Multimodal Transportation最新文献_第5页

STG-LSTM: Spatial-temporal graph-based long short-term memory for vehicle trajectory prediction STG-LSTM：基于时空图的车辆轨迹预测长短期记忆

Multimodal Transportation

Pub Date : 2025-03-14 DOI: 10.1016/j.multra.2025.100222

Daniela Daniel Ndunguru , Fan Xing , Chrispus Zacharia Oroni , Arsenyan Ani , Chao Li

Vehicle trajectory prediction plays a crucial role in enhancing the safety, efficiency, and effectiveness of intelligent transportation systems. Accurate predictions of future vehicle movements are essential for applications such as autonomous driving, traffic management, and collision avoidance systems. However, many existing methods either focus solely on spatial or temporal dimensions, neglecting the dynamic interactions between vehicles, which reduces prediction accuracy, especially in complex traffic scenarios. To address these limitations, the study proposes a Spatial-Temporal Graph-Based Long Short-Term Memory model, which integrates graph convolutional networks with long short-term memory networks to effectively capture both spatial relationships and temporal dependencies in vehicle trajectories. The proposed model employs a proximity-based method to construct dynamic adjacency matrices that represent real-time vehicle interactions. To capture spatial dependencies between vehicles, the study uses graph convolutional networks to model the relationships between neighboring vehicles. The long short-term memory network is then applied to capture temporal dynamics by learning the sequential dependencies in vehicle movement patterns. The output from the long short-term memory network is passed through a fully connected layer, which generates trajectory predictions for each vehicle. The study experimental results demonstrate that the proposed model outperforms existing state-of-the-art models across various prediction metrics. Specifically, at 3s and 4s prediction horizons, the model reduces the root mean square error by 22.4 % and 25.5 %, respectively, compared to the best performing interaction-aware long short-term memory model. At the 5s prediction horizon, the model achieves a significant root mean square error reduction of 26.6 %. These findings highlight the model's potential to improve safety and decision-making in autonomous driving systems and traffic management applications.

车辆轨迹预测在提高智能交通系统的安全性、效率和有效性方面起着至关重要的作用。准确预测未来车辆的运动对于自动驾驶、交通管理和防撞系统等应用至关重要。然而，现有的许多方法只关注空间或时间维度，忽略了车辆之间的动态相互作用，从而降低了预测的准确性，特别是在复杂的交通场景中。为了解决这些限制，该研究提出了一个基于时空图的长短期记忆模型，该模型将图卷积网络与长短期记忆网络集成在一起，以有效地捕捉车辆轨迹中的空间关系和时间依赖性。该模型采用基于接近度的方法构建动态邻接矩阵，表示实时车辆交互。为了捕捉车辆之间的空间依赖关系，该研究使用图形卷积网络对相邻车辆之间的关系进行建模。然后应用长短期记忆网络通过学习车辆运动模式中的顺序依赖关系来捕捉时间动态。长短期记忆网络的输出通过一个完全连接的层，该层为每辆车生成轨迹预测。研究实验结果表明，所提出的模型在各种预测指标上优于现有的最先进模型。具体来说，与表现最好的交互感知长短期记忆模型相比，在第3和第4个预测阶段，该模型分别将均方根误差降低了22.4%和25.5%。在5s的预测范围内，该模型实现了显著的均方根误差减小26.6%。这些发现突出了该模型在提高自动驾驶系统和交通管理应用的安全性和决策方面的潜力。

{"title":"STG-LSTM: Spatial-temporal graph-based long short-term memory for vehicle trajectory prediction","authors":"Daniela Daniel Ndunguru , Fan Xing , Chrispus Zacharia Oroni , Arsenyan Ani , Chao Li","doi":"10.1016/j.multra.2025.100222","DOIUrl":"10.1016/j.multra.2025.100222","url":null,"abstract":"<div><div>Vehicle trajectory prediction plays a crucial role in enhancing the safety, efficiency, and effectiveness of intelligent transportation systems. Accurate predictions of future vehicle movements are essential for applications such as autonomous driving, traffic management, and collision avoidance systems. However, many existing methods either focus solely on spatial or temporal dimensions, neglecting the dynamic interactions between vehicles, which reduces prediction accuracy, especially in complex traffic scenarios. To address these limitations, the study proposes a Spatial-Temporal Graph-Based Long Short-Term Memory model, which integrates graph convolutional networks with long short-term memory networks to effectively capture both spatial relationships and temporal dependencies in vehicle trajectories. The proposed model employs a proximity-based method to construct dynamic adjacency matrices that represent real-time vehicle interactions. To capture spatial dependencies between vehicles, the study uses graph convolutional networks to model the relationships between neighboring vehicles. The long short-term memory network is then applied to capture temporal dynamics by learning the sequential dependencies in vehicle movement patterns. The output from the long short-term memory network is passed through a fully connected layer, which generates trajectory predictions for each vehicle. The study experimental results demonstrate that the proposed model outperforms existing state-of-the-art models across various prediction metrics. Specifically, at 3s and 4s prediction horizons, the model reduces the root mean square error by 22.4 % and 25.5 %, respectively, compared to the best performing interaction-aware long short-term memory model. At the 5s prediction horizon, the model achieves a significant root mean square error reduction of 26.6 %. These findings highlight the model's potential to improve safety and decision-making in autonomous driving systems and traffic management applications.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 3","pages":"Article 100222"},"PeriodicalIF":0.0,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714709","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 learning-to-rank method to identify crash hotspots based on large-scale ride-hailing crash data 基于大规模网约车事故数据识别事故热点的学习排序方法

