Physica A: Statistical Mechanics and its Applications最新文献

Large language model-driven bi-level game framework for connected and automated vehicle pair at mixed unsignalized intersections 混合无信号交叉口网联和自动车辆对的大语言模型驱动双层博弈框架

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-29 DOI: 10.1016/j.physa.2026.131327

Tianwen Yan , Maode Yan , Lei Zuo , Mingren Fu

As Connected and Automated Vehicles (CAVs) become core Internet of Things (IoT) terminals in modern transportation, growing research has focused on safe driving strategy at unsignalized intersections. However, existing studies often neglect how multiple CAVs can cooperate to improve safety, and rarely address traffic fairness under stochastic Human-Driven Vehicle (HDV) behaviors. To address these issues, we propose a novel large language model (LLM)-driven bi-level game framework for CAV Pair at mixed unsignalized intersections, namely BiG-LLM. This framework combines semantic scene understanding via LLM with verifiable game-theoretic decision-making, effectively alleviating the hallucinations caused by pure LLM. A CAV Pair-based mixed platoon is introduced to exploit multi-CAV synergy. In the bi-level game strategy, the upper-level game enables the lead CAV to optimize the allocation of right-of-way. The tail CAV activates the lower-level fairness game, using the waiting anxiety model based on prospect theory to quantify frustration and discourage extreme waiting. Extensive simulations are conducted under multiple traffic flow conditions. The results demonstrate that BiG-LLM consistently achieves a high success rate, improves safety by increasing the minimum Post-Encroachment Time (PET), and reduces the maximum waiting time compared to baseline methods, while maintaining competitive efficiency. These results verify the effectiveness of BiG-LLM in balancing efficiency, safety, and system-level traffic fairness at unsignalized intersections.

随着互联和自动驾驶汽车（cav）成为现代交通的核心物联网（IoT）终端，越来越多的研究关注于无信号交叉口的安全驾驶策略。然而，现有的研究往往忽视了多辆自动驾驶汽车如何协同提高安全性，很少涉及随机人类驾驶车辆（HDV）行为下的交通公平问题。为了解决这些问题，我们提出了一种新的大型语言模型（LLM）驱动的混合无信号交叉口CAV对双层博弈框架，即BiG-LLM。该框架将基于LLM的语义场景理解与可验证的博弈论决策相结合，有效缓解了纯LLM带来的幻觉。引入基于CAV对的混合排，实现多CAV协同。在双层博弈策略中，上层博弈使领先CAV能够优化路权的分配。尾部CAV激活下层公平博弈，利用基于前景理论的等待焦虑模型量化挫败感，抑制极端等待。在多种交通流条件下进行了大量的仿真。结果表明，与基线方法相比，BiG-LLM在保持竞争效率的同时，通过增加最小入侵后时间（PET）来提高安全性，并减少最大等待时间，从而始终保持较高的成功率。这些结果验证了BiG-LLM在无信号交叉口平衡效率、安全性和系统级交通公平性方面的有效性。

{"title":"Large language model-driven bi-level game framework for connected and automated vehicle pair at mixed unsignalized intersections","authors":"Tianwen Yan , Maode Yan , Lei Zuo , Mingren Fu","doi":"10.1016/j.physa.2026.131327","DOIUrl":"10.1016/j.physa.2026.131327","url":null,"abstract":"<div><div>As Connected and Automated Vehicles (CAVs) become core Internet of Things (IoT) terminals in modern transportation, growing research has focused on safe driving strategy at unsignalized intersections. However, existing studies often neglect how multiple CAVs can cooperate to improve safety, and rarely address traffic fairness under stochastic Human-Driven Vehicle (HDV) behaviors. To address these issues, we propose a novel large language model (LLM)-driven bi-level game framework for CAV Pair at mixed unsignalized intersections, namely BiG-LLM. This framework combines semantic scene understanding via LLM with verifiable game-theoretic decision-making, effectively alleviating the hallucinations caused by pure LLM. A CAV Pair-based mixed platoon is introduced to exploit multi-CAV synergy. In the bi-level game strategy, the upper-level game enables the lead CAV to optimize the allocation of right-of-way. The tail CAV activates the lower-level fairness game, using the waiting anxiety model based on prospect theory to quantify frustration and discourage extreme waiting. Extensive simulations are conducted under multiple traffic flow conditions. The results demonstrate that BiG-LLM consistently achieves a high success rate, improves safety by increasing the minimum Post-Encroachment Time (PET), and reduces the maximum waiting time compared to baseline methods, while maintaining competitive efficiency. These results verify the effectiveness of BiG-LLM in balancing efficiency, safety, and system-level traffic fairness at unsignalized intersections.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131327"},"PeriodicalIF":3.1,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cumulant expansions and the large deviation rate function 累积展开和大偏差率函数

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-28 DOI: 10.1016/j.physa.2026.131332

J.M. Rickman , M.L. Comer , J. Xu

We show that that, by analogy with results from extreme-value theory, the rare events in a Gibbs distribution, as quantified by the exceedance of the distribution, may be expressed in terms of tail joint cumulants that systematically describe rare fluctuations in the tail of the distribution. A joint cumulant expansion based on histogram reweighting is then used to obtain the rate function from large deviation theory, a quantity that characterizes the exponential dependence of the tail of the distribution, for two prototypical systems, namely a 1-D ferromagnet in an external field and an atomistic model of a solid in the isobaric-isothermal ensemble. It is shown that this expansion permits one to capture the behavior of the rate function over a relatively wide range of parameter space from calculations done at one point in this space. Finally, we discuss the extension of this approach to multivariate distributions and suggest strategies for characterizing infrequent events described by the tails of such distributions.

