International Journal of Transportation Science and Technology最新文献_第9页

Unravelling route choices of large trucks using trajectory clustering and conditional Logit models 利用轨迹聚类和条件 Logit 模型解读大型卡车的路线选择

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.04.007

Yue Ma, Jan-Dirk Schmöcker, Wenzhe Sun, Satoshi Nakao

The mobility of sizable trucks is often limited by their large size. They thus may have additional requirements on road types, road widths, and the turning radius at the intersection when travelling. Therefore, this study explores the unique needs and preferences of large truck drivers’ route choice with a focus on trip and road network characteristics. Global positioning system (GPS) trajectory data from the central Kansai area of Japan with numerous ports and freight terminals are used. Trajectories are considered to have the same origin (destination) if their starting (ending) coordinates are in the same 500 m × 500 m mesh. For the trajectories of the same pair of origin-destination (OD) meshes, several route clusters are obtained based on geographical configuration using a QuickBundles algorithm. Sampling techniques are employed to equalize the number of input points for each vehicle trajectory and the optimal number of clusters is determined automatically by our algorithm based on the silhouette coefficient. By taking the clusters as route choice options for an OD pair, a conditional logit model is used to identify the factors that influence the route choice considering both vehicle- and trip-specific attributes. The results quantify the preference of trucks for wider roads and toll routes, as well as aversion to long distances and turns. The heterogeneity in route choice based on vehicle type, trip time (date), and trip purpose is also evident. The findings of this study can provide insights for freight road network design and optimization.

大型卡车的机动性往往受到其大尺寸的限制。因此，他们可能对道路类型、道路宽度和行驶时十字路口的转弯半径有额外的要求。因此，本研究以行程和路网特征为重点，探讨了大型货车驾驶员路线选择的独特需求和偏好。使用的是日本关西中部地区的全球定位系统（GPS）轨迹数据，该地区有许多港口和货运码头。如果轨迹的起始（结束）坐标在相同的500米× 500米网格中，则认为轨迹具有相同的原点（目的地）。对于同一对OD （origin-destination）网格的轨迹，采用QuickBundles算法基于地理配置获得多个路由簇。采用采样技术均衡每条车辆轨迹的输入点数量，并根据轮廓系数自动确定最优簇数。通过将集群作为OD对的路径选择选项，使用条件logit模型来识别考虑车辆和行程特定属性的影响路径选择的因素。结果量化了卡车对较宽道路和收费路线的偏好，以及对长距离和转弯的厌恶。基于车辆类型、出行时间（日期）和出行目的的路径选择异质性也很明显。研究结果可为货运路网设计与优化提供参考。

{"title":"Unravelling route choices of large trucks using trajectory clustering and conditional Logit models","authors":"Yue Ma, Jan-Dirk Schmöcker, Wenzhe Sun, Satoshi Nakao","doi":"10.1016/j.ijtst.2024.04.007","DOIUrl":"10.1016/j.ijtst.2024.04.007","url":null,"abstract":"<div><div>The mobility of sizable trucks is often limited by their large size. They thus may have additional requirements on road types, road widths, and the turning radius at the intersection when travelling. Therefore, this study explores the unique needs and preferences of large truck drivers’ route choice with a focus on trip and road network characteristics. Global positioning system (GPS) trajectory data from the central Kansai area of Japan with numerous ports and freight terminals are used. Trajectories are considered to have the same origin (destination) if their starting (ending) coordinates are in the same 500 m × 500 m mesh. For the trajectories of the same pair of origin-destination (OD) meshes, several route clusters are obtained based on geographical configuration using a QuickBundles algorithm. Sampling techniques are employed to equalize the number of input points for each vehicle trajectory and the optimal number of clusters is determined automatically by our algorithm based on the silhouette coefficient. By taking the clusters as route choice options for an OD pair, a conditional logit model is used to identify the factors that influence the route choice considering both vehicle- and trip-specific attributes. The results quantify the preference of trucks for wider roads and toll routes, as well as aversion to long distances and turns. The heterogeneity in route choice based on vehicle type, trip time (date), and trip purpose is also evident. The findings of this study can provide insights for freight road network design and optimization.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 238-250"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140758354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A weakly-supervised deep learning model for end-to-end detection of airfield pavement distress 用于端到端检测机场路面状况的弱监督深度学习模型

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.02.010

Zefeng Tao, Hongren Gong, Liming Liu, Lin Cong, Haimei Liang

Accurate and timely surveying of airfield pavement distress is crucial for cost-effective airport maintenance. Deep learning (DL) approaches, leveraging advancements in computer science and image acquisition techniques, have become the mainstream for automated airfield pavement distress detection. However, fully-supervised DL methods require a large number of manually annotated ground truth labels to achieve high accuracy. To address the challenge of limited high-quality manual annotations, we propose a novel end-to-end distress detection model called class activation map informed weakly-supervised distress detection (WSDD-CAM ). Based on YOLOv5, WSDD-CAM consists of an efficient backbone, a classification branch, and a localization network. By utilizing class activation map (CAM) information, our model significantly reduces the need for manual annotations, automatically generating pseudo bounding boxes with a 71% overlap with the ground truth. To evaluate WSDD-CAM, we tested it on a self-made dataset and compared it with other weakly-supervised and fully-supervised models. The results show that our model achieves 49.2% mean average precision (mAP), outperforming other weakly-supervised methods and even approaching state-of-the-art fully-supervised methods. Additionally, ablation experiments confirm the effectiveness of our architecture design. In conclusion, our WSDD-CAM model offers a promising solution for airfield pavement distress detection, reducing manual annotation time while maintaining high accuracy. This efficient and effective approach can significantly contribute to cost-effective airport maintenance management.

