Journal of Transportation Systems Engineering and Information Technology

Combined Flexible Working Arrangements and Time-dependent Parking Charges

LI Hao, NING Xu-qi, YU Lu, CHEN Hao

2020, 20(2): 1-7.

Abstract ( )

PDF (444KB) ( )

A combined strategy, flexible working arrangement (FWA) and time-dependent parking charges (TPC), is proposed to alleviate traffic congestion. The departure patterns and optimal parking charges are investigated with the combined strategy.A bi-level programming model is established to optimize the parking charges dynamically constrained to the dynamic user equilibrium. A gradient descent algorithm based on a sensitivity analysis is developed to solve the bi-level problem. The multi-user departure patterns and the optimal dynamic parking charges under different flexible user scales are demonstrated by a simulation approach. The results show that there are obvious staggered departures between flexible and non- flexible users. There is significant interacting effects between the two strategies, which yields better effects on congestion alleviation than implementing a single one separately. This study is expected to provide a new solution method for the traffic management optimization.

References | Related Articles | Metrics

An Evaluation Method of CO2 Emissions Reduction in Urban Traffic Congestion Mitigation

LI Zhen-yu, LIAO Kai, CUI Zhan-wei, LIU Yang

2020, 20(2): 8-12.

Abstract ( )

PDF (516KB) ( )

Urban traffic congestion has been one of the important reasons for the rapid growth of carbon emissions of urban transport. Firstly, this paper analyzes the key factors affecting the carbon emissions of urban transport. Secondly, by collecting the existing carbon emission factors and conducting vehicle emission tests, a database of urban transport carbon emission factors for different road types and service levels has been established. Based on the classification method of urban traffic congestion, the ASIF method is improved. And a meso-scale assessment method and model are proposed for the benefit evaluation on carbon emission reduction from traffic congestion mitigation. Lastly, taking Chengdu as an example, the city survey, data collection and analysis, and the calculation are carried out using the model. The total carbon emissions of urban transport in Chengdu are evaluated throughout the year, of which 48.2% were carbon emissions from traffic congestion. By setting four positive development scenarios and three negative development scenarios, and the carbon emission effects under different scenarios are analyzed.

References | Related Articles | Metrics

Modelling Pre-peak Discount Price Plan for Urban Rail Transit

ZOU Qing-ru, YAO Xiang-ming, ZHAO Peng

2020, 20(2): 13-19.

Abstract ( )

PDF (412KB) ( )

An optimization model for establishing a pre-peak discount price plan to relieve peak congestion in urban rail transit is proposed. The optimization objective is to maximize the balance between transport capacities and passenger flows, based on the analysis of peak demand changes due to pre-peak discount price, and acceptable capacities of the operating company for revenue loss. Then, the detailed pricing plan (discount stations, discount rates, and discount deadlines) can be determined scientifically. The model is verified on the BT line of Beijing subway, and the results show that the discount deadline of the optimization scheme is much later than the practical one. The key reason for limiting the effect of the current scheme is unreasonable discount deadlines; The effect of peak congestion relief can be improved by optimizing pricing schemes under the premise of non-increase revenue loss.

References | Related Articles | Metrics

Evaluation of Emergency Guarantee Capability of Key Nodes in Maritime Transport Based on Improved Copula Function

LI Bao-de, LU Jing, LI Jing

2020, 20(2): 20-25.

Abstract ( )

PDF (381KB) ( )

Emergency guarantee of key nodes in maritime transport is an important aspect to minimize the losses caused by emergencies. This study developed an evaluation model of emergency guarantee capability of key nodes in maritime transport based on factor analysis and Copula function. The modeling considered the following major elements like correlation between the influential factors, factor distributions, and the difficulty of solving Copula function because of multi- dimensional variables. The proposed model was used to evaluate the emergency guarantee capability of key nodes of the existing maritime transport, and the impact of different factors. The results show that the Straits of Gibraltar, Malacca, and Panama Canal have high emergency guarantee capability, and the Suez Canal has relatively low emergency guarantee capability from an overall comparison. However, Malacca Strait has a relatively high FA1 (Factor Analysis) value and Suez Canal has a relatively low FA1 score. Comparing the joint distribution of different variables, the distributions of different variables have varied impacts on the results.

References | Related Articles | Metrics

Fluctuation Patterns of China Export Containerized Freight Index Based on Complex Network Theory

TANG Xia, KUANG Hai-bo, GUO Yuan-yuan, LAN Xian-gang

2020, 20(2): 26-32.

Abstract ( )

PDF (418KB) ( )

This study developed a complex network model for the China Export Containerized Freight Index (CCFI) fluctuation analysis, based on the symbolic dynamic method. The CCFI fluctuation patterns were analyzed by the network dynamic topological properties, such as mode strength, strength distribution, weighted clustering coefficient, average shortest path length, and mode betweenness. The results indicate that the container freight fluctuation network has the small- world and scale- free characteristics. The CCFI fluctuation is characterized by clustering, periodicity, continuity, and graduality. The conversion cycle between linkage modes within the clustering subgroups is less than 2 months, and the average conversion cycle of linkage modes between clustering subgroups lasts 3.8 months. The CCFI clustering fluctuation carries out conversion by the modes with low strength and high betweenness. The CCFI fluctuation pattern study based on complex network theory provides a new perspective for governments and shipping enterprises to understand fluctuation characteristics and reduce risks.

