Computers in Industry最新文献_第9页

Evaluating calibration of deep fault diagnostic models under distribution shift 分布移位下深断层诊断模型的标定评价

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-07-04 DOI: 10.1016/j.compind.2025.104334

Yiming Xiao , Haidong Shao , Bin Liu

Current intelligent fault diagnosis studies focus on improving model accuracy. While accuracy is crucial, an exclusive emphasis on this metric can leave users oblivious to potentially untrustworthy decisions made by the model. This underscores the importance of confidence estimation and brings the model miscalibration problem to the forefront, i.e., the softmax probability, which is supposed to indicate the likelihood of the predicted label being correct but fails to reflect the true probability accurately. Addressing this issue is imperative for several reasons. Firstly, a well-calibrated model can provide users with an assessment of the risk associated with prediction failures, thereby withholding decision-making when the confidence is low and mitigating the risk of erroneous outputs. Especially in situations involving out-of-distribution (OOD) and distribution-shifted inputs, where the risk of model failure increases, the calibration property becomes even more critical. Secondly, well-calibrated confidence estimates can enhance users’ trust in today’s many black-box models. However, there have been limited fault diagnosis studies that specifically explore model calibration. The effectiveness of existing calibration methods in handling OOD and distribution-shifted inputs also remains unclear. Therefore, this paper evaluates multiple calibration methods and discusses their advantages and limitations, providing insights for subsequent studies. The results suggest that a deep ensemble method, which derives predictive expectations using multiple models with significantly different structures or parameters, has the potential to be the best calibration method. Code used in this paper is available at https://github.com/xiaoyiming1999/Calibration_for_RMFD.

目前的智能故障诊断研究主要集中在提高模型精度上。虽然准确性是至关重要的，但只强调这个指标可能会让用户忽略模型做出的可能不可信的决策。这强调了置信度估计的重要性，并将模型误校准问题带到了最前沿，即softmax概率，它应该表明预测标签正确的可能性，但不能准确反映真实的概率。出于几个原因，解决这个问题是必要的。首先，一个校准良好的模型可以为用户提供与预测失败相关的风险评估，从而在置信度较低时保留决策并降低错误输出的风险。特别是在涉及分布外（OOD）和分布移位输入的情况下，模型失效的风险增加，校准特性变得更加关键。其次，校准良好的置信度估计可以增强用户对当今许多黑盒模型的信任。然而，专门探讨模型校准的故障诊断研究有限。现有的校准方法在处理OOD和分布移位输入方面的有效性仍然不清楚。因此，本文对多种校准方法进行了评估，并讨论了它们的优点和局限性，为后续研究提供见解。结果表明，利用结构或参数显著不同的多个模型推导预测期望的深度集成方法有可能成为最佳的校准方法。本文中使用的代码可在https://github.com/xiaoyiming1999/Calibration_for_RMFD上获得。

{"title":"Evaluating calibration of deep fault diagnostic models under distribution shift","authors":"Yiming Xiao , Haidong Shao , Bin Liu","doi":"10.1016/j.compind.2025.104334","DOIUrl":"10.1016/j.compind.2025.104334","url":null,"abstract":"<div><div>Current intelligent fault diagnosis studies focus on improving model accuracy. While accuracy is crucial, an exclusive emphasis on this metric can leave users oblivious to potentially untrustworthy decisions made by the model. This underscores the importance of confidence estimation and brings the model miscalibration problem to the forefront, i.e., the softmax probability, which is supposed to indicate the likelihood of the predicted label being correct but fails to reflect the true probability accurately. Addressing this issue is imperative for several reasons. Firstly, a well-calibrated model can provide users with an assessment of the risk associated with prediction failures, thereby withholding decision-making when the confidence is low and mitigating the risk of erroneous outputs. Especially in situations involving out-of-distribution (OOD) and distribution-shifted inputs, where the risk of model failure increases, the calibration property becomes even more critical. Secondly, well-calibrated confidence estimates can enhance users’ trust in today’s many black-box models. However, there have been limited fault diagnosis studies that specifically explore model calibration. The effectiveness of existing calibration methods in handling OOD and distribution-shifted inputs also remains unclear. Therefore, this paper evaluates multiple calibration methods and discusses their advantages and limitations, providing insights for subsequent studies. The results suggest that a deep ensemble method, which derives predictive expectations using multiple models with significantly different structures or parameters, has the potential to be the best calibration method. Code used in this paper is available at <span><span>https://github.com/xiaoyiming1999/Calibration_for_RMFD</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104334"},"PeriodicalIF":8.2,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144563666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing retrieval-augmented generation for interoperable industrial knowledge representation and inference toward cognitive digital twins 面向认知数字孪生的可互操作工业知识表示和推理的检索增强生成

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-06-26 DOI: 10.1016/j.compind.2025.104330

