Loading...

Archive

    25 August 2022, Volume 22 Issue 4 Previous Issue    Next Issue

    For Selected: Toggle Thumbnails
    Game Pricing of Container Road and Multimodal Transport Considering Railway Discount
    PIAN Feng, CHEN Yang, PANG Shi-hua, SU Min
    2022, 22(4): 1-10.  DOI: 10.16097/j.cnki.1009-6744.2022.04.001
    Abstract ( )   PDF (1341KB) ( )  
    Considering the economic cost, time cost, and carbon emission, this paper uses the generalized cost function with road transport and combined road-rail transport for the analysis. The bi-level programming model is developed with the upper layer maximizing the carrier's profits and the lower layer minimizing generalized costs. The equilibrium price of containers transported by road and by combined road-rail through a dry port is analyzed in consideration of the discount scale of railway transportation. A sensitivity-based heuristic algorithm is used, and the sensitivity of shipper's utility sensitive weight coefficient is analyzed. The results show that the price game between road and multimodal transport reduces freight rates for both parties. The railway transport discount scale allows multimodal transport to increase freight sharing rate through freight rate reduction, thus the multimodal transport profit will be increased, both the generalized transport cost and the carbon emission will be reduced. A 20% railway transportation discount increases the freight sharing rate of multimodal transportation from 29.44% to 45.37% , increases profits by 14.20% , reduces generalized costs of all container transportation by 2.71%, and reduces carbon emissions by 16.70%. The study also found that direct rail access to seaports through dry ports increases multimodal transport's freight sharing rate by 13.82%, increases profits by 33.27%, reduces generalized cost of all container transportation by 4.24%, and reduces carbon emissions by 14.16%. The shipper's transportation service preference will affect the dry port's pricing strategy.
    References | Related Articles | Metrics
    Pricing Game Model of Travel Service Chain Under Social Preference View
    LI Wan-ying, GUAN Hong-zhi, HAN Yan , CAO Yang-liu
    2022, 22(4): 11-22.  DOI: 10.16097/j.cnki.1009-6744.2022.04.002
    Abstract ( )   PDF (2262KB) ( )  
    To analyze the pricing strategy game problem from the perspective of social preference, a travel service chain alliance was constructed to compose service providers and MaaS (Mobility as a Service) platform. Three Stackelberg game models were established based on the social preference theory, i.e., the inequality-averse decision model, the altruistic decision model, and the social welfare comprehensive decision model. The results were compared with two basic decentralized and centralized decision models. The pricing strategy of travel service products and service chain profits were explored with the social preferences of decision-makers under MaaS scenarios, and the validity of the model was verified by numerical analysis. The results show that the dispersion of decision-making power leads to a decrease in the sales price of products and the overall optimal profit of the service chain. It is proved that the optimal price of the product is not affected and the overall profit of the service chain remains unchanged in the inequality-averse decision model. The overall optimal profit of the service chain increases continuously in the altruistic decision model, and the optimal decision variables are more influenced by the altruistic preference of service providers in the comprehensive social welfare decision model. It is necessary to pay attention to the inequality-averse preferences of decision-makers in the service chain and implement altruistic behaviors, which can increase the demand and overall profit of travel service products, facilitate the coordinated operation of the travel service chain, and realize the integrated development of transportation and tourism industry.
    References | Related Articles | Metrics
    Intercity Train Time-sharing Pricing Based on Cumulative Prospect Theory
    YANG Yun , ZHANG Xiao-qiang, XU Xin-hao
    2022, 22(4): 23-29.  DOI: 10.16097/j.cnki.1009-6744.2022.04.003
    Abstract ( )   PDF (1375KB) ( )  
    Intercity passenger flow is characterized by an uneven temporal distribution, which leads to insufficient capacity during peak periods and a low seating rate during non-peak periods. To balance the passenger flow and improve the income of the intercity high-speed railway, differential pricing in the passenger corridor is set in different periods. In this paper, the limited rational choice behavior of passengers is considered, and the cumulative prospect value is used to describe the travel utility value of passengers to parallel trains. Considering the difference and bounded rationality of passengers' choice behavior, the latent class analysis is used for passenger classification. By selecting two influencing factors of fare and period value, a product utility model is established for parallel passenger trains with double reference points, and the travel utility is described using the cumulative prospect value of heterogeneous passengers on parallel trains. Based on the cumulative prospect value, a time-sharing pricing bi-level programming model with the maximum profit of railway enterprises. and the minimum generalized travel cost of passengers is constructed. A heuristic algorithm based on sensitivity analysis is designed to solve the model. Finally, the NanningBeihai railway is taken as an example to analyze the peak and non-peak hours. The results show that the proposed timesharing pricing method can increase revenue by about 2.5%, which makes the passenger flow distributed more evenly.
    References | Related Articles | Metrics
    A Methodological Study of Multimodal Freight Transportation Models for Large Regions Based on an Integrated Modelling Framework
    DENG Gui-hua , ZHONG Ming , RAZA Asif , HUNT John-douglas , ZHOU Yong
    2022, 22(4): 30-42.  DOI: 10.16097/j.cnki.1009-6744.2022.04.004
    Abstract ( )   PDF (3592KB) ( )  
    With an accelerated regional development and integration in China, an increasing number of regions have begun to consider how to coordinate the development of economy, land/space, transportation, and environment toward a sustainable development through an integrated modeling/planning process. With this, this study proposes a methodology for designing and developing a large-scale, integrated, multi-commodity, multimodal freight transportation model, through systematically considering various interactions between economy, land/space, transportation, and environment of a region. First, the structure of the model is developed by creating an Aggregate Economic Flows (AEFs) table embedded in the PECAS (Production, Exchange and Consumption Allocation System)modeling framework primarily based on the majority of types of goods transported by the three modes (waterway, railway and highway) and corresponding producing industries or sectors, with a target to provide a full economic representation for regional freight demand analysis. With such a representation, the producers, consumers, exchanged commodities, the categories of land and space and the transportation modes are fully represented and their interactive relationships are quantitatively represented. Next, the four individual modules of the model are developed, including a Economic and Demographic (ED) module, an Activity Allocation (AA) module, a Space Development (SD) module, and a multimodal Transportation Systems (TS) module, which predicts and simulates the total of socio-economic activities (by sector), allocates the total activites into different land use zones (LUZs) or traffic analysis zones (TAZs), simulates land/space development over the zones and estimates corresponding freight demand to be transportated between those zones over a multimodal supernetwork (consisting of the above three networks and connecting/transfer links), respectively. Given the constraints based on the socio-economic development target, land/space available, level of supply of transportation infrastructure and environmental capacity in the regional development, the designed modeling approach can achieve a much more objective simulation of temporal and spatial distribution characteristics of socioeconomic activites, the interactions between them, the corresponding freight demand and multi-commodity flows over the multimodal transportation network, which may provide a good decision support capacity for an integrated planning of regional allocation of socioeconomic activities, and a land use and transportation system with a high level of service. The proposed modelling method is applied to the Yangtze River Economic Belt (YREB) and an integrated, multi-commodity, multimodal freight transportation model of the region is developed as a case study. The activity totals, land use/space patterns and corresponding freight demand and multimodal transportation systems are modelled from the year 2015 to 2035. The modeling results clearly show that the goodness of fit of the predicted annual average daily tonnage is higher than 85%, and the errors of mode share between predicted and observed values are less than 1%, which proves that the proposed modeling method is accurate and practical for policy analysis at regional scale.
