Pub Date : 2025-03-29DOI: 10.1016/j.eswa.2025.127212
Javier González-Alonso , Paula Martín-Tapia, David González-Ortega , Míriam Antón-Rodríguez , Francisco Javier Díaz-Pernas , Mario Martínez-Zarzuela
This study presents ME-WARD (Multimodal Ergonomic Workplace Assessment and Risk from Data), a novel system for ergonomic assessment and musculoskeletal risk evaluation that implements the Rapid Upper Limb Assessment (RULA) method. ME-WARD is designed to process joint angle data from motion capture systems, including inertial measurement unit (IMU)-based setups, and deep learning human body pose tracking models. The tool’s flexibility enables ergonomic risk assessment using any system capable of reliably measuring joint angles, extending the applicability of RULA beyond proprietary setups. To validate its performance, the tool was tested in an industrial setting during the assembly of conveyor belts, which involved high-risk tasks such as inserting rods and pushing conveyor belt components. The experiments leveraged gold standard IMU systems alongside a state-of-the-art monocular 3D pose estimation system. The results confirmed that ME-WARD produces reliable RULA scores that closely align with IMU-derived metrics for flexion-dominated movements and comparable performance with the monocular system, despite limitations in tracking lateral and rotational motions. This work highlights the potential of integrating multiple motion capture technologies into a unified and accessible ergonomic assessment pipeline. By supporting diverse input sources, including low-cost video-based systems, the proposed multimodal approach offers a scalable, cost-effective solution for ergonomic assessments, paving the way for broader adoption in resource-constrained industrial environments.
{"title":"ME-WARD: A multimodal ergonomic analysis tool for musculoskeletal risk assessment from inertial and video data in working places","authors":"Javier González-Alonso , Paula Martín-Tapia, David González-Ortega , Míriam Antón-Rodríguez , Francisco Javier Díaz-Pernas , Mario Martínez-Zarzuela","doi":"10.1016/j.eswa.2025.127212","DOIUrl":"10.1016/j.eswa.2025.127212","url":null,"abstract":"<div><div>This study presents ME-WARD (<em>Multimodal Ergonomic Workplace Assessment and Risk from Data</em>), a novel system for ergonomic assessment and musculoskeletal risk evaluation that implements the Rapid Upper Limb Assessment (RULA) method. ME-WARD is designed to process joint angle data from motion capture systems, including inertial measurement unit (IMU)-based setups, and deep learning human body pose tracking models. The tool’s flexibility enables ergonomic risk assessment using any system capable of reliably measuring joint angles, extending the applicability of RULA beyond proprietary setups. To validate its performance, the tool was tested in an industrial setting during the assembly of conveyor belts, which involved high-risk tasks such as inserting rods and pushing conveyor belt components. The experiments leveraged gold standard IMU systems alongside a state-of-the-art monocular 3D pose estimation system. The results confirmed that ME-WARD produces reliable RULA scores that closely align with IMU-derived metrics for flexion-dominated movements and comparable performance with the monocular system, despite limitations in tracking lateral and rotational motions. This work highlights the potential of integrating multiple motion capture technologies into a unified and accessible ergonomic assessment pipeline. By supporting diverse input sources, including low-cost video-based systems, the proposed multimodal approach offers a scalable, cost-effective solution for ergonomic assessments, paving the way for broader adoption in resource-constrained industrial environments.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127212"},"PeriodicalIF":7.5,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724874","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}
Pub Date : 2025-03-26DOI: 10.1016/j.eswa.2025.127332
George Papazafeiropoulos
This study introduces a new clustering technique to boost Artificial Neural Networks’ (ANNs’) performance. The term “Cluster-Boosted Artificial Neural Networks” (“CBANNs”) is coined for ANNs using this technique. By adding cluster identifiers as extra input features, CBANNs enhance conventional ANNs and improve the model’s ability to identify underlying patterns in complicated data landscapes. This method offers a solution to some limitations of standard ANNs, which often struggle with high-dimensional data, local minima, and nonlinear relationships. Without the need for manual feature engineering or in-depth domain knowledge, CBANNs greatly increase prediction accuracy by employing unsupervised clustering, using k-medoids, to build a more structured input space. Various numerical results are presented which validate the superior predictive ability of CBANNs across nine benchmark functions, including De Jong’s 5th, Griewank, and Rastrigin functions. Compared to conventional ANNs with identical hyperparameters, CBANNs achieve error reductions of up to 98%, consistently demonstrating higher performance on functions with intricate geometries and multiple minima. Furthermore, CBANNs are applied to a terrain modeling problem, which proved that CBANNs can reduce the prediction error by up to 95% compared to standard ANNs, indicating their potential for high-precision applications. These findings underscore the CBANN’s ability to generalize effectively in challenging datasets, suggesting its broader applicability in fields that demand accuracy in the presence of complex data distributions.