我们证明，通过类比极值理论的结果，吉布斯分布中由分布的超出量量化的罕见事件可以用尾部联合累积量来表示，尾部联合累积量系统地描述了分布尾部的罕见波动。然后使用基于直方图重加权的联合累积展开从大偏差理论中获得速率函数，这是一个表征分布尾部指数依赖性的量，用于两个原型系统，即外场中的一维铁磁体和等压-等温系综中的固体原子模型。结果表明，这种扩展允许人们从在该空间中某一点所做的计算中获得速率函数在相对广泛的参数空间范围内的行为。最后，我们讨论了该方法在多变量分布中的扩展，并提出了描述由这些分布的尾部描述的罕见事件的策略。

引用次数: 0

Excitation of Kármán vortex street in Bose–Einstein condensate with moiré lattice 具有莫尔晶格的玻色-爱因斯坦凝聚中Kármán涡旋街的激发

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-28 DOI: 10.1016/j.physa.2026.131333

Kaihua Shao , Baolong Xi , Zhonghong Xi , Pu Tu , Xi Zhao , Yuren Shi

The excitation of various wake patterns in a Bose–Einstein condensate (BEC) with moiré lattice is investigated numerically. The effects of physical parameters such as the velocity of the obstacle potential and the depth of the moiré lattice on vortex shedding are systematically investigated using numerical simulation. Appropriate parameters lead to the shedding of periodic vortex pairs and the formation of different wake patterns. Interestingly, when the vortex is situated at the potential well, its size is smaller than at the potential barrier of moiré lattice, indicating that its size is reduced in the higher-density region. The formation of Kármán vortex street (Kvs) becomes less favorable as the moiré lattice deepens. We also find that if a single vortex is shed behind the obstacle potential, the wake pattern eventually exhibits Kvs. It is worth noting that the distance between a pair of vortices and the angular frequency of their rotation oscillates periodically with time evolution. Additionally, as more vortices are shed from the obstacle potential, the kinetic energy, interaction energy and total energy in the system increase, and the potential energy exhibits oscillatory behavior. The direction of motion of the obstacle potential in the moiré lattice affects the vortex distribution. Furthermore, we propose a possible implementation of the Kvs in a BEC with moiré lattice.

用数值方法研究了具有莫尔格的玻色-爱因斯坦凝聚体（BEC）中各种尾迹的激发。采用数值模拟的方法，系统地研究了障碍势速度和莫尔栅格深度等物理参数对涡流脱落的影响。适当的参数会导致周期性涡对的脱落和不同尾迹型的形成。有趣的是，当涡旋位于势阱处时，其尺寸小于莫尔维尔晶格势垒处，表明其尺寸在高密度区域减小。随着摩尔晶格的加深，Kármán涡街（Kvs）的形成变得越来越不利。我们还发现，如果在障碍势后面有一个单一的涡流，尾迹模式最终表现为kv。值得注意的是，一对涡旋之间的距离及其旋转的角频率随时间的变化而周期性地振荡。此外，随着障碍势释放出更多的涡流，系统的动能、相互作用能和总能量增加，势能呈现振荡行为。涡流点阵中障碍势的运动方向影响着涡流的分布。此外，我们还提出了一种在具有莫尔格的BEC中实现Kvs的可能方法。

{"title":"Excitation of Kármán vortex street in Bose–Einstein condensate with moiré lattice","authors":"Kaihua Shao , Baolong Xi , Zhonghong Xi , Pu Tu , Xi Zhao , Yuren Shi","doi":"10.1016/j.physa.2026.131333","DOIUrl":"10.1016/j.physa.2026.131333","url":null,"abstract":"<div><div>The excitation of various wake patterns in a Bose–Einstein condensate (BEC) with moiré lattice is investigated numerically. The effects of physical parameters such as the velocity of the obstacle potential and the depth of the moiré lattice on vortex shedding are systematically investigated using numerical simulation. Appropriate parameters lead to the shedding of periodic vortex pairs and the formation of different wake patterns. Interestingly, when the vortex is situated at the potential well, its size is smaller than at the potential barrier of moiré lattice, indicating that its size is reduced in the higher-density region. The formation of Kármán vortex street (Kvs) becomes less favorable as the moiré lattice deepens. We also find that if a single vortex is shed behind the obstacle potential, the wake pattern eventually exhibits Kvs. It is worth noting that the distance between a pair of vortices and the angular frequency of their rotation oscillates periodically with time evolution. Additionally, as more vortices are shed from the obstacle potential, the kinetic energy, interaction energy and total energy in the system increase, and the potential energy exhibits oscillatory behavior. The direction of motion of the obstacle potential in the moiré lattice affects the vortex distribution. Furthermore, we propose a possible implementation of the Kvs in a BEC with moiré lattice.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131333"},"PeriodicalIF":3.1,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Facilitating cooperative behavior through reinforcement learning with age-driven state transitions in structured populations 结构群体中年龄驱动状态转换的强化学习促进合作行为