    References | Related Articles | Metrics
    Knowledge Mapping Analysis on China's Ship Emission Reduction Technology Innovation Under "Double Carbon" Goal
    DONG Gang , GUAN Min
    2022, 22(4): 43-52.  DOI: 10.16097/j.cnki.1009-6744.2022.04.005
    Abstract ( )   PDF (1790KB) ( )  
    As ship exhaust emission has become a significant source of air pollution and to improve China's ship emission reduction through technology innovation under the "double carbon" goal, this paper systematically analyzes the research on technological innovation of ship emission reduction in China based on the knowledge mapping method. The study comprehensively uses the visualization tools such as CiteSpace and SATI, and reviews the literatures on ship emission reduction technology innovation in both domestic and abroad. The results show that the research on ship emission reduction technology innovation in China has overall experienced four stages: initial stage (2001- 2007), growing stage (2008-2014), stabilized stage (2015-2018), and prosperous stage (2019-2021). The research hot topics include ship energy- saving and emission reduction technologies and measures, green ship and green shipping, ship energy efficiency and efficiency management. The future research will develop towards carbon neutralization, alternative fuels, shore power, low sulfur fuel, green ship design, etc. Through the systematic combing and in-depth analysis of the existing core literature, the paper reveals the current research status and future direction in the field of ship emission reduction technology innovation in China, which provides reference for the practices of extensive and profound "economic and social systematic change" brought by China's ship emission reduction technology innovation under the "double carbon" goal.
    References | Related Articles | Metrics
    Synergistic Benefit Analysis of CO2 and NOx Emissions in Civil Aviation of China Under Dual-carbon Target
    HAN Bo , DENG Zhi-qiang , YU Jing-lei, SHI Yi-lin , YU Jian
    2022, 22(4): 53-62.  DOI: 10.16097/j.cnki.1009-6744.2022.04.006
    Abstract ( )   PDF (1989KB) ( )   PDF(English version) (711KB) ( 41 )  
    As an important part of transportation, the civil aviation industry is of great significance to the implementation of the national carbon peak and carbon neutral policy. In this study, a comprehensive prediction model of carbon emission and air pollutant emission is constructed to fit the characteristics of civil aviation, and the growth of civil aviation aircraft and CO2 and NOx emissions in 2019-2050 is predicted. The synergistic control coordinate system and synergistic emission elasticity coefficient are used to evaluate the synergistic benefits of emission reduction. The research shows that the future annual increment of civil aviation aircraft will show a trend of continuous growth, which is closely related to the development of GDP, potential output, and labor efficiency. The improvement of fuel efficiency cannot change the current situation of the continuous growth of CO2 and NOx emissions in civil aviation. The development of sustainable aviation fuel will make the CO2 emission of civil aviation peak at 3.18×108 tons in 2045 and promote the continuous growth of NOx emission. This effect can be eliminated through technological advancement and the introduction of new power aircraft, and the peak time of CO2 emission in civil aviation will be advanced to 2040, in which the emission can be reduced to 2.65×108 tons. On this basis, accelerating the application of sustainableaviation fuel can make civil aviation CO2 emissions peak at 2.47×108 tons in 2037. Civil aviation cannot achieve a carbon peak in 2030. The proper introduction of sustainable aviation fuel, the acceleration of civil aviation technology improvement, and the application of new power aircraft are the best choices to strengthen the coordinated emission reduction of CO2 and NOx in civil aviation. Finally, some suggestions are provided that, the best way for the civil aviation industry to achieve dual carbon is to focus on improving aircraft fuel economy in the short term, accelerating the application of sustainable aviation fuel in the medium term, and realizing zero-carbon flight by relying on new powered aircraft in the long term.
    References | Related Articles | Metrics
    Vehicle Lane Change Intention Recognition Driven by Trajectory Data
    ZHAO Jian-dong, ZHAO Zhi-min , QU Yun-chao , XIE Dong-fan , SUN Hui-jun
    2022, 22(4): 63-71.  DOI: 10.16097/j.cnki.1009-6744.2022.04.007
    Abstract ( )   PDF (1998KB) ( )   PDF(English version) (832KB) ( 80 )  
    In order to accurately identify the vehicle's lane-changing intention and improve the driving safety of the vehicle, I comprehensively considered the spatiotemporal characteristics of the vehicle's lane-changing process and the influence of different characteristics on the vehicle, and proposed a lane-changing intention recognition model with attention mechanism, which is based on the combination of Convolutional Neural Network (CNN) and Gated Recurrent Unit Neural Network (GRU). Firstly, I filtered and smoothed the vehicle trajectory data, and divided the vehicle trajectory data into three categories: left lane change, right lane change, and straight driving, so as to construct a sample set of lane change intention. Secondly, I built a CNN_GRU model that integrates attention mechanism to identify the sample set of lane change intention. Considering the interaction between vehicles during driving, I utilized the position, the speed information of the predicted vehicle and surrounding vehicles as the input of the model. After the CNN layer feature extraction, I then chose the extracted features as the input of GRU layer. And I also added different weight coefficients to different features through the attention mechanism layer, and leveraged the Softmax layer to identify the lane change intention. Finally, I verified the performance of CNN_GRU model with fused attention mechanism by using the trajectory data of US-101 dataset in NGSIM, and at the same time, compared and analyzed it with LSTM, GRU, CNN_GRU and CNN_LSTM_Att models. The results showed that the proposed model achieves an overallaccuracy of 97.37% for vehicle lane change intention recognition with an iteration time of 6.66 s, which is at most 9.89% and at least 2.1% improvement in accuracy compared to other models. By analyzing the accuracy of intention recognition at different pre-determination times, we know that the intention to change lanes can be accurately recognized within 2 s before the vehicle changes lanes, and the accuracy rate is above 89%, so the model has good recognition performance.
    References | Related Articles | Metrics
    Analyzing Error Bounds of Highway Traffic State Estimation Via Kalman Filter Fusion
    CHEN Xi-qun , CAO Zhen, MO Dong
    2022, 22(4): 72-78.  DOI: 10.16097/j.cnki.1009-6744.2022.04.008
    Abstract ( )   PDF (2268KB) ( )  
    This paper analyzes the data quality of highway traffic flow and proposes a decision-level fusion model based on Squared Flow Error Bound (SFEB) and Extended Kalman Filter (EKF), namely SFEB-EKF. It is used to calculate the error bounds of traffic state estimation for the road sections with and without detectors when the detection spatial coverage is insufficient. Compared with the SFEB algorithm, the fusion model uses the EKF traffic state estimation model to estimate traffic states of the whole highway section. The lower bound of traffic state estimation error is calculated based on the obtained estimated samples. At the same time, the Nearest Neighbor Method (NNM) is used to calculate the upper bound of traffic state estimation error of the whole road sections. The model is tested with an open-source highway dataset, and results show that, compared with the SFEB algorithm that needs to input real samples, the fusion model SFEB-EKF can achieve similar results in the absence of real samples and the gap is kept within 5%. The model shows good stability under different detection coverage experiments. The research results can give the estimation boundary of the traffic state of the road section without detectors, and provide a reference for the layout plan of highway traffic detectors.