准确、及时地测量机场路面破损情况对机场维修的经济效益至关重要。利用计算机科学和图像采集技术的进步，深度学习（DL）方法已经成为机场路面破损自动检测的主流。然而，完全监督的深度学习方法需要大量手动标注的地面真值标签来实现高精度。为了解决高质量手工标注有限的问题，我们提出了一种新的端到端遇险检测模型，称为类激活图通知弱监督遇险检测（WSDD-CAM）。基于YOLOv5的wsdl - cam由高效主干、分类分支和定位网络组成。通过使用类激活图（CAM）信息，我们的模型显著减少了对手动注释的需求，自动生成与地面事实重叠71%的伪边界框。为了评估WSDD-CAM，我们在一个自制数据集上对其进行了测试，并将其与其他弱监督和完全监督的模型进行了比较。结果表明，我们的模型达到49.2%的平均精度（mAP），优于其他弱监督方法，甚至接近最先进的全监督方法。此外，烧蚀实验也证实了结构设计的有效性。总之，我们的wsdl - cam模型为机场路面破损检测提供了一个很有前景的解决方案，减少了人工标注时间，同时保持了较高的准确性。这种高效率和有效的方法可以大大提高机场维修管理的成本效益。

{"title":"A weakly-supervised deep learning model for end-to-end detection of airfield pavement distress","authors":"Zefeng Tao, Hongren Gong, Liming Liu, Lin Cong, Haimei Liang","doi":"10.1016/j.ijtst.2024.02.010","DOIUrl":"10.1016/j.ijtst.2024.02.010","url":null,"abstract":"<div><div>Accurate and timely surveying of airfield pavement distress is crucial for cost-effective airport maintenance. Deep learning (DL) approaches, leveraging advancements in computer science and image acquisition techniques, have become the mainstream for automated airfield pavement distress detection. However, fully-supervised DL methods require a large number of manually annotated ground truth labels to achieve high accuracy. To address the challenge of limited high-quality manual annotations, we propose a novel end-to-end distress detection model called class activation map informed weakly-supervised distress detection (WSDD-CAM ). Based on YOLOv5, WSDD-CAM consists of an efficient backbone, a classification branch, and a localization network. By utilizing class activation map (CAM) information, our model significantly reduces the need for manual annotations, automatically generating pseudo bounding boxes with a 71% overlap with the ground truth. To evaluate WSDD-CAM, we tested it on a self-made dataset and compared it with other weakly-supervised and fully-supervised models. The results show that our model achieves 49.2% mean average precision (mAP), outperforming other weakly-supervised methods and even approaching state-of-the-art fully-supervised methods. Additionally, ablation experiments confirm the effectiveness of our architecture design. In conclusion, our WSDD-CAM model offers a promising solution for airfield pavement distress detection, reducing manual annotation time while maintaining high accuracy. This efficient and effective approach can significantly contribute to cost-effective airport maintenance management.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 67-78"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140274711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Measurement of public acceptance of TDM policies using public policy acceptance (PPA) and value belief norm (VBN) combination approach 利用公共政策接受度（PPA）和价值信念规范（VBN）相结合的方法衡量公众对 TDM 政策的接受程度

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.02.006

Nindyo Cahyo Kresnanto

One strategy to overcome the problem of imbalance between the demand and supply of road transportation that causes congestion is to apply transport demand management (TDM). TDM is a series of transportation policies aimed at achieving sustainable transportation by reducing the use of private vehicles and prioritizing public transport and/or non-motorized vehicles. The level of public acceptance of TDM largely determines the success of TDM implementation. Through the value belief norm (VBN) theory approach, it can be seen that public acceptance of TDM policies will be influenced by how high the norm of community partiality towards the environment. The level of public acceptance of a TDM regulation can also be measured by the public policy acceptance (PPA) model. The results show that the acceptance of TDM strategy implementation was significantly influenced by the pro-environment attitude of the community. The PPA model result showed that people tend to be skeptical of the implementation of TDM policies.

克服道路运输供需不平衡导致交通拥挤的一种策略是应用运输需求管理（TDM）。TDM是一系列交通政策，旨在通过减少私人车辆的使用和优先考虑公共交通和/或非机动车辆来实现可持续交通。公众对TDM的接受程度在很大程度上决定了TDM实施的成功程度。通过价值信仰规范（VBN）理论方法可以看出，社区环境偏好规范的高低会影响公众对TDM政策的接受程度。公众对TDM监管的接受程度也可以通过公共政策接受（PPA）模型来衡量。结果表明，社区亲环境态度对TDM战略实施的接受程度有显著影响。PPA模型结果显示，人们对TDM政策的实施往往持怀疑态度。

引用次数: 0

Examining causal factors of traffic conflicts at intersections using vehicle trajectory data 利用车辆轨迹数据研究交叉路口交通冲突的成因