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-27 DOI: 10.1016/j.physa.2026.131319

Ran Zhang , Tianbo An , Jian Zhao , Zhen Wang

The Q-learning algorithm in reinforcement learning closely parallels human experience based decision making, and has been widely applied to evolutionary game theory to study the emergence of cooperation. Previous studies have expanded the definition of the state space to some extent, but they have not captured the continuity of states in the temporal dimension. To address this gap, we propose a multistate transition mechanism driven by strategy age, where an agent’s state is defined by the time steps it persists with the same strategy. As the strategy age increases, the agent transitions into higher states until reaching the maximum threshold. For comparison, we also design a bistate mechanism that distinguishes states only between ages below and above the threshold. Simulation results show that both multistate and bistate mechanisms promote cooperation significantly better than memoryless and self-regarding Q-learning, with the multistate mechanism performing best. The key reason is that high age defectors see their Q-value of choosing defection drop below that of choosing cooperation and thus occasionally switch to cooperate under prolonged exposure to defectors. These switches periodically seed new low age cooperators, continually replenishing the cooperative pool. Raising the strategy age threshold expands the state space, giving defectors more chances to switch to cooperation and further boosting cooperation. By contrast, the bistate mechanism partitions the space too coarsely, limiting such transitions and yielding weaker outcomes. We also find that cooperation is most likely to emerge under moderate learning rates

α

and higher discount factors

γ

.

强化学习中的Q-learning算法与人类基于经验的决策非常相似，已被广泛应用于进化博弈论中研究合作的出现。以往的研究在一定程度上扩展了状态空间的定义，但没有捕捉到状态在时间维度上的连续性。为了解决这一差距，我们提出了一种由策略年龄驱动的多状态转换机制，其中智能体的状态由其坚持相同策略的时间步长定义。随着策略年龄的增加，智能体过渡到更高的状态，直到达到最大阈值。为了进行比较，我们还设计了一种双状态机制，仅区分年龄低于和高于阈值的状态。仿真结果表明，多状态和双状态机制对合作的促进作用都明显优于无记忆和自相关q学习，其中多状态机制表现最好。关键原因是，高年龄的叛逃者在长时间接触叛逃者的情况下，其选择叛逃的q值低于选择合作的q值，因此偶尔会转向合作。这些开关周期性地播种新的低年龄的合作者，不断补充合作池。提高战略年龄门槛扩大了状态空间，使叛逃者有更多的机会转向合作，进一步促进合作。相比之下，双态机制将空间划分得过于粗糙，限制了这种转变，并产生了较弱的结果。我们还发现，在中等学习率α和较高的折扣因子γ下，合作最有可能出现。

{"title":"Facilitating cooperative behavior through reinforcement learning with age-driven state transitions in structured populations","authors":"Ran Zhang , Tianbo An , Jian Zhao , Zhen Wang","doi":"10.1016/j.physa.2026.131319","DOIUrl":"10.1016/j.physa.2026.131319","url":null,"abstract":"<div><div>The Q-learning algorithm in reinforcement learning closely parallels human experience based decision making, and has been widely applied to evolutionary game theory to study the emergence of cooperation. Previous studies have expanded the definition of the state space to some extent, but they have not captured the continuity of states in the temporal dimension. To address this gap, we propose a multistate transition mechanism driven by strategy age, where an agent’s state is defined by the time steps it persists with the same strategy. As the strategy age increases, the agent transitions into higher states until reaching the maximum threshold. For comparison, we also design a bistate mechanism that distinguishes states only between ages below and above the threshold. Simulation results show that both multistate and bistate mechanisms promote cooperation significantly better than memoryless and self-regarding Q-learning, with the multistate mechanism performing best. The key reason is that high age defectors see their Q-value of choosing defection drop below that of choosing cooperation and thus occasionally switch to cooperate under prolonged exposure to defectors. These switches periodically seed new low age cooperators, continually replenishing the cooperative pool. Raising the strategy age threshold expands the state space, giving defectors more chances to switch to cooperation and further boosting cooperation. By contrast, the bistate mechanism partitions the space too coarsely, limiting such transitions and yielding weaker outcomes. We also find that cooperation is most likely to emerge under moderate learning rates <span><math><mi>α</mi></math></span> and higher discount factors <span><math><mi>γ</mi></math></span>.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131319"},"PeriodicalIF":3.1,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146049116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Novel approach to accounting for correlations in evolution of an open quantum system and the Lindblad equation 解释开放量子系统和林德布莱德方程演化中相关关系的新方法