    References | Related Articles | Metrics
    Modeling and Simulation of Pedestrian Crossing Behavior at Uncontrolled Mid-block Crosswalk
    CHEN Peng , TANG Peng , YAN Wei-xi , SUN Qiu-yue
    2022, 22(4): 79-88.  DOI: 10.16097/j.cnki.1009-6744.2022.04.009
    Abstract ( )   PDF (2279KB) ( )  
    To describe the pedestrian crossing decision-making and micro motion behavior at uncontrolled mid-block crosswalk, this paper proposes a two-layer pedestrian crossing model from the tactical and operational perspectives. At the tactical level, considering the step-by-step decision-making characteristics of pedestrians, the study uses the binary logistic regression model to develop the pedestrian crossing decision-making model for pedestrians on the roadside and in the middle of the road. At the operational level, considering the limitations of the traditional social force model in describing pedestrian crossing, the study introduces the pedestrian active avoidance force and the force of crosswalk on pedestrians to develop the pedestrian crossing micro motion model. The pedestrian crossing double-layer model is simulated by the AnyLogic platform, and the effectiveness of the model is analyzed with the actual data and simulation results. The results show that compared with the roadside decision-making model, pedestrian gender, the number of vehicles within the parking sight distance and the distance between pedestrians and potential conflict points have a more significant impact on the decision-making of pedestrians in the middle of the road. The improved social force model of pedestrian active avoidance force and pedestrian crossing force on pedestrians can better reflect pedestrian behavior characteristics at uncontrolled mid-block crosswalk. The effective pedestrian crossing behavior simulation model for uncontrolled mid-block crossing can provide a support for subsequent pedestrian crossing safety improvement studies and polices.
    References | Related Articles | Metrics
    An Optimization Model of Road Traffic Flow Path Identification Based on RFID Data
    REN Qi-liang , XU Tao , CHENG Long-chun
    2022, 22(4): 89-95.  DOI: 10.16097/j.cnki.1009-6744.2022.04.010
    Abstract ( )   PDF (1682KB) ( )  
    To reduce the error of traffic flow path identification on non-RFID (Radio Frequency Identification) covered roads, an optimization model of traffic flow path identification on RFID roads is proposed based on Floating Car Data (FCD)verification. Firstly, the initial RFID data is categorized into traceable traffic flow, non-traceable traffic flow, and random items by TIDWT. According to the number of floating cars in the statistical section, the road is classified into three categories: Full, Defect, and Null. A FCD-RFID tracing path model is established to identify the composition of traceable traffic flow paths. At the same time, a comprehensive cost function is proposed that considers travel time, road grade, and driving preference. The non-traceable traffic flow and random items path are estimated by RPL-OSUE. Finally, the final path composition of road section traffic flow is identified by path superposition. The results show that, compared with the single RFID traffic flow path recognition, the combined model has higher accuracy. The Mean Absolute Error (MAE) is 72 vehicles, which is 62.5% lower than the single RFID algorithm, and the Mean Relative Error (MRE) reaches 9.5%, which is 72.2% lower than the single RFID algorithm. In the non-RFID covered roads, the MRE of the combined model is 13.3% , which is 82.0% lower than that of the single RFID algorithm, showing feasibility and applicability of the model.
    References | Related Articles | Metrics
    Simulation Modeling of Additional Channelization Island Type Exit Based on Cellular Automata
    CHEN Yong-heng , LI Hao-nan, WU Chang-jian, LI Wan-ning
    2022, 22(4): 96-105.  DOI: 10.16097/j.cnki.1009-6744.2022.04.011
    Abstract ( )   PDF (3483KB) ( )  
    To study the influence of the additional channelization island type exit on expressway traffic flow, this paper establishes a cellular automata model for it. Based on the KKW (Kerner-Klenov-Wolf) model, three lane-changing rules are introduced and set by sections to describe the lane-changing behavior of vehicles under the influence of the additional channelization island type exit. According to the geometric characteristics of the direct connection between the main and side roads, the process of vehicles on the main road passing through the exit is equivalently simplified, and the rules of vehicles driving out are defined under the additional channelization island type exit. The numerical simulation results show that the channelization island can make the inside lane of the downstream side road play the role of the auxiliary lane by guiding vehicles from the upstream of the channelization island to merge with the outside lane. Under its setting, the direct interference caused by side road traffic flow on vehicles driving out is relieved, and the traffic efficiency of the main road is significantly improved. However, the mandatory lane changing of side road vehicles after passing through the channelization island will cause disturbance to the stable outgoing traffic flow and induce local congestions, which will affect the further evacuation of outgoing traffic flow and lead to the decrease of saturated flow and average speed of the main road. In addition, the channelization island causes the road reduction bottleneck to form on the side road, and the total saturated flow of the side road decreases significantly under its influence. The occupation of the channelization island on the inner lane of the side road makes its flow rate and average speed lower than those of the outer lane, and the traffic efficiency is more affected.
    References | Related Articles | Metrics
    Refined Classification of Urban Rail Transit Stations Based on Clustered Station's Passenger Traffic Flow Features
    JIANG Yang-sheng , YU Gao-shang, HU Lu, LI Yan
    2022, 22(4): 106-112.  DOI: 10.16097/j.cnki.1009-6744.2022.04.012
    Abstract ( )   PDF (1689KB) ( )  
    The existing classification of urban rail transit stations is mostly based on qualitative analysis, which cannot meet the needs of refined design and operation. This paper proposes a refined station classification method based on clustering station public features. First of all, the entry flow data from AFC (Automatic Fare Collection) is processed into time series data, and each station is clustered by the data based on K-Means++ algorithm; A fitting equation between the passenger flow clustering and the multi-dimensional parameters of land use characteristics is established to calculate the public characteristics of stations in five major categories, such as residential station, working station, and regional center station. On this basis, considering the segmentation characteristics of different stations belonging to the same broad category of stations, this paper proposes a refined description of specific station types using a combination of the public features of five types of passenger flow. The result of case study shows that the mean absolute percentage error between the real passenger flow value and the fitted passenger flow value calculated using the proposed refined classification method for each station falls within 14% , indicating a good feasibility of the proposed classification method.
    References | Related Articles | Metrics
    Accessibility to Urban Rail Transit Based on Enhanced Two-step Floating Catchment Area Method
    YUAN Hong-xia , XU Ling , YAN Yu-song , JIANG Wen-hui
    2022, 22(4): 113-119.  DOI: 10.16097/j.cnki.1009-6744.2022.04.013
    Abstract ( )   PDF (2117KB) ( )  
    The research on the accessibility to urban rail transit is very significant to elevate the service level of rail transit. From the perspective of departure locations, this study calculates and analyzes the accessibility to rail transit stations by three connection modes (walking, shared bicycle, and bus) by using the enhanced two-step floating catchment area method (E2SFCA) based on a distance decay function. Also, taking the areas along Metro Line 2 and Line 6 outside the Third Ring Road in Chengdu as an example, the accessibility differences in spatial locations and the accessibility differences caused by different modes are analyzed by ArcGIS. By comparing the accessibility and the population distribution, the key areas of rail transit travel are sorted out, and the inaccessible areas are found. The results show that the areas that are much closer to the rail transit station and the city center have higher average accessibility of the three connection modes. The accessibility by walking is the smallest in general, and the accessibility difference in spatial distribution is the most obvious. Moreover, the areas with high accessibility are within 0.8, 1.5, and 3.0 km of rail transit stations, for walking, shared bicycle, and bus, respectively, and the accessibility along bus lines is higher than that not. Finally, the different levels of inaccessible areas by rail transit are identified by comparing the accessibility and population spatial distribution.