{"title":"Cluster-Boosted Artificial Neural Networks: Theory, implementation, and performance evaluation","authors":"George Papazafeiropoulos","doi":"10.1016/j.eswa.2025.127332","DOIUrl":"10.1016/j.eswa.2025.127332","url":null,"abstract":"<div><div>This study introduces a new clustering technique to boost Artificial Neural Networks’ (ANNs’) performance. The term “Cluster-Boosted Artificial Neural Networks” (“CBANNs”) is coined for ANNs using this technique. By adding cluster identifiers as extra input features, CBANNs enhance conventional ANNs and improve the model’s ability to identify underlying patterns in complicated data landscapes. This method offers a solution to some limitations of standard ANNs, which often struggle with high-dimensional data, local minima, and nonlinear relationships. Without the need for manual feature engineering or in-depth domain knowledge, CBANNs greatly increase prediction accuracy by employing unsupervised clustering, using k-medoids, to build a more structured input space. Various numerical results are presented which validate the superior predictive ability of CBANNs across nine benchmark functions, including De Jong’s 5th, Griewank, and Rastrigin functions. Compared to conventional ANNs with identical hyperparameters, CBANNs achieve error reductions of up to 98%, consistently demonstrating higher performance on functions with intricate geometries and multiple minima. Furthermore, CBANNs are applied to a terrain modeling problem, which proved that CBANNs can reduce the prediction error by up to 95% compared to standard ANNs, indicating their potential for high-precision applications. These findings underscore the CBANN’s ability to generalize effectively in challenging datasets, suggesting its broader applicability in fields that demand accuracy in the presence of complex data distributions.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127332"},"PeriodicalIF":7.5,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724872","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}
Pub Date : 2025-03-26DOI: 10.1016/j.eswa.2025.127217
Zhong Wang , Jia-Xuan Jiang , Hao-Ran Wang , Ling Zhou , Yuee Li
Medical image translation synthesizes missing modalities to aid clinical diagnoses, yet Generative Adversarial Networks (GANs) often overfit in limited data scenarios. This work introduces Localized Adaptive Style Mixing (LASM), a novel regularization strategy addressing this challenge. Unlike global statistical mixing, LASM segments discriminator feature maps into grids and blends localized high-order statistics (mean, variance, skewness, kurtosis) from reference and input images. This forces the discriminator to focus on structural content rather than style, effectively mitigating overfitting. Experiments on brain T1- to-CT, pelvic T1-to-CT, and T2-FLAIR synthesis tasks demonstrate that LASM-equipped GANs outperform state-of-the-art methods, achieving 54.84 FID (vs. 131.54 baseline) with only 10% training data. Notably, LASM requires no transfer learning and integrates seamlessly into existing frameworks. Our approach advances data-efficient medical image translation, particularly for rare diseases with scarce datasets. Code is available at here.