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.02.011

Xiaoyan Xu , Xuesong Wang , Ruolin Shi

Conflict severity results from the complex interactions between the roadway and environmental characteristics and the vehicle motion. Understanding how and to what extent a vehicle is influenced by roadway and surrounding road users during a conflict is helpful in analyzing the causal mechanisms of collisions, thus providing insights into roadway safety improvement countermeasures. This study utilized the NGSIM vehicle trajectory datasets to investigate the causal factors in conflicts at intersections by analyzing roadway-to-vehicle and vehicle-to-vehicle interactions. In order to remove the outliers and white noise existing in the raw data, vehicle trajectories were reconstructed by applying discrete wavelet transform and Kalman filtering (KF). Generalized time-to-collision was adopted to detect and measure the severity of conflicts, by which 1127 conflict events were extracted. Path analysis (PA) models were then established to determine in exactly which ways the roadway-to-vehicle and vehicle-to-vehicle interactions were related to conflict severity. Various roadway and environmental characteristics such as traffic flow average speed, percentage of trucks, and intersection skew angle were included in the models. The results indicate that the roadway and environmental characteristics have both direct and indirect effects on conflict severity. In the indirect effects, the kinematics of conflicting vehicles such as the average and standard deviation of speed, plays an intermediate role in linking roadway factors and conflict outcome. The framework of this study can be used to assess roadway readiness for both human-driven and automated vehicles.

冲突严重程度是道路、环境特征与车辆运动之间复杂相互作用的结果。了解车辆在冲突中如何以及在多大程度上受到道路和周围道路使用者的影响，有助于分析碰撞的因果机制，从而为道路安全改善对策提供见解。本研究利用NGSIM车辆轨迹数据集，通过分析道路与车辆以及车辆与车辆之间的相互作用，来调查十字路口冲突的原因。为了去除原始数据中存在的异常值和白噪声，采用离散小波变换和卡尔曼滤波（KF）对车辆轨迹进行重构。采用广义碰撞时间方法检测和度量冲突的严重程度，提取了1127个冲突事件。然后建立路径分析（PA）模型，以准确确定道路与车辆以及车辆与车辆之间的相互作用与冲突严重程度之间的关系。模型中包含了各种道路和环境特征，如交通流平均速度、卡车百分比和十字路口倾斜角。结果表明，道路和环境特征对冲突严重程度既有直接影响，也有间接影响。在间接影响中，冲突车辆的运动学，如速度的平均值和标准差，在连接道路因素和冲突结果方面起着中间作用。本研究的框架可用于评估人类驾驶和自动驾驶车辆的道路准备情况。

{"title":"Examining causal factors of traffic conflicts at intersections using vehicle trajectory data","authors":"Xiaoyan Xu , Xuesong Wang , Ruolin Shi","doi":"10.1016/j.ijtst.2024.02.011","DOIUrl":"10.1016/j.ijtst.2024.02.011","url":null,"abstract":"<div><div>Conflict severity results from the complex interactions between the roadway and environmental characteristics and the vehicle motion. Understanding how and to what extent a vehicle is influenced by roadway and surrounding road users during a conflict is helpful in analyzing the causal mechanisms of collisions, thus providing insights into roadway safety improvement countermeasures. This study utilized the NGSIM vehicle trajectory datasets to investigate the causal factors in conflicts at intersections by analyzing roadway-to-vehicle and vehicle-to-vehicle interactions. In order to remove the outliers and white noise existing in the raw data, vehicle trajectories were reconstructed by applying discrete wavelet transform and Kalman filtering (KF). Generalized time-to-collision was adopted to detect and measure the severity of conflicts, by which 1127 conflict events were extracted. Path analysis (PA) models were then established to determine in exactly which ways the roadway-to-vehicle and vehicle-to-vehicle interactions were related to conflict severity. Various roadway and environmental characteristics such as traffic flow average speed, percentage of trucks, and intersection skew angle were included in the models. The results indicate that the roadway and environmental characteristics have both direct and indirect effects on conflict severity. In the indirect effects, the kinematics of conflicting vehicles such as the average and standard deviation of speed, plays an intermediate role in linking roadway factors and conflict outcome. The framework of this study can be used to assess roadway readiness for both human-driven and automated vehicles.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 79-94"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140470257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quantifying the impacts of right-turn-on-red, exclusive turn lanes, and pedestrian movements on the efficiency of urban transportation networks 量化红灯右转、专用转弯车道和行人通行对城市交通网络效率的影响