    References | Related Articles | Metrics
    An Auxiliary Model for Preliminary Planning of Rail Transit Networks and Lines with Given Station Locations
    LIU Ming-min, HE Hong-jie, JIN An
    2022, 22(4): 120-128.  DOI: 10.16097/j.cnki.1009-6744.2022.04.014
    Abstract ( )   PDF (1938KB) ( )  
    To address the lack of quantitative network generation techniques and high planning complexity in the case of large numbers of stations, this paper establishes a bi-level model composed of a network design model and a line selection model, and minimizes the total length of the network and maximizes the operational efficiency. The model considers the contiguity of stations and the presence of a line on the edge, represented by binaries variables in the planning of urban rail transit networks and lines given stations. Under the constraints of station accessibility and engineering feasibility, this paper designs a minimum spanning tree algorithm based on clustering and network repair strategies, to solve network design model and improve the computational speed of network generation. Under the constraints of operation feasibility, this paper designs an algorithm for solving the bi-objective programming based on the Tabu Search, which solves the line selection model and transforms the bi-objective programming problem into a single-objective programming problem. The bi-level model can automatically generate the optimal network and line alignments when using the stations in operation and passenger OD as input, and the results can be reasonable when compared with the opening transit network. The model results show that they are similar to the real network and lines, but have less the total length of the network and higher operational efficiency. The ratio of direct passenger flow to the total increases by 0.7% and the ratio of once- transfer passenger flow to the total increases by 4.2% when the total length of the network decreases by 1.7%.
    References | Related Articles | Metrics
    Propagation Model for Hazard Causes of Intelligent Driving Assistance System Based on Complex Network
    ZHANG Shi-jie , TANG Tao , LIU Jin-tao , LI Chen-ling
    2022, 22(4): 129-136.  DOI: 10.16097/j.cnki.1009-6744.2022.04.015
    Abstract ( )   PDF (2595KB) ( )  
    To study the safety of the intended functionality (SOTIF) of the Intelligent Awareness-based Train Driving Assistance System (IATDAS) and to improve the hazard control, this paper proposes a propagation model for hazard causes of the IATDAS based on the complex network. With the SOTIF- related hazard causes network, this model provides an overall load-capacity propagation mechanism to describe the propagation of hazard causes. The case study shows that the proposed model can overcome the deficiency of the existing models in matching the actual situation under complex causal relationships. For example, this model can represent the feature that the causes with longer subsequent propagation paths are more difficult to cause a hazard. The results also show that hazard control strategies gained from this model can significantly reduce the propagating speed. For example, in the control of causes with a large and rapid influence on the network, the average propagation speed can be reduced by 68%, which is 58% lower than that of the random control strategy. Thus, our propagation model can effectively support the decision-making in hazard control.
    References | Related Articles | Metrics
    Driving Style Recognition and Quantification for Heavy-duty Truck Drivers
    QIN Wen-wen , YAN Qi-yang , GU Jin-jing , LI Wu , JI Xiao-feng
    2022, 22(4): 137-148.  DOI: 10.16097/j.cnki.1009-6744.2022.04.016
    Abstract ( )   PDF (2625KB) ( )  
    Aggressive driving styles of heavy-duty truck drivers may lead to serious traffic accidents with mass casualties, since aggressive styles have strong habitual characteristics and potential behavior risks, which are difficult to correct. However, most of the studies focus on driving style recognition of passenger cars rather than those of heavyduty trucks. In this paper, based on low- frequency trajectory data of heavy- duty trucks in Yunnan Province obtained from a national road freight platform, a framework of driving style analysis for heavy-duty truck drivers consideringfatigue driving and speeding driving characteristics is developed from the perspectives of style clustering, recognition, and evaluation. First, considering the driving behavior patterns conveyed by trajectory data, fatigue driving and speeding driving features were constructed to characterize the driving styles of heavy-duty truck drivers. Then, factor analysis for feature reduction was employed, and K-means clustering was used to classify the driving styles of heavyduty truck drivers. In addition, driving style recognition models based on support vector machine were constructed and compared with the recognition results of gradient boosting decision trees. Finally, based on the cumulative distributions of fatigue and speeding features, a quantitative assessment model for driving styles of heavy-duty truck drivers using the CRITIC (Criteria Importance Though Intercriteria Correlation) assignment method was proposed. The results show that the extracted fatigue factor and speeding factor can contain 80.838% information from the mentioned above two types of features when dimension reduction has been conducted. After that, according to the two types of features, the driving style can be divided into four categories: robust, speeding, fatigue, and dangerous. The proportion of the corresponding heavy-load truck drivers is 62.60%, 25.02%, 7.40% and 4.98%, respectively. Besides, the accuracy of the support vector machine-based driving style recognition model for heavy goods vehicle drivers is greater than 97% for all different styles, and the overall performance is better than the gradient boosting decision tree. Moreover, the proposed assessment model with the CRITIC assignment method can effectively quantify risk score for the driving style of each heavy- duty truck driver, in which the robust drivers performed the best in terms of more than 75% of drivers having a total style risk score higher than 60, whereas the risky drivers performed the worst with more than 75% of them scoring less than 20. This study could provide a theoretical basis and technical support for risky behavior monitoring, intervention, and management.
    References | Related Articles | Metrics
    Calculation of Carbon Footprint of Electric Bus Operation Based on Real World Driving Cycles
    TIAN Shun, ZHENG Bo-wen , SUN Jian , LIU Jing-yu
    2022, 22(4): 149-157.  DOI: 10.16097/j.cnki.1009-6744.2022.04.017
    Abstract ( )   PDF (1956KB) ( )  
    To monitor the carbon footprint of electric bus operations, this paper proposes a method to estimate the converted carbon emissions of electric buses based on the actual driving conditions. Taking the bus route 609 operating in the urban area of Xi'an City and the bus route 362 in the new built urban area as examples, based on real world driving data, a developmentscheme of localized driving cycles for urban electric bus operation lines is proposed. The T- distributed stochastic neighbor embedding (T- SNE) method is used for data dimensionality reduction. The Birch clustering method is used for classification, and then the driving cycles of the two study bus routes were constructed based on the principle of highest similarity and the ratio of each speed interval. The carbon emission is obtained through the conversion formula based on the power consumption calculated in cruise simulation environment. The results indicate that the electric bus routes 609 and 362 show obvious different energy consumption per 100 kilometers, which are respectively 121.71 kilowatt hourand 144.46 kilowatt hour. The result proves the necessity of constructing driving cycles for different bus routes. The carbon footprints of the routes 609 and 362 in November 2019 were also calculated based on the proposed method, which are respectively 114.099 ton and 117.863 ton. The proposed carbon footprint calculation method for electric bus is helpful for urban transportation carbon emission monitoring and management.