{"title":"Localized Adaptive Style Mixing for feature statistics manipulation in medical image translation with limited Data","authors":"Zhong Wang , Jia-Xuan Jiang , Hao-Ran Wang , Ling Zhou , Yuee Li","doi":"10.1016/j.eswa.2025.127217","DOIUrl":"10.1016/j.eswa.2025.127217","url":null,"abstract":"<div><div>Medical image translation synthesizes missing modalities to aid clinical diagnoses, yet Generative Adversarial Networks (GANs) often overfit in limited data scenarios. This work introduces Localized Adaptive Style Mixing (LASM), a novel regularization strategy addressing this challenge. Unlike global statistical mixing, LASM segments discriminator feature maps into grids and blends localized high-order statistics (mean, variance, skewness, kurtosis) from reference and input images. This forces the discriminator to focus on structural content rather than style, effectively mitigating overfitting. Experiments on brain T1- to-CT, pelvic T1-to-CT, and T2-FLAIR synthesis tasks demonstrate that LASM-equipped GANs outperform state-of-the-art methods, achieving 54.84 FID (vs. 131.54 baseline) with only 10% training data. Notably, LASM requires no transfer learning and integrates seamlessly into existing frameworks. Our approach advances data-efficient medical image translation, particularly for rare diseases with scarce datasets. Code is available at <span><span>here</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"277 ","pages":"Article 127217"},"PeriodicalIF":7.5,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697238","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}
Modeling sequential dependencies plays a significant role in simulating the dynamic changes in users’ interests, and the introduction of deep learning can address long sequence data to some extent, thereby enabling more precise capture of these changes. However, most existing models still struggle to train sequences with insufficient interaction information or overly long sequences, and they also fail to capture the genuine intentions of users reflected by the interaction behaviors. Additionally, they overlook the characteristic that items interacted with by users are not strictly ordered and are highly homogeneous within a certain period, while the items between different periods are likely to be heterogeneous. In this paper, we propose a sequential recommendation model based on Multi-session Transformers and multi-attribute integration of items (MTMISRec), which enriches the missing interaction information of sparse data by integrating items’ attributes with users’ historical interaction sequences and distinguishes the true intentions of users under similar interactions. Furthermore, we set a time threshold to partition items with interaction intervals within this threshold into a session, thereby capturing homogeneous relationships within each session. We employ the dual attention mechanism to perform local attention within each session and introduce the learned type weights of each session into the complete interaction sequence to perform global attention, thereby blurring the sequential relationships within sessions and integrating global relevance with local details to handle overly long sequences precisely. We conducted extensive experiments on four datasets, and the results demonstrate that MTMISRec surpasses advanced sequential models on sparse and dense datasets.
{"title":"Multi-session transformers and multi-attribute integration of items for sequential recommendation","authors":"Jiahao Hu , Ruizhen Chen , Yihao Zhang , Yong Zhou","doi":"10.1016/j.eswa.2025.127266","DOIUrl":"10.1016/j.eswa.2025.127266","url":null,"abstract":"<div><div>Modeling sequential dependencies plays a significant role in simulating the dynamic changes in users’ interests, and the introduction of deep learning can address long sequence data to some extent, thereby enabling more precise capture of these changes. However, most existing models still struggle to train sequences with insufficient interaction information or overly long sequences, and they also fail to capture the genuine intentions of users reflected by the interaction behaviors. Additionally, they overlook the characteristic that items interacted with by users are not strictly ordered and are highly homogeneous within a certain period, while the items between different periods are likely to be heterogeneous. In this paper, we propose a sequential recommendation model based on Multi-session Transformers and multi-attribute integration of items (MTMISRec), which enriches the missing interaction information of sparse data by integrating items’ attributes with users’ historical interaction sequences and distinguishes the true intentions of users under similar interactions. Furthermore, we set a time threshold to partition items with interaction intervals within this threshold into a session, thereby capturing homogeneous relationships within each session. We employ the dual attention mechanism to perform local attention within each session and introduce the learned type weights of each session into the complete interaction sequence to perform global attention, thereby blurring the sequential relationships within sessions and integrating global relevance with local details to handle overly long sequences precisely. We conducted extensive experiments on four datasets, and the results demonstrate that MTMISRec surpasses advanced sequential models on sparse and dense datasets.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127266"},"PeriodicalIF":7.5,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724079","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}