Dachuan Shi , Jianzhang Li , Olga Meyer , Thomas Bauernhansl

The escalating volume and complexity of digital data within the manufacturing sector highlight an urgent need for an efficient knowledge representation and inference solution. Traditional approaches, which often rely on ontologies, knowledge graphs, or digital twins (DTs) for knowledge representation, and rule-based algorithms for inference, are becoming insufficient. The emergence of generative AI, particularly large language models (LLM) and retrieval-augmented generation (RAG), offers a more efficient and intelligent alternative. However, the performance of an RAG system is heavily dependent on the quality of retrieval results, which can be compromised by domain-specific knowledge and retrieval distractors. To address this challenge, we propose to enhance RAG systems tailored for the manufacturing industry in two aspects. First, we utilize the Asset Administration Shell (AAS), which represents the German industrial perspective on cognitive DTs, to create a representation of assets and knowledge in standardized information models. This establishes a robust foundation for the retrieval sources. Second, we propose a contrastive selection loss (CSL) to fine-tune an open-source LLM to refine the retrieval results. Fine-tuned LLMs possess higher efficiency and accuracy on task- and domain-specific datasets, while the CSL further enhances the model's ability to distinguish true positives from similar distractors. The enhanced RAG system is demonstrated in a robotic work cell integration use case and evaluated through a novel evaluation protocol. Additionally, the retrieval effectiveness of the RAG system, specifically the LLM fine-tuned with CSL, is extensively validated through statistical experiments. The results confirm its superior performance over state-of-the-art methods, including GPT-4 with in-context learning prompts and other fine-tuned models.

制造业中不断增长的数字数据量和复杂性凸显了对高效知识表示和推理解决方案的迫切需求。传统的方法通常依赖于本体、知识图或数字孪生（DTs）来表示知识，以及基于规则的推理算法，这些方法已经变得不够用了。生成式人工智能的出现，尤其是大型语言模型（LLM）和检索增强生成（RAG），提供了一种更高效、更智能的替代方案。然而，RAG系统的性能在很大程度上依赖于检索结果的质量，这可能会受到领域特定知识和检索干扰因素的影响。为了应对这一挑战，我们建议从两个方面加强为制造业量身定制的RAG系统。首先，我们利用资产管理外壳（AAS），它代表了德国工业对认知dt的看法，在标准化信息模型中创建资产和知识的表示。这为检索源建立了坚实的基础。其次，我们提出了一种对比选择损失（CSL）来微调开源LLM以优化检索结果。微调llm在任务和领域特定数据集上具有更高的效率和准确性，而CSL进一步增强了模型区分真实阳性和类似干扰物的能力。增强的RAG系统在机器人工作单元集成用例中进行了演示，并通过一种新的评估协议进行了评估。此外，RAG系统的检索有效性，特别是与CSL微调的LLM，通过统计实验得到了广泛的验证。结果证实了它优于最先进的方法，包括具有上下文学习提示和其他微调模型的GPT-4。

{"title":"Enhancing retrieval-augmented generation for interoperable industrial knowledge representation and inference toward cognitive digital twins","authors":"Dachuan Shi , Jianzhang Li , Olga Meyer , Thomas Bauernhansl","doi":"10.1016/j.compind.2025.104330","DOIUrl":"10.1016/j.compind.2025.104330","url":null,"abstract":"<div><div>The escalating volume and complexity of digital data within the manufacturing sector highlight an urgent need for an efficient knowledge representation and inference solution. Traditional approaches, which often rely on ontologies, knowledge graphs, or digital twins (DTs) for knowledge representation, and rule-based algorithms for inference, are becoming insufficient. The emergence of generative AI, particularly large language models (LLM) and retrieval-augmented generation (RAG), offers a more efficient and intelligent alternative. However, the performance of an RAG system is heavily dependent on the quality of retrieval results, which can be compromised by domain-specific knowledge and retrieval distractors. To address this challenge, we propose to enhance RAG systems tailored for the manufacturing industry in two aspects. First, we utilize the Asset Administration Shell (AAS), which represents the German industrial perspective on cognitive DTs, to create a representation of assets and knowledge in standardized information models. This establishes a robust foundation for the retrieval sources. Second, we propose a contrastive selection loss (CSL) to fine-tune an open-source LLM to refine the retrieval results. Fine-tuned LLMs possess higher efficiency and accuracy on task- and domain-specific datasets, while the CSL further enhances the model's ability to distinguish true positives from similar distractors. The enhanced RAG system is demonstrated in a robotic work cell integration use case and evaluated through a novel evaluation protocol. Additionally, the retrieval effectiveness of the RAG system, specifically the LLM fine-tuned with CSL, is extensively validated through statistical experiments. The results confirm its superior performance over state-of-the-art methods, including GPT-4 with in-context learning prompts and other fine-tuned models.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104330"},"PeriodicalIF":8.2,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Vision-based hand pose estimation methods for Augmented Reality in industry: Crowdsourced evaluation on HoloLens 2 工业增强现实中基于视觉的手姿估计方法：HoloLens众包评估

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-06-24 DOI: 10.1016/j.compind.2025.104328

Kamil Żywanowski , Mikołaj Łysakowski , Michał R. Nowicki , Jason T. Jacques , Sławomir K. Tadeja , Thomas Bohné , Piotr Skrzypczyński

Gestural input based on hand pose estimation is a common interaction method for augmented reality (AR). This interaction technique has gained more popularity with the emergence of novel AR-supporting devices such as Microsoft HoloLens 2 (HL2) and advancements in computer vision research underpinning hand-tracking and gesture recognition methods. In our work, we focus on challenging cases where the AR interface is facilitated with a state-of-the-art HL2 headset for unconstrained execution of tasks requiring simultaneous hand movement and tracking. When using this headset, AR users might bimanually interact with digital and physical objects that are visible in the user’s field of view (FoV) through the see-through visor. Due to the limiting in-built capabilities, we investigated a range of hand pose estimation functionalities from different domains. To ensure a fair comparison, we asked several participants to carry out tasks requiring interactions with real-world objects and record the performance of various hand-tracking solutions. Next, we evaluated the performance of these algorithms through crowdsourcing, often used to provide ground truth for machine learning training. Our results provide a guideline for AR developers in selecting appropriate hand-tracking solutions for a given deployment context.