    References | Related Articles | Metrics
    Urban Passenger Transportation Mode Carbon Emission Efficiency Difference: A Case Study of Xiangyang City
    ZHU Shun-ying , LIAO Ling-yun, WU Jing-an, CHANG Hong-guang, WANG Hong
    2022, 22(4): 158-166.  DOI: 10.16097/j.cnki.1009-6744.2022.04.018
    Abstract ( )   PDF (2151KB) ( )  
    To accurately measure the difference of carbon emission efficiency of urban passenger transportation modes and explore the improvement possibility of multi-mobility and carbon emission reduction, this paper proposes an index framework considering vehicle stock, road, land, energy, and environment. The three-stage super-efficient SlacksBased Measure (SBM) and Stochastic Frontier Approach (SFA) are used to measure the carbon emission efficiency and clarify the influence of policy and exogenous environmental factors on the redundancy of input factors. This paper also develops an efficiency contribution model and a marginal effect model to analyze the efficiency contribution of input and output factors and their marginal effects. The case study of Xiangyang City indicates that: The carbon emission efficiency of seven modes of passenger transportation is different, such as traditional public transportation mode (including cruisers) shows higher efficiency than shared transportation which is more efficient than private transportation. Among the exogenous environmental factors, optimization of energy structure can significantly reduce the redundant consumption of energy resources, and the convenience of transportation has a significant positive effect on the elimination of redundant inputs. The efficiency contribution results reflect the heterogeneity of carbon emission efficiency improvement paths: excess vehicle size is the main constraint on carbon emission efficiency of shared transportation, with each 10% reduction in size, the efficiency increase by 12.0% for ride-hailing and increase by 19.8% for shared electric bicycles. The promotion of clean energy plays an important role on improving the carbon emission efficiency of private vehicles. Every 10% increase in clean ratio, there will be an average increase of 8.9% of car emission efficiency. At the current level of bus priority development, the carbon emission efficiency will increase by 0.201 for every 1 km∙h-1 increase in the average bus operating speed. The paper also proposes the suggestions andmeasures to improve carbon emission efficiency from technical and management aspects
    References | Related Articles | Metrics
    Impact of Urban Traffic Operations on Vehicle Carbon Dioxide Emission
    FENG Hai-xia , WANG Xing-yu , XIAN Hua-cai , LIU Xin-hua , LI Jian , NING Er-wei
    2022, 22(4): 167-175.  DOI: 10.16097/j.cnki.1009-6744.2022.04.019
    Abstract ( )   PDF (2328KB) ( )   PDF(English version) (545KB) ( 52 )  
    To quantitatively analyze the impact of traffic operation conditions in different cities on motor vehicle carbon emission, this study first used the congestion delay index (CDI) data provided by a map big data platform to analyze the spatial distribution characteristics and CDI characteristics traffic congestion of major cities in China. A speed-based CO2 emission factor was constructed, and VISSIM software was used to simulate different traffic operation conditions and the corresponding motor vehicle carbon emissions. The results indicate: the distribution of traffic congestion cities is spatially dependent and aggregated, and two high aggregation centers have been found in the Yangtze River Delta economic zone and the Pearl River Delta economic zone; the CDI has an obvious 7-day periodic fluctuation law, which is greatly affected by weather and human activities and broken by the epidemic situation; Urban traffic conditions have a significant impact on CO2 emission of motor vehicles. When the traffic is in mild congestion (CDI is 1.582), the total annual CO2 emission of urban motor vehicles in China during rush hours is about 77 million tons, which is 4.51 times that in unimpeded conditions; When the traffic can keep basically unimpeded (CDI is 1.35), the total annual CO2 emission can be reduced by 29 million tons; When the traffic reaches moderate congestion (CDI is 1.909), the total annual CO2 emission of urban motor vehicles in China during rush hours can be increased by 22 million tons;traffic is in serious congestion (CDI up to 2.394), the total annual CO2 emission of urban motor vehicles in China can reach 133 million tons during rush hours. Improving urban traffic operation can significantly reduce CO2 emission of motor vehicle emissions.
    References | Related Articles | Metrics
    Collaborative Optimization of Multimodal Evacuation Traffic Fleet Configuration and Lane Allocation
    LIU Jia-lin , JIA Bin , LIU Zheng , JIANG Rui , LI Xin-gang
    2022, 22(4): 176-185.  DOI: 10.16097/j.cnki.1009-6744.2022.04.020
    Abstract ( )   PDF (2478KB) ( )  
    Considering the dynamics and risk of evacuation traffic, this paper focuses on the joint optimization of fleet configuration and lane allocations for multimodal evacuation traffic. Based on different free- flow speed, the study discretized the road network into multi-size cell network to load multimodal traffic flow via the cell transmission model (CTM). The multimodal traffic cooperative dynamic evacuation problem is formulated as a mixed integer linear programming (MILP) model to minimize the total travel risk of evacuees. The "vehicle holding-back" problem caused by model relaxation is eliminated by adding a penalty item. Numerical experiments are conducted in Nguyen-Dupuis network to analyze fleet configuration, lane allocation, evacuation efficiency, and routes under different evacuation demands. The results show that using multimodal fleet can further reduce the total risk of evacuation compared with single-mode fleet under certain evacuation demand, and the optimal bus proportion changes stepwise against evacuation demand; there is an upper boundary of road network utilization due to the limited road network capacity; the total evacuation risk is more sensitive to the change of evacuation demand than the network clearance time; the exit sections near the risk source are usually shared by multimodal traffic, and allocating the shortest path to high-capacity buses can improve the system efficiency.
    References | Related Articles | Metrics
    A Method for Calculating Speed of Non-motorized Transportation Based on Fusion Network
    LI Xi-ying , CHEN Li-juan
    2022, 22(4): 186-193.  DOI: 10.16097/j.cnki.1009-6744.2022.04.021
    Abstract ( )   PDF (2085KB) ( )  
    Speed is an important parameter for non-motorized transportation. The current method of extracting the speed from video through target detection cannot ensure the accuracy of the detection and the stability of the target bounding box simultaneously, and the selected speed calculation datum point (abbreviated as datum point) fluctuates greatly, and thus the acquired speed is inaccurate and unstable. To solve this problem, this paper proposes a fusion detection tracking network and speed calculation method based on YOLOv5(You Only Look Once), which can obtain a more stable and accurate speed. First, we use the target detection and target tracking unit to obtain the bounding box and ID information of the detection target. We can obtain the interest region of the target according to the bounding box and send it to the head detection unit. We then obtain the matching head bounding box by the head detection unit. Secondly, according to the target features in the scene, we determine whether the head detection bounding box belongs to the current target or not. According to the different results, we provide two datum point calculation methods. Finally, the three-dimensional coordinates are obtained by mapping the two-dimensional datum point coordinates, and theresults are substituted into the speed calculation formula to obtain the speed. In addition, we proposed two evaluation indicators, accuracy ( MA ) and stability ( MS ) to quantify the evaluation methods. The detection and tracking effects of the fusion network are verified on the public datasets PETS09-S2L1 and TUD-Stadtmitte, and the effects of the datum point calculation and speed calculation methods are verified on the self-built dual-view collaborative dataset. The experimental results show that the target detection and tracking accuracy (MOTA) of the fusion network is more than 25% higher than that of using a single network. The speed calculation method of this paper is 30% more accurate and 6.28% more stable than the commonly used speed calculation method. In conclusion, the method in this paper can ensure both the detection accuracy and the stability of the target bounding box, and the fluctuation of the selected datum points is less, and the speed obtained by the method in this paper is more accurate and stable.
