2016 No. 5

  Previous Issue | Next Issue 
Display Method:
2016, 34(5) doi: 10.3963/j.issn.1674-4861.2016.05.018
Abstract(150) PDF(0)
Abstract:
2016, 34(5) doi: 10.3963/j.issn.1674-4861.2016.05.019
Abstract(184) PDF(0)
Abstract:
A Real-hime Prediction Model for Rear-end Crash on Two-lane Freeway
2016, 34(5): 1-7,22. doi: 10.3963/j.issn1674-4861.2016.05.001
Abstract(332) PDF(1)
Abstract:
The existing models for real-time crash prediction are difficult to be applied in the freeway management system that high-resolution traffic data cannot be collected.In this study, a real-time prediction model for rear-end crash is proposed based on the traffic data collected using a single detector.Based on the traffic data collected by ultrasonic detectors on Qiyang Freeway in Yangzhou, Jiangsu Province, China, the methods of matched case-control and binary logistic regression are used to develop a real-time prediction model for rear-end crash for a two-lane freeway.Three spatio-temporal matrixes, including a flow matrix, a speed matrix and an average space headway matrix, are extracted from the traffic data 5-20 minutes before crashes.Eigenvalues of matrixes are introduced to simplify the modeling process and avoid a strong correlation among the parameters.Results show that overall accuracy of this proposed model is 85.7%, and accuracy of prediction for crash rate is 33.3%, with a corresponding false alarm rate less than 2%.Thus the performance and effectiveness of this proposed model is verified.
A Linear Forecasting Model and Algorithm for Running Time of Urban Rail Transit
CHEN Zhiguo, LI Wenming, LI Wenfeng
2016, 34(5): 8-15. doi: 10.3963/j.issn1674-4861.2016.05.002
Abstract(306) PDF(0)
Abstract:
A modified linear forecasting model is established in order to accurately forecast the running time of trains of urban rail transit.This proposed model computes the linear prediction coefficients based on orthogonal function and the rule of integrating minimum mean square error (MMSE).Sample data with different lengths is used to construct forecasting models.The effect of length of sample data and order of prediction on forecasting accuracy is analyzed by comparing computed results.The adjustment mechanism of sequential iteration-based data is then incorporated into this model to improve the accuracy of data in computing coefficients.The effects of the inter-station distances on forecasting accuracy are analyzed by comparing the results of this model before and after the transformation of distance under the unequal inter-station distances scenario.A linear transformation method of inter-station distances is incorporated into this proposed model to improve the precision of forecast.The results show that the average accuracy of forecast of this proposed model is 95.43% while which of the original model is 92.53%, increases by 3.13%;the accuracy of this model can be slightly improved by increasing the order of prediction, predictive accuracy of running time can be obviously improved by using the modified model in contrast with the original model.This proposed model is used to forecast the running time of trains of Shanghai Metro Line 2 as a case study, and the forecast error of this model is 17.4% less than the train′s motion model, which shows the applicability and high accuracy of this model.
A Study on an Index System of Performance Management for Safety Information in Civil Aviation
CAI Yueru, CUI Zhenxin, SUN He
2016, 34(5): 16-22. doi: 10.3963/j.issn1674-4861.2016.05.003
Abstract(378) PDF(0)
Abstract:
In order to modify the regulations of safety management and improve the quality of safety information in civil aviation, the balance scored card (BSC) method is modified to develop an index system of performance management for safety information.The logical relation of four levels in BSC is adjusted from traditional progressive to parallel progressive.To determine the weights of indices, the methods of analytic hierarchy process (AHP) and variable precision rough set (VPRS) are used to obtain subjective and objective weights, respectively.The subjective and objective weights are optimized by principle of minimum relative entropy (PMRE).This new index system is validated by an improved TOPSIS method.The results show that, the quality of safety information is the key factor to evaluate the performance.Its optimal weight is 0.407 and accounts for 40.66% of the total weight.One of the secondary indices, the specification degree of safety information weight, contributes the largest ρi.Its optimal weight is 0.288 and accounts for 11.97% of the total weight, indicating the largest reliability and validity for the performance management system.Hence, the proposed index system can be applied to actual processes of performance management for safety information in order to improving the level of management as well as the quality of safety information in civil aviation.