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-26 DOI: 10.1016/j.physa.2026.131326

Victor F. Los

A projection operator is introduced which exactly transforms the inhomogeneous Nakajima–Zwanzig generalized master equation for the relevant part of a system +bath statistical operator, containing the inhomogeneous irrelevant term comprising the initial correlations, into the homogeneous equation accounting for initial correlations in the kernel governing its evolution. No ”factorizing initial state” approximation has been used. The obtained equation is equivalent to completely closed (homogeneous) equation for the statistical operator of a system of interest interacting with a bath. In the Born approximation (weak system–bath interaction) this equation can be presented as the time-local Redfield-like equation with additional terms caused by initial correlations. As an application, a quantum oscillator, interacting with a Boson field and driven from the Gibbs initial equilibrium system+bath state by a weak external force, is considered. All terms determining the oscillator evolution over time are explicitly calculated at all timescales. They show how initial correlations influence the evolution process. It is also demonstrated that at the large timescale this influence vanishes, and the evolution equation for the quantum oscillator statistical operator acquires the Lindblad form.

引入了一种投影算子，将系统相关部分的非齐次Nakajima-Zwanzig广义主方程+bath统计算子（包含包含初始关联项的非齐次无关项）精确地转化为控制系统演化的核中包含初始关联项的齐次方程。没有使用“分解初始状态”近似。所得方程等价于与浴体相互作用的感兴趣系统的统计算子的完全闭（齐次）方程。在玻恩近似（弱系统浴相互作用）中，该方程可以表示为带有由初始关联引起的附加项的时间局部类红场方程。作为一个应用，考虑了一个量子振荡器，它与玻色子场相互作用，并被弱外力从吉布斯初始平衡系统+浴态驱动。所有决定振荡器随时间演变的项都在所有时间尺度上明确计算。它们显示了最初的相关性如何影响进化过程。在大时间尺度上，这种影响消失，量子振荡器统计算符的演化方程获得Lindblad形式。

{"title":"Novel approach to accounting for correlations in evolution of an open quantum system and the Lindblad equation","authors":"Victor F. Los","doi":"10.1016/j.physa.2026.131326","DOIUrl":"10.1016/j.physa.2026.131326","url":null,"abstract":"<div><div>A projection operator is introduced which exactly transforms the inhomogeneous Nakajima–Zwanzig generalized master equation for the relevant part of a system +bath statistical operator, containing the inhomogeneous irrelevant term comprising the initial correlations, into the homogeneous equation accounting for initial correlations in the kernel governing its evolution. No ”factorizing initial state” approximation has been used. The obtained equation is equivalent to completely closed (homogeneous) equation for the statistical operator of a system of interest interacting with a bath. In the Born approximation (weak system–bath interaction) this equation can be presented as the time-local Redfield-like equation with additional terms caused by initial correlations. As an application, a quantum oscillator, interacting with a Boson field and driven from the Gibbs initial equilibrium system+bath state by a weak external force, is considered. All terms determining the oscillator evolution over time are explicitly calculated at all timescales. They show how initial correlations influence the evolution process. It is also demonstrated that at the large timescale this influence vanishes, and the evolution equation for the quantum oscillator statistical operator acquires the Lindblad form.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131326"},"PeriodicalIF":3.1,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146049117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Metro passenger alighting flow prediction for real-time crowding information: A deep learning approach based on sequential images 基于实时拥挤信息的地铁乘客下车流量预测：基于序列图像的深度学习方法

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-25 DOI: 10.1016/j.physa.2026.131322

Xiling Lin , Qun Chen , Yan Wang

With the continued increase in metro ridership, real-time crowding information (RTCI) has become essential for improving travel experience and making better use of carriage space. By accurately predicting the number of passengers alighting from each carriage and estimating available space in real time, it is possible to guide waiting passengers to better boarding choices and reduce congestion. Most existing RTCI studies focus on total passenger flow prediction or instantaneous crowd level estimation, while paying little attention to how alighting behavior of passengers directly affects carriage space availability. To address this issue, this study presents three main contributions. (1) A feature vector is constructed to represent the dynamic state of a carriage, along with a mapping function that describes the evolution of available space, forming a theoretical basis for RTCI generation.(2) A deep learning model based on image sequences is developed by integrating EfficientNet-B0 with LSTM (Long Short-Term Memory), enabling accurate prediction of the number of passengers alighting at the carriage level. (3) Systematic experiments using a dataset from the Changsha metro demonstrate that the method achieves high prediction accuracy while maintaining computational efficiency. Comparative and ablation studies confirm the importance of each core component in improving model performance. The findings offer technical support for optimizing carriage selection and enhancing metro service operations.