基于手部姿态估计的手势输入是增强现实（AR）中常用的一种交互方法。随着新型ar支持设备（如Microsoft HoloLens 2 (HL2)）的出现以及支持手部跟踪和手势识别方法的计算机视觉研究的进步，这种交互技术越来越受欢迎。在我们的工作中，我们专注于具有挑战性的案例，其中AR界面与最先进的HL2耳机相结合，可以不受约束地执行需要同时进行手部运动和跟踪的任务。当使用这种头戴式耳机时，AR用户可以通过透明遮阳板与用户视野（FoV）中可见的数字和物理对象进行手动交互。由于有限的内置功能，我们研究了一系列来自不同领域的手姿估计功能。为了确保公平的比较，我们要求几位参与者执行需要与现实世界物体交互的任务，并记录各种手部跟踪解决方案的性能。接下来，我们通过众包来评估这些算法的性能，众包通常用于为机器学习训练提供基础事实。我们的研究结果为AR开发人员在给定的部署环境中选择合适的手部跟踪解决方案提供了指导。

{"title":"Vision-based hand pose estimation methods for Augmented Reality in industry: Crowdsourced evaluation on HoloLens 2","authors":"Kamil Żywanowski , Mikołaj Łysakowski , Michał R. Nowicki , Jason T. Jacques , Sławomir K. Tadeja , Thomas Bohné , Piotr Skrzypczyński","doi":"10.1016/j.compind.2025.104328","DOIUrl":"10.1016/j.compind.2025.104328","url":null,"abstract":"<div><div>Gestural input based on hand pose estimation is a common interaction method for augmented reality (AR). This interaction technique has gained more popularity with the emergence of novel AR-supporting devices such as Microsoft HoloLens 2 (HL2) and advancements in computer vision research underpinning hand-tracking and gesture recognition methods. In our work, we focus on challenging cases where the AR interface is facilitated with a state-of-the-art HL2 headset for unconstrained execution of tasks requiring simultaneous hand movement and tracking. When using this headset, AR users might bimanually interact with digital and physical objects that are visible in the user’s field of view (FoV) through the see-through visor. Due to the limiting in-built capabilities, we investigated a range of hand pose estimation functionalities from different domains. To ensure a fair comparison, we asked several participants to carry out tasks requiring interactions with real-world objects and record the performance of various hand-tracking solutions. Next, we evaluated the performance of these algorithms through crowdsourcing, often used to provide ground truth for machine learning training. Our results provide a guideline for AR developers in selecting appropriate hand-tracking solutions for a given deployment context.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104328"},"PeriodicalIF":8.2,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144365666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy consumption forecasting in non-stationary machining processes based on a physics-guided transformer 基于物理导向变压器的非平稳加工过程能耗预测

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-06-17 DOI: 10.1016/j.compind.2025.104321

Meihang Zhang , Ruiping Wang

Predicting energy consumption in non-stationary machining processes is challenging due to data complexity, dynamic variations, and real-time requirements. This paper proposes a novel physics-guided Transformer model incorporating supervisory-compensatory mechanisms. Firstly, several data preprocessing and feature extraction techniques, including Lagrange interpolation, wavelet transform, and a combined approach of principal component analysis and correlation analysis, are employed to enhance data quality and identify key physical variable. Secondly, modeling accuracy and efficiency are improved by integrating physics-guided variables into the traditional Transformer model, leading to advancements in the existing de-stationary attention module. Finally, distinct training and prediction models are established, incorporating a self-supervised, self-compensation mechanism in the training phase. This mechanism utilizes ground truth for training convergence, with optimized model parameters subsequently applied to the prediction model, significantly enhancing predictive efficacy. Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches, achieving improvements in energy consumption prediction accuracy of over 76 % for carbon fiber processing, 30 % for plastic, 32.7 % for aluminum, and 54.5 % for 45steel. The integration of physical principles with sequence modeling enables precise energy consumption forecasting in non-stationary machining, advancing both predictive accuracy and industrial energy management.

由于数据复杂性、动态变化和实时要求，预测非平稳加工过程中的能耗具有挑战性。本文提出了一种新的物理导向变压器模型，该模型包含了监督-补偿机制。首先，采用拉格朗日插值、小波变换、主成分分析与相关分析相结合的方法对数据进行预处理和特征提取，提高数据质量，识别关键物理变量；其次，通过将物理导向变量集成到传统的Transformer模型中，提高了建模的精度和效率，从而改进了现有的去平稳注意力模块。最后，建立了不同的训练和预测模型，并在训练阶段引入了自我监督、自我补偿机制。该机制利用ground truth进行训练收敛，并将优化后的模型参数应用到预测模型中，显著提高了预测效果。实验结果表明，所提出的方法优于最先进的方法，碳纤维加工的能耗预测精度提高了76% %，塑料加工的能耗预测精度提高了30% %，铝加工的能耗预测精度提高了32.7% %，钢加工的能耗预测精度提高了54.5 %。将物理原理与序列建模相结合，可以在非平稳加工中实现精确的能耗预测，从而提高预测精度和工业能源管理。