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.02.007

Hao Liu, Zecheng Xiong, Vikash V. Gayah

Previous studies demonstrated that restricting left turning movements can enhance transportation network efficiency. However, this strategy can lead to significant increases in the volume of right-turn movements. While these right-turn movements do not conflict with opposing through traffic, they still must interact with pedestrians in adjacent crosswalks. Further, their movement is influenced by the presence of right-turn-on-red (RTOR), which is commonly applied at signalized intersections to improve intersection capacity, and the presence of exclusive right-turn lanes. This paper examines the influence of these three factors (pedestrian activity, RTOR, and exclusive right-turn lanes) on vehicular operational performance at a network-wide level. Simple grid network structures are considered due to their generalizability and the performances of three network types are tested: two-way streets that accommodate left turns, two-way streets that prohibit left turns, and one-way streets. The results reveal that when there are no pedestrians, right-ROTR can improve the operational performance regardless of the existence of exclusive lanes, especially for the networks restricting left-turn movements, and the presence of exclusive turn lanes increases the benefits obtained by allowing RTPR. The results also suggest that allowing RTPR is more important than providing exclusive lanes when the traffic load is light; however, under heavier traffic, exclusive turn lanes become more important. The presence of pedestrians reduces overall network performance and the benefits provided by RTPR for most scenarios, as expected. This decrease in performance is larger for networks made up of two-way streets compared to those made up of one-way streets. Exclusive lanes are also found to be critical for two-way streets with left turns protected to maintain network efficiency. Overall, prohibiting left turns on two-way streets still provides the largest operational performance of all networks with these features considered.

以往的研究表明，限制左转可以提高交通网络的效率。然而，这种策略可能导致右转动作的数量显著增加。虽然这些右转动作不会与对面的交通发生冲突，但它们仍然必须与相邻人行横道上的行人相互作用。此外，它们的运动受到右转红灯（RTOR）的影响，RTOR通常用于信号交叉口以提高交叉口的通行能力，并且存在专用右转车道。本文在网络层面上考察了这三个因素（行人活动、RTOR和专用右转车道）对车辆运行性能的影响。考虑简单的网格网络结构，因为它们具有普遍性，并且测试了三种网络类型的性能：允许左转的双向街道，禁止左转的双向街道和单行道。研究结果表明，在没有行人的情况下，无论是否存在专用车道，右转右转都能提高交通网络的运行性能，特别是对于限制左转运动的网络，而专用车道的存在增加了允许右转右转的效益。结果还表明，在交通负荷较轻时，允许RTPR比提供专用车道更重要；然而，在交通繁忙的情况下，专用转弯车道变得更加重要。正如预期的那样，行人的存在降低了整体网络性能和RTPR在大多数情况下提供的好处。与由单行道组成的网络相比，由双行道组成的网络的性能下降幅度更大。对于保护左转弯以保持网络效率的双行道来说，专用车道也至关重要。总的来说，在考虑到这些特征后，禁止在双向街道上左转仍然提供了所有网络中最大的运行性能。

{"title":"Quantifying the impacts of right-turn-on-red, exclusive turn lanes, and pedestrian movements on the efficiency of urban transportation networks","authors":"Hao Liu, Zecheng Xiong, Vikash V. Gayah","doi":"10.1016/j.ijtst.2024.02.007","DOIUrl":"10.1016/j.ijtst.2024.02.007","url":null,"abstract":"<div><div>Previous studies demonstrated that restricting left turning movements can enhance transportation network efficiency. However, this strategy can lead to significant increases in the volume of right-turn movements. While these right-turn movements do not conflict with opposing through traffic, they still must interact with pedestrians in adjacent crosswalks. Further, their movement is influenced by the presence of right-turn-on-red (RTOR), which is commonly applied at signalized intersections to improve intersection capacity, and the presence of exclusive right-turn lanes. This paper examines the influence of these three factors (pedestrian activity, RTOR, and exclusive right-turn lanes) on vehicular operational performance at a network-wide level. Simple grid network structures are considered due to their generalizability and the performances of three network types are tested: two-way streets that accommodate left turns, two-way streets that prohibit left turns, and one-way streets. The results reveal that when there are no pedestrians, right-ROTR can improve the operational performance regardless of the existence of exclusive lanes, especially for the networks restricting left-turn movements, and the presence of exclusive turn lanes increases the benefits obtained by allowing RTPR. The results also suggest that allowing RTPR is more important than providing exclusive lanes when the traffic load is light; however, under heavier traffic, exclusive turn lanes become more important. The presence of pedestrians reduces overall network performance and the benefits provided by RTPR for most scenarios, as expected. This decrease in performance is larger for networks made up of two-way streets compared to those made up of one-way streets. Exclusive lanes are also found to be critical for two-way streets with left turns protected to maintain network efficiency. Overall, prohibiting left turns on two-way streets still provides the largest operational performance of all networks with these features considered.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 52-66"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139828623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-informed deep learning with Kalman filter mixture for traffic state prediction 利用卡尔曼滤波混合物进行交通状态预测的物理信息深度学习

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.04.002

Niharika Deshpande, Hyoshin (John) Park

Accurate traffic forecasting is crucial for understanding and managing congestion for efficient transportation planning. However, conventional approaches often neglect epistemic uncertainty, which arises from incomplete knowledge across different spatiotemporal scales. This study addresses this challenge by introducing a novel methodology to establish dynamic spatiotemporal correlations that captures the unobserved heterogeneity in travel time through distinct peaks in probability density functions, guided by physics-based principles. We propose an innovative approach to modifying both prediction and correction steps of the Kalman filter (KF) algorithm by leveraging established spatiotemporal correlations. Central to our approach is the development of a novel deep learning (DL) model called the physics informed-graph convolutional gated recurrent neural network (PI-GRNN). Functioning as the state-space model within the KF, the PI-GRNN exploits established correlations to construct dynamic adjacency matrices that utilize the inherent structure and relationships within the transportation network to capture sequential patterns and dependencies over time. Furthermore, our methodology integrates insights gained from correlations into the correction step of the KF algorithm that helps in enhancing its correctional capabilities. This integrated approach proves instrumental in alleviating the inherent model drift associated with data-driven methods, as periodic corrections through update step of KF refine the predictions generated by the PI-GRNN. To the best of our knowledge, this study represents a pioneering effort in integrating DL and KF algorithms in this unique symbiotic manner. Through extensive experimentation with real-world traffic data, we demonstrate the superior performance of our model compared to the benchmark approaches.