随着地铁客流量的不断增加，实时拥挤信息（RTCI）对于改善出行体验和更好地利用车厢空间变得至关重要。通过准确预测每节车厢下车的乘客数量，实时估计可用空间，可以指导等待的乘客更好的登机选择，减少拥堵。现有的RTCI研究大多集中在总客流预测或瞬时人群水平估计上，很少关注乘客下车行为如何直接影响车厢空间可用性。为了解决这个问题，本研究提出了三个主要贡献。(1)构造表征车厢动态状态的特征向量，以及描述可用空间演化的映射函数，为生成RTCI提供理论基础。(2)将EfficientNet-B0与LSTM （Long - Short-Term Memory，长短期记忆）技术相结合，建立了基于图像序列的深度学习模型，实现了对车厢下车人数的准确预测。(3)基于长沙地铁数据集的系统实验表明，该方法在保持计算效率的前提下实现了较高的预测精度。对比和消融研究证实了每个核心部件在提高模型性能方面的重要性。研究结果为优化车厢选择和提高地铁服务运营水平提供了技术支持。

{"title":"Metro passenger alighting flow prediction for real-time crowding information: A deep learning approach based on sequential images","authors":"Xiling Lin , Qun Chen , Yan Wang","doi":"10.1016/j.physa.2026.131322","DOIUrl":"10.1016/j.physa.2026.131322","url":null,"abstract":"<div><div>With the continued increase in metro ridership, real-time crowding information (RTCI) has become essential for improving travel experience and making better use of carriage space. By accurately predicting the number of passengers alighting from each carriage and estimating available space in real time, it is possible to guide waiting passengers to better boarding choices and reduce congestion. Most existing RTCI studies focus on total passenger flow prediction or instantaneous crowd level estimation, while paying little attention to how alighting behavior of passengers directly affects carriage space availability. To address this issue, this study presents three main contributions. (1) A feature vector is constructed to represent the dynamic state of a carriage, along with a mapping function that describes the evolution of available space, forming a theoretical basis for RTCI generation.(2) A deep learning model based on image sequences is developed by integrating EfficientNet-B0 with LSTM (Long Short-Term Memory), enabling accurate prediction of the number of passengers alighting at the carriage level. (3) Systematic experiments using a dataset from the Changsha metro demonstrate that the method achieves high prediction accuracy while maintaining computational efficiency. Comparative and ablation studies confirm the importance of each core component in improving model performance. The findings offer technical support for optimizing carriage selection and enhancing metro service operations.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"685 ","pages":"Article 131322"},"PeriodicalIF":3.1,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pedestrian evacuation in fire environments: Scientometric analysis and systematic evaluation 火灾环境下的行人疏散：科学计量分析和系统评价

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-25 DOI: 10.1016/j.physa.2026.131323

Xinyue Qi , Xiaolian Li , Jun Zhang , Weiguo Song

Understanding pedestrian evacuation behaviors in fire is critical to mitigate human casualties in emergencies. Benefiting from the rapid progress in fire safety engineering and computer simulation technologies, research on pedestrian evacuation in fire environments has attracted increasing attention from the academic community. To have a comprehensive understanding on the hotspots and development trends in this field, in this study a total of 673 relevant literatures from 2000 to 2024 have been collected from the database Web of Science and Scopus. The keywords co-occurrence network, authors' collaborative network, journal productivity, regional cooperation and research themes are analyzed. The results show that the amount of literature related to fire evacuation exhibit an exponential increase on the whole since 2015. The most productive journal is Physica A-statistical Mechanics and Its Applications, with 76 published articles, accounting for 11.3 % of the total research publications. The topic analysis shows that the research hotspots focus on the movement behavior and characteristics of the crowd in the fire environment, the simulation and optimization of the evacuation process, and the individual's risk perception and decision-making behavior. Further, the study identifies the key challenges for improving pedestrian emergency evacuation capabilities and suggests possible research directions.

了解火灾中行人的疏散行为对减轻突发事件中的人员伤亡至关重要。随着消防安全工程和计算机仿真技术的飞速发展，火灾环境下行人疏散的研究越来越受到学术界的重视。为了全面了解该领域的热点和发展趋势，本研究从Web of Science和Scopus数据库中收集了2000 - 2024年的673篇相关文献。分析了共现网络、作者协同网络、期刊生产力、区域合作和研究主题等关键词。结果表明：2015年以来，与火灾疏散相关的文献数量总体呈指数增长。产量最高的期刊是《物理学a -统计力学及其应用》，发表了76篇论文，占研究出版物总数的11.3%。课题分析表明，研究热点集中在火灾环境下人群的运动行为与特征、疏散过程的模拟与优化、个体的风险感知与决策行为等方面。此外，该研究还确定了提高行人紧急疏散能力的主要挑战，并提出了可能的研究方向。