Multimodal Transportation

Pub Date : 2025-03-10 DOI: 10.1016/j.multra.2025.100219

Xiang Wen , Pengfei Cui , Yuanwei Luo , Runbo Hu , Yanyong Guo

Machine learning have been widely used in crash hotspot identification due to its superior prediction accuracy. Existing studies mainly treat hotspot identification as a classification or regression problem. This paper proposed a learning-to-rank(LTR) method to identify hotspots on a single trip and deviced a risk warning system based on the method to verify its effectiveness in crash mitigation. Ride-hailing crashes for a year in China were used as training and testing data. Three kinds of features were extracted to describe the safety level of each road segments, namely, road design features, time-related features, and traffic features. LambdaMART, a pairwise LTR algorism was applied to rank the road segments based on the extracted features. The experiment results suggested that the proposed LTR model outperforms three traditional machine learning models in terms of NDCG@10. The proposed LTR risk warning system integrated with Didi's ride-hailing service outperforms traditional zone-based warning system and bring a significant drop in Average Death Rate per Billion Kilometers.

机器学习以其优越的预测精度在碰撞热点识别中得到了广泛的应用。现有研究主要将热点识别作为分类或回归问题。本文提出了一种LTR （learning-to-rank）方法来识别单次行程的热点，并基于该方法设计了一个风险预警系统，验证了该方法在碰撞缓解中的有效性。中国一年的网约车事故被用作训练和测试数据。提取三种特征来描述每个路段的安全等级，即道路设计特征、时间相关特征和交通特征。基于提取的特征，采用双LTR算法LambdaMART对路段进行排序。实验结果表明，提出的LTR模型在NDCG@10方面优于三种传统的机器学习模型。与滴滴网约车服务相结合的LTR风险预警系统优于传统的基于区域的预警系统，并显著降低了每十亿公里的平均死亡率。

{"title":"A learning-to-rank method to identify crash hotspots based on large-scale ride-hailing crash data","authors":"Xiang Wen , Pengfei Cui , Yuanwei Luo , Runbo Hu , Yanyong Guo","doi":"10.1016/j.multra.2025.100219","DOIUrl":"10.1016/j.multra.2025.100219","url":null,"abstract":"<div><div>Machine learning have been widely used in crash hotspot identification due to its superior prediction accuracy. Existing studies mainly treat hotspot identification as a classification or regression problem. This paper proposed a learning-to-rank(LTR) method to identify hotspots on a single trip and deviced a risk warning system based on the method to verify its effectiveness in crash mitigation. Ride-hailing crashes for a year in China were used as training and testing data. Three kinds of features were extracted to describe the safety level of each road segments, namely, road design features, time-related features, and traffic features. LambdaMART, a pairwise LTR algorism was applied to rank the road segments based on the extracted features. The experiment results suggested that the proposed LTR model outperforms three traditional machine learning models in terms of NDCG@10. The proposed LTR risk warning system integrated with Didi's ride-hailing service outperforms traditional zone-based warning system and bring a significant drop in Average Death Rate per Billion Kilometers.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 3","pages":"Article 100219"},"PeriodicalIF":0.0,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143654639","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 life-saving impact of seatbelt usage: A random forest analysis of unobserved heterogeneity and latent risk factors in vehicular fatalities 量化安全带使用对生命的影响：车辆死亡中未观察到的异质性和潜在风险因素的随机森林分析

Multimodal Transportation

Pub Date : 2025-03-07 DOI: 10.1016/j.multra.2025.100221

Ittirit Mohamad

Seatbelt use significantly reduces the severity of injuries and fatalities in vehicular accidents. This study leverages the Random Forest algorithm to evaluate the impact of seatbelt usage on fatality probabilities in Thailand, with a novel focus on drivers who caused the accidents. The model demonstrated high accuracy, correctly identifying 95.10 % of non-fatal cases and 91.60 % of fatal cases, though some misclassifications were observed. A key contribution of this research is the identification of hidden risk factors influencing fatality rates, including temporal patterns that revealed a surge in fatalities after 17:00, with increased risks associated with non-seatbelt use during late evening and early morning hours. Younger drivers, particularly active at night, were found to exhibit higher rates of non-seatbelt usage and were more likely to be involved in severe accidents. These findings highlight the critical need for targeted seatbelt enforcement and safety interventions during high-risk periods, especially among younger drivers who are at fault in accidents.