Pub Date : 2025-03-26DOI: 10.1016/j.eswa.2025.127326
Hao-Yu Zhu , Ming-Zhi Guo , Yan Zhang
The constitutive modelling is the only method to describe the triaxial stress–strain behavior of cement-based materials while its theoretical deduction, modelling parameters determination and numerical calibrations made it difficult to be further applied. To overcome the limitation of constitutive modelling, a comprehensive machine learning (ML) approach, including Artificial Neural Network (ANN), Gaussian Process (GP), Gradient Boosting (GB) and Optimized Gaussian Process (OGP) was firstly proposed to predict triaxial mechanical behavior of recycled aggregate concrete (RAC). The data augment technology was employed to increase the training data size from 249 to 580, effectively improving the generalization performance. The performance statistics of the aforementioned ML models were compared and validated by R2, MAE, RMSE, and Taylor diagram, showing that the OGP had the best study ability and prediction accuracy. The 99 % prediction results generated by the OGP model concentrated within the ± 10 % confidence interval (R2 = 0.991, MAE = 1.04, RMSE = 0.122). Furthermore, to address the black box nature of ML models, the shapley additive explanation and partial dependence analysis were employed to elucidate the underlying arithmetic mechanism. Finally, the best OGP model was compared with previous constitutive method and further utilized to validate its applicability. Unlike classical constitutive modeling, which requires specialized expertise, the proposed ML approach, available as open source at https://doi.org/10.13140/RG.2.2.15784.89608, offered an accessible and effective solution for predicting triaxial behavior with experimental data.
{"title":"Reliable forecasting non-linear triaxial mechanical response of recycled aggregate concrete by knowledge-enhanced, modified, explainable and replicable machine learning algorithms","authors":"Hao-Yu Zhu , Ming-Zhi Guo , Yan Zhang","doi":"10.1016/j.eswa.2025.127326","DOIUrl":"10.1016/j.eswa.2025.127326","url":null,"abstract":"<div><div>The constitutive modelling is the only method to describe the triaxial stress–strain behavior of cement-based materials while its theoretical deduction, modelling parameters determination and numerical calibrations made it difficult to be further applied. To overcome the limitation of constitutive modelling, a comprehensive machine learning (ML) approach, including Artificial Neural Network (ANN), Gaussian Process (GP), Gradient Boosting (GB) and Optimized Gaussian Process (OGP) was firstly proposed to predict triaxial mechanical behavior of recycled aggregate concrete (RAC). The data augment technology was employed to increase the training data size from 249 to 580, effectively improving the generalization performance. The performance statistics of the aforementioned ML models were compared and validated by R<sup>2</sup>, MAE, RMSE, and Taylor diagram, showing that the OGP had the best study ability and prediction accuracy. The 99 % prediction results generated by the OGP model concentrated within the ± 10 % confidence interval (R<sup>2</sup> = 0.991, MAE = 1.04, RMSE = 0.122). Furthermore, to address the black box nature of ML models, the shapley additive explanation and partial dependence analysis were employed to elucidate the underlying arithmetic mechanism. Finally, the best OGP model was compared with previous constitutive method and further utilized to validate its applicability. Unlike classical constitutive modeling, which requires specialized expertise, the proposed ML approach, available as open source at <span><span>https://doi.org/10.13140/RG.2.2.15784.89608</span><svg><path></path></svg></span>, offered an accessible and effective solution for predicting triaxial behavior with experimental data.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127326"},"PeriodicalIF":7.5,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724877","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}
Pub Date : 2025-03-25DOI: 10.1016/j.eswa.2025.127236
Kai Zhang , Hongbo Gang , Feng Hu , Runlong Yu , Qi Liu
Keyphrase extraction (KE) refers to the process of identifying words or phrases that signify the primary themes of a document. Although keyphrase extraction is important in many downstream applications, including scientific document indexing, search, and question answering, the challenge lies in executing this extraction both adaptively and effectively. To this end, we propose a novel Distillation-based Adaptive Ensemble Learning (DAEL) method specifically designed for efficient keyphrase extraction, encompassing diagnosis, aggregation, and distillation processes. Specifically, we initiate with a Cognitive Diagnosis Module (CDM) to evaluate the diverse capabilities of individual KE models. Following this, an Adaptive Aggregation Module (AAM) is employed to create a weight distribution uniquely suited to each data instance. The process concludes with a Knowledge Distillation Module (KDM) to distill the superior performance of the ensemble model into a single model, thereby refining its efficiency and reducing computational cost. Extensive testing on real-world datasets highlights the superior performance of the proposed model. In comparison with leading-edge methods, our approach notably excels in processing text with complex structures or significant noise, marking a substantial advancement in KE effectiveness.