{"title":"Pedestrian evacuation in fire environments: Scientometric analysis and systematic evaluation","authors":"Xinyue Qi , Xiaolian Li , Jun Zhang , Weiguo Song","doi":"10.1016/j.physa.2026.131323","DOIUrl":"10.1016/j.physa.2026.131323","url":null,"abstract":"<div><div>Understanding pedestrian evacuation behaviors in fire is critical to mitigate human casualties in emergencies. Benefiting from the rapid progress in fire safety engineering and computer simulation technologies, research on pedestrian evacuation in fire environments has attracted increasing attention from the academic community. To have a comprehensive understanding on the hotspots and development trends in this field, in this study a total of 673 relevant literatures from 2000 to 2024 have been collected from the database Web of Science and Scopus. The keywords co-occurrence network, authors' collaborative network, journal productivity, regional cooperation and research themes are analyzed. The results show that the amount of literature related to fire evacuation exhibit an exponential increase on the whole since 2015. The most productive journal is Physica A-statistical Mechanics and Its Applications, with 76 published articles, accounting for 11.3 % of the total research publications. The topic analysis shows that the research hotspots focus on the movement behavior and characteristics of the crowd in the fire environment, the simulation and optimization of the evacuation process, and the individual's risk perception and decision-making behavior. Further, the study identifies the key challenges for improving pedestrian emergency evacuation capabilities and suggests possible research directions.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131323"},"PeriodicalIF":3.1,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on the macroscopic characteristics of heterogeneous traffic flow under the influence of autonomous vehicle takeover behavior 自主车辆接管行为影响下的异构交通流宏观特征研究

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-25 DOI: 10.1016/j.physa.2026.131325

Lili Yang, Shicheng Ma, Hongfei Jia, Tianze Ma, Zixuan Mao

The study of macroscopic characteristics in heterogeneous traffic flows incorporating L3 automated vehicle(TOV) is of significant importance for advancing traffic flow theory, enhancing the robustness of cooperative control strategies for connected and automated vehicles, and supporting high precision traffic management. To address the research gap concerning takeover behavior in macroscopic fundamental diagram(MFD) studies, this paper develops an integrated methodology combining theoretical and experimental modeling. The theoretical framework incorporates the dual phase operation of TOV, heterogeneous car following modes, and cooperative adaptive cruise control(CACC) degradation mechanisms. For experimental framework, we implemented a two phase, multi cycle experimental framework through Sumo and Python cosimulation, where an enhanced underwood model was calibrated against experimental data to establish the experimental MFD. The validation demonstrated strong consistency between theoretical and experimental models. The analysis reveals that while increased penetration of TOV substantially improves road capacity (achieving up to 184.1 % enhancement in takeover free scenarios), takeover behavior significantly undermines this benefit (causing up to 62.3 % capacity reduction under high takeover rates). These findings provide valuable insights for developing effective traffic management strategies and designing resilient automated driving systems.

研究L3自动驾驶汽车（TOV）的异构交通流宏观特征，对于推进交通流理论，增强车联网和自动驾驶车辆协同控制策略的鲁棒性，支持高精度交通管理具有重要意义。为了解决宏观基本图（MFD）研究中关于收购行为的研究空白，本文发展了一种理论与实验相结合的方法。该理论框架结合了TOV的双相位运行、异质跟车模式和协同自适应巡航控制（CACC）退化机制。对于实验框架，我们通过Sumo和Python联合仿真实现了一个两阶段，多周期的实验框架，其中根据实验数据校准了增强的underwood模型，以建立实验MFD。验证表明理论模型与实验模型具有较强的一致性。分析表明，虽然TOV的普及大大提高了道路容量（在无接管的情况下，可提高184.1 %），但接管行为显著破坏了这一效益（在高接管率下，可导致高达62.3 %的容量减少）。这些发现为制定有效的交通管理策略和设计弹性自动驾驶系统提供了有价值的见解。

{"title":"Research on the macroscopic characteristics of heterogeneous traffic flow under the influence of autonomous vehicle takeover behavior","authors":"Lili Yang, Shicheng Ma, Hongfei Jia, Tianze Ma, Zixuan Mao","doi":"10.1016/j.physa.2026.131325","DOIUrl":"10.1016/j.physa.2026.131325","url":null,"abstract":"<div><div>The study of macroscopic characteristics in heterogeneous traffic flows incorporating L3 automated vehicle(TOV) is of significant importance for advancing traffic flow theory, enhancing the robustness of cooperative control strategies for connected and automated vehicles, and supporting high precision traffic management. To address the research gap concerning takeover behavior in macroscopic fundamental diagram(MFD) studies, this paper develops an integrated methodology combining theoretical and experimental modeling. The theoretical framework incorporates the dual phase operation of TOV, heterogeneous car following modes, and cooperative adaptive cruise control(CACC) degradation mechanisms. For experimental framework, we implemented a two phase, multi cycle experimental framework through Sumo and Python cosimulation, where an enhanced underwood model was calibrated against experimental data to establish the experimental MFD. The validation demonstrated strong consistency between theoretical and experimental models. The analysis reveals that while increased penetration of TOV substantially improves road capacity (achieving up to 184.1 % enhancement in takeover free scenarios), takeover behavior significantly undermines this benefit (causing up to 62.3 % capacity reduction under high takeover rates). These findings provide valuable insights for developing effective traffic management strategies and designing resilient automated driving systems.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"686 ","pages":"Article 131325"},"PeriodicalIF":3.1,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Behavioral dynamics of a modified car-following model integrating multiple delays and driver reaction to the preceding vehicle taillight signals 集成多重延迟和驾驶员对前车尾灯信号反应的改进汽车跟随模型的行为动力学