References | Related Articles | Metrics

Calculation and Prediction Analysis of Air Pollutant of Civil Aviation Aircraft in China

FEIWen-peng, XIONG Ling-li, OUYANG Bin, BIAN Xue-hang, SONG Guo-hua

2020, 20(2): 33-40.

Abstract ( )

PDF (529KB) ( )

Although civil aviation is an important transportation mode in China, there are relatively few studies on calculation and prediction of air pollutant emissions from the civil aviation aircrafts. Based on the international fuel flow and LTO (Landing and Take-off) cycle method, this paper calculates and predicts the air pollutant emissions such as HC, CO, NOx, SO2 and fine particulate matter for China civil aviation. The empirical analysis results show that the fuel consumption and air pollutant emissions of civil aviation aircraft in China are increasing gradually, and the fuel consumption is expected to reach a peak of 118.26×106 ton in 2040. The emissions of HC, CO, NOx, SO2, PM2.5 and PM10 are respectively 11 167 ton, 86 785 ton, 1 260 131 ton, 63 264 ton, 11 149 ton and 11 359 ton in 2018. Moreover, the corresponding emission intensities of HC, CO, NOx, SO2, PM2.5 and PM10 are respectively 0.09, 0.72, 10.44, 0.52, 0.09 and 0.09 gram/(ton · kilometer). The HC emissions reached a peak of 15 433 ton in 2010. The CO emissions are expected to reach a peak of 141 038 ton in 2025. The NOx and SO2 emissions are expected to reach the peak of 2.353 5×106 ton and 0.118 3×106 ton in 2040, respectively.

References | Related Articles | Metrics

Scheduling Design for Container Ship Along Yangtze River Considering Transshipment

WANG Qing-bin, XIAO Qin-fei, LI Xiu-ying, JI Ming-jun

2020, 20(2): 41-47.

Abstract ( )

PDF (392KB) ( )

This paper focused on the ship-scheduling problem in which ship allocation, port selection, and freight assignment were considered simultaneously. A nonlinear programming model aiming at the minimum total transportation cost was established in consideration of transshipment. Given the properties of the model, an improved genetic algorithm (IGA) was designed to solve the problem, and the optimal solution for the real case was obtained. The optimal solutions in different cases were compared with the solutions without transshipment. Experimental results show that the ship scheduling with transshipment could reduce total transportation costs, improve the comprehensive utilization of ships, strengthen the competitiveness and provide a reference for shipping companies in ship scheduling.

References | Related Articles | Metrics

Fast Traffic Sign Detection Based on Weighted Densely Connected Convolutional Network

SHAO Yi-ming, QU Zhi-hua, DENG Tian-min, SONG Xiao-hua

2020, 20(2): 48-54.

Abstract ( )

PDF (488KB) ( )

The current traffic sign detection algorithms appear to have poor detection effect and low practicability under complex traffic environments. This paper proposed a fast traffic sign detection algorithm based on dynamic weighted densely connected convolutional network. The purpose is to enhance the detection accuracy of the algorithm and identify traffic signs with improved speed. YOLOv2 was selected as the basic network, and the weight of each layer's feature map was adjusted by adding dynamic weighted densely blocks. The deep highsemantic information and shallow low-semantic information was integrated subsequently. The lightweight network structure in MobileNet and the separable convolution operation reduced the network computing costs effectively. To resolve the image feature loss problem in pooling operation, the study used the Convolutional Block Attention Module (CBAM) to enhance the performance of key features by using channel attention and spatial attention information. The experimental results confirmed that the proposed algorithm achieved a detection accuracy of 96.14%, and the detection speed was 139 frame/s on the German Traffic Sign Detection Benchmark (GTSDB) dataset. The algorithm effectively improved the detection efficiency and practicability, and maintained good detection accuracy.

References | Related Articles | Metrics

Superhighway Virtual Track System Based on Intelligent Road Button

HE Yong-ming, PEI Yu-long, RAN Bin

2020, 20(2): 55-60.

Abstract ( )

PDF (406KB) ( )

This paper focuses on the virtual track system of superhighway and uses intelligent knobs to improve the safety through structural analysis and mathematical modeling. The virtual track system consists of the subsystems of roadway, vehicle, and service center. When the vehicle that equipped with onboard system get close to the defined road button, the virtual track system will be activated. The coordinates and road alignment of the relevant road buttons will be extracted by the reader. The data processing module reads the linear parameters and obtains the tangent angle of roadway and the angle of the vehicle body. Using the front wheel angle, vehicle speed and the distance between the adjacent two button labels, the system can calculate vehicle steering wheel rotation velocity between the adjacent two road buttons and send the control parameters to the steering motor of the vehicle. The study results show that when the design speeds of superhighway are 140 km/h, 160 km/h, and 180 km/h, the distances between the road buttons are less than 1.33 m, 1.50 m, and 1.50 m, the distance of the vehicle off from the centerline would be less than 0.5 m. The virtual track system based on intelligent road knobs could keep vehicles traveling on the virtual track and improve the safety of superhighways.