{"title":"Energy consumption forecasting in non-stationary machining processes based on a physics-guided transformer","authors":"Meihang Zhang , Ruiping Wang","doi":"10.1016/j.compind.2025.104321","DOIUrl":"10.1016/j.compind.2025.104321","url":null,"abstract":"<div><div>Predicting energy consumption in non-stationary machining processes is challenging due to data complexity, dynamic variations, and real-time requirements. This paper proposes a novel physics-guided Transformer model incorporating supervisory-compensatory mechanisms. Firstly, several data preprocessing and feature extraction techniques, including Lagrange interpolation, wavelet transform, and a combined approach of principal component analysis and correlation analysis, are employed to enhance data quality and identify key physical variable. Secondly, modeling accuracy and efficiency are improved by integrating physics-guided variables into the traditional Transformer model, leading to advancements in the existing de-stationary attention module. Finally, distinct training and prediction models are established, incorporating a self-supervised, self-compensation mechanism in the training phase. This mechanism utilizes ground truth for training convergence, with optimized model parameters subsequently applied to the prediction model, significantly enhancing predictive efficacy. Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches, achieving improvements in energy consumption prediction accuracy of over 76 % for carbon fiber processing, 30 % for plastic, 32.7 % for aluminum, and 54.5 % for 45steel. The integration of physical principles with sequence modeling enables precise energy consumption forecasting in non-stationary machining, advancing both predictive accuracy and industrial energy management.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104321"},"PeriodicalIF":8.2,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MiniMaxAD: A lightweight autoencoder for feature-rich anomaly detection MiniMaxAD：一个轻量级的自动编码器，用于功能丰富的异常检测

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-06-05 DOI: 10.1016/j.compind.2025.104315

Fengjie Wang, Chengming Liu, Lei Shi, Haibo Pang

Previous industrial anomaly detection (IAD) methods often struggle to handle the extensive diversity in training sets, particularly when they contain stylistically diverse and feature-rich samples, which we categorize as feature-rich anomaly detection datasets (FRADs). This challenge is evident in applications such as multi-view and multi-class scenarios. To address this challenge, we developed MiniMaxAD, a efficient autoencoder designed to efficiently compress and memorize extensive information from normal images. Our model employs a technique that enhances feature diversity, thereby increasing the effective capacity of the network. It also utilizes large kernel convolution to extract highly abstract patterns, which contribute to efficient and compact feature embedding. Moreover, we introduce an Adaptive Contraction Hard Mining Loss (ADCLoss), specifically tailored to FRADs. In our methodology, any dataset can be unified under the framework of feature-rich anomaly detection, in a way that the benefits far outweigh the drawbacks. Our approach has achieved state-of-the-art performance in multiple challenging benchmarks. Code is available at: https://github.com/WangFengJiee/MiniMaxAD.

以前的工业异常检测（IAD）方法通常难以处理训练集的广泛多样性，特别是当它们包含风格多样化和特征丰富的样本时，我们将其分类为特征丰富的异常检测数据集（FRADs）。这一挑战在多视图和多类场景等应用程序中很明显。为了应对这一挑战，我们开发了MiniMaxAD，这是一种高效的自动编码器，旨在有效地压缩和记忆来自正常图像的大量信息。我们的模型采用了一种增强特征多样性的技术，从而提高了网络的有效容量。它还利用大核卷积提取高度抽象的模式，有助于高效紧凑的特征嵌入。此外，我们还引入了一种专门为FRADs量身定制的自适应收缩硬挖掘损失（ADCLoss）。在我们的方法中，任何数据集都可以统一在特征丰富的异常检测框架下，以一种利远大于弊的方式。我们的方法在多个具有挑战性的基准测试中取得了最先进的性能。代码可从https://github.com/WangFengJiee/MiniMaxAD获得。

{"title":"MiniMaxAD: A lightweight autoencoder for feature-rich anomaly detection","authors":"Fengjie Wang, Chengming Liu, Lei Shi, Haibo Pang","doi":"10.1016/j.compind.2025.104315","DOIUrl":"10.1016/j.compind.2025.104315","url":null,"abstract":"<div><div>Previous industrial anomaly detection (IAD) methods often struggle to handle the extensive diversity in training sets, particularly when they contain stylistically diverse and feature-rich samples, which we categorize as feature-rich anomaly detection datasets (FRADs). This challenge is evident in applications such as multi-view and multi-class scenarios. To address this challenge, we developed MiniMaxAD, a efficient autoencoder designed to efficiently compress and memorize extensive information from normal images. Our model employs a technique that enhances feature diversity, thereby increasing the effective capacity of the network. It also utilizes large kernel convolution to extract highly abstract patterns, which contribute to efficient and compact feature embedding. Moreover, we introduce an Adaptive Contraction Hard Mining Loss (ADCLoss), specifically tailored to FRADs. In our methodology, any dataset can be unified under the framework of feature-rich anomaly detection, in a way that the benefits far outweigh the drawbacks. Our approach has achieved state-of-the-art performance in multiple challenging benchmarks. Code is available at: <span><span>https://github.com/WangFengJiee/MiniMaxAD</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104315"},"PeriodicalIF":8.2,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144212331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Autonomous vehicle crash risk modeling by integrating data augmentation and two-layer stacking 集成数据增强和两层叠加的自动驾驶汽车碰撞风险建模