安全带的使用大大降低了车辆事故中受伤和死亡的严重程度。本研究利用随机森林算法来评估安全带使用对泰国死亡概率的影响，重点关注造成事故的司机。该模型显示出较高的准确率，正确识别了95.10%的非致命病例和91.60%的致命病例，尽管存在一些错误分类。这项研究的一个关键贡献是确定了影响死亡率的潜在风险因素，包括时间模式，揭示了17:00之后死亡人数激增，深夜和清晨不使用安全带的风险增加。研究发现，年轻司机，尤其是夜间活跃的司机，使用非安全带的比例更高，更有可能发生严重事故。这些发现强调了在高风险时期，特别是在事故中有过错的年轻司机中，有针对性地实施安全带和安全干预的迫切需要。

引用次数: 0

Autonomous fleet management system in smart ports: Practical design and analytical considerations 智能港口的自主船队管理系统：实际设计和分析考虑

Multimodal Transportation

Pub Date : 2025-03-05 DOI: 10.1016/j.multra.2025.100211

Rui Chen , Jing Zhang , Hua Wang

Automated horizontal transportation in container terminals represents a significant advancement in the field of autonomous commercial vehicles. Traditionally, these systems rely on the individual intelligence of each vehicle, similar to autonomous passenger vehicles. However, recent uses of automated technology in select container terminals have demonstrated the benefits of integrating vehicles with a centralized Autonomous Fleet Management System (AFMS). This collaboration not only mitigates information silos but also enhances operational efficiency, safety, and fosters fleet cluster intelligence. This study examines current applications in automated container terminals, analyses practical scenarios, and identifies the essential characteristics of an effective AFMS to support horizontal transportation management. The insights from this comprehensive analysis assist port operators in designing and operating their systems and help scholars better understand and define research questions in this field.

集装箱码头的自动水平运输代表了自动商用车领域的重大进步。传统上，这些系统依赖于每辆车的个人智能，类似于自动驾驶乘用车。然而，最近在特定集装箱码头使用的自动化技术已经证明了将车辆与集中式自动车队管理系统（AFMS）集成的好处。这种合作不仅减轻了信息孤岛，还提高了运营效率、安全性，并促进了舰队集群智能。本研究考察了自动化集装箱码头的当前应用，分析了实际场景，并确定了有效的AFMS的基本特征，以支持横向运输管理。这一综合分析的见解有助于港口运营商设计和操作他们的系统，并帮助学者更好地理解和定义该领域的研究问题。

引用次数: 0

Fuzzy reliability theory analysis of traffic signal lamp performance 交通信号灯性能的模糊可靠性理论分析

Multimodal Transportation

Pub Date : 2025-03-01 DOI: 10.1016/j.multra.2025.100195

Mason D. Gemar , Shidong Pan , Zhanmin Zhang , Randy B. Machemehl

Over the past decade, many municipalities have begun to replace incandescent lamps in their traffic signals with light emitting diode (LED) arrays. While LED technology boasts longer lifetimes and superior performance over their counterparts, there are many limitations involved in both testing and evaluating their reliability. As such, the methodology and subsequent analysis procedures used to evaluate the reliability of traffic signal lamps along a corridor is proposed. To accomplish this task, the progression of the reliability assessment from individual lamp to the entire signal light system for a corridor is demonstrated. Furthermore, due to the nature of these systems and reliability assessment strategies, it is suggested that fuzzy, or more specifically, profust reliability theory could be applied to effectively analyze LED arrays, as well as corridor-wide signal light systems. Preliminary case study results, coupled with field observations of partially burned-out LED arrays, support this hypothesis.

在过去的十年里，许多城市已经开始用发光二极管（LED）阵列取代交通信号灯中的白炽灯。虽然LED技术拥有比同类产品更长的使用寿命和更优越的性能，但在测试和评估其可靠性方面存在许多限制。因此，本文提出了用于评估走廊沿线交通信号灯可靠性的方法和后续分析程序。为了完成这一任务，演示了从单个灯到整个走廊信号灯系统可靠性评估的过程。此外，由于这些系统的性质和可靠性评估策略，建议模糊，或者更具体地说，信任可靠性理论可以应用于有效地分析LED阵列，以及全走廊信号灯系统。初步的案例研究结果，加上对部分烧毁的LED阵列的现场观察，支持了这一假设。