References | Related Articles | Metrics

Railway Perimeter Intrusion Detection Algorithms Based on Video Deep Learning

WANG Rui, LI Xiao-feng, SHI Tian-yun, ZOU Qi

2020, 20(2): 61-68.

Abstract ( )

PDF (466KB) ( )

Compared with radar and vibration optical fiber, the railway perimeter intrusion detection algorithm based on video intelligent analysis has the advantages of low cost and low false alarm rate. Aiming at the simultaneous detection of large and small targets in videos, this paper proposes an improved Cascade Mask RCNN (CMR) model. The model uses the cascaded structure to locate the target accurately. At the same time, a multi- scale feature extraction model is added to the original model to enhance the expressive ability for small targets. Multi-scale module is implemented by feature pyramid networks (FPN). The spatial context enhancement module is implemented through an atrous spatial pyramid pooling (ASPP) subnetwork. The effectiveness of the proposed model was verified by the actual railway scene videos. The results show that the new model improves the F-measure of small targets detection by up to 0.24 compared with the original model. The proposed model enhances the detection capability of railway perimeter intrusion in different scenarios, and improves the accuracy of video detection for small targets.

References | Related Articles | Metrics

Automatic Train Operation Algorithm Based on Adaptive Iterative Learning Control Theory

HE Zhi-yu, XU Ning

2020, 20(2): 69-75.

Abstract ( )

PDF (392KB) ( )

To study the automatic control of high-speed trains with time-varying exterior disturbances and state saturation, this paper proposes an adaptive iterative learning control algorithm. Based on Lyapunov function, the control law and the parameter of updating law are deduced by considering the state error during the operating process. Then the Lyapunov-like composite energy function is established. The differential negative definiteness and robustness of the proposed function are verified. The proposed adaptive iterative learning control algorithm has been applied to computational simulation and real case study to verify the tracking performance. The results show that the proposed algorithm improves tracking accuracy and convergence speed. It was able to accurately track the desired profile with less iterative times than before.

References | Related Articles | Metrics

Distributed Signal Control of Multi-agent Reinforcement Learning Based on Game

QU Zhao-wei, PAN Zhao-tian, CHEN Yong-heng, LI Hai-tao,WANG Xin

2020, 20(2): 76-82.

Abstract ( )

PDF (473KB) ( )

The difficulty of distributed signal control is increasing due to the unbalance and fluctuation of traffic demand. Since the decision-making of existing independent action multi-agent reinforcement learning (IA-MARL) is based on its own historical experience, the distributed signal control based on IA-MARL is difficult to timely alleviate the impact of unbalanced and fluctuating traffic demand. In this paper, the framework of multi- agent reinforcement learning based on the game (G-MARL) was proposed by improving the decision- making of IAMARL with integrating the mixed strategy Nash- equilibrium, which is a concept in game theory. In the grid network with the Poisson arrival rate, the distributed control methods based on IA-MARL and G-MARL were simulated to obtain the unit travel time and the unit vehicle delay curves. The results show that, the unit travel time and the unit vehicle average delay obtained by G-MARL are reduced by 59.94% and 81.45% compared with IAMARL respectively. It is proved that G-MARL is suitable for distributed signal control when there are unbalances and fluctuations in traffic demand with the unsaturated state.

References | Related Articles | Metrics

Adaptive Foreground Object Detection in Railway Scene

LI Xing-xin, ZHU Li-qiang, YU Zu-jun

2020, 20(2): 83-90.

Abstract ( )

PDF (528KB) ( )

In modern railway systems, intelligent video analysis has been widely applied in foreign body intrusion monitoring, and foreground object detection is an essential step for intrusion detection. Background subtraction is commonly used to detect foreground objects. However, existing threshold- segmentation- based methods and deep-learning-based methods cannot meet the requirements in a complex railway scene, which contains dynamic background and unknown objects. In this paper, we proposed an adaptive- threshold- based foreground detection algorithm, which utilizes the temporal dynamic of pixel intensity, feedback information of detection result and spatial information of super- pixel to determine a factor, and then automatically adjusts the threshold by the factor to follow scene change. In addition, we also proposed a flexible and reliable background model initialization method that eliminates the ghost problem and flexibly switches from one-frame initialization to multiple- frame initialization. Experimental results show that the proposed algorithm achieves better accuracy and wrong classification rate in railway scenes, and also gets a better trade-off between accuracy and speed.

References | Related Articles | Metrics

Real-time Detection of Rail Dynamic Foreign Object Intrusion Based on Improved MOG-LRMF

HOU Tao, WU Hai-ping, NIU Hong-xia

2020, 20(2): 91-100.