{"title":"Adaptive ensemble learning for efficient keyphrase extraction: Diagnosis, aggregation, and distillation","authors":"Kai Zhang , Hongbo Gang , Feng Hu , Runlong Yu , Qi Liu","doi":"10.1016/j.eswa.2025.127236","DOIUrl":"10.1016/j.eswa.2025.127236","url":null,"abstract":"<div><div>Keyphrase extraction (KE) refers to the process of identifying words or phrases that signify the primary themes of a document. Although keyphrase extraction is important in many downstream applications, including scientific document indexing, search, and question answering, the challenge lies in executing this extraction both adaptively and effectively. To this end, we propose a novel <em><strong>D</strong>istillation-based <strong>A</strong>daptive <strong>E</strong>nsemble <strong>L</strong>earning (<strong>DAEL</strong>)</em> method specifically designed for efficient keyphrase extraction, encompassing diagnosis, aggregation, and distillation processes. Specifically, we initiate with a <em>Cognitive Diagnosis Module (CDM)</em> to evaluate the diverse capabilities of individual KE models. Following this, an <em>Adaptive Aggregation Module (AAM)</em> is employed to create a weight distribution uniquely suited to each data instance. The process concludes with a <em>Knowledge Distillation Module (KDM)</em> to distill the superior performance of the ensemble model into a single model, thereby refining its efficiency and reducing computational cost. Extensive testing on real-world datasets highlights the superior performance of the proposed model. In comparison with leading-edge methods, our approach notably excels in processing text with complex structures or significant noise, marking a substantial advancement in KE effectiveness.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127236"},"PeriodicalIF":7.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704060","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}
Pub Date : 2025-03-25DOI: 10.1016/j.eswa.2025.127288
Muhammad Rahim , Shah Zeb Khan , Adel M. Widyan , A. Almutairi , Hamiden Abd El-Wahed Khalifa
Sustainable urban development (SUD) projects aim to enhance infrastructure, services, and facilities in cities to improve residents’ quality of life, promote economic growth, and ensure long-term sustainability. As urbanization accelerates globally, decision-makers face significant challenges in selecting projects that balance environmental, economic, social, and technological factors while aligning with strategic urban planning goals. The complexity of these decisions is further heightened by uncertainties in stakeholder opinions, evolving policy frameworks, and real-world constraints. To address these challenges, this study introduces a multi-criteria group decision-making (MCGDM) framework designed specifically for evaluating SUD projects. The proposed methodology leverages complex spherical fuzzy sets ( SFSs) to provide a more flexible and adaptive decision-making structure. These fuzzy sets allow decision-makers to model varying degrees of membership with greater adaptability, ensuring a more precise and comprehensive evaluation of alternatives. The primary contribution of this study lies in its parametric approach, which enhances the dynamism and adaptability of decision-making in complex urban development scenarios. To achieve this, the study is structured into three phases. First, we introduce the fundamental notations and operational laws of SFSs, followed by the development of aggregation operators to handle uncertainty in expert evaluations. In the second phase, we construct a TOPSIS-based approach utilizing these aggregation operators, enabling systematic ranking of SUD project alternatives. The effectiveness of the proposed approach is demonstrated through a numerical example evaluating five alternatives across seven criteria, capturing key factors influencing sustainable urban planning. Finally, the results are compared with existing decision-making methodologies to validate the robustness, effectiveness, and applicability of the proposed framework. By providing a structured, data-driven, and adaptable approach, this study aims to assist urban planners and policymakers in making more informed, balanced, and sustainable decisions for future urban development.