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-06-03 DOI: 10.1016/j.compind.2025.104320

Leipeng Zhu , Zhiqing Zhang , Yongnan Zhang , Jingyang Yu , Hongjia Wang

Autonomous vehicle (AV) technology aims to eliminate traffic crashes caused by driver errors, but its adoption has introduced new types of crashes. Due to the high dimensionality and limited sample size of AV crash data, identifying underlying risk factors remains challenging, and crash predictive performance is often suboptimal. To address these issues, this study develops an interpretable data augmentation strategy and the optimized two-layer stacking algorithm, further integrating them into a unified framework that accurately identifies key crash contributing factors and significantly improves predictive performance. The findings reveal that: 1) AV crashes show significant variation in their temporal distributions but follow consistent spatial agglomeration patterns. 2) AV reliability significantly decreases in high-interaction scenarios, with peak travel times and uncertain road conditions identified as key contributing factors. 3) The data augmentation algorithm enhances on key contributing factors and the feature crosses, enhances the model’s ability to capture nonlinear relationships in crash data and improves predictive accuracy in small-sample scenarios, particularly for injury-related crashes. 4) The optimized two-layer stacking algorithm integrates the heterogeneous learning capabilities of models such as LightGBM and Random Forest, significantly improving the ability to recognize complex crash patterns. When combined with data augmentation, the framework achieves strong predictive performance, with both precision and recall reaching 0.92 and the area under the receiver operating characteristic curve at 0.96. Compared to existing machine learning approaches, this framework shows notable advantages in handling high-dimensional small-sample AV crash data. The framework provides an effective solution for AV crash risk modeling and safety design, contributing to the development and implementation of safer intelligent transportation systems.

自动驾驶汽车（AV）技术的目的是消除驾驶员失误造成的交通事故，但它的采用引入了新的交通事故类型。由于自动驾驶汽车碰撞数据的高维度和有限的样本量，识别潜在的风险因素仍然具有挑战性，并且碰撞预测性能通常不是最优的。为了解决这些问题，本研究开发了一种可解释的数据增强策略和优化的两层叠加算法，并将它们进一步整合到一个统一的框架中，从而准确识别关键的崩溃因素，显著提高预测性能。结果表明：1)AV崩溃在时间分布上存在显著差异，但在空间集聚上具有一致性；2)在高交互场景下，自动驾驶汽车的可靠性显著下降，高峰出行时间和不确定路况是主要影响因素。3)数据增强算法对关键影响因素和特征交叉进行了增强，增强了模型捕捉碰撞数据非线性关系的能力，提高了小样本场景下，特别是伤害相关碰撞的预测精度。4)优化后的两层叠加算法融合了LightGBM、Random Forest等模型的异构学习能力，显著提高了对复杂碰撞模式的识别能力。结合数据增强，该框架具有较强的预测性能，准确率和召回率均达到0.92，接收者工作特征曲线下面积达到0.96。与现有的机器学习方法相比，该框架在处理高维小样本自动驾驶汽车碰撞数据方面显示出显著的优势。该框架为自动驾驶汽车碰撞风险建模和安全设计提供了有效的解决方案，有助于开发和实施更安全的智能交通系统。

{"title":"Autonomous vehicle crash risk modeling by integrating data augmentation and two-layer stacking","authors":"Leipeng Zhu , Zhiqing Zhang , Yongnan Zhang , Jingyang Yu , Hongjia Wang","doi":"10.1016/j.compind.2025.104320","DOIUrl":"10.1016/j.compind.2025.104320","url":null,"abstract":"<div><div>Autonomous vehicle (AV) technology aims to eliminate traffic crashes caused by driver errors, but its adoption has introduced new types of crashes. Due to the high dimensionality and limited sample size of AV crash data, identifying underlying risk factors remains challenging, and crash predictive performance is often suboptimal. To address these issues, this study develops an interpretable data augmentation strategy and the optimized two-layer stacking algorithm, further integrating them into a unified framework that accurately identifies key crash contributing factors and significantly improves predictive performance. The findings reveal that: 1) AV crashes show significant variation in their temporal distributions but follow consistent spatial agglomeration patterns. 2) AV reliability significantly decreases in high-interaction scenarios, with peak travel times and uncertain road conditions identified as key contributing factors. 3) The data augmentation algorithm enhances on key contributing factors and the feature crosses, enhances the model’s ability to capture nonlinear relationships in crash data and improves predictive accuracy in small-sample scenarios, particularly for injury-related crashes. 4) The optimized two-layer stacking algorithm integrates the heterogeneous learning capabilities of models such as LightGBM and Random Forest, significantly improving the ability to recognize complex crash patterns. When combined with data augmentation, the framework achieves strong predictive performance, with both precision and recall reaching 0.92 and the area under the receiver operating characteristic curve at 0.96. Compared to existing machine learning approaches, this framework shows notable advantages in handling high-dimensional small-sample AV crash data. The framework provides an effective solution for AV crash risk modeling and safety design, contributing to the development and implementation of safer intelligent transportation systems.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"171 ","pages":"Article 104320"},"PeriodicalIF":8.2,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Industrial Internet of Things: Implementations, challenges, and potential solutions across various industries 工业物联网：跨不同行业的实现、挑战和潜在解决方案

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-05-28 DOI: 10.1016/j.compind.2025.104317