准确的交通预测对于理解和管理拥堵、制定有效的交通规划至关重要。然而，传统的方法往往忽略了认知的不确定性，这种不确定性源于不同时空尺度上的不完整知识。本研究通过引入一种新的方法来解决这一挑战，该方法通过基于物理原理的概率密度函数的不同峰值来捕获旅行时间中未观察到的异质性，从而建立动态时空相关性。我们提出了一种创新的方法，通过利用已建立的时空相关性来修改卡尔曼滤波（KF）算法的预测和校正步骤。我们方法的核心是开发一种新的深度学习（DL）模型，称为物理通知图卷积门控递归神经网络（PI-GRNN）。作为KF中的状态空间模型，PI-GRNN利用已建立的相关性来构建动态邻接矩阵，该矩阵利用运输网络中的固有结构和关系来捕获随时间变化的顺序模式和依赖关系。此外，我们的方法将从相关性中获得的见解整合到KF算法的校正步骤中，有助于增强其校正能力。这种集成方法有助于缓解与数据驱动方法相关的固有模型漂移，因为通过KF的更新步骤进行周期性修正可以改进PI-GRNN生成的预测。据我们所知，这项研究代表了以这种独特的共生方式整合DL和KF算法的开创性努力。通过对真实世界流量数据的大量实验，我们证明了与基准方法相比，我们的模型具有优越的性能。

{"title":"Physics-informed deep learning with Kalman filter mixture for traffic state prediction","authors":"Niharika Deshpande, Hyoshin (John) Park","doi":"10.1016/j.ijtst.2024.04.002","DOIUrl":"10.1016/j.ijtst.2024.04.002","url":null,"abstract":"<div><div>Accurate traffic forecasting is crucial for understanding and managing congestion for efficient transportation planning. However, conventional approaches often neglect epistemic uncertainty, which arises from incomplete knowledge across different spatiotemporal scales. This study addresses this challenge by introducing a novel methodology to establish dynamic spatiotemporal correlations that captures the unobserved heterogeneity in travel time through distinct peaks in probability density functions, guided by physics-based principles. We propose an innovative approach to modifying both prediction and correction steps of the Kalman filter (KF) algorithm by leveraging established spatiotemporal correlations. Central to our approach is the development of a novel deep learning (DL) model called the physics informed-graph convolutional gated recurrent neural network (PI-GRNN). Functioning as the state-space model within the KF, the PI-GRNN exploits established correlations to construct dynamic adjacency matrices that utilize the inherent structure and relationships within the transportation network to capture sequential patterns and dependencies over time. Furthermore, our methodology integrates insights gained from correlations into the correction step of the KF algorithm that helps in enhancing its correctional capabilities. This integrated approach proves instrumental in alleviating the inherent model drift associated with data-driven methods, as periodic corrections through update step of KF refine the predictions generated by the PI-GRNN. To the best of our knowledge, this study represents a pioneering effort in integrating DL and KF algorithms in this unique symbiotic manner. Through extensive experimentation with real-world traffic data, we demonstrate the superior performance of our model compared to the benchmark approaches.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 161-174"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140795854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The economic burden of road traffic accidents and injuries: A small island perspective 道路交通事故和伤害的经济负担：小岛屿视角

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.03.002

Verena Tandrayen-Ragoobur

The existing evidence on the economic burden of road accidents on gross domestic product (GDP) focused mainly on developed countries. This paper addresses an important gap in the literature by investigating into the impacts of road traffic accidents and injuries on GDP in the small island of Mauritius. Mauritius, having perceived an important structural transformation over the past decades, is witnessing a rise in road accidents and injuries, which is a concern in terms of economic costs. In addition, it is a small island nation with limited road infrastructure, making it vulnerable to traffic congestion and accidents. The paper provides important insights for other island countries with similar geographic challenges. The Vector error correction model (VECM) approach is used to assessing the existence of a long-run relationship between road traffic accidents and GDP in Mauritius from 1980 to 2020. In addition to road crashes, different levels of injury severity linked to road accidents are evaluated. The results reveal that on average a 1% rise in road accidents leads to a 0.42% fall in real GDP. Further, a 1% rise in casualties linked to road accidents is likely to cause a 0.18% decline in GDP. Although a long-run relationship is established, there is no statistically significant influence of road accidents on GDP in the short-run.