引用次数: 0

Analyzing feature importance for older pedestrian crash severity: A comparative study of DNN models, emphasizing road and vehicle types with SHAP interpretation 分析特征对老年行人碰撞严重程度的重要性：DNN模型的比较研究，强调道路和车辆类型与SHAP解释

Multimodal Transportation

Pub Date : 2025-02-25 DOI: 10.1016/j.multra.2025.100203

Rocksana Akter , Susilawati Susilawati , Hamza Zubair , Wai Tong Chor

Recognizing the importance of road safety modeling, the study explores Deep Neural Networks (DNN) with features like hidden layers, batch normalization, Rectified Linear Unit (ReLU) activation, and dropout to predict crash severity, interpreting decisions using SHapley Additive exPlanations (SHAP) for crashes involving older pedestrians. The objective is to understand features influencing crashes involving older pedestrians, including vehicle attributes, road and environmental conditions, and temporal parameters. The analysis focused on 1808 pedestrian crashes involving individuals aged 65 and over at intersections in Victoria, Australia. This dataset comprises 6.14% fatalities, 52.38% serious injuries, and 41.48% incidents with other injuries. The study evaluated three DNN models for crash severity prediction, with the two hidden layers DNN model excelling in precision metrics and achieving a perfect Area Under the Receiver Operating Characteristics curve for fatalities. Compared to XGBoost, the DNN models demonstrated superior performance in predicting severe outcomes. SHAP analysis on the two hidden layers DNN model highlighted key factors influencing crash severity, offering insights into the nuanced relationships between features and predictions. The analysis highlighted the significance of variables like Traffic Control, Vehicle Type, and Movement in predicting fatalities and serious injuries. This study emphasizes the importance of considering Road and Vehicle Types to understand their roles in accident severity and identify interventions to reduce risks. Neglecting these factors may lead to incomplete or biased conclusions about crash outcomes. This research provides valuable insights for improving road safety, highlighting the effectiveness of SHAP force plots, bars, beeswarm plots, and dependency plots in enhancing clarity and understanding of DNN model predictions. These tools help identify the impact of features on crash severity.

认识到道路安全建模的重要性，该研究探索了具有隐藏层、批归一化、校正线性单元（ReLU）激活和dropout等特征的深度神经网络（DNN），以预测碰撞严重程度，并使用SHapley加性解释（SHAP）解释涉及老年行人的碰撞的决策。目标是了解影响涉及老年行人的碰撞的特征，包括车辆属性、道路和环境条件以及时间参数。该分析集中在澳大利亚维多利亚州十字路口发生的1808起涉及65岁及以上老年人的行人事故。该数据集包括6.14%的死亡，52.38%的严重伤害和41.48%的其他伤害事件。该研究评估了三种DNN模型的碰撞严重程度预测，其中两个隐藏层DNN模型在精度指标方面表现出色，并实现了完美的接收器操作特征曲线下的区域。与XGBoost相比，DNN模型在预测严重后果方面表现出更好的性能。对两个隐藏层DNN模型的SHAP分析突出了影响碰撞严重程度的关键因素，为特征和预测之间的微妙关系提供了见解。该分析强调了交通控制、车辆类型和运动等变量在预测死亡和严重伤害方面的重要性。本研究强调了考虑道路和车辆类型的重要性，以了解它们在事故严重程度中的作用，并确定干预措施以降低风险。忽视这些因素可能会导致关于坠机结果的结论不完整或有偏见。该研究为改善道路安全提供了有价值的见解，突出了SHAP力图、条形图、蜂群图和依赖图在提高DNN模型预测清晰度和理解方面的有效性。这些工具有助于确定功能对崩溃严重程度的影响。

{"title":"Analyzing feature importance for older pedestrian crash severity: A comparative study of DNN models, emphasizing road and vehicle types with SHAP interpretation","authors":"Rocksana Akter , Susilawati Susilawati , Hamza Zubair , Wai Tong Chor","doi":"10.1016/j.multra.2025.100203","DOIUrl":"10.1016/j.multra.2025.100203","url":null,"abstract":"<div><div>Recognizing the importance of road safety modeling, the study explores Deep Neural Networks (DNN) with features like hidden layers, batch normalization, Rectified Linear Unit (ReLU) activation, and dropout to predict crash severity, interpreting decisions using SHapley Additive exPlanations (SHAP) for crashes involving older pedestrians. The objective is to understand features influencing crashes involving older pedestrians, including vehicle attributes, road and environmental conditions, and temporal parameters. The analysis focused on 1808 pedestrian crashes involving individuals aged 65 and over at intersections in Victoria, Australia. This dataset comprises 6.14% fatalities, 52.38% serious injuries, and 41.48% incidents with other injuries. The study evaluated three DNN models for crash severity prediction, with the two hidden layers DNN model excelling in precision metrics and achieving a perfect Area Under the Receiver Operating Characteristics curve for fatalities. Compared to XGBoost, the DNN models demonstrated superior performance in predicting severe outcomes. SHAP analysis on the two hidden layers DNN model highlighted key factors influencing crash severity, offering insights into the nuanced relationships between features and predictions. The analysis highlighted the significance of variables like Traffic Control, Vehicle Type, and Movement in predicting fatalities and serious injuries. This study emphasizes the importance of considering Road and Vehicle Types to understand their roles in accident severity and identify interventions to reduce risks. Neglecting these factors may lead to incomplete or biased conclusions about crash outcomes. This research provides valuable insights for improving road safety, highlighting the effectiveness of SHAP force plots, bars, beeswarm plots, and dependency plots in enhancing clarity and understanding of DNN model predictions. These tools help identify the impact of features on crash severity.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 2","pages":"Article 100203"},"PeriodicalIF":0.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143696674","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