Shaila Afrin , Sabiha Jannat Rafa , Maliha Kabir , Tasfia Farah , Md. Sakib Bin Alam , Aiman Lameesa , Shams Forruque Ahmed , Amir H. Gandomi

The Industrial Internet of Things (IIoT) has emerged as a potent catalyst for transformation across many industries as a part of Industry 4.0. This review thoroughly examines IIoT applications, demonstrating how it enhances operational efficiency, informed decision-making, cost optimization, innovation, and workplace safety. While prior research has often concentrated on technical dimensions such as fog and edge computing, network protocols, or big data integration, several emerging and high-impact application areas remain underexplored. This study addresses that gap by systematically reviewing IIoT implementations in critical yet often overlooked domains, including environmental monitoring, agriculture, construction, healthcare, robotics, smart grids, and predictive maintenance. It offers fresh insights into how IIoT is being adapted to meet real-world challenges in these sectors. In addition to outlining the current landscape, the review identifies core barriers such as data security, interoperability, and system scalability. It underscores the importance of cross-sector collaboration and strategic alignment to fully leverage the transformative potential of IIoT. The paper concludes by outlining key research gaps and future opportunities to guide continued innovation and scholarly investigation.

作为工业4.0的一部分，工业物联网（IIoT）已经成为许多行业转型的有力催化剂。本文全面考察了工业物联网的应用，展示了它如何提高运营效率、明智决策、成本优化、创新和工作场所安全。虽然之前的研究通常集中在雾和边缘计算、网络协议或大数据集成等技术层面，但一些新兴和高影响力的应用领域仍未得到充分探索。本研究通过系统地回顾工业物联网在关键但经常被忽视的领域的实施情况，包括环境监测、农业、建筑、医疗保健、机器人、智能电网和预测性维护，解决了这一差距。它为如何适应工业物联网以应对这些领域的现实挑战提供了新的见解。除了概述当前形势外，审查还确定了核心障碍，如数据安全性、互操作性和系统可伸缩性。它强调了跨部门合作和战略协调的重要性，以充分利用工业物联网的变革潜力。论文最后概述了关键的研究差距和未来的机会，以指导持续的创新和学术研究。

{"title":"Industrial Internet of Things: Implementations, challenges, and potential solutions across various industries","authors":"Shaila Afrin , Sabiha Jannat Rafa , Maliha Kabir , Tasfia Farah , Md. Sakib Bin Alam , Aiman Lameesa , Shams Forruque Ahmed , Amir H. Gandomi","doi":"10.1016/j.compind.2025.104317","DOIUrl":"10.1016/j.compind.2025.104317","url":null,"abstract":"<div><div>The Industrial Internet of Things (IIoT) has emerged as a potent catalyst for transformation across many industries as a part of Industry 4.0. This review thoroughly examines IIoT applications, demonstrating how it enhances operational efficiency, informed decision-making, cost optimization, innovation, and workplace safety. While prior research has often concentrated on technical dimensions such as fog and edge computing, network protocols, or big data integration, several emerging and high-impact application areas remain underexplored. This study addresses that gap by systematically reviewing IIoT implementations in critical yet often overlooked domains, including environmental monitoring, agriculture, construction, healthcare, robotics, smart grids, and predictive maintenance. It offers fresh insights into how IIoT is being adapted to meet real-world challenges in these sectors. In addition to outlining the current landscape, the review identifies core barriers such as data security, interoperability, and system scalability. It underscores the importance of cross-sector collaboration and strategic alignment to fully leverage the transformative potential of IIoT. The paper concludes by outlining key research gaps and future opportunities to guide continued innovation and scholarly investigation.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"170 ","pages":"Article 104317"},"PeriodicalIF":8.2,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SRLFormer: Single Retinex-based and low-light image guidance Transformer for low-light image enhancement SRLFormer：基于单一视黄醇的低光图像引导变压器，用于低光图像增强

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-05-27 DOI: 10.1016/j.compind.2025.104314

Bin Wang, Bini Zhang, Jinfang Sheng

In image enhancement for low-illumination images, deep learning methods based on the Retinex theory typically decompose the image into illumination and reflectance, followed by iterative optimization or the use of prior custom enhancements. The reflectance map is then approximated as the enhanced image by dividing the radiance by the illumination map. However, this approach does not account for the noise hidden in low-illumination images or introduced during the enhancement of illumination. Additionally, it may cause computational overflow and amplify noise when the illumination in certain regions approaches ”0”. Moreover, these methods often require cumbersome multi-stage training and rely solely on convolutional neural networks, indicating limitations in capturing long-range dependencies. This paper proposes an efficient single-stage framework named SRF(Retinex-based single-retinex-based framework based on Retinex). SRF first estimates the inverse illumination image, then enhances the image by multiplying the inverse illumination with the low-illumination image, resulting in an image with improved brightness but still containing noise. Finally, we design a low-illumination guided Transformer network, LGF (Low-Illumination Guided Transformer), which utilizes the low-illumination image to guide denoising, thus more comprehensively considering the edge and detail information of the enhanced image. By integrating the LGT into SRF, we obtain the proposed algorithm SRLFormer. Experimental results show that SRLFormer significantly outperforms state-of-the-art methods in both qualitative and quantitative experiments, and its potential practical value is also demonstrated in downstream tasks and applications.