关于道路事故对国内生产总值的经济负担的现有证据主要集中在发达国家。本文通过调查道路交通事故和伤害对毛里求斯小岛国内生产总值的影响，解决了文献中的一个重要差距。毛里求斯在过去几十年中经历了重要的结构转变，但它正在目睹道路事故和伤害的增加，这在经济成本方面令人关切。此外，它是一个小岛屿国家，道路基础设施有限，使其容易受到交通拥堵和事故的影响。本文为其他面临类似地理挑战的岛国提供了重要的见解。矢量误差修正模型（VECM）方法用于评估毛里求斯1980年至2020年道路交通事故与GDP之间存在的长期关系。除了道路交通事故外，还评估了与道路交通事故有关的不同程度的伤害严重程度。结果显示，道路交通事故平均上升1%，实际GDP就会下降0.42%。此外，与交通事故相关的伤亡人数每增加1%，就可能导致GDP下降0.18%。虽然建立了长期的关系，但在短期内，道路交通事故对GDP的影响在统计上并不显著。

{"title":"The economic burden of road traffic accidents and injuries: A small island perspective","authors":"Verena Tandrayen-Ragoobur","doi":"10.1016/j.ijtst.2024.03.002","DOIUrl":"10.1016/j.ijtst.2024.03.002","url":null,"abstract":"<div><div>The existing evidence on the economic burden of road accidents on gross domestic product (GDP) focused mainly on developed countries. This paper addresses an important gap in the literature by investigating into the impacts of road traffic accidents and injuries on GDP in the small island of Mauritius. Mauritius, having perceived an important structural transformation over the past decades, is witnessing a rise in road accidents and injuries, which is a concern in terms of economic costs. In addition, it is a small island nation with limited road infrastructure, making it vulnerable to traffic congestion and accidents. The paper provides important insights for other island countries with similar geographic challenges. The Vector error correction model (VECM) approach is used to assessing the existence of a long-run relationship between road traffic accidents and GDP in Mauritius from 1980 to 2020. In addition to road crashes, different levels of injury severity linked to road accidents are evaluated. The results reveal that on average a 1% rise in road accidents leads to a 0.42% fall in real GDP. Further, a 1% rise in casualties linked to road accidents is likely to cause a 0.18% decline in GDP. Although a long-run relationship is established, there is no statistically significant influence of road accidents on GDP in the short-run.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 109-119"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140273411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Investigating the contributing factors of crashes on interstate bridges in Louisiana using latent class clustering and association rule mining 利用潜类聚类和关联规则挖掘调查路易斯安那州州际桥梁上的碰撞诱因

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.04.011

M. Ashifur Rahman , Elisabeta Mitran , Julius Codjoe , Kofi K. Ampofo-Twumasi

Drivers on long interstate bridges often encounter unique challenges, including restricted lane widths, inadequate shoulders, and a lack of clear zones for safe recovery. Studies on understanding the factors that contribute to crash severity on such high-risk sections of interstates are limited. This research study applies latent class clustering (LCC) to detect homogeneous clusters while accounting for unobserved heterogeneity in a dataset of 10 036 crashes that occurred over a 6-year period (2015–2020) on eight selected bridges. Utilizing the LCC method, the research identifies four optimal clusters in bridge crashes, characterized by attributes such as ′4-lane′, ′6-lane′, ′single-vehicle crashes′, and ′unknown driver′. The association rule mining (ARM) approach is used to identify the important collective factors to visible injury (KAB – fatal, severe, and moderate) and property damage only (PDO or no injury). In Cluster 1 (4-lane), KAB and PDO crashes differ in collision type and visibility conditions, with rear-end crashes linked to KAB and sideswipe crashes to PDO. Cluster 2 (6-lane) shows similar distinctions but lacks specific lighting associations for PDO. In Cluster 3 (single-vehicle crashes), KAB involves moderate traffic and low visibility, while PDO has lower speed limits and non-dry surfaces. Cluster 4 (unknown driver), despite overrepresenting hit-and-run cases, underscores challenges in injury crash data collection in high-volume mobility scenarios. The discussions of the findings on the severity factors in this study are expected to help traffic safety engineers, policymakers, and planners to identify effective safety countermeasures on major elevated sections.

长州际桥梁上的司机经常遇到独特的挑战，包括车道宽度有限，肩部不足，以及缺乏安全恢复的清晰区域。在州际公路这样的高风险路段上，对导致撞车严重程度的因素的研究是有限的。本研究应用潜在类聚类（LCC）来检测同质聚类，同时考虑到在6年（2015-2020年）期间发生在8座选定桥梁上的10036起事故的数据集中未观察到的异质性。利用LCC方法，研究确定了桥梁碰撞的四个最优集群，其特征属性为“4车道”、“6车道”、“单车辆碰撞”和“未知驾驶员”。关联规则挖掘（ARM）方法用于识别可见伤害（KAB -致命，严重和中度）和仅财产损害（PDO或无伤害）的重要集体因素。在第1组（4车道）中，KAB和PDO碰撞在碰撞类型和能见度条件上有所不同，追尾事故与KAB有关，侧滑事故与PDO有关。集群2（6车道）显示出类似的区别，但缺乏特定的PDO照明关联。在集群3（单车辆碰撞）中，KAB涉及中度交通和低能见度，而PDO涉及较低的速度限制和非干燥路面。第4组（未知驾驶员），尽管肇事逃逸案例占比过高，但也凸显了在高容量机动场景中收集伤害碰撞数据的挑战。本研究结果对严重因素的讨论有望帮助交通安全工程师、政策制定者和规划者确定主要高架路段的有效安全对策。