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-24 DOI: 10.1016/j.physa.2026.131293

Md. Zakir Hosen , Md. Anowar Hossain , Jun Tanimoto

Within the Intelligent Transportation System (ITS) applications, numerous studies have been conducted using various analytical frameworks to examine traffic flow dynamics. By adopting these technologies, drivers can achieve intelligent driving by receiving enhanced traffic information. Our proposed modified car-following model incorporates several factors, such as multiple time delays and human reaction to the preceding vehicle taillight signals. This study aims to present an improved traffic model that incorporates three crucial features, accounting for multiple time delays in sensing both headway and velocity information and human reaction delay to the preceding vehicle taillight signals to optimize traffic flow and diminish traffic instability, building upon the Taillight Adaptive Model (TAM) and the Full Velocity Difference (FVD) model. To investigate the influence of these multiple delays, we perform linear analysis, nonlinear analysis, and numerical simulations of our proposed model. Neutral stability conditions have been derived using linear stability analysis theory, demonstrating stable, metastable, and unstable regions. Employing nonlinear theory, three distinct nonlinear wave equations are obtained, namely the Burgers’ equation, Korteweg–de Vries (KdV) equation, and modified Korteweg–de Vries (mKdV) equation, which characterize soliton waves and kink-antikink wave solutions in the stable, metastable, and unstable regions, respectively. Finally, comprehensive numerical simulations have been conducted to validate our proposed model and depict the dynamic evolution of traffic flow under various parameter configurations. The obtained outcomes demonstrate that reducing distance-sensing delay, along with decreasing human reaction delay to the preceding vehicle’s taillight signals, substantially suppresses traffic congestion, while the velocity-sensing delay exhibits an opposite effect. The analytical and numerical simulation results demonstrate strong mutual consistency, validating the theoretical framework.

在智能交通系统（ITS）的应用中，已经进行了许多研究，使用各种分析框架来检查交通流量动态。通过采用这些技术，驾驶员可以通过接收增强的交通信息实现智能驾驶。我们提出的改进的汽车跟随模型包含了多个因素，如多重时间延迟和人类对前车尾灯信号的反应。在尾灯自适应模型（TAM）和全速度差（FVD）模型的基础上，提出了一种改进的交通模型，该模型结合了三个关键特征，考虑了车头时距和速度信息的多重时间延迟以及人类对前车尾灯信号的反应延迟，以优化交通流并减少交通不稳定性。为了研究这些多重延迟的影响，我们对我们提出的模型进行了线性分析、非线性分析和数值模拟。中性稳定条件已导出使用线性稳定性分析理论，证明稳定，亚稳和不稳定区域。利用非线性理论，得到了三个不同的非线性波动方程，即Burgers方程、Korteweg-de Vries （KdV）方程和修正Korteweg-de Vries （mKdV）方程，分别表征了稳定、亚稳和不稳定区域的孤子波和扭结-反扭结波解。最后进行了全面的数值模拟，验证了模型的有效性，并描述了不同参数配置下交通流的动态演化过程。研究结果表明，减少距离感知延迟，以及人类对前车尾灯信号的反应延迟，可以显著抑制交通拥堵，而速度感知延迟则表现出相反的效果。分析结果与数值模拟结果具有较强的一致性，验证了理论框架。

{"title":"Behavioral dynamics of a modified car-following model integrating multiple delays and driver reaction to the preceding vehicle taillight signals","authors":"Md. Zakir Hosen , Md. Anowar Hossain , Jun Tanimoto","doi":"10.1016/j.physa.2026.131293","DOIUrl":"10.1016/j.physa.2026.131293","url":null,"abstract":"<div><div>Within the Intelligent Transportation System (ITS) applications, numerous studies have been conducted using various analytical frameworks to examine traffic flow dynamics. By adopting these technologies, drivers can achieve intelligent driving by receiving enhanced traffic information. Our proposed modified car-following model incorporates several factors, such as multiple time delays and human reaction to the preceding vehicle taillight signals. This study aims to present an improved traffic model that incorporates three crucial features, accounting for multiple time delays in sensing both headway and velocity information and human reaction delay to the preceding vehicle taillight signals to optimize traffic flow and diminish traffic instability, building upon the Taillight Adaptive Model (TAM) and the Full Velocity Difference (FVD) model. To investigate the influence of these multiple delays, we perform linear analysis, nonlinear analysis, and numerical simulations of our proposed model. Neutral stability conditions have been derived using linear stability analysis theory, demonstrating stable, metastable, and unstable regions. Employing nonlinear theory, three distinct nonlinear wave equations are obtained, namely the Burgers’ equation, Korteweg–de Vries (KdV) equation, and modified Korteweg–de Vries (mKdV) equation, which characterize soliton waves and kink-antikink wave solutions in the stable, metastable, and unstable regions, respectively. Finally, comprehensive numerical simulations have been conducted to validate our proposed model and depict the dynamic evolution of traffic flow under various parameter configurations. The obtained outcomes demonstrate that reducing distance-sensing delay, along with decreasing human reaction delay to the preceding vehicle’s taillight signals, substantially suppresses traffic congestion, while the velocity-sensing delay exhibits an opposite effect. The analytical and numerical simulation results demonstrate strong mutual consistency, validating the theoretical framework.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"685 ","pages":"Article 131293"},"PeriodicalIF":3.1,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Graph machine learning for flight delay prediction due to holding maneuver 基于图机学习的等待机动飞行延误预测