Abstract ( )

PDF (562KB) ( )

To address the issues of low detection accuracy and poor anti- interference ability for the dynamic intrusion of foreign objects in complex rail environments, a real-time detection method for foreign object intrusion in railway track based on improved MOG- LRMF algorithm is proposed in this paper. Firstly, the affine transformation is used to pre-correct video sequences. Then, the background of the frame in a video sequence is predicted with the background knowledge learned in the previous frame to improve the MOG-LRMF model by analyzing the characteristics of the MOG-LRMF model. Finally, the EM algorithm is used to solve the parameters of the MOG-LRMF model, and it can realize the online real-time update of the background. The experiment results show that the improved MOG-LRMF algorithm can greatly enhance the target detection accuracy under sufficient illumination, weak light, camera jitter, complex environment, and multiple targets. Moreover, the improved MOGLRMF algorithm has better anti-interference, robustness, and rapidity.

References | Related Articles | Metrics

Hierarchical and Distributed Optimal Control Research on Urban Freeway Network

JIANG Zhu, LIN Hao, LI Shu-bin, LEI Zhen-yu

2020, 20(2): 101-106.

Abstract ( )

PDF (445KB) ( )

In order to solve the problem that the current urban freeway control system cannot make a timely and effective response to the real-time road traffic situation. Based on information of time-varying traffic state estimation and OD matrix prediction, according to the intelligent transportation and the distributed and hierarchy idea, a novel optimal control strategy on urban freeway is proposed. Firstly, the total traffic demands in a long time was predicted in advance by polynomial trend filtering, and the upper bound of the future queue length was made; Secondly, the future traffic state was predicted and the harmonious restrictions for each ramp were built based on global optimum to make a reference for ramp rate. The simulation results show that the hierarchical and distributed optimal control system can effectively alleviate the congestion by coordinating the interests of each ramp. It has certain practical significance to optimize the overall performance of urban expressway network.

References | Related Articles | Metrics

Optimization Model and Algorithm of Collaborative Layout for Hierarchical Nodes Considering Nodes and Links Capacity Constrains

LIU Jie, LI Hao-dong, ZHANG Meng

2020, 20(2): 107-113.

Abstract ( )

PDF (474KB) ( )

Hierarchical layout of logistics nodes has significant impact on the cost and benefit of the logistics system. This paper proposed the logistics network simplification method. The logistics nodes network was built from actual physical networks, and the factors of shortest path, shared segment and links capacity were considered in the process. Then, the collaborative layout model for hierarchical nodes was developed considering nodes and links capacity constraints. The service radius, capacity and cost of hierarchical nodes and the transport corridor were factored in the modeling. The improved harmony search algorithm was used to resolve this model. The realworld example was used to verify the effectiveness of the proposed model and algorithm. The result shows the strong adaptability of the model and algorithm, which can be used to solve the actual hierarchical nodes layout problem in practices.

References | Related Articles | Metrics

Evaluation of Road Transportation System Resilience and Link Importance

LV Biao, GAO Zi-qiang, LIU Yi-liu

2020, 20(2): 114-121.

Abstract ( )

PDF (408KB) ( )

Resilience comprehensively describes the ability of a roadway system to absorb and recover from a disruptive event. The existing resilience metrics can be improved to measure the system performance more accurately and the impact of traffic flow should be considered in the evaluation. This study first examined the roadway capacity degradation and recovery processes, then used the network efficiency as a measure of system performance to develop a resilience metric for the entire disruptive event duration period. The study proposed the resilience achievement worth (RAW) and resilience reduction worth (RRW) indices to reflect the link importance respectively from the optimistic and pessimistic perspectives, and the heuristic algorithm was used to identify the importance degrees of roadway links. The results from the examples show that the proposed resilience metric is able to measure the average cumulative performance of the roadway system during the entire degradation and recovery processes after a disruptive event, which is more consistent with the function of resilience metric. The indices of RAW and RRW are also able to effectively identify the link importance degree. The importance degrees of most of the roadway links change dynamically with time.

References | Related Articles | Metrics

Impact of Metro's In-vehicle Crowding and Commuters' Heterogeneity on Value of Time

LIU Jian-rong, HUANG Ling

2020, 20(2): 122-126.

Abstract ( )

PDF (291KB) ( )

There is significant heterogeneity in travelers' choice preferences. Previous studies divided travelers into different subgroups based on travelers' demographic characteristics and then analyzed the travel behavior for the specific subgroups. However, this classification method doesn't ground on solid theoretical basis. This paper analyzed the impacts of commuters' heterogeneity and in-vehicle crowding on the value of time of the metro, using the latent- class conditional Logit model. The latent- class conditional Logit model normally analyzes travelers' choice behavior first and then factor in travelers' demographic characteristics on the classification. Through analysis, travelers were divided into two subgroups: class 1 and class 2. Travelers in class 1 didn't care much about the degree of crowding inside the metro. Travelers in class 2 attached great importance to the in-vehicle crowding of the metro. The value of time for the sitter and the standee of the class 1 group were found to be 14.7 RMB/h and 16.9 RMB/h, respectively. As to the standee in class 2 group, the value of time increased 8.6 RMB/h when the standee density increased by 1 person/m2. Travelers' requirement of comfortability had a significant impact on the classification. The traveler with higher requirement on comfortability is more likely belong to class 2 group.