An Improved Model of Accident Prediction on Freeways Based on Fuzzy Logic
MENG Xianghai, HE Shali, ZHENG Lai
2016, 34(5): 23-30. doi: 10.3963/j.issn1674-4861.2016.05.004
Abstract(276) PDF(0)
Abstract:
AADT, length of segments, number of lanes, proportion of oversize vehicles, and road alignment of each road section are collected from a freeway network in Liaoning Province, which are set as the inputs of a basic fuzzy logic model for accident prediction, and the frequency of accidents per kilometer is set as the outputs.In consideration of that structures of Fuzzy Sets and rules of Fuzzy Control may have possible negative impacts on prediction results, a method to improve the fuzzy logic model is established by resizing structures of Fuzzy Sets and setting up rules of Fuzzy Control with prior knowledge, respectively.The portability of the basic fuzzy logic model and the improved model is analyzed.This improved model and a negative binomial prediction model are both used in a case study of Yuegan and Kaiyang freeway as a comparison.The results show that, subdividing Fuzzy Sets can increase the accuracy of prediction.Compared with the basic model, the average relative error in total (Zt) of the subdivided model decreases by 8.3%, and model goodness (MC) increases by 0.357.With prior knowledge, the accuracy of prediction can be increased.Compared with the basic model, the Zt of the basic prior model decreases by 1.9%, and MC increases by 0.164, respectively.Prior knowledge also increases the portability of this model, and improves the average accuracy of prediction, Ra decreases by 3.8%, overall error decreases by 3.4%, Zt decreases by 4.1%, and MC increases by 0.385, respectively.However, compare with the model of rough sets and basic model, the portability of this model decreases when apply subdividing Fuzzy Sets.The basic fuzzy logic model and negative binomial prediction model for crash prediction are almost at the same level in accuracy of prediction and portability
Identification of Overspeed Vehicles in Broadside Collision with Electric Bicycles
BAO Xu, CHEN Jinwen
2016, 34(5): 31-37,67. doi: 10.3963/j.issn1674-4861.2016.05.005
Abstract(258) PDF(0)
Abstract:
In order to identify whether a vehicle is overspeed at braking moment in broadside collision with an electronic bicycle, a case study at one urban intersection is used to develop a method to identify the speed of a vehicle in a broadside side collision with an electric bicycle.According to the moment of a collision, it is divided into two stages: pre-collision and post-collision.Use the method for reverse reduction of accidents, the thrown distance and empirical formula of impact velocity are combined to propose a model to compute vehicles′ speed in collisions.The model of vehicles′ speed at braking moment is proposed by using the Work Energy Theorem based on the length of the brake trace.The simulations in PC-Crash software indicate that when the speed of vehicles at braking moment is 34-38 km/h, the error of identification is 1.47%;when the speed is 42-70 km/h, the error is less than 1.3%.The accuracy of identification is improved by 15.94% in the maximum when use the proposed method.Compare with GIDAS, this method can accurately analyze broadside collisions between vehicles and electronic bicycles, and identify whether the vehicle is overspeed or not.However, it is still necessary to improve the accuracy of identification when the speed of vehicle is 34-38 km/h.
A Study on Differences of Physiological Characteristics of Drivers Driving Through Urban Tunnels
WU Sunan, CHEN Xin, GUO Tangyi
2016, 34(5): 38-45. doi: 10.3963/j.issn1674-4861.2016.05.006
Abstract(342) PDF(1)
Abstract:
Variations of physiological characteristics of drivers have been frequently accounted for safety issues at the entrances and exits of tunnels.In order to better understand it and improve traffic safety, a field experiment is conducted to collect the illumination, real-time data of speed, and physiological data of drivers including heart rate and pupil diameter when they driving through urban tunnels.The variations and differences on physiological characteristics at the entrances and exits of urban tunnels are comparatively analyzed using mathematical statistics.Partial correlation analyses for heart rate, illumination and driving speed are conducted.A multiple linear regression model is established to study the influences of external factors on heart rate at the entrances of tunnels.The results show that heart rate has significant differences before/after entering an urban tunnel;while has no significant difference before/after leaving an urban tunnel.Pupil diameter has significant differences when entering and leaving an urban tunnel, and the value of its significance test is close to 0.The partial correlation coefficients between heart rate and illumination, heart rate and speed are -0.65 and -0.74, respectively.The degree of a fitted regression model is 0.66 when only take speed and illumination into consideration.