{"title":"A novel complex (p,q,r)- spherical fuzzy TOPSIS framework for sustainable urban development assessment","authors":"Muhammad Rahim , Shah Zeb Khan , Adel M. Widyan , A. Almutairi , Hamiden Abd El-Wahed Khalifa","doi":"10.1016/j.eswa.2025.127288","DOIUrl":"10.1016/j.eswa.2025.127288","url":null,"abstract":"<div><div>Sustainable urban development (SUD) projects aim to enhance infrastructure, services, and facilities in cities to improve residents’ quality of life, promote economic growth, and ensure long-term sustainability. As urbanization accelerates globally, decision-makers face significant challenges in selecting projects that balance environmental, economic, social, and technological factors while aligning with strategic urban planning goals. The complexity of these decisions is further heightened by uncertainties in stakeholder opinions, evolving policy frameworks, and real-world constraints. To address these challenges, this study introduces a multi-criteria group decision-making (MCGDM) framework designed specifically for evaluating SUD projects. The proposed methodology leverages complex <span><math><mrow><mo>(</mo><mi>p</mi><mo>,</mo><mi>q</mi><mo>,</mo><mi>r</mi><mo>)</mo><mo>-</mo></mrow></math></span> spherical fuzzy sets (<span><math><msub><mrow><mi>Com</mi></mrow><mrow><mo>(</mo><mi>p</mi><mo>,</mo><mi>q</mi><mo>,</mo><mi>r</mi><mo>)</mo></mrow></msub></math></span> SFSs) to provide a more flexible and adaptive decision-making structure. These fuzzy sets allow decision-makers to model varying degrees of membership with greater adaptability, ensuring a more precise and comprehensive evaluation of alternatives. The primary contribution of this study lies in its parametric approach, which enhances the dynamism and adaptability of decision-making in complex urban development scenarios. To achieve this, the study is structured into three phases. First, we introduce the fundamental notations and operational laws of <span><math><msub><mrow><mi>Com</mi></mrow><mrow><mo>(</mo><mi>p</mi><mo>,</mo><mi>q</mi><mo>,</mo><mi>r</mi><mo>)</mo></mrow></msub></math></span> SFSs, followed by the development of aggregation operators to handle uncertainty in expert evaluations. In the second phase, we construct a TOPSIS-based approach utilizing these aggregation operators, enabling systematic ranking of SUD project alternatives. The effectiveness of the proposed approach is demonstrated through a numerical example evaluating five alternatives across seven criteria, capturing key factors influencing sustainable urban planning. Finally, the results are compared with existing decision-making methodologies to validate the robustness, effectiveness, and applicability of the proposed framework. By providing a structured, data-driven, and adaptable approach, this study aims to assist urban planners and policymakers in making more informed, balanced, and sustainable decisions for future urban development.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127288"},"PeriodicalIF":7.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724875","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}
Pub Date : 2025-03-25DOI: 10.1016/j.eswa.2025.127316
Yan-ze Yu , Guo-zheng Quan , Yu-qing Zhang , Ying-ying Liu , Li-he Jiang , Wei Xiong , Jiang Zhao
It is a great challenge to design the multi-stage loading path of variable parameters to obtain the component with smooth shape and fine-grained microstructures during the electric upsetting process of large-scale valves. To achieve this, an intelligent design methodology was developed and applied in an electric upsetting process of Ni80A alloy. The methodology integrates backpropagation neural network (BP neural network), case-based reasoning (CBR), and parameter self-feedback adjustment coupling with finite element (FE). Firstly, a BP neural network model was developed based on the basic database to predict the initial processing parameters of components (upsetting force, current, and pre-heating time). Secondly, utilizing the CBR method, the suitable design schemes were identified by retrieving similar components, and then the multi-stages loading paths for upsetting force and current were devised. Thirdly, a subroutine of self-feedback adjustment to fine-tune the loading paths was developed and implanted into the multi-field and multi-scale coupling FE model. Finally, the optimal loading paths was obtained using the FE model until the deformed component meets the requirements of shape and grain size. The results indicated that the surface contour of component was smoother and without macroscopic defects under the optimal loading paths, with the maximum grain size refined to 103.9 μm. To further improve the automation level of the parameters design process, an expert system was developed based on the designed methodology. This work contributes to the intelligent design of processing parameters for the electric upsetting process, which provides a design framework of processing parameter in other manufacturing technologies.