References | Related Articles | Metrics

Bi-layer BPNN Prediction Model for Bus Arrival Time Considering Preceding Segment State

MIAO Xu,WANG Zhong-yu,WU Bing,YANG Hang,WANG Yan-li

2020, 20(2): 127-133.

Abstract ( )

PDF (482KB) ( )

An integrated prediction model is proposed based on the bilevel BPNN and the state of the presequence road section in order to improve the prediction accuracy of bus arrival time. Based on the static variables and the upper-level BPNN model, the initial travel time of the vehicle to each station is predicted. The K-means clustering method and the Markov chain model with the state of the pre- sequence road section are adopted to predict the travel time at the targeted road section. Taking the predicted values of the above two models and the travel time of the previous vehicle as input variables, the travel time of the vehicle on the targeted road segment is predicted based on the underlying BPNN model, and then the arrival time of the vehicle at each station is dynamically adjusted. Taking the travel time of Bus Route 791 in Shanghai in the morning and evening peaks as an example, the model test was performed and compared with the other four models. The results indicate that the proposed model shows higher prediction accuracy. Especially on rainy days, it improves the prediction accuracy by 57.25% compared with the traditional BPNN model.

References | Related Articles | Metrics

Multilevel Bus Cooperative Scheduling Considering Public Bike

GAO Hua,SHEN Guo-jiang

2020, 20(2): 134-138.

Abstract ( )

PDF (255KB) ( )

Using public bike for bus transfer can be an effective way to expend the service area of buses. The cooperative scheduling for multilevel bus lines would provide benefit for a comprehensive public transportation system. This study describes the concept of public bike, and then investigates the cooperative scheduling for multilevel bus lines considering the public-transportation bike. The cooperative scheduling model was developed with minimal weighted waiting time of passenger transfers. The genetic algorithm was used to solve the model. The effectiveness of the proposed model was verified by simulations with actual bus lines. The result indicates that passenger transfer waiting time was reduced significantly by using the cooperative scheduling model.

References | Related Articles | Metrics

Multi-attributes Lane-changing Decision Model Based on Entropy Weight with Driving Styles

FENG Huan-huan, DENG Jian-hua, GE Ting

2020, 20(2): 139-144.

Abstract ( )

PDF (401KB) ( )

AHP lane-changing model needs subjective weight. In order to address the problem, the lane-changing decision mechanism and attributes are analyzed, and a multi-attribute lane change decision model based on entropy weight with driving styles is proposed. The model is simulated with different space availability D . The average lane-changing motivation probability and success probability are obtained under different driving styles and different lane demarcation patterns. The result shows that driving styles have impacts on the average probability of lane-changing motivation and lane-changing success in the range of (0.10, 0.85) and (0.30,0.85), in which the probability of radical driving style is the largest while the probability of conservative driving style is the smallest. It shows that the model can respond to the influence of driving style preference. And the entropy weight has better applicability in the multi-lane cellular automaton lane-changing decision model.

References | Related Articles | Metrics

Uncertain Bi-level Programming Model for Vehicle Allocation Problem of Bus Lines

XUE Yun-qiang, GUO Jun, AN Jing, XUE Luo-wei, SANG Zi

2020, 20(2): 145-150.

Abstract ( )

PDF (358KB) ( )

In order to rationally optimize the bus line allocation, this paper considers the uncertainty in passenger demand at bus stops and introduces the uncertainty theory to construct an uncertain bi-level programming model for bus line allocation. Among them, the upper- level goal is to maximize the revenue of the bus operation enterprise, and the lower-level goal is to minimize the total travel time and the cost of passengers. The constraints include the service level and the ride rate required by the government. The model is solved through Matlab programming. Taking Bus Line No. 210 in Nanchang as an example, the uncertain bi-level programming model is used to optimize the vehicle allocation in the morning peak 07:00-08:00, and under the ride rate limit of 80%. The fleet size of vehicles is reduced from 26 to 23 by 11.5%; the total weighted cost of passengers after peak hours slightly increased by 0.5%; and the profit increased by 112 yuan with a rate of 29.6% after optimization. The results show that the optimization effectiveness of the uncertain bi-level model for vehicle allocation is significant in the early peak hours. The study provides theoretical support for bus operators to rationally optimize vehicle allocation considering uncertain factors in reality.

References | Related Articles | Metrics

Defensive Driving Behavior Analysis Model with Follow-up Surveys

GAO Hong-li, GAO Li-ying, YANGWei-cheng, FAN Shuang-shuang

2020, 20(2): 151-156.