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2026-01-23 DOI: 10.1016/j.physa.2026.131318

Jorge L. Franco , Manoel V. Machado Neto , Filipe A.N. Verri , Diego R. Amancio

Flight delays due to holding maneuvers are a critical and costly phenomenon in aviation, driven by the need to manage air traffic congestion and ensure safety. Holding maneuvers occur when aircraft are instructed to circle in designated airspace, often due to factors such as airport congestion, adverse weather, or air traffic control restrictions. This study models the prediction of flight delays due to holding maneuvers as a graph problem, leveraging advanced Graph Machine Learning (Graph ML) techniques to capture complex interdependencies in air traffic networks. Holding maneuvers, while crucial for safety, cause increased fuel usage, emissions, and passenger dissatisfaction, making accurate prediction essential for operational efficiency. Traditional machine learning models, typically using tabular data, often overlook spatial–temporal relations within air traffic data. To address this, we model the problem of predicting holding as edge feature prediction in a directed (multi)graph where we apply both CatBoost, enriched with graph features capturing network centrality and connectivity, and Graph Attention Networks (GATs), which excel in relational data contexts. Our results indicate that CatBoost outperforms GAT in this imbalanced dataset, effectively predicting holding events and offering interpretability through graph-based feature importance. Additionally, we discuss the model’s potential operational impact through a web-based tool that allows users to simulate real-time delay predictions. This research underscores the viability of graph-based approaches for predictive analysis in aviation, with implications for enhancing fuel efficiency, reducing delays, and improving passenger experience.

由于需要管理空中交通拥堵和确保安全，由于等待演习导致的航班延误是航空领域一个重要且代价高昂的现象。当飞机被指示在指定空域盘旋时，通常是由于机场拥堵、恶劣天气或空中交通管制限制等因素。本研究将等待机动导致的航班延误预测建模为一个图问题，利用先进的图机器学习（graph ML）技术来捕获空中交通网络中复杂的相互依赖关系。保持机动虽然对安全至关重要，但会增加燃料使用、排放和乘客不满，因此准确的预测对运营效率至关重要。传统的机器学习模型通常使用表格数据，经常忽略空中交通数据中的时空关系。为了解决这个问题，我们将预测持有的问题建模为有向（多）图中的边缘特征预测，其中我们应用CatBoost（丰富了捕获网络中心性和连接性的图特征）和图注意网络（GATs），它们在关系数据环境中表现出色。我们的结果表明，CatBoost在这种不平衡数据集中优于GAT，有效地预测了持有事件，并通过基于图的特征重要性提供了可解释性。此外，我们通过一个基于网络的工具讨论了该模型的潜在操作影响，该工具允许用户模拟实时延迟预测。这项研究强调了基于图表的方法在航空预测分析中的可行性，对提高燃油效率、减少延误和改善乘客体验具有重要意义。

{"title":"Graph machine learning for flight delay prediction due to holding maneuver","authors":"Jorge L. Franco , Manoel V. Machado Neto , Filipe A.N. Verri , Diego R. Amancio","doi":"10.1016/j.physa.2026.131318","DOIUrl":"10.1016/j.physa.2026.131318","url":null,"abstract":"<div><div>Flight delays due to holding maneuvers are a critical and costly phenomenon in aviation, driven by the need to manage air traffic congestion and ensure safety. Holding maneuvers occur when aircraft are instructed to circle in designated airspace, often due to factors such as airport congestion, adverse weather, or air traffic control restrictions. This study models the prediction of flight delays due to holding maneuvers as a graph problem, leveraging advanced Graph Machine Learning (Graph ML) techniques to capture complex interdependencies in air traffic networks. Holding maneuvers, while crucial for safety, cause increased fuel usage, emissions, and passenger dissatisfaction, making accurate prediction essential for operational efficiency. Traditional machine learning models, typically using tabular data, often overlook spatial–temporal relations within air traffic data. To address this, we model the problem of predicting holding as edge feature prediction in a directed (multi)graph where we apply both CatBoost, enriched with graph features capturing network centrality and connectivity, and Graph Attention Networks (GATs), which excel in relational data contexts. Our results indicate that CatBoost outperforms GAT in this imbalanced dataset, effectively predicting holding events and offering interpretability through graph-based feature importance. Additionally, we discuss the model’s potential operational impact through a web-based tool that allows users to simulate real-time delay predictions. This research underscores the viability of graph-based approaches for predictive analysis in aviation, with implications for enhancing fuel efficiency, reducing delays, and improving passenger experience.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"685 ","pages":"Article 131318"},"PeriodicalIF":3.1,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0