Multimodal integration in India: Opportunities, challenges, and strategies for sustainable urban mobility 印度的多式联运一体化：可持续城市交通的机遇、挑战和战略

Multimodal Transportation

Pub Date : 2025-02-20 DOI: 10.1016/j.multra.2025.100210

Rahul Tanwar, Pradeep Kumar Agarwal

This study explores the opportunities and challenges of advancing multimodal integration for sustainable urban mobility in India. With rapid urbanization and increasing motorization, Indian cities face issues of congestion, air pollution, and social inequity. Multimodal integration, the seamless integration of different transportation modes, is a promising approach to address these challenges. The study assesses the current state of urban mobility in India, examines the concepts and benefits of multimodal integration, and identifies key opportunities, including supportive policies, technological advancements, and public-private partnerships. It also discusses challenges such as institutional barriers, financial constraints, and the need for behavioral change. Case studies of successful initiatives in Delhi and Ahmedabad demonstrate the potential benefits of integrated transport systems. The study proposes recommendations for advancing multimodal integration, focusing on policy reforms, infrastructure development, capacity building, and stakeholder engagement. It concludes by summarizing key findings and identifying future research directions, emphasizing the need for further investigation into long-term impacts, innovative funding mechanisms, emerging technologies, comparative policy analysis, and social and behavioral aspects of sustainable urban mobility. This research contributes to the growing knowledge on multimodal integration and sustainable urban mobility in India, providing valuable insights for policymakers, urban planners, and transportation professionals working towards creating more sustainable, efficient, and inclusive cities.

本研究探讨了印度推进多式联运一体化以实现可持续城市交通的机遇和挑战。随着快速城市化和机动车化程度的提高，印度城市面临着拥堵、空气污染和社会不平等等问题。多式联运，即不同运输方式的无缝整合，是解决这些挑战的一种很有前途的方法。该研究评估了印度城市交通的现状，考察了多式联运一体化的概念和好处，并确定了关键机遇，包括支持性政策、技术进步和公私合作伙伴关系。它还讨论了诸如制度障碍、财政约束和行为改变的需要等挑战。对德里和艾哈迈达巴德成功举措的案例研究显示了综合运输系统的潜在效益。该研究为推进多式联运一体化提出了建议，重点关注政策改革、基础设施建设、能力建设和利益相关者参与。最后，总结了主要发现并确定了未来的研究方向，强调需要进一步研究可持续城市交通的长期影响、创新融资机制、新兴技术、比较政策分析以及社会和行为方面。这项研究有助于增加印度多式联运一体化和可持续城市交通的知识，为政策制定者、城市规划者和交通专业人士提供有价值的见解，以创造更可持续、更高效、更包容的城市。

{"title":"Multimodal integration in India: Opportunities, challenges, and strategies for sustainable urban mobility","authors":"Rahul Tanwar, Pradeep Kumar Agarwal","doi":"10.1016/j.multra.2025.100210","DOIUrl":"10.1016/j.multra.2025.100210","url":null,"abstract":"<div><div>This study explores the opportunities and challenges of advancing multimodal integration for sustainable urban mobility in India. With rapid urbanization and increasing motorization, Indian cities face issues of congestion, air pollution, and social inequity. Multimodal integration, the seamless integration of different transportation modes, is a promising approach to address these challenges. The study assesses the current state of urban mobility in India, examines the concepts and benefits of multimodal integration, and identifies key opportunities, including supportive policies, technological advancements, and public-private partnerships. It also discusses challenges such as institutional barriers, financial constraints, and the need for behavioral change. Case studies of successful initiatives in Delhi and Ahmedabad demonstrate the potential benefits of integrated transport systems. The study proposes recommendations for advancing multimodal integration, focusing on policy reforms, infrastructure development, capacity building, and stakeholder engagement. It concludes by summarizing key findings and identifying future research directions, emphasizing the need for further investigation into long-term impacts, innovative funding mechanisms, emerging technologies, comparative policy analysis, and social and behavioral aspects of sustainable urban mobility. This research contributes to the growing knowledge on multimodal integration and sustainable urban mobility in India, providing valuable insights for policymakers, urban planners, and transportation professionals working towards creating more sustainable, efficient, and inclusive cities.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 2","pages":"Article 100210"},"PeriodicalIF":0.0,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579851","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