Abstract ( )

PDF (351KB) ( )

This study mainly deals with the problem of quantitatively measuring psychological process of defensive driving behavior. The defensive driving behavior scale is designed by defining the concept of defensive driving behavior. The analysis model was developed based on the theory of planned behavior (TPB). In the model, the prior behaviors, behavioral attitude, subjective norms, and perceived control are predictive variables; the behavioral intention is intermediary variable, and the subsequent behaviors are outcome variables. Questionnaire data were collected from non- professional drivers at two time points separated by three months. TPB variables, demographic information, and prior behavior were measured at the first survey and subsequent behavior was measured at the second survey. 213 valid and matching questionnaires were obtained from two surveys. The analysis results show that, subjective norms have no significant impact on defensive driving intention. The perceived control (0.40), behavior attitude (0.29), and past driving behavior (0.26) have positive impact on behavior intention. Behavioral attitude, perceived control, and prior behaviors can explain subsequent behaviors to some extent through the behavioral intention (0.36).

References | Related Articles | Metrics

Collaborative Decision-making of Emergency Vehicle Scheduling and Traffic Evacuation Based on Bi-level Bat Algorithm

DUAN Xiao-hong, WU Jia-xin, ZHOU Zhi-qing

2020, 20(2): 157-165.

Abstract ( )

PDF (425KB) ( )

In the emergency rescue of multiple accidents in an urban road network, the phenomenon of emergency vehicle detention caused by traffic congestion often occurs, which seriously affects the rescue efficiency of road traffic accidents. This paper improves the reliability of the rescue path through traffic evacuation, and constructs a bi-level programming model for collaborative decision-making of emergency vehicle scheduling and traffic evacuation. A bi-level bat algorithm is designed to solve the model. In the upper level, the dispatching strategy with the shortest response time is got under the constraints of emergency vehicle demand, accident time window, and available vehicles. In the lower level, the traffic evacuation strategies of multiple shortest paths are obtained under the constraints of road capacity and evacuation demand, and then the paths of the shortest time are selected. The results of the numerical example show that the model can effectively improve the efficiency of emergency rescue, and the algorithm shows excellent optimization ability and running speed.

References | Related Articles | Metrics

Traffic Congestion Identification of Air Route Network Segment Based on Ensemble Learning Algorithms

LI Gui-yi, GUO Ming-yu, LUO Yi-fan

2020, 20(2): 166-173.

Abstract ( )

PDF (547KB) ( )

This study uses four indicators for traffic congestion evaluations: air traffic flow, density, saturation, and approaching rate, based on the aircraft ADS-B track data in air route network. The level of traffic congestion on air route segments was defined through the fuzzy C-means clustering algorithm and the parameters of historical traffic congestion evaluation indicators. Based on ensemble learning algorithm, a traffic congestion status identification model was developed to identify the traffic congestion status of air route network. The empirical analysis results show that the accuracy rate of the proposed model is 98.34% in traffic congestion identifications. Decision tree- based learner performs better than k - nearest neighbor- based learner. Increasing the number of ensemble base learner could improve the identification accuracy of the model. The ensemble model performs better than the BP neural network model in terms of the identification performances. The identification method proposed in this study is practical and has certain application value.

References | Related Articles | Metrics

Train Stop Plan Optimization of High-speed Rail for Improving Passenger Travel Efficiency

XU Ruo-xi, NIE Lei, FU Hui-ling

2020, 20(2): 174-180.

Abstract ( )

PDF (454KB) ( )

When optimizing train stop plan of long distance high-speed rail (HSR) lines, the relationship between increasing the number of fast trains between major stations and enhancing the train service frequency at non-major stations should be appropriately balanced. Based on using strategies of eliminating stops at non-major stations, and adding all- stop trains running between two adjacent major stations, this paper proposes a mixed integer programming model to optimize the train stop plan. The goal is to minimize the total passenger time loss generated both from train stops and transfers, and a genetic algorithm is designed to solve the model. The approach was applied to improve a practical train stop plan of Beijing-Guangzhou HSR. Comparing the optimal plan with the actual one, the number of fast trains was increased by 94.4%, which enhanced the competitiveness of HSR. The frequency of all- stop trains and the train frequency of non-major stations were at least 1 train/(3 h), which benefited passengers who need to transfer. The time loss of passengers was reduced by 40.08%. The overall travel efficiency of passengers was improved.

References | Related Articles | Metrics

Finding Shortest Path for Battery Electric Vehicles Considering Loops

HE Fang, LUO Zhi-xiong, YANGYan-ni, LI Meng

2020, 20(2): 181-187.

Abstract ( )

PDF (414KB) ( )

Battery electric vehicles enjoy fast- growing adoption in recent years. However, there are still some problems including limited driving range, insufficient charging infrastructure and long battery charging time that cannot be ignored and result in electric vehicle drivers detouring to recharge. In this situation, there exist loops on the traveling paths between an OD pair. To address this problem, the road network was reconstructed. Then considering the driver's queuing and choice behaviors with different level of charging stations, a mixed integer programming model for finding the shortest path of EV was put forward, which can be solved by mature programming software. In order to improve the solving efficiency under large- scale road network, an improved label-setting algorithm based on dynamic programming was proposed, which can be employed to efficiently solve the shortest path problem of electric vehicles with loops. Finally, a numerical example was illustrated to verify the rationality and effectiveness of the proposed model and algorithm.