{"title":"Investigating the contributing factors of crashes on interstate bridges in Louisiana using latent class clustering and association rule mining","authors":"M. Ashifur Rahman , Elisabeta Mitran , Julius Codjoe , Kofi K. Ampofo-Twumasi","doi":"10.1016/j.ijtst.2024.04.011","DOIUrl":"10.1016/j.ijtst.2024.04.011","url":null,"abstract":"<div><div>Drivers on long interstate bridges often encounter unique challenges, including restricted lane widths, inadequate shoulders, and a lack of clear zones for safe recovery. Studies on understanding the factors that contribute to crash severity on such high-risk sections of interstates are limited. This research study applies latent class clustering (LCC) to detect homogeneous clusters while accounting for unobserved heterogeneity in a dataset of 10 036 crashes that occurred over a 6-year period (2015–2020) on eight selected bridges. Utilizing the LCC method, the research identifies four optimal clusters in bridge crashes, characterized by attributes such as ′4-lane′, ′6-lane′, ′single-vehicle crashes′, and ′unknown driver′. The association rule mining (ARM) approach is used to identify the important collective factors to visible injury (KAB – fatal, severe, and moderate) and property damage only (PDO or no injury). In Cluster 1 (4-lane), KAB and PDO crashes differ in collision type and visibility conditions, with rear-end crashes linked to KAB and sideswipe crashes to PDO. Cluster 2 (6-lane) shows similar distinctions but lacks specific lighting associations for PDO. In Cluster 3 (single-vehicle crashes), KAB involves moderate traffic and low visibility, while PDO has lower speed limits and non-dry surfaces. Cluster 4 (unknown driver), despite overrepresenting hit-and-run cases, underscores challenges in injury crash data collection in high-volume mobility scenarios. The discussions of the findings on the severity factors in this study are expected to help traffic safety engineers, policymakers, and planners to identify effective safety countermeasures on major elevated sections.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 293-311"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141040553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How to detect occluded crosswalks in overview images? Comparing three methods in a heavily occluded area 如何检测概览图像中被遮挡的人行横道？比较严重遮挡区域中的三种方法

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.04.001

Yuanyuan Zhang , Joseph Luttrell IV , Chaoyang Zhang

Crosswalk presence data are crucial for pedestrian safety and urban planning. However, obtaining such data at a large scale is often challenging due to the high cost associated with traditional collection methods. While automated methods based on computer vision have been explored to detect crosswalks from aerial images, a major obstacle to their application is the handling of candidate crosswalks occluded by objects or shadows in the aerial imagery. To address this challenge, this study explores different deep learning-based solutions, including the aerial-view method (AVM) and street-view method (SVM), which are commonly used, and a combination of them, i.e., the dual-perspective method (DPM). Deep learning models based on convolutional neural networks (CNNs) with the VGG16 architecture were trained using 16 815 images to automatically detect crosswalks from both aerial and street view images. To compare the performance of these methods in handling occlusions, 1 378 images from a heavily occluded area were processed separately by the three methods. The results showed that the AVM suffered the most when dealing with images from a heavily occluded area, resulting in the lowest accuracy, precision, recall, and F1 score among the three methods. On the other hand, the SVM outperformed the AVM significantly. The DPM demonstrated the highest accuracy and precision values, indicating its superiority in accurately predicting the location of a crosswalk. However, the SVM exhibited the highest recall value, highlighting its superior ability to recover an occluded crosswalk among all methods.

人行横道存在数据对行人安全和城市规划至关重要。然而，由于与传统收集方法相关的高成本，大规模获取此类数据通常具有挑战性。虽然已经探索了基于计算机视觉的自动方法来检测航空图像中的人行横道，但其应用的主要障碍是处理航空图像中被物体或阴影遮挡的候选人行横道。为了应对这一挑战，本研究探索了不同的基于深度学习的解决方案，包括常用的鸟瞰图方法（AVM）和街景方法（SVM），以及它们的组合，即双视角方法（DPM）。基于VGG16架构的卷积神经网络（cnn）的深度学习模型使用16815张图像进行训练，自动检测航拍和街景图像中的人行横道。为了比较这三种方法处理遮挡的性能，分别对1 378幅重度遮挡区域的图像进行了处理。结果表明，AVM在处理严重遮挡区域的图像时受到的影响最大，导致三种方法的准确率、精密度、召回率和F1分数最低。另一方面，SVM的性能明显优于AVM。DPM具有最高的准确度和精度值，表明其在准确预测人行横道位置方面具有优势。然而，支持向量机的召回值最高，突出了其在所有方法中恢复闭塞人行横道的能力。