Explainable artificial intelligence visions on incident duration using eXtreme Gradient Boosting and SHapley Additive exPlanations 使用极端梯度增强和SHapley加性解释解释事件持续时间的可解释人工智能视觉

Multimodal Transportation

Pub Date : 2025-02-20 DOI: 10.1016/j.multra.2025.100209

Khaled Hamad , Emran Alotaibi , Waleed Zeiada , Ghazi Al-Khateeb , Saleh Abu Dabous , Maher Omar , Bharadwaj R.K. Mantha , Mohamed G. Arab , Tarek Merabtene

Efficient management of traffic incidents is a focal point in traffic management, with direct implications for road safety, congestion, and the environment. Traditional models have grappled with the unpredictability inherent in traffic incidents, often failing to capture the multifaceted influences on incident durations. This study introduces an application of Explainable Artificial Intelligence (XAI) using eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP) to analyze the complexities of traffic incident duration prediction. Utilizing a substantial dataset of over 366,000 records from the Houston traffic management center, the study innovates in the domain of traffic analytics by predicting incident durations and revealing the contribution of each predictive variable. The XGBoost algorithm's ability to handle multi-dimensional datasets was employed to identify crucial variables affecting incident durations. Meanwhile, SHAP values offered transparency into the model's decision-making process, clarifying the roles of over fifty parameters. The study's results demonstrate that variables such as the involvement of heavy trucks and blockage of main lanes are essential in influencing incident durations, aligning with findings from previous literature. The SHAP analysis further revealed time-sensitive patterns, with time of day and day of the week exhibiting considerable effects on predictions. The beeswarm plots of SHAP provided a detailed visualization of these effects, differentiating between high and low values effects for each variable. The model's high accuracy, with a coefficient of determination (R²) of 0.72 and a root mean square error (RMSE) of 21.2 min, indicates the potential of XAI in enhancing traffic management systems.

有效管理交通事故是交通管理的重点，对道路安全、交通拥堵和环境都有直接影响。传统模型一直在努力解决交通事故固有的不可预测性问题，但往往无法捕捉到事故持续时间的多方面影响因素。本研究介绍了一种可解释人工智能（XAI）的应用，即使用极梯度提升（XGBoost）和SHAPLEY Additive exPlanations（SHAP）来分析交通事故持续时间预测的复杂性。该研究利用休斯顿交通管理中心超过 366,000 条记录的大量数据集，通过预测事故持续时间和揭示每个预测变量的贡献，在交通分析领域进行了创新。XGBoost 算法具有处理多维数据集的能力，可用于识别影响事故持续时间的关键变量。同时，SHAP 值为模型的决策过程提供了透明度，明确了 50 多个参数的作用。研究结果表明，重型卡车的参与和主要车道的堵塞等变量在影响事故持续时间方面至关重要，这与以往文献的研究结果一致。SHAP 分析进一步揭示了对时间敏感的模式，一天中的时间和一周中的某一天对预测有相当大的影响。SHAP 的蜂群图提供了这些影响的详细直观图，区分了每个变量的高值和低值影响。该模型的准确度很高，决定系数 (R2) 为 0.72，均方根误差 (RMSE) 为 21.2 分钟，这表明 XAI 在增强交通管理系统方面具有潜力。

{"title":"Explainable artificial intelligence visions on incident duration using eXtreme Gradient Boosting and SHapley Additive exPlanations","authors":"Khaled Hamad , Emran Alotaibi , Waleed Zeiada , Ghazi Al-Khateeb , Saleh Abu Dabous , Maher Omar , Bharadwaj R.K. Mantha , Mohamed G. Arab , Tarek Merabtene","doi":"10.1016/j.multra.2025.100209","DOIUrl":"10.1016/j.multra.2025.100209","url":null,"abstract":"<div><div>Efficient management of traffic incidents is a focal point in traffic management, with direct implications for road safety, congestion, and the environment. Traditional models have grappled with the unpredictability inherent in traffic incidents, often failing to capture the multifaceted influences on incident durations. This study introduces an application of Explainable Artificial Intelligence (XAI) using eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP) to analyze the complexities of traffic incident duration prediction. Utilizing a substantial dataset of over 366,000 records from the Houston traffic management center, the study innovates in the domain of traffic analytics by predicting incident durations and revealing the contribution of each predictive variable. The XGBoost algorithm's ability to handle multi-dimensional datasets was employed to identify crucial variables affecting incident durations. Meanwhile, SHAP values offered transparency into the model's decision-making process, clarifying the roles of over fifty parameters. The study's results demonstrate that variables such as the involvement of heavy trucks and blockage of main lanes are essential in influencing incident durations, aligning with findings from previous literature. The SHAP analysis further revealed time-sensitive patterns, with time of day and day of the week exhibiting considerable effects on predictions. The beeswarm plots of SHAP provided a detailed visualization of these effects, differentiating between high and low values effects for each variable. The model's high accuracy, with a coefficient of determination (R<sup>2</sup>) of 0.72 and a root mean square error (RMSE) of 21.2 min, indicates the potential of XAI in enhancing traffic management systems.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 2","pages":"Article 100209"},"PeriodicalIF":0.0,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143621291","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 evolving dynamics of airport ground access: A multinomial logit analysis of mode choice at Guwahati Airport, India 机场地面通道的动态演化：印度古瓦哈提机场模式选择的多项逻辑分析