References | Related Articles | Metrics

Taxi Pooling Method of Urban Integrated Passenger Transport Hub with Trajectory Similarity

WU Yue-lin, YUAN Zhen-zhou, CHEN Qiu-fang, XIAO Qing-yu,WANGWen-cheng,WEI Lai

2020, 20(2): 188-195.

Abstract ( )

PDF (506KB) ( )

For the problem of passenger queue stranded in integrated passenger transport hubs, a taxi pooling model considering trajectory similarity is proposed in this paper. The objectives are to minimize the number of taxis and the total mileage. A trajectory similarity indicator based on boundary area is introduced to morphologically restrict the driving trajectory after taxi pooling. A two-phase algorithm is designed to solve this NP-hard problem. In the first phase, the k-medoids is used to clustering the demands; and in the second phase, the ant colony algorithm is designed to obtain the passenger matching schemes and driving routes. Finally, the results based on survey data prove that the method can decrease the number of taxis and mileage and reduce the waiting time of passengers. Besides, the JAC value of route after taxi pooling is improved because of the trajectory similarity constraint, which satisfies the passengers' expectations of the least difference between the taxi pooling route and the original route.

References | Related Articles | Metrics

BP Neural Network Model for Short-time Traffic Flow Forecasting Based on Transformed Grey Wolf Optimizer Algorithm

ZHANGWen-sheng, HAO Zi-qi, ZHU Ji-jun, DU Tian-tian, HAO Hui-min

2020, 20(2): 196-203.

Abstract ( )

PDF (451KB) ( )

Accurate short-time traffic flow forecasting is the basis of traffic control and traffic induction. In this paper, a short time traffic flow forecasting model (TGWO- BP) is proposed based on transformed grey wolf optimizer algorithm (TGWO) and BP neural network, which can effectively improve the accuracy of short- time traffic flow forecast. Firstly, due to the drawbacks that the standard gray wolf algorithm converges slowly and tends to fall into the local extremum, an adaptive decreasing convergence factor is proposed, so that the grey wolf algorithm can distinguish the global search from the local search. Secondly, the position renewal formula of the gray wolf individual is improved by introducing the inertial weight. By adjusting the size of the inertial weight, the grey wolf algorithm has the ability to jump out of the local extremum. Finally, four short- time traffic flow forecasting models of TGWO-BP, GWO-BP, PSO-BP and BP are constructed, and the results show that the error of the short-time traffic flow forecasting model of TGWO-BP is 10.03%, and the accuracy of the prediction is better.

References | Related Articles | Metrics

Short-term Traffic Flow Forecast Based on Improved Wavelet Packet and Long Short-term Memory Combination Model

ZHANG Yang, YANG Shu-min，XIN Dong-rong

2020, 20(2): 204-210.

Abstract ( )

PDF (415KB) ( )

This paper proposes a short- term traffic flow prediction method based on improved wavelet packet analysis and long short-term memory neural network combination (IWPA-LSTM) to overcome the shortcomings of short-term traffic flow prediction under the condition of unsteady traffic flow, such as low precision and overreliance on large sample historical data. First, the wavelet packet analysis algorithm was improved by the idea of power spectrum refinement. The wavelet packet algorithm was used to perform multi- scale decomposition and single-branch reconstruction of small sample traffic time series. Next, the phase space of low-frequency sequence and high-frequency sequence is heavy. At the same time, the layer-by-layer construction of the long and short-term memory model was completed, local preservation was performed and adaptively updated according to prediction accuracy. Then the reconstructed subsequence was input into the model for training and prediction. In the last step, the predicted values of each subsequence were superimposed to output the final predicted value of IWPA-LSTM. It was proved that the proposed IWPA-LSTM model produced higher prediction accuracy than the classical deep learning model for small sample size problems, and the practicability of the model was improved significantly.

References | Related Articles | Metrics

An UGV Swarm Formation Reconfiguration Method with Space-time Constraints

SU Zhi-yuan, LI Yong-le, XU You-chun，ZHANG Yong-jin

2020, 20(2): 211-217.

Abstract ( )

PDF (414KB) ( )

This study proposes a framework for formation reconfiguration of unmanned ground vehicle (UGV) swarm with space-time constraints. The graph theory method was used to describe the coordination relationship of swarm, and the swarm reconfiguration problem was decomposed into the formation transformation problem of the minimum transformation unit. To meet the space constraints of the formation transformation, the study used the model-based predictive path generation algorithm to produce the travelable path that satisfies the initial and final constrains for the UGV. The curve interpolation method was used to create a speed curve that connects the initial and final state smoothly and meets the time constraints. The boundary conditions of speed planning were then proposed based on the collision detection results. Both simulation analysis and vehicle experiments were performed to verify the effectiveness of the proposed swarm reconfiguration method.