The Effects of Chevron Alignment Signs on Characteristics of Drivers′ Eye Movement: An Experimental Study on Prairie Highways
GE Jingfang, QI Chunhua, ZHU Shoulin, YANG Feng, Zhang Guiman
2016, 34(5): 46-52. doi: 10.3963/j.issn1674-4861.2016.05.007
Abstract(326) PDF(0)
Abstract:
Chevron alignment signs on prairie highways play a significant role in directing changes of road alignment, inducing sight of drivers, relieving visual monotony, and improving driving safety.Different curve sections on a prairie highway are selected, and indicators of eye movement are collected from seven drivers using eye tracking system.The data of five drivers are recognized as valid.A One-way ANOVA analysis is applied to filter sensitive indices considering the influence of different factors.The results indicate that pupil diameter is sensitive to the existence of chevron alignment signs and gender of drivers;while saccade amplitude is sensitive to the positions of chevron alignment signs.When no chevron alignment sign exists, the average pupil diameter is 0.09 mm, higher than that with chevron alignment signs.The pupil diameter changes in a wide range with variance of the average rate of change about 0.5%.When chevron alignment signs are positioned on the left side, the saccade amplitude is 2.59° higher than on the right side.And variance of the average rate of change is about 1.3%.Female drivers are found to be more concentrated, with a larger (0.25 mm) pupil diameter, and variance of the average rate of change is 0.7%.
A Detection Algorithm of Parking Occupancy via Geomagnetic Field Based on Vector Operation and Multilevel Threshold
WANG Weifeng, WAN Jian, XIE Bin, ZHOU Yuncheng, TAN Ting
2016, 34(5): 53-60. doi: 10.3963/j.issn1674-4861.2016.05.008
Abstract(308) PDF(0)
Abstract:
An algorithm is proposed to deal with the issues that the existing detection algorithms cannot exclude the interference of ortho parking spaces and the corresponding judgment criteria fail to accommodate multiple types of vehicles and parking scenarios.Based on the analysis of the magnetic dipole model, the geomagnetic disturbance vector of targeted parking space is extracted through a vector operation during parking process, which can effectively eliminate the negative impact on geomagnetic field caused by the parking process from the ortho parking spaces.With a collection of large number of real geomagnetic disturbance data, the intensity of geomagnetic disturbances and characteristics of direction changes are statistically deduced and the multilevel thresholds are thus determined as the detection criterion, which ensures that the detection algorithm is suitable to various vehicle types and parking application scenarios.With a three-month field test, it is found that the detection precision of the proposed algorithm achieves 99.3%, superior to the algorithms with a single-feature using either pure intensity or certain direction.Moreover, the method could detect whether the parking is standard.The algorithm has not been applied to large passenger cars, trucks or special vehicles.
A Risk Analysis of Fatal Factors in Traffic Accidents Based on Inherent Matching Pairs
ZHANG Hao, MA Lu, YU Hongxi, CHEN Dalin, YAN Xuedong
2016, 34(5): 61-67. doi: 10.3963/j.issn1674-4861.2016.05.009
Abstract(285) PDF(1)
Abstract:
The risk ratio of occupants′ death in traffic accidents is affected by many factors.In order to understand the nonlinear effects, the data of fatal crashes that obtained from the Fatality Analysis Reporting System (FARS) are adopted.After matching and filtering the data, according to death of occupants, a dataset of inherent matching pairs is obtained.The relative risk of death is set as an indicator in this paper.A model based on nonparametric logistic regression is developed to estimate the influential mechanisms of gender, age, belt usage, and seating position on fatalities.The results are expected to provide an important basis for the development of policy and the implementation of measures for traffic safety.Besides, the model is able to reveal significant influence of nonlinear effects when age is regarded as a continuous variable.The results show that the four factors of gender, age, belt usage, and seating position can significantly affect the risk of occupants′ death in traffic accidents, and the structures of inherently matched pairs are able to exclude the interference of external factors in this model.Compared with male occupants, the risk of death of female is 15.9% higher (the logarithmic value);the usage of seat belt can reduce the risk of death by 74.8% (the logarithmic value);the risk of death increases with age;and the middle and left of rear seats are the safest seating positions in a car.
