real-time freeway rear-end crash potential predictors using support vector machine(SVM) technique. The relationship between rear-end crash occurrences and traffic conditions were explored using historical loop detector data from Interstate-894 in Milwaukee, Wisconsin, USA. The extracted loop detection data were aggregated over different stations and time intervals to produce explanatory features. A feature selection process, which addresses the interaction between SVM classifiers and explanatory features, was adopted to identify the features that significantly influence rear-end crashes. Afterwards, the identified significant explanatory features over three separate time levels were used to train three SVM models. In the end, the multi-layer perceptron(MLP) artificial neural network models were used as benchmarks to evaluate the performance of SVM models. The results show that the proposed feature selection procedure greatly enhances the accuracy and generalization capability of SVM models. Moreover, the optimal SVM classifier achieves 81.1% overall prediction precision rate. In comparison with MLP artificial neural networks, SVM models provide better results in terms of crash prediction accuracy and false positive rate, which confirms the superior performance of SVM technique in rear-end crash potential prediction analysis.展开更多
Although either absolute speed or speed difference can be considered as a measure for speed consistency, few researches consider both in practice. The factor analysis method was introduced to extract an optimal number...Although either absolute speed or speed difference can be considered as a measure for speed consistency, few researches consider both in practice. The factor analysis method was introduced to extract an optimal number of factors from numerous original measures. The freeway diverging zone was divided into four elements, namely the upstream, the diverge area, the downstream and the exit ramp. Operating speeds together with individual vehicle speeds were collected at each element with radar guns. Following the factor analysis procedure, two factors, which explain 96.722% of the variance in the original data, were retained from the initial seven speed measures. According to the loadings after Varimax rotation, the two factors are clearly classified into two categories. The first category is named "speed scale" reflecting the absolute speed, and the other one is named "speed dispersion" interpreting speed discreteness. Then, the weighted score of speed consistency for each diverge area is given in terms of linear combination of the two retained factors. To facilitate the level classification of speed consistency, the weighted scores are normalized in the range of (0, 1.0). The criterion for speed consistency classification is given as 0≤F N <0.30, good consistency; 0.30≤F N <0.60, fair consistency; 0.60≤ F N ≤1.00, poor consistency. The validation by comparing with previously developed measures shows that the proposed measure is acceptable in evaluating speed consistency.展开更多
A control strategy of variable speed limits(VSL)was developed to reduce the travel time at freeway recurrent bottleneck areas.The proposed control strategy particularly focused on preventing the capacity drop and incr...A control strategy of variable speed limits(VSL)was developed to reduce the travel time at freeway recurrent bottleneck areas.The proposed control strategy particularly focused on preventing the capacity drop and increasing the discharge flow.A cell transmission model(CTM)was developed to evaluate the effects of the proposed VSL control strategy on the traffic operations.The results show that the total travel time is reduced by 25.5% and the delay is reduced by 56.1%.The average travel speed is increased by 34.3% and the queue length is reduced by 31.0%.The traffic operation is improved by the proposed VSL control strategy.The way to use the proposed VSL control strategy in different types of freeway bottlenecks was also discussed by considering different traffic flow characteristics.It is concluded that the VSL control strategy is effective for merge bottlenecks but is less effective for diverge bottlenecks.展开更多
基金Project(BK20160685)supported by the Science Foundation of Jiangsu Province,ChinaProject(61620106002)supported by the National Natural Science Foundation of China
文摘This study develops new real-time freeway rear-end crash potential predictors using support vector machine(SVM) technique. The relationship between rear-end crash occurrences and traffic conditions were explored using historical loop detector data from Interstate-894 in Milwaukee, Wisconsin, USA. The extracted loop detection data were aggregated over different stations and time intervals to produce explanatory features. A feature selection process, which addresses the interaction between SVM classifiers and explanatory features, was adopted to identify the features that significantly influence rear-end crashes. Afterwards, the identified significant explanatory features over three separate time levels were used to train three SVM models. In the end, the multi-layer perceptron(MLP) artificial neural network models were used as benchmarks to evaluate the performance of SVM models. The results show that the proposed feature selection procedure greatly enhances the accuracy and generalization capability of SVM models. Moreover, the optimal SVM classifier achieves 81.1% overall prediction precision rate. In comparison with MLP artificial neural networks, SVM models provide better results in terms of crash prediction accuracy and false positive rate, which confirms the superior performance of SVM technique in rear-end crash potential prediction analysis.
基金Project(2012CB725400) supported by the National Key Basic Research Program of ChinaProject(2012AA112304) supported by the National High Technology Research and Development Program of ChinaProject(2009BAG13A07-5) supported by National Science and Technology Plan of Action of China for Traffic Safety
文摘Although either absolute speed or speed difference can be considered as a measure for speed consistency, few researches consider both in practice. The factor analysis method was introduced to extract an optimal number of factors from numerous original measures. The freeway diverging zone was divided into four elements, namely the upstream, the diverge area, the downstream and the exit ramp. Operating speeds together with individual vehicle speeds were collected at each element with radar guns. Following the factor analysis procedure, two factors, which explain 96.722% of the variance in the original data, were retained from the initial seven speed measures. According to the loadings after Varimax rotation, the two factors are clearly classified into two categories. The first category is named "speed scale" reflecting the absolute speed, and the other one is named "speed dispersion" interpreting speed discreteness. Then, the weighted score of speed consistency for each diverge area is given in terms of linear combination of the two retained factors. To facilitate the level classification of speed consistency, the weighted scores are normalized in the range of (0, 1.0). The criterion for speed consistency classification is given as 0≤F N <0.30, good consistency; 0.30≤F N <0.60, fair consistency; 0.60≤ F N ≤1.00, poor consistency. The validation by comparing with previously developed measures shows that the proposed measure is acceptable in evaluating speed consistency.
基金Project(2012CB725400)supported by the National Key Basic Research Program of ChinaProject(2011AA110303)supported by the National High Technology Research and Development Program of ChinaProject(YBPY1211)supported by the Scientific Research Foundation of the Graduate School of Southeast University,China
文摘A control strategy of variable speed limits(VSL)was developed to reduce the travel time at freeway recurrent bottleneck areas.The proposed control strategy particularly focused on preventing the capacity drop and increasing the discharge flow.A cell transmission model(CTM)was developed to evaluate the effects of the proposed VSL control strategy on the traffic operations.The results show that the total travel time is reduced by 25.5% and the delay is reduced by 56.1%.The average travel speed is increased by 34.3% and the queue length is reduced by 31.0%.The traffic operation is improved by the proposed VSL control strategy.The way to use the proposed VSL control strategy in different types of freeway bottlenecks was also discussed by considering different traffic flow characteristics.It is concluded that the VSL control strategy is effective for merge bottlenecks but is less effective for diverge bottlenecks.