{"title":"How to detect occluded crosswalks in overview images? Comparing three methods in a heavily occluded area","authors":"Yuanyuan Zhang , Joseph Luttrell IV , Chaoyang Zhang","doi":"10.1016/j.ijtst.2024.04.001","DOIUrl":"10.1016/j.ijtst.2024.04.001","url":null,"abstract":"<div><div>Crosswalk presence data are crucial for pedestrian safety and urban planning. However, obtaining such data at a large scale is often challenging due to the high cost associated with traditional collection methods. While automated methods based on computer vision have been explored to detect crosswalks from aerial images, a major obstacle to their application is the handling of candidate crosswalks occluded by objects or shadows in the aerial imagery. To address this challenge, this study explores different deep learning-based solutions, including the aerial-view method (AVM) and street-view method (SVM), which are commonly used, and a combination of them, i.e., the dual-perspective method (DPM). Deep learning models based on convolutional neural networks (CNNs) with the VGG16 architecture were trained using 16 815 images to automatically detect crosswalks from both aerial and street view images. To compare the performance of these methods in handling occlusions, 1 378 images from a heavily occluded area were processed separately by the three methods. The results showed that the AVM suffered the most when dealing with images from a heavily occluded area, resulting in the lowest accuracy, precision, recall, and F1 score among the three methods. On the other hand, the SVM outperformed the AVM significantly. The DPM demonstrated the highest accuracy and precision values, indicating its superiority in accurately predicting the location of a crosswalk. However, the SVM exhibited the highest recall value, highlighting its superior ability to recover an occluded crosswalk among all methods.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 148-160"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140793411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimizing roads for sustainability: Inverted pavement design with life cycle cost analysis and carbon footprint estimation 优化道路，实现可持续性：采用生命周期成本分析和碳足迹估算的反向路面设计

IF 4.3 Q2 TRANSPORTATION

International Journal of Transportation Science and Technology

Pub Date : 2025-03-01 DOI: 10.1016/j.ijtst.2024.04.008

Uppuluri Siva Rama Krishna , Mohan Badiger , Yatin Chaudhary , Turumella Vijaya Gowri , Esamsetti Jahnavi Devi

Inverted pavements have proven performance across the world, and there is a need to optimize the layer thickness and material properties of the pavement addressing the critical failures in the mechanistic-empirical pavement design. The present study is made on bituminous concrete (BC) pavement with traffic up to 50 samples per second (MSA) and studying the critical factors affecting the pavement performance. The Minitab’s response surface methodology (RSM) − box behnken method, was used for the design of experiments which includes critical factors and responses obtained from ANSYS finite element modeling of the inverted pavement. The critical factors and responses are normally distributed and indicate a linear relationship with the least error. The composite desirability for minimum stress and strains in the pavement layers was found to be 0.89. The optimized pavement thickness and layer material properties were validated with two pavement field cross sections of different Indian national highways, and it is observed that the optimized cross-section is safe. Further, this research paper carried out life cycle cost analysis (LCCA) and life cycle assessment (LCA) of inverted pavement with optimized pavement cross-section obtained including the carbon footprint during the vehicle operation phase. The study demonstrated the benefits of inverted pavement with reduced costs and carbon emissions. Thus, this approach paves the way towards sustainable and long-lasting pavements.

倒立路面的性能已在世界范围内得到了验证，为了解决机械经验路面设计中的关键失效问题，有必要对倒立路面的层厚和材料性能进行优化。本研究以流量高达50个样本/秒（MSA）的沥青混凝土路面为研究对象，研究了影响路面性能的关键因素。采用Minitab的响应面法(RSM) - box behnken法设计试验，其中包括ANSYS有限元模拟的倒路面的关键因素和响应。关键因素和响应呈正态分布，并与最小误差呈线性关系。路面层内最小应力应变的复合理想值为0.89。利用印度不同国道的两个路面场截面对优化后的路面厚度和层材性能进行了验证，结果表明优化后的路面场截面是安全的。在此基础上，对倒置路面进行了全生命周期成本分析（LCCA）和全生命周期评价（LCA），得到了优化后的路面截面，包括车辆运行阶段的碳足迹。该研究证明了倒置路面在降低成本和碳排放方面的好处。因此，这种方法为可持续和持久的路面铺平了道路。

{"title":"Optimizing roads for sustainability: Inverted pavement design with life cycle cost analysis and carbon footprint estimation","authors":"Uppuluri Siva Rama Krishna , Mohan Badiger , Yatin Chaudhary , Turumella Vijaya Gowri , Esamsetti Jahnavi Devi","doi":"10.1016/j.ijtst.2024.04.008","DOIUrl":"10.1016/j.ijtst.2024.04.008","url":null,"abstract":"<div><div>Inverted pavements have proven performance across the world, and there is a need to optimize the layer thickness and material properties of the pavement addressing the critical failures in the mechanistic-empirical pavement design. The present study is made on bituminous concrete (BC) pavement with traffic up to 50 samples per second (MSA) and studying the critical factors affecting the pavement performance. The Minitab’s response surface methodology (RSM) − box behnken method, was used for the design of experiments which includes critical factors and responses obtained from ANSYS finite element modeling of the inverted pavement. The critical factors and responses are normally distributed and indicate a linear relationship with the least error. The composite desirability for minimum stress and strains in the pavement layers was found to be 0.89. The optimized pavement thickness and layer material properties were validated with two pavement field cross sections of different Indian national highways, and it is observed that the optimized cross-section is safe. Further, this research paper carried out life cycle cost analysis (LCCA) and life cycle assessment (LCA) of inverted pavement with optimized pavement cross-section obtained including the carbon footprint during the vehicle operation phase. The study demonstrated the benefits of inverted pavement with reduced costs and carbon emissions. Thus, this approach paves the way towards sustainable and long-lasting pavements.</div></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"17 ","pages":"Pages 251-275"},"PeriodicalIF":4.3,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140795730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0