Multimodal Transportation

Pub Date : 2025-02-13 DOI: 10.1016/j.multra.2025.100208

Lalit Swami, Mokaddes Ali Ahmed, Suprava Jena

As shared mobility options like ridesourcing services continue to reshape urban transportation systems globally, their impact on airport ground access has become increasingly significant. This study investigates the changing dynamics of airport access at Lokpriya Gopinath Bordoloi International Airport (LGBI) in Guwahati, India, amidst the growing presence of ridesourcing services. A total of 700 air passengers were surveyed using a random sampling technique over 15 consecutive days, providing comprehensive data for the analysis. A multinomial logit (MNL) model was employed to examine factors influencing mode choice, considering variables such as age, residential status, group size, car ownership, luggage, safety, and convenience. The model explains 48.2 % to 57.2 % of the variation in mode choice. The results reveal that younger passengers (aged 21–30) are 2.14 times more likely to choose ridesourcing services. Additionally, visitors are significantly more inclined to use ridesourcing services compared to locals, with an odds ratio of 2.56. While passengers with car ownership are 5.43 times more likely to prefer private vehicles. The study underscores the growing significance of ridesourcing services in airport ground access and highlights the need for transportation planning and policymaking to adapt to these evolving trends.

随着拼车服务等共享出行选择继续重塑全球城市交通系统，它们对机场地面通道的影响变得越来越大。本研究调查了印度古瓦哈蒂Lokpriya Gopinath Bordoloi国际机场（LGBI）在打车服务日益增长的背景下，机场通道的变化动态。通过连续15天的随机抽样调查，共对700名航空乘客进行了调查，为分析提供了全面的数据。考虑年龄、居住状况、群体规模、汽车拥有量、行李、安全性和便利性等因素，采用多项logit （MNL）模型考察影响模式选择的因素。该模型解释了48.2%至57.2%的模式选择变化。结果显示，年轻乘客（21-30岁）选择约车服务的可能性高出2.14倍。此外，与当地人相比，游客更倾向于使用拼车服务，优势比为2.56。而拥有汽车的乘客选择私家车的可能性是前者的5.43倍。该研究强调了约车服务在机场地面通道中的重要性，并强调了交通规划和政策制定的必要性，以适应这些不断变化的趋势。

{"title":"The evolving dynamics of airport ground access: A multinomial logit analysis of mode choice at Guwahati Airport, India","authors":"Lalit Swami, Mokaddes Ali Ahmed, Suprava Jena","doi":"10.1016/j.multra.2025.100208","DOIUrl":"10.1016/j.multra.2025.100208","url":null,"abstract":"<div><div>As shared mobility options like ridesourcing services continue to reshape urban transportation systems globally, their impact on airport ground access has become increasingly significant. This study investigates the changing dynamics of airport access at Lokpriya Gopinath Bordoloi International Airport (LGBI) in Guwahati, India, amidst the growing presence of ridesourcing services. A total of 700 air passengers were surveyed using a random sampling technique over 15 consecutive days, providing comprehensive data for the analysis. A multinomial logit (MNL) model was employed to examine factors influencing mode choice, considering variables such as age, residential status, group size, car ownership, luggage, safety, and convenience. The model explains 48.2 % to 57.2 % of the variation in mode choice. The results reveal that younger passengers (aged 21–30) are 2.14 times more likely to choose ridesourcing services. Additionally, visitors are significantly more inclined to use ridesourcing services compared to locals, with an odds ratio of 2.56. While passengers with car ownership are 5.43 times more likely to prefer private vehicles. The study underscores the growing significance of ridesourcing services in airport ground access and highlights the need for transportation planning and policymaking to adapt to these evolving trends.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 2","pages":"Article 100208"},"PeriodicalIF":0.0,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527555","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