    References | Related Articles | Metrics
    Relationship Between Built Environment and Elderly Active Travel of Based on Group Heterogeneity
    WU Jing-xian, QIAN Yi-nan, HAN Yin
    2022, 22(4): 194-201.  DOI: 10.16097/j.cnki.1009-6744.2022.04.022
    Abstract ( )   PDF (1687KB) ( )  
    In the context of healthy aging, active travel is an effective way to improve the physical and mental health and quality of life of the elderly. The built environment is an important factor affecting the active travel of the elderly. Using data from 2013 Nanjing Household Travel Survey and built environment data, this study analyzed the heterogeneity of elderly groups based on latent class clustering, and then used a random forest method to scrutinize the importance magnitude of built environment indicators on the active travel time of different elderly groups and the nonlinear relationship between the build environment and elderly active travel. The results show that the five category variables of gender, whether retired, education, household income, and whether has preschool children in the family can be used as the exogenous variables to further classify the elderly groups into two main categories. The overall goodness of fit of the classified random forest model were significantly improved. The major built environment variables that significantly affect the first identified elderly group are the density of the active road network, the number of intersections, and the proximity to subway stations. The factors that significantly affect the active travel time of the second identified elderly group are the proximity to subway stations, the density of the active road network, and the density of residential buildings. The non-linear relationships of the same built environment indicator to the active travel time of the two elderly groups are different. The result of this study provides a theoretical reference for the policy making to facilitate the elderly travel environment and ultimately advance the healthy aging.
    References | Related Articles | Metrics
    A Fuzzy Temporal Network Model for Identifying Critical Intersections in Urban Road Network
    LI Jun-xian , SHEN Zhou-biao , TONG Wen-cong , WU Zhi-zhou
    2022, 22(4): 202-209.  DOI: 10.16097/j.cnki.1009-6744.2022.04.023
    Abstract ( )   PDF (2168KB) ( )  
    To incorporate the temporal characteristics of an urban road network, rather than considering its topology and kinetics features only, a fuzzy temporal network model is proposed to accommodate the task of recognizing key nodes in the urban road network. Firstly, the general description of the temporal network is presented, and the temporal network described with the supra-adjacency matrix is introduced. The pros and cons of their applications in traffic analysis are discussed. Then improved methods are put forward correspondingly. On the one hand, regarding the functions of the road network, indexes with fuzziness calculated from dynamic traffic parameters are suggested to depict the interaction intensity between intersections in one interval layer. On the other hand, the layer coupling coefficient is referred to and fuzzed to differentiate the interlayer correlation of intersections between two adjacent interval networks. After that, the improved intralayer and interlayer matrices are integrated to build the Fuzzy Supraadjacency Matrix (FSAM) temporal network model (FSAM model). Finally, the model's effectiveness is illustrated with the data of a busy local road network, including 147 intersections. The results show that it is necessary to analyze the importance of intersections with temporal network models. It is more reliable to define the importance-ranking of an intersection with the median. The ranking series reported by the FSAM model holds persistence for a period, and the proposed model is more comprehensive than identifying critical intersections based on a single index concerning only one isolated interval. Moreover, with different time granularity, the FSAM model exhibits good consistency in rankingthe intersections by their importance, and the results are stable. The model can be referred to identify critical intersections in urban road networks.
    References | Related Articles | Metrics
    Survival Analysis of Vehicle Lane-changing Duration on Road Segment Adjacent to Freeway Tunnels
    CHEN Zheng, WEN Hui-ying
    2022, 22(4): 210-217.  DOI: 10.16097/j.cnki.1009-6744.2022.04.024
    Abstract ( )   PDF (1984KB) ( )  
    This paper investigates vehicles' lane-changing behaviorson road segment adjacent to freeway tunnels and aims to improve vehicle's driving safety. Naturalistic driving tests were carried out on the three freeway tunnels' adjacent sections in Guangdong Province. The driving trajectories of the lane-changing vehicle and the surrounding vehicles are collected. Considering the heterogeneity of different drivers' perception of lane-changing risk level, a random parameter accelerated failure time (AFT) model was developed to describe the influence of environment and vehicle movement status on lane-changing duration in the tunnel adjacent sections. The results show that the random parameter in the AFT model has better goodness of fit compared to the fixed parameter AFT model. The significant influencing factors of lane-changing duration include distance to the tunnel, the speed difference to the leading vehicle, lane change direction, and distance to the leading vehicle in the target lane. The closer the position of lane-changing vehicle is to the tunnel and the closer to the vehicle in front of the target lane, the shorter of the lane-changing duration. Compared to the case that the speed of the lane-changing vehicle is greater than the leading vehicle in the original lane, the lane-chancing duration under a non-following state would increase by 7%. When the speed of the lane-changing vehicle is lower than the leading vehicle in the original lane, the lane-changing duration would increase by 20% compared to the case that the speed of the lane-changing vehicle is greater than the leading vehicle in the original lane. The study provides theoretical basis and method guidance for the improvement of traffic safety facilities on freeway tunnel adjacent sections and the development of microscopic driving behavior models.
    References | Related Articles | Metrics
    A Heuristic Close Contact Tracing Method for Urban Rail Transit
    XIE Liang-hui , ZHANG Zhen-ji, GONG Da-qing
    2022, 22(4): 218-227.  DOI: 10.16097/j.cnki.1009-6744.2022.04.025
    Abstract ( )   PDF (1928KB) ( )  
    In complex rail transit networks, there are always passengers whose trips are relatively fixed. These passengers can act as heuristic "witnesses" and prove the feasibility of a certain trip for other groups of passengers and help to identify potential close contacts with a guaranteed recall rate. This paper aims to develop a method to trace the close contacts in the urban rail transit system. A heuristic tree search method was used to form the feasible trip chains of target passengers by leveraging a limited number of witnesses. It could be determined whether the target passenger was a close contact by identifying whether there were any overlaps between the target trip chain and the infected trip chain. Taking Beijing urban rail transit as an example, volunteers were recruited to ride on specific lines and the information of infected persons was assumed for the study purpose. The effectiveness of the method was verified by extracting relevant Automatic Fare Collection (AFC) data to identify close contacts. In the experimental scenario, the recall rate reached 100% and the accuracy rate was 92.7% using the proposed method, indicating the feasibility of the method. The proposed method is helpful for the relevant department to take appropriate countermeasures to prevent the spread and transmission of the epidemic.