An Optimized Method for Reducing Traffic Emissions in Arterial Road Using Coordinated Signal Control
YAO Ronghan, WANG Xiaoyu, XU Hongfeng, ZHAO Shengchuan
2016, 34(5): 68-74,101. doi: 10.3963/j.issn1674-4861.2016.05.010
Abstract(274) PDF(3)
Abstract:
To reduce traffic emissions in an arterial road system, an improved calibration method on the basis of specific power of vehicle is proposed to compute the emission factors during the period of traffic light.An optimization model for allocation of spatial-temporal resources is developed for one arterial intersection group.The decision variable is the effective green time in each phase, and the objective function is to minimize the total emissions for all intersections.By analyzing the relationship between the delay for one platoon and the phase difference between adjacent intersections, this optimized model aiming at minimizing the delay for one platoon, is amended.A case study is carried out to verify this model.A control scheme for traffic signals is firstly obtained as a reference by using traditional methods.The emission factors during the period of traffic light are calibrated by Vissim software.A control scheme for traffic signals is optimized by this proposed method.The results indicate that the emission factors of each pollutant during a green phase are explicitly greater than which during a red phase.Compared with referenced control schemes for traffic signals, both delay of vehicles and traffic emissions decrease by 8-11% at each intersection when the optimized control scheme is adopted.It shows that this proposed model can be applied to decrease both the delay of vehicles and traffic emissions in an arterial road system.Thus, it can generate a coordinated control scheme for traffic signals, and relieve traffic jam.
A Study on Evaluation of Emergency Response Capability of Air Traffic Controllers
WANG Yanqing, LI Miao
2016, 34(5): 75-81,101. doi: 10.3963/j.issn1674-4861.2016.05.011
Abstract(272) PDF(0)
Abstract:
In order to improve capability of emergency response of air traffic controllers, and to ensure safety operation of aircrafts in various emergencies, a "controller′s onion model" is established based on an "onion model", introducing two indices: personality and awareness of controllers, and in combination with their characteristics at work.In accordance with the process of emergency response, a set of indices for evaluating controller′s emergency response capability is proposed, which includes 4 first-class indices (knowledge, skill, awareness, and personality of controllers) and 16 second-class indices.The weight of each index is computed by adopting the method of analytic hierarchy process (AHP).By using a fuzzy comprehensive evaluation method, the emergency response capability of controllers at one air traffic control unit is evaluated as a case study.The score of this unit is 3.689, which in medium level.The quantitative method proposed in this study can be applied for an accurate evaluation of emergency response capability of air traffic controllers.
A Method for Extraction of Keywords from Safety Information in Civil Aviation
CUI Zhenxin, LU Haowen
2016, 34(5): 82-86,101. doi: 10.3963/j.issn1674-4861.2016.05.012
Abstract(236) PDF(2)
Abstract:
Keywords in civil aviation can reflect synopsis of safety information.It is significant for security officers to extract and call information.An academic review of technologies to extract keywords is conducted in this paper.The features of keywords in civil aviation are analyzed.And a naive Bayes model for extraction of keywords is proposed.The selected features of this model are length of keywords;part of speech;frequency of words (including span of paragraph and position of words);and Term Frequency-Inverse Document Frequency (TF-IDF) value;which reflect the basic attributes of each candidate word.This model is trained by the safety information which has been manually labeled;in order to obtain the probability of each feature for extracting keywords.The probability of features is used to compute the score of all alternative words.The words with top three scores are regarded as keywords.Compared with the traditional TF-IDF algorithm and KEA algorithm;this method improves the accuracy by 18% and 11.9%;respectively.The recognition rate of words is also improved by 15.3% and 12.3%;respectively.The results show that;compared with other general methods;the accuracy and capability to recognize special words in civil aviation can be significantly improved by the method proposed in this study.
An Application of Heuristic Selection Sampling Method Based on Genetic Algorithm in Detection of Traffic Incidents
LI Miaohua, CHEN Shuyan, LAO Yechun, GU Jian
2016, 34(5): 87-92. doi: 10.3963/j.issn1674-4861.2016.05.013
Abstract(213) PDF(0)
Abstract:
In order to improve the comprehensive performance in detection of traffic incidents for an imbalanced dataset.Two automatic incident detection (AID) algorithms based on GA-based heuristic sampling method are proposed.The method of GA-based Instance Selection (GA-IS) is proposed to settle the issue of instability caused by manual setting of sampling rate in non-heuristic sampling method.The method of GA-based Support vectors Selection (GA-SS) is proposed to improve efficiency of detection under a condition of large learning datasets.In a case study, a simulation database of Ayer Rajah Expressway (AYE) in Singapore is used, and support vector machine (SVM) is adopted as a classifier to detect incidents.The results show that the detection rate in GA-IS SVM AID model is 94%, the average time to detect incidents is 1.413 3 min, and the performance index (PI) is 0.157.Meanwhile, the decision time in GA-SS SVM AID model is 4.55 s, and the PI is 0.151.The decision time in SMOTE SVM AID model is 35.21 s, and the PI is 0.329.Compared with SMOTE, the proposed methods can provide better comprehensive performance in detection of traffic incident for imbalanced datasets.