在低照度图像的图像增强中，基于Retinex理论的深度学习方法通常将图像分解为照度和反射率，然后进行迭代优化或使用先前的自定义增强。然后将反射率图近似为增强图像，方法是将亮度除以照度图。然而，这种方法没有考虑到低照度图像中隐藏的噪声或在增强照度过程中引入的噪声。此外，当某些区域的照明接近“0”时，可能会导致计算溢出和放大噪声。此外，这些方法通常需要繁琐的多阶段训练，并且仅依赖于卷积神经网络，这表明在捕获远程依赖关系方面存在局限性。本文提出了一种高效的单阶段框架SRF（Retinex-based single-retinex-based framework based on Retinex）。SRF首先对逆照度图像进行估计，然后将逆照度与低照度图像相乘对图像进行增强，得到亮度有所提高但仍含有噪声的图像。最后，我们设计了一种低照度引导变压器网络LGF (low-illumination guided Transformer)，它利用低照度图像来引导去噪，从而更全面地考虑了增强图像的边缘和细节信息。通过将LGT积分到SRF中，得到了提出的SRLFormer算法。实验结果表明，SRLFormer在定性和定量实验中都明显优于目前最先进的方法，并且在下游任务和应用中也证明了其潜在的实用价值。

{"title":"SRLFormer: Single Retinex-based and low-light image guidance Transformer for low-light image enhancement","authors":"Bin Wang, Bini Zhang, Jinfang Sheng","doi":"10.1016/j.compind.2025.104314","DOIUrl":"10.1016/j.compind.2025.104314","url":null,"abstract":"<div><div>In image enhancement for low-illumination images, deep learning methods based on the Retinex theory typically decompose the image into illumination and reflectance, followed by iterative optimization or the use of prior custom enhancements. The reflectance map is then approximated as the enhanced image by dividing the radiance by the illumination map. However, this approach does not account for the noise hidden in low-illumination images or introduced during the enhancement of illumination. Additionally, it may cause computational overflow and amplify noise when the illumination in certain regions approaches ”0”. Moreover, these methods often require cumbersome multi-stage training and rely solely on convolutional neural networks, indicating limitations in capturing long-range dependencies. This paper proposes an efficient single-stage framework named SRF(Retinex-based single-retinex-based framework based on Retinex). SRF first estimates the inverse illumination image, then enhances the image by multiplying the inverse illumination with the low-illumination image, resulting in an image with improved brightness but still containing noise. Finally, we design a low-illumination guided Transformer network, LGF (Low-Illumination Guided Transformer), which utilizes the low-illumination image to guide denoising, thus more comprehensively considering the edge and detail information of the enhanced image. By integrating the LGT into SRF, we obtain the proposed algorithm SRLFormer. Experimental results show that SRLFormer significantly outperforms state-of-the-art methods in both qualitative and quantitative experiments, and its potential practical value is also demonstrated in downstream tasks and applications.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"170 ","pages":"Article 104314"},"PeriodicalIF":8.2,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A multiscale process-aware retention network for fault prediction in mixed-model production 混合模型生产故障预测的多尺度过程感知保持网络

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-05-26 DOI: 10.1016/j.compind.2025.104313

Mingda Chen , Ruiyun Yu , Zhiyuan Liang , Kun Li , Haifei Qi

In the manufacturing industry, the demand for fault-prediction solutions is increasing to prevent unexpected downtimes and reduce maintenance costs. Although deep-learning methods have demonstrated excellent performance in this domain, the current methods typically overlook the analysis of variable and random processes within mixed-model production, which is a manufacturing strategy that offers flexibility and efficiency in satisfying diverse consumer demands. Hence, we propose the multiscale process-aware retention network (MPRNet), which segments a time series into multiscale patches, thus enabling the model to focus on local information within each production process and correlations across all production processes. Furthermore, the network incorporates a cross-channel interaction module designed to dynamically capture the interactions between various sensor data types using a graph attention network, as well as transmit fault information across processes using state equations. We validate our proposed model on the BBA stud welding gun dataset and four additional open case studies. Compared with other established fault-prediction and time-series models, the MPRNet demonstrates improved F1-score by 13.1% in the BBA case and consistently achieves the best or near-best results in the open case studies.

在制造业中，对故障预测解决方案的需求正在增加，以防止意外停机并降低维护成本。尽管深度学习方法在这一领域表现出色，但目前的方法通常忽略了混合模型生产中对变量和随机过程的分析，而混合模型生产是一种为满足不同消费者需求提供灵活性和效率的制造策略。因此，我们提出了多尺度过程感知保留网络（MPRNet），它将时间序列分割成多尺度补丁，从而使模型能够关注每个生产过程中的局部信息和所有生产过程之间的相关性。此外，该网络还集成了一个跨通道交互模块，旨在使用图关注网络动态捕获各种传感器数据类型之间的交互，并使用状态方程跨过程传输故障信息。我们在BBA螺柱焊枪数据集和另外四个开放案例研究上验证了我们提出的模型。与其他已建立的故障预测和时间序列模型相比，MPRNet在BBA情况下的f1得分提高了13.1%，在开放情况下的结果始终是最好或接近最好的。