    References | Related Articles | Metrics
    Decision Making of Opening Gated Community Considering Traffic Impact on Environment
    WANG Xiao-ning , CUI Zi-yu , ZHANG Yu
    2022, 22(4): 228-235.  DOI: 10.16097/j.cnki.1009-6744.2022.04.026
    Abstract ( )   PDF (1801KB) ( )  
    Opening gated community can extend urban road network to some extent and help to alleviate traffic congestion. However, the decision-making approach in the current gated community opening plan is single layer and has not consider the vehicle emission and traffic noise pollution brought to the community after the opening. This paper used the minimum cost function value as the optimization goal and developed a bi-level decision-making model with upper-layer as optimal system cost and lower-layer for user balance. Three decision-making methods were introduced in the model: whether the gated community is open, one-way or two-way travel, and speed limit on the road. The genetic algorithm was used to solve the upper-layer model and the Frank-Wolfe algorithm was used to solve the lowerlayer model. The validation analysis of the model shows that the relative deviation of the optimal solution cost value obtained from the model is 0.67%, and the average year-on-year cost saving is 11.8%. The comparative analysis shows that the reasonable setting of three decision-making methods for the opening of gated community can reduce the travel time and detour distance of vehicles, thereby reduce the travel cost and the additional cost considering the impact on the environment. A relatively high posted speed limit helps to reduce the total travel cost.
    References | Related Articles | Metrics
    Optimization of Final "100 Meters" Drone Delivery in High-rise Residential Buildings
    JIANG Li , YANG Lu, LIANG Chang-yong, DONG Jun-feng
    2022, 22(4): 236-245.  DOI: 10.16097/j.cnki.1009-6744.2022.04.027
    Abstract ( )   PDF (2062KB) ( )  
    This paper focuses on optimization of door-to-door distribution path for high-rise residence. A high-rise residential drone door-to-door distribution model is constructed to minimize the flight cost and energy consumption cost of drone, to solve the problem of "distribution in the last 100 meters of vertical position". Considering the heterogeneity of packages and the energy consumption of drone, the constraints of the model involve drone capacity, battery pack capacity, and other factors. Based on this model, a hybrid ant colony optimization with VND (HACOVND) is designed, and four operators are introduced for variable neighborhood descent (VND) search. To improve the solution performance of the algorithm, two local search operator combinations are proposed, and different operator combinations are used for different number of customer points. The experimental results show that the HACO-VND algorithm performs better than CPLEX in terms of calculation accuracy and time, especially for large and mediumsized examples. Parameter analysis shows that the higher of the high- rise residential buildings, the greater the energy consumption utilization rate in a single flight for the drone. The drone capacity and battery pack capacity jointly affect the distribution scheme. The result of the study provides reference and ideas for the drone door-to-door service research in the future.
    References | Related Articles | Metrics
    A Risk Assessment Method of Logistics Drones on Ground
    ZHONG Gang , LI Jin, ZHANG Xiao-wei, ZHANG Hong-hai
    2022, 22(4): 246-254.  DOI: 10.16097/j.cnki.1009-6744.2022.04.028
    Abstract ( )   PDF (2127KB) ( )  
    With the improvement of the intelligent transportation system and the expansion of UAV application scenarios, logistics UAVs came into being. Logistics drones are considered to an effective transportation means for the "last-mile" express delivery. However, falls caused by air collisions and self- failures may cause injuries to ground personnel and cause certain economic losses. This paper focuses on the logistics and transportation scenarios of drones, studies the number of deaths and economic losses caused by the falling of logistics drones, and uses the risk matrix to comprehensively evaluate the risks. The main contents include: analyzing the separation of aircraft and cargo after the collision of logistics drones in the air, modeling the number of deaths on the ground caused by logistics drones and cargo falling to the ground; classifying drones and cargo, and modeling economic losses, including direct economic losses caused by loss of drone and cargo value and costs by social service agencies indirect economic loss due to consumption; integrating the number of deaths and economic losses on the ground to establish a risk assessment matrix. The calculation example indicates that the number of deaths on the ground is between 3.74×10- 12 and 1.87×10- 7 per flight hour, which is in line with the equivalent safety level; the economic loss is between 6734.8 and 33619 yuan per accident, which is also within a reasonable range. The method provided in this paper can provide some reference and guiding significance for the ground risk assessment and control of logistics drones.
    References | Related Articles | Metrics
    Airplane Boarding Efficiency Analysis Considering Passenger Overtaking Behaviors
    MA Jian , LIAO Wei-yi , CHEN Juan , WANG Qiao , TANG Tie-qiao , YUAN Zhi-lu
    2022, 22(4): 255-267.  DOI: 10.16097/j.cnki.1009-6744.2022.04.029
    Abstract ( )   PDF (3353KB) ( )  
    Improving boarding efficiency by reducing boarding time is one of the most important methods to reduce the airline turnaround costs. Passenger boarding time is determined by a combination of boarding strategy and individual behavior. When boarding with a specific boarding strategy, if blocking occurs and the queue moves slowly, passengers in the queue may adopt overtaking behavior to increase their own speed, which then affects boarding efficiency. Considering the passenger overtaking behavior, this paper proposes a microscopic simulation model for the boarding efficiency analysis. With the proposed cellular automata model, simulation of the boarding process under different boarding strategies have been performed to compare the impact of overtaking behavior on the efficiency of different boarding strategies, and to quantify the efficiency of boarding strategies under the scenarios with varying passenger luggage ratio, boarding release interval, and passenger load factor when considering passenger overtaking behavior. The results show that after considering the overtaking behavior, the longer the boarding time strategy when the overtaking behavior is not originally considered, the higher the percentage of boarding time reduction, and the reduction in passenger walking interference delays is the main reason for the reduction in boarding time. In the scenario with large number of passengers with luggage or the passenger release interval is short, the strategy of boarding sequentially from the window to the aisle is more efficient than the Reverse Pyramid strategy. In the scenario where the occupancy rate is 70%, the strategy of boarding in four sequences from back to front is the best strategy for boarding in column order.
    References | Related Articles | Metrics
    Robust Model Predictive Control Method for Connected Vehicle Platoon
    ZHANG Pei-yu, ZHOU Jian-shan, TIAN Da-xin
    2022, 22(4): 268-274.  DOI: 10.16097/j.cnki.1009-6744.2022.04.030
    Abstract ( )   PDF (1851KB) ( )  
    This paper focuses on the connected vehicle platoon control problem with structural uncertain disturbances and proposes a model predictive platoon control method based on robust equivalence transformation. First, the traditional vehicle platoon control model is described based on the vehicle kinematics state equation. Then, the affine structure uncertainty matrix is introduced into the state equation to develop the vehicle platoon control model under the minimum maximization paradigm. Based on the robust equivalence theory, the model is transformed into a computationally tractable epigraph optimization model under box uncertainty set. The sequential quadratic programming algorithm is then used to obtain the optimal platoon control scheme, and the control scheme is compared with the traditional vehicle platoon control scheme through simulation experiments to verify the effectiveness of the model. The results indicate that the proposed optimal platoon control method can resist the uncertain interference of system structureand ensure the vehiclesafety while achieving the goal of the stable platoon.