{"title":"An intelligent design methodology for multi-stage loading paths of variable parameters during large-scale electric upsetting process","authors":"Yan-ze Yu , Guo-zheng Quan , Yu-qing Zhang , Ying-ying Liu , Li-he Jiang , Wei Xiong , Jiang Zhao","doi":"10.1016/j.eswa.2025.127316","DOIUrl":"10.1016/j.eswa.2025.127316","url":null,"abstract":"<div><div>It is a great challenge to design the multi-stage loading path of variable parameters to obtain the component with smooth shape and fine-grained microstructures during the electric upsetting process of large-scale valves. To achieve this, an intelligent design methodology was developed and applied in an electric upsetting process of Ni80A alloy. The methodology integrates backpropagation neural network (BP neural network), case-based reasoning (CBR), and parameter self-feedback adjustment coupling with finite element (FE). Firstly, a BP neural network model was developed based on the basic database to predict the initial processing parameters of components (upsetting force, current, and pre-heating time). Secondly, utilizing the CBR method, the suitable design schemes were identified by retrieving similar components, and then the multi-stages loading paths for upsetting force and current were devised. Thirdly, a subroutine of self-feedback adjustment to fine-tune the loading paths was developed and implanted into the multi-field and multi-scale coupling FE model. Finally, the optimal loading paths was obtained using the FE model until the deformed component meets the requirements of shape and grain size. The results indicated that the surface contour of component was smoother and without macroscopic defects under the optimal loading paths, with the maximum grain size refined to 103.9 μm. To further improve the automation level of the parameters design process, an expert system was developed based on the designed methodology. This work contributes to the intelligent design of processing parameters for the electric upsetting process, which provides a design framework of processing parameter in other manufacturing technologies.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127316"},"PeriodicalIF":7.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724082","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}
Pub Date : 2025-03-25DOI: 10.1016/j.eswa.2025.127325
Miaorui Yang , Kun Zhang , Yanping Zhu , Long Zhang , Yonggang Xu
The slewing bearing is a critical component in large equipment like shield machines and wind turbines. Because slewing bearings operate in complex situations with fluctuating speed and load on a regular basis, the vibration signal they produce contains several interferences, making fault features difficult to identify. The specific objective of this study is to provide a new fault diagnosis method, named difference optimization bispectrum, for slewing bearing signals under strong noise interference. The method designs a convex optimization bispectrum model by the convex optimization theory, covering the shortage of traditional decomposition by differentiating features. Based on the model, a two-dimensional weight coefficient is constructed to calculate the difference optimization bispectrum, which reduces the noise and enhances the features in positive and negative bispectrum-domain. This study offers a fresh perspective on extraction of fault information from the signal under strong noise interference, making an original contribution for the fault diagnosis of the slewing bearing. The experiment work presented here provides the practical effect of the method for the slewing bearing signals.