References | Related Articles | Metrics

Traffic Characteristics and Capacity of Overtaking Lane Partly Occupied Traffic Control Zone of Four-lane Freeway

MENG Xiang-hai, ZHANG Long-zhao, LI Sheng-long

2020, 20(2): 218-224.

Abstract ( )

PDF (501KB) ( )

This paper studies the lane utilization rate, traffic flow distribution on the lane, lane changing characteristics, speed characteristics and road capacity of the four-lane freeway traffic control zone which overtaking lane is partly occupied. The flow distribution curve, lane changing direction and lane changing rate curve, and velocity distribution curve of each main section in the carriageway and overtaking lane in the traffic control zone were obtained. The Greenshields model describing the relationship between velocity and flow was calibrated, and the capacity of each main section was determined accordingly. The results show that when the traffic flow is at a low level, the utilization rate of overtaking lane is low, only about 20% of the normal level. At this time, lane division, lane setting and traffic control mode should be optimized and adjusted. At the low level of traffic flow, the average speed and running speed of the traffic control zone are both high, so the running speed should be used as the value of the speed limit value. The capacity of the activity area in the traffic control zone is the lowest, only about 89% of that of the normal section. On the premise of ensuring traffic safety, efforts should be made to improve the capacity of the bottleneck section, and the section capacity should be taken as the basis for determining whether to carry out forced diversion.

References | Related Articles | Metrics

Standing Passenger Density of Urban Rail Transit Based on Tolerance

CHENWei, LI Zong-ping, YU Da-ben, JU Yan-ni, YIN Jia-cheng

2020, 20(2): 225-230.

Abstract ( )

PDF (417KB) ( )

In order to determine the seat density of urban rail transit reasonably, this paper discusses the influence of seat density on the tolerance from the perspective of seat passenger tolerance. From the perspective of time and space perception, this paper discusses the classification of influencing factors of comprehensive tolerance, including three main factors: standing passenger density, standing time and waiting time. Through the introduction of interval fuzzy numbers, the fuzzy processing of tolerance perception is realized, and the regression accuracy of parameters is improved. Regression analysis is carried out on the model, and the parameter estimation of different types of passenger tolerance is obtained, and the contour map of tolerance threshold with respect to factor superposition is drawn. According to the simulation results, the tolerance threshold of seat density is modified. The results show that under normal conditions, the seat density that ordinary passengers can accept are 6.57~6.92 people/m2. The bounds of seat density that commuters can accept are 7.21~7.63 people /m2. The results of the model can provide a reference for determining the reasonable stand density in planning and design.

References | Related Articles | Metrics

Bus Accessibility in Urban Areas Based on Hot Spot Detection

QUANWei, SUN Chao

2020, 20(2): 231-236.

Abstract ( )

PDF (403KB) ( )

An evaluation method was proposed for bus accessibility in urban areas based on taxi trajectory data. First, kernel density analysis and hot spot detection were used to identify the travel sensitive areas of taxi users, and the travel heat was represented as the travel demand in the different areas for studying the public transport accessibility in the sensitive areas. Secondly, key indicators affecting accessibility were selected to model for evaluating bus accessibility by regarding the correlation between each indicator and the heat of hotspot as the weight index. Finally, the variation coefficient of public transport supply and demand, the Moran scatter map of supply and demand, as well as the distribution figure of the gap between public transport supply and demand, were introduced to evaluate the coupling between supply and demand of public transport accessibility for the whole city and identify the regional imbalance degree in order to help the public transport system optimize the supply of bus. Xi'an, a large city was selected as a case study for methodology verification. The variation coefficient of regional bus supply and demand is 1.89, which indicated that there is a certain imbalance. According to the distribution of the gap between supply and demand as well as scatter plot, it is effective to identify the priority-processing areas and take actions to improve the accessibility of them to optimize the coupling degree for the whole city.

References | Related Articles | Metrics

Delay Extraction Based on Vehicle's Trajectory Reconstruction at Signalized Intersection

ZHANG Hui-ling, LIU Xiao-xiao, XU Yu-dong

2020, 20(2): 237-243.

Abstract ( )

PDF (414KB) ( )

Taking the first f- th vehicles' information under the monitoring camera facility at the signalized intersection as the data input, this paper reconstructs the trajectory of the vehicles, and extracts the delay information. The calculation method on the key time points in the vehicle's trajectory reconstruction, including the initial deceleration time, stopping time, start-up time and recovery time, is described as follows. With the first f-th vehicles' stopping time and the vehicle's arrival distribution function, the following vehicle's stopping time can be deduced; with the vehicles' deceleration procedure fitted as the hyperbolic sine function, using this function and the stopping time, the deceleration time can be calculated; using the first f-th vehicles' start-up time and the saturation headway, the following vehicle's start- up time can be determined; and using the uniform variable acceleration procedure, the recovery time can be calculated. Lastly, the accuracy of the method has been verified through the real vehicle's trajectory data.

References | Related Articles | Metrics

Archive