Urban intersection traffic flow prediction: A physics-guided stepwise framework utilizing spatio-temporal graph neural network algorithms 城市交叉口交通流量预测：利用时空图神经网络算法的物理引导逐步框架

Multimodal Transportation

Pub Date : 2025-02-12 DOI: 10.1016/j.multra.2025.100207

Yuyan Annie Pan , Fuliang Li , Anran Li , Zhiqiang Niu , Zhen Liu

Accurate traffic flow forecasting at urban intersections is critical for optimizing transportation infrastructure and reducing congestion. This manuscript introduces a novel framework, the Physics-Guided Spatio-Temporal Graph Neural Network (PG-STGNN), specifically designed for traffic flow prediction. By integrating the principles of traffic flow physics with advanced spatio-temporal graph neural network algorithms, the framework captures complex spatio-temporal dependencies in traffic networks. PG-STGNN adopts a stepwise approach, addressing key performance metrics like queue formation and signal timing complexities at intersections. To validate its effectiveness, the model was applied to real-world traffic data from the Yizhuang District of Beijing. Compared to traditional models such as ARIMA, KNN, and Random Forest, PG-STGNN significantly improves prediction accuracy, achieving MAPE reductions of 19.9 %, 18.6 %, 6.1 %, 20.7 %, 5.0 %, 1.8 %, and 1.1 % against KNN, ARIMA, RF, BP, T-GCN, STGCN, and ST-ED-RMGC, respectively. With the lowest MAPE (9.452 %), MAE (2.485), and RMSE (4.364), PG-STGNN demonstrates superior prediction performance. These results underscore its potential to provide reliable short-term traffic forecasts, offering essential insights for the strategic planning and management of urban intelligent transportation systems.

准确预测城市交叉口的交通流量对于优化交通基础设施和减少拥堵至关重要。本手稿介绍了一种新颖的框架，即物理引导时空图神经网络（PG-STGNN），专门用于交通流预测。通过将交通流物理学原理与先进的时空图神经网络算法相结合，该框架可捕捉交通网络中复杂的时空依赖关系。PG-STGNN 采用循序渐进的方法，解决了交叉口队列形成和信号配时复杂性等关键性能指标。为验证其有效性，该模型被应用于北京亦庄地区的实际交通数据。与 ARIMA、KNN 和随机森林等传统模型相比，PG-STGNN 显著提高了预测精度，与 KNN、ARIMA、RF、BP、T-GCN、STGCN 和 ST-ED-RMGC 相比，MAPE 分别降低了 19.9%、18.6%、6.1%、20.7%、5.0%、1.8% 和 1.1%。PG-STGNN 的 MAPE (9.452 %)、MAE (2.485) 和 RMSE (4.364) 最低，显示出卓越的预测性能。这些结果凸显了 PG-STGNN 在提供可靠的短期交通预测方面的潜力，为城市智能交通系统的战略规划和管理提供了重要见解。

{"title":"Urban intersection traffic flow prediction: A physics-guided stepwise framework utilizing spatio-temporal graph neural network algorithms","authors":"Yuyan Annie Pan , Fuliang Li , Anran Li , Zhiqiang Niu , Zhen Liu","doi":"10.1016/j.multra.2025.100207","DOIUrl":"10.1016/j.multra.2025.100207","url":null,"abstract":"<div><div>Accurate traffic flow forecasting at urban intersections is critical for optimizing transportation infrastructure and reducing congestion. This manuscript introduces a novel framework, the Physics-Guided Spatio-Temporal Graph Neural Network (PG-STGNN), specifically designed for traffic flow prediction. By integrating the principles of traffic flow physics with advanced spatio-temporal graph neural network algorithms, the framework captures complex spatio-temporal dependencies in traffic networks. PG-STGNN adopts a stepwise approach, addressing key performance metrics like queue formation and signal timing complexities at intersections. To validate its effectiveness, the model was applied to real-world traffic data from the Yizhuang District of Beijing. Compared to traditional models such as ARIMA, KNN, and Random Forest, PG-STGNN significantly improves prediction accuracy, achieving MAPE reductions of 19.9 %, 18.6 %, 6.1 %, 20.7 %, 5.0 %, 1.8 %, and 1.1 % against KNN, ARIMA, RF, BP, T-GCN, STGCN, and ST-ED-RMGC, respectively. With the lowest MAPE (9.452 %), MAE (2.485), and RMSE (4.364), PG-STGNN demonstrates superior prediction performance. These results underscore its potential to provide reliable short-term traffic forecasts, offering essential insights for the strategic planning and management of urban intelligent transportation systems.</div></div>","PeriodicalId":100933,"journal":{"name":"Multimodal Transportation","volume":"4 2","pages":"Article 100207"},"PeriodicalIF":0.0,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479107","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