{"title":"A new difference feature extraction method of slewing bearings in wind turbines via optimization bispectrum domain model","authors":"Miaorui Yang , Kun Zhang , Yanping Zhu , Long Zhang , Yonggang Xu","doi":"10.1016/j.eswa.2025.127325","DOIUrl":"10.1016/j.eswa.2025.127325","url":null,"abstract":"<div><div>The slewing bearing is a critical component in large equipment like shield machines and wind turbines. Because slewing bearings operate in complex situations with fluctuating speed and load on a regular basis, the vibration signal they produce contains several interferences, making fault features difficult to identify. The specific objective of this study is to provide a new fault diagnosis method, named difference optimization bispectrum, for slewing bearing signals under strong noise interference. The method designs a convex optimization bispectrum model by the convex optimization theory, covering the shortage of traditional decomposition by differentiating features. Based on the model, a two-dimensional weight coefficient is constructed to calculate the difference optimization bispectrum, which reduces the noise and enhances the features in positive and negative bispectrum-domain. This study offers a fresh perspective on extraction of fault information from the signal under strong noise interference, making an original contribution for the fault diagnosis of the slewing bearing. The experiment work presented here provides the practical effect of the method for the slewing bearing signals.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127325"},"PeriodicalIF":7.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724565","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}
Pub Date : 2025-03-25DOI: 10.1016/j.eswa.2025.127379
Jungmin Lee , Jiyoon Lee , Seoung Bum Kim
As time series data continues to be collected across various fields, the importance of automated anomaly detection systems is steadily increasing. A key challenge in anomaly detection lies in setting an optimal threshold for anomaly scores to distinguish anomalies from normal data. Most existing studies use a fixed threshold, often resulting in misclassification of ambiguous data. Therefore, defining a dynamic and optimal threshold is crucial for improving detection performance. We aim to quantify uncertainty as a metric that determines the degree of ambiguity in the data. Because our models are trained only on normal data, anomalies exhibiting patterns divergent from the normal data entail higher uncertainty. Accordingly, in this study, we propose a dynamic thresholding method that better aligns with the nature of the data through uncertainty quantification. Through experimentation with synthetic datasets and five benchmark datasets for time series anomaly detection, we demonstrate the efficacy of our proposed method. Our proposed method outperforms both the fixed threshold and existing dynamic thresholding methods, achieving an average F1-score improvement of over 0.06 across benchmark datasets. In particular, the performance improvement is more significant when the distributions of normal data and anomalies are more similar. The source code can be accessed at https://github.com/jungminkr9195/UDT.
{"title":"Uncertainty-informed dynamic threshold for time series anomaly detection","authors":"Jungmin Lee , Jiyoon Lee , Seoung Bum Kim","doi":"10.1016/j.eswa.2025.127379","DOIUrl":"10.1016/j.eswa.2025.127379","url":null,"abstract":"<div><div>As time series data continues to be collected across various fields, the importance of automated anomaly detection systems is steadily increasing. A key challenge in anomaly detection lies in setting an optimal threshold for anomaly scores to distinguish anomalies from normal data. Most existing studies use a fixed threshold, often resulting in misclassification of ambiguous data. Therefore, defining a dynamic and optimal threshold is crucial for improving detection performance. We aim to quantify uncertainty as a metric that determines the degree of ambiguity in the data. Because our models are trained only on normal data, anomalies exhibiting patterns divergent from the normal data entail higher uncertainty. Accordingly, in this study, we propose a dynamic thresholding method that better aligns with the nature of the data through uncertainty quantification. Through experimentation with synthetic datasets and five benchmark datasets for time series anomaly detection, we demonstrate the efficacy of our proposed method. Our proposed method outperforms both the fixed threshold and existing dynamic thresholding methods, achieving an average F1-score improvement of over 0.06 across benchmark datasets. In particular, the performance improvement is more significant when the distributions of normal data and anomalies are more similar. The source code can be accessed at https://github.com/jungminkr9195/UDT.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127379"},"PeriodicalIF":7.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724624","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号