    References | Related Articles | Metrics
    Capacity Analysis Method of Mixed Flow with Connected and Automated Truck Platooning
    QIN Yan-yan , ZHU Yi-wen, ZHU Li, TANG Hong-hui
    2022, 22(4): 275-282.  DOI: 10.16097/j.cnki.1009-6744.2022.04.031
    Abstract ( )   PDF (1724KB) ( )   PDF(English version) (506KB) ( 44 )  
    Connected and automated truck platooning is expected to be one of the first scenarios for the application of connected and automated driving. This paper studies the traffic capacity of mixed traffic flow of connected and automated truck platooning, where the random mixed traffic flow is composed of connected and automated trucks, manual trucks, and cars. Firstly, this paper analyzes 10 types of car- following behavior in the mixed traffic flow considering the spatial distribution characteristics of the scale of connected and automated truck platooning, develops its probability expression, and then constructs a general capacity analysis method of the mixed traffic flow. Then, considering the randomness of truck distribution in the actual traffic flow operation, the mixed traffic flow of connected and automated truck platooning is divided into three situations: dominant flow, random flow, and inferior flow to improve the universality of the mixed traffic flow capacity analysis method. Finally, the connected and automated truckfollowing model calibrated by the measured data is selected for case analysis to verify the effectiveness of the theoretical analysis method. The results show that the increase of the proportion of connected and automated trucks, or the increase of its platooning size, are conducive to the reduction of vehicle conversion coefficient and relative entropy in the mixed traffic flow of the three situations, which can effectively improve the traffic capacity of the mixed traffic flow. Under different conditions such as the proportion of connected and automated trucks, the optimal platooning size of connected and automated truck platooning randomly distributed is 2~4 vehicles. At the same time, the trafficcapacity of three mixed traffic flows, i.e., dominant flow, random flow, and inferior flow, decreases in turn. The research results reveal the internal mechanism of improving the capacity of mixed traffic flow of connected and automated truck platooning and provide methodological support for the operation and management of intelligent network truck platooning in the future.
    References | Related Articles | Metrics
    Multi-drivers Risk Evaluation Based Proactive Intervention of Drivers' Risky Behavior Under Connected Transportation Contexts
    BAO Qiong, QU Qi-kai, TANG Han-run, CHEN Jian-ming, SHEN Yong-jun
    2022, 22(4): 283-292.  DOI: 10.16097/j.cnki.1009-6744.2022.04.032
    Abstract ( )   PDF (2251KB) ( )  
    A proactive intervention framework based on a comprehensive evaluation of risky driving behavior among multiple drivers under intelligent and connected transportation contexts is proposed in this study. By selecting driving behavior parameters, different types of drivers' risky behavior are defined and identified based on their driving data. Then, an area method is utilized to obtain an integrated score considering the frequency, duration, and amplitude of each risky driving behavior, and the relationship between this score and crash risk is established by using variable weights. Based on the mechanism of data envelopment analysis (DEA), a modelling approach for risky driving behavior evaluation is proposed. Next, a microsimulation scenario based on Simulation of Urban Mobility(SUMO) is built to simulate the intelligent and connected transportation environment, and the historical driving data of multiple drivers are extracted by a time window. The comprehensive evaluation method of risky driving behavior is applied to identify risky drivers. Two intervention methods based on a cumulative window approach and a sliding window approach are proposed, respectively, and each method implements two strategies, i.e., single-vehicle intervention and multi-vehicle intervention. Finally, based on the micro-simulation experiment, the effect of different intervention methods is analyzed, and the impacts of window size, driver acceptance rate, and the number of intervention vehiclesare discussed. The results showed that the strategy of multi-vehicle intervention has achieved better results. The total scores of drivers' risky behavior under the cumulative window approach and the sliding window method are decreased by 22.80% and 10.50% , respectively. Compared with the situation without intervention, when the intervention acceptance rate is 50% , the total score of drivers' risky behavior in the scenario still decreases. Relative to the cumulative window approach, the sliding window approach appears to be a more reasonable way for practical application. The proposed framework can provide technical support for driving behavior monitoring.
    References | Related Articles | Metrics
    Motion Planning Method for Intelligent Vehicles Under Constrained Dynamic and Static Environment
    QI Yao, ZHU Yan-qi, LI Yong-le , XU You-chun
    2022, 22(4): 293-301.  DOI: 10.16097/j.cnki.1009-6744.2022.04.033
    Abstract ( )   PDF (2919KB) ( )  
    Focusing on intelligent vehicle motion planning in an environment with dynamic and static obstacles, a search-based method in 3D space is proposed. The method establishes the spatiotemporal grid by adding a time dimension to the Cartesian coordinate system and constructs vehicle motion primitives at different speeds. And an intelligent vehicle is transformed into a line segment composed of multiple circle centers. The method then uses the expanding static obstacle method for rapid collision detection of irregular obstacles and simplifies dynamic obstacle collision detection to line segment intersection detection by the time interval method. For guiding the search tree to reach the target position and velocity quickly in spatiotemporal space, a velocity heuristic function constrained by maximum velocity and acceleration is constructed. Finally, the motion trajectory of integrating time information and “stop-wait”decision information is obtained by searching in the spatiotemporal grid, which is based on a heuristic method and partial motion planning. Experimental results show that the proposed motion planning method can guide the intelligent vehicle to drive safely in a constrained dynamic and static environment. The average success rate of safe driving is increased by 23% compared with velocity obstacles. Compared with hybrid state A*, the average success rate is increased by 19%, and the total driving time is reduced by 21%.
    References | Related Articles | Metrics
    Layout Optimization of Smart Intersections Under Novel Mixed Traffic Flow
    LI Tong-fei, CAO Ya-ning, DOU Xue-ping , XIONG Jie, XU Yan, ZHOU Wen-han
    2022, 22(4): 302-312.  DOI: 10.16097/j.cnki.1009-6744.2022.04.034
    Abstract ( )   PDF (2056KB) ( )  
    Due to the advanced Internet of vehicles and autonomous driving technology, connected autonomous vehicles (CAVs) can adopt a new mode of traffic organization (i.e., smart intersections) to significantly improve the efficiency of intersections. To reduce the total travel cost of urban traffic systems under mixed CAVs and human-driven vehicles (HVs), the layout optimization problem of smart intersections in the urban traffic network is proposed. A mathematical optimization model is established and used to solve the problem. First, based on the analysis of the driving characteristics of the two types of vehicles, a mixed user equilibrium model is established to formulate the path choice behavior of CAVs and HVs. Second, from the perspective of traffic planners, with the integration of a mixed user equilibrium model, a spatial layout optimization model of smart intersections is established for the urban traffic network under the novel mixed traffic flow. It takes system optimization as the optimization objective and is solved by an improved genetic algorithm. Finally, a set of numerical experiments based on the Sioux-Falls network is conducted to verify the validity of the model and algorithm. Besides, the influence of the penetration rate of CAVs on the optimization results is also analyzed. The results show that the rational planning of smart intersections in the urban traffic network can significantly improve the travel efficiency in the novel mixed traffic scenario. Moreover, it greatly reduces the gaps in travel efficiency between CAVs and conventional HVs due to CAVs' unilaterally technological advantages, which further improves traffic fairness.
    References | Related Articles | Metrics