{"title":"A multiscale process-aware retention network for fault prediction in mixed-model production","authors":"Mingda Chen , Ruiyun Yu , Zhiyuan Liang , Kun Li , Haifei Qi","doi":"10.1016/j.compind.2025.104313","DOIUrl":"10.1016/j.compind.2025.104313","url":null,"abstract":"<div><div>In the manufacturing industry, the demand for fault-prediction solutions is increasing to prevent unexpected downtimes and reduce maintenance costs. Although deep-learning methods have demonstrated excellent performance in this domain, the current methods typically overlook the analysis of variable and random processes within mixed-model production, which is a manufacturing strategy that offers flexibility and efficiency in satisfying diverse consumer demands. Hence, we propose the multiscale process-aware retention network (MPRNet), which segments a time series into multiscale patches, thus enabling the model to focus on local information within each production process and correlations across all production processes. Furthermore, the network incorporates a cross-channel interaction module designed to dynamically capture the interactions between various sensor data types using a graph attention network, as well as transmit fault information across processes using state equations. We validate our proposed model on the BBA stud welding gun dataset and four additional open case studies. Compared with other established fault-prediction and time-series models, the MPRNet demonstrates improved F1-score by 13.1% in the BBA case and consistently achieves the best or near-best results in the open case studies.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"170 ","pages":"Article 104313"},"PeriodicalIF":8.2,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-style adversarial variational self-distillation in randomized domains for single-domain generalized fault diagnosis 面向单域广义故障诊断的随机域多风格对抗变分自蒸馏

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2025-05-24 DOI: 10.1016/j.compind.2025.104319

Fan Yang, Xiaofeng Liu, Chunbing Zhang, Lin Bo

As rotating machinery often operates under complex and variable harsh conditions, domain generalization-based fault diagnosis has been adopted to tackle the challenge of distribution shifts and unseen data in target domains. However, most existing methods depend on fully labeled data from multiple source domains to learn domain-invariant representations. In practice, collecting comprehensive labeled data across diverse working conditions is often impractical, resulting in data insufficiency and distribution inconsistencies. To address the challenging scenario in which only a single fully labeled source domain is available, this article proposes a multi-style adversarial variational self-distillation (MSAVSD) framework based on domain randomization for single-domain generalized fault diagnosis. First, a domain-randomized generation module is developed to adaptively generate samples following randomized distributions by integrating adaptive noise and multi-scale style learning, thereby enriching the synthetic data with diverse and informative fault representations. Next, a scale-enhanced feature extraction module is introduced to extract rich domain-invariant features, thereby maximizing the utilization of fault-related information under limited training conditions. The method suppresses task-irrelevant noise and redundancy via variational self-distillation and employs contrastive learning to enhance the discriminability and consistency of task-relevant features. Extensive diagnostic experiments on three datasets, two self-collected and one publicly available, demonstrate that the proposed method outperforms state-of-the-art methods.

由于旋转机械经常在复杂多变的恶劣条件下运行，基于域泛化的故障诊断被用于解决目标域中分布变化和未知数据的挑战。然而，大多数现有方法依赖于来自多个源域的完全标记数据来学习域不变表示。在实践中，在不同的工作条件下收集全面的标记数据通常是不切实际的，从而导致数据不足和分布不一致。为了解决只有一个完全标记的源域可用的具有挑战性的场景，本文提出了一种基于域随机化的多风格对抗变分自蒸馏（MSAVSD）框架，用于单域广义故障诊断。首先，通过集成自适应噪声和多尺度风格学习，开发了域随机生成模块，根据随机分布自适应生成样本，从而使合成数据丰富多样、信息丰富；其次，引入尺度增强特征提取模块，提取丰富的域不变特征，从而在有限的训练条件下最大限度地利用故障相关信息。该方法通过变分自蒸馏来抑制与任务无关的噪声和冗余，并利用对比学习来增强任务相关特征的可辨别性和一致性。在三个数据集上进行了广泛的诊断实验，两个数据集是自己收集的，一个数据集是公开的，表明所提出的方法优于最先进的方法。

{"title":"Multi-style adversarial variational self-distillation in randomized domains for single-domain generalized fault diagnosis","authors":"Fan Yang, Xiaofeng Liu, Chunbing Zhang, Lin Bo","doi":"10.1016/j.compind.2025.104319","DOIUrl":"10.1016/j.compind.2025.104319","url":null,"abstract":"<div><div>As rotating machinery often operates under complex and variable harsh conditions, domain generalization-based fault diagnosis has been adopted to tackle the challenge of distribution shifts and unseen data in target domains. However, most existing methods depend on fully labeled data from multiple source domains to learn domain-invariant representations. In practice, collecting comprehensive labeled data across diverse working conditions is often impractical, resulting in data insufficiency and distribution inconsistencies. To address the challenging scenario in which only a single fully labeled source domain is available, this article proposes a multi-style adversarial variational self-distillation (MSAVSD) framework based on domain randomization for single-domain generalized fault diagnosis. First, a domain-randomized generation module is developed to adaptively generate samples following randomized distributions by integrating adaptive noise and multi-scale style learning, thereby enriching the synthetic data with diverse and informative fault representations. Next, a scale-enhanced feature extraction module is introduced to extract rich domain-invariant features, thereby maximizing the utilization of fault-related information under limited training conditions. The method suppresses task-irrelevant noise and redundancy via variational self-distillation and employs contrastive learning to enhance the discriminability and consistency of task-relevant features. Extensive diagnostic experiments on three datasets, two self-collected and one publicly available, demonstrate that the proposed method outperforms state-of-the-art methods.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"170 ","pages":"Article 104319"},"PeriodicalIF":8.2,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144131011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0