A Study on a Quantitative Evaluation System for Navigation APPs
ZHANG Zheng, MA Siyong, WENG Jiancheng, LIU Wentao, CAO Hehong
2016, 34(5): 93-101. doi: 10.3963/j.issn1674-4861.2016.05.014
Abstract(232) PDF(5)
Abstract:
Taking the most popular evaluation systems and standards for APPs all around the world as references, an evaluation index system for road navigation APPs is proposed.Quantitative measurements of functions, degrees of user experience, and self-evaluation by developers are set as indices.APPs are evaluated from both subjective and objective aspects.Four quantitative measurements along with correspondence algorithms are constructed, which are the accuracy of estimated travel time, coverage rate of information, accuracy of localization, and grading accuracy of traffic information.Six indicators are considered to assess the degrees of user experience for navigation APPs, including operational convenience, accuracy, fluency, and timeliness of information, stability, and satisfaction.Self-evaluation by developers mainly focuses on the quality of data source, transmission mode of information, and software features.The evaluation results for APPs are finally determined based on the most unfavorable rating in the multi-level evaluation.A case study is carried out by assessing three most popular APPs for road navigation in China.The results of four indicators for APP (A) are 76.42%, 76.24%, 100% and 76.53%, respectively.The score of user experience is 4.54, which indicates that APP (A) rating in the first level.The results are consistent with the market share of it as well as the results from market investigation, which verifies the usability and dependability of this proposed evaluation system.
Risk Assessment of Traffic Facility on Freeway Based on a Modified Bayesian Network Model
HU Xiaowei, ZHANG Daoyu
2016, 34(5): 102-107. doi: 10.3963/j.issn1674-4861.2016.05.015
Abstract(283) PDF(4)
Abstract:
Traffic facilities play an important role in traffic safety.Assessing the risk of facilities on freeways is one of the effective methods to prevent traffic accidents.In order to evaluate the risk, this study analyzes the effects of facilities on freeway safety.The nodes and network are extracted by factor analysis.According to the relationship of nodes in a network, combined with probability theory (such as the conditional probability), this paper modifies the Bayesian formula after rounding off unrelated variables and no-effect constants, and establishes a modified Bayesian model for assessing facilities on freeways.With the traffic data of freeways and the result of a survey, the risk of facilities can be assessed.The results computed by the modified model show that the smaller the value, the higher the risk of freeway facilities.Using the data of Huiwu freeway to verify this proposed model, it shows that when the value is smaller than 0.5, the number of crashes is relatively larger, correspondingly, the risk of facilities is higher;otherwise, the risk of facilities is lower.
A Model of Decision Process of Travel Modes Based on DAG-SVM
CAO Xiongjiu, JIA Hongfei, WU Sufeng, ZHANG Yang, KANG Hao
2016, 34(5): 108-114. doi: 10.3963/j.issn1674-4861.2016.05.016
Abstract(235) PDF(3)
Abstract:
Improving the accuracy of prediction of travel modes of residents is of a great importance to evaluate the effect of traffic planning and transport strategy.Based on psychology and behavior science, the decision process of travel modes is analyzed.With a structuralized process of decision, a library of travel scenarios is established.A principal component analysis is used to analyze the main factors which have impacts on the decision process of travel modes.The factors are regarded as the inputs of support vector machine (SVM).The differences between SVM and neural network in principles of modeling are analyzed by a statistical learning theory.Then a directed acyclic graph support vector machine (DAG-SVM) model is developed.The results of prediction from different kernel functions are evaluated, and the parameters are optimized by the grid method and genetic algorithm.The results show that among several kernel functions, the radial basis function is the best for prediction.The genetic algorithm is better than the grid method in parameter optimization.The overall accuracy of prediction from the DAG-SVM model is 82.3%, which is nearly 9% higher than that from the neural network model.However the accuracy of prediction for travel by taxi is slightly lower than other ones.This is mainly due to the fact that travel by taxi is an alternative way for residents in particular circumstances, not as regular as other travel modes.
2016, 34(5): 115-117. doi: 10.3963/j.issn.1674-4861.2016.05.017
Abstract(197) PDF(0)
Abstract:

Copyright Editorial Office of Transport Information and Safety鄂ICP备05003336号-2

Address：No. 1178，Heping Avenue, Wuchang, Wuhan, CHINA (430063)

Tel：027-86580355 Email：jtjsj@vip.163.com

Supported by: Beijing Renhe Information Technology Co. Ltd