Pub Date : 2024-11-19DOI: 10.1016/j.compind.2024.104213
Feng Liu, Yingjie Lu, Debiao Li, Raymond Chiong
This paper proposes a Wasserstein distributionally robust learning (WDRL) model to predict the production cycle time of highly mixed printed circuit board (PCB) orders on multiple production lines. The PCB production cycle time is essential for optimizing production plans. However, the design of the PCB, production line configuration, order combinations, and personnel factors make the prediction of the PCB production cycle time difficult. In addition, practical production situations with significant disturbances in feature data make traditional prediction models inaccurate, especially when there is new data. Therefore, we establishe a WDRL model, derive a tight upper bound for the expected loss function, and reformulate a tractable equivalent model based on the bound. To demonstrate the effectiveness of this method, we collected data related to surface mounted technology (SMT) production lines from a leading global display manufacturer for our computational experiments. In addition, we also designed experiments with perturbations in the training and testing datasets to verify the WDRL model’s ability to handle perturbations. This proposed method has also been compared with other machine learning methods, such as the support vector regression combined with symbiotic organism search, decision tree, and kernel extreme learning machine, among others. Experimental results indicate that the WDRL model can make robust predictions of PCB cycle time, which helps to effectively plan production capacity in uncertain situations and avoid overproduction or underproduction. Finally, we implement the WDRL model for the actual SMT production to predict the production cycle time and set it as the target for production. We observed a 98–103 % achievement rate in the last 20 months since the implementation in September 2022.
{"title":"Wasserstein distributionally robust learning for predicting the cycle time of printed circuit board production","authors":"Feng Liu, Yingjie Lu, Debiao Li, Raymond Chiong","doi":"10.1016/j.compind.2024.104213","DOIUrl":"https://doi.org/10.1016/j.compind.2024.104213","url":null,"abstract":"This paper proposes a Wasserstein distributionally robust learning (WDRL) model to predict the production cycle time of highly mixed printed circuit board (PCB) orders on multiple production lines. The PCB production cycle time is essential for optimizing production plans. However, the design of the PCB, production line configuration, order combinations, and personnel factors make the prediction of the PCB production cycle time difficult. In addition, practical production situations with significant disturbances in feature data make traditional prediction models inaccurate, especially when there is new data. Therefore, we establishe a WDRL model, derive a tight upper bound for the expected loss function, and reformulate a tractable equivalent model based on the bound. To demonstrate the effectiveness of this method, we collected data related to surface mounted technology (SMT) production lines from a leading global display manufacturer for our computational experiments. In addition, we also designed experiments with perturbations in the training and testing datasets to verify the WDRL model’s ability to handle perturbations. This proposed method has also been compared with other machine learning methods, such as the support vector regression combined with symbiotic organism search, decision tree, and kernel extreme learning machine, among others. Experimental results indicate that the WDRL model can make robust predictions of PCB cycle time, which helps to effectively plan production capacity in uncertain situations and avoid overproduction or underproduction. Finally, we implement the WDRL model for the actual SMT production to predict the production cycle time and set it as the target for production. We observed a 98–103 % achievement rate in the last 20 months since the implementation in September 2022.","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"18 1","pages":""},"PeriodicalIF":10.0,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673318","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 : 2024-11-18DOI: 10.1016/j.compind.2024.104207
Seungeun Lim, Changmo Yeo, Byung Chul Kim, Kyung Cheol Bae, Duhwan Mun
Topological elements form the basis for tasks such as geometric calculations, feature analysis, and direct modeling in 3D CAD systems. Handling these elements is also essential in various automated systems. This study proposes a method to search for topological elements within a boundary representation (B-rep) model by employing topological queries. To address complex scenarios that are difficult to handle using a single query, a topological query procedure that sequentially executes a predefined set of topological queries is used. To verify the effectiveness of the proposed method, experiments were conducted on Test Cases 1, 2, and 3, confirming the successful search of all target topological elements. Furthermore, tests on modified Snap-fit hook A and Bridge B models demonstrated that the same queries remained effective, provided the topological relationships and geometric constraints expressed in the query were preserved. In addition, a search time comparison showed that the proposed method reduced search time by over 90 % compared to manual processes. Finally, in an experiment involving participants with varying levels of programming proficiency, the results indicated that, for a developer with high programming skills, writing topological queries reduced the time required to search for a single topological element by more than 95 % compared to writing the program code.
拓扑元素是三维 CAD 系统中几何计算、特征分析和直接建模等任务的基础。在各种自动化系统中，处理这些元素也是必不可少的。本研究提出了一种通过拓扑查询在边界表示（B-rep）模型中搜索拓扑元素的方法。为解决单一查询难以处理的复杂情况，本研究采用了拓扑查询程序，该程序可按顺序执行一组预定义的拓扑查询。为了验证所提方法的有效性，对测试案例 1、2 和 3 进行了实验，证实成功搜索到了所有目标拓扑元素。此外，对修改后的 Snap-fit 挂钩 A 和桥梁 B 模型进行的测试表明，只要保留查询中表达的拓扑关系和几何约束，同样的查询仍然有效。此外，搜索时间比较显示，与人工处理相比，建议的方法减少了 90% 以上的搜索时间。最后，在一项由不同编程能力水平的参与者参与的实验中，结果表明，对于编程能力较高的开发人员来说，编写拓扑查询与编写程序代码相比，可将搜索单个拓扑元素所需的时间减少 95% 以上。
{"title":"BRepQL: Query language for searching topological elements in B-rep models","authors":"Seungeun Lim, Changmo Yeo, Byung Chul Kim, Kyung Cheol Bae, Duhwan Mun","doi":"10.1016/j.compind.2024.104207","DOIUrl":"https://doi.org/10.1016/j.compind.2024.104207","url":null,"abstract":"Topological elements form the basis for tasks such as geometric calculations, feature analysis, and direct modeling in 3D CAD systems. Handling these elements is also essential in various automated systems. This study proposes a method to search for topological elements within a boundary representation (B-rep) model by employing topological queries. To address complex scenarios that are difficult to handle using a single query, a topological query procedure that sequentially executes a predefined set of topological queries is used. To verify the effectiveness of the proposed method, experiments were conducted on Test Cases 1, 2, and 3, confirming the successful search of all target topological elements. Furthermore, tests on modified Snap-fit hook A and Bridge B models demonstrated that the same queries remained effective, provided the topological relationships and geometric constraints expressed in the query were preserved. In addition, a search time comparison showed that the proposed method reduced search time by over 90 % compared to manual processes. Finally, in an experiment involving participants with varying levels of programming proficiency, the results indicated that, for a developer with high programming skills, writing topological queries reduced the time required to search for a single topological element by more than 95 % compared to writing the program code.","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"251 1","pages":""},"PeriodicalIF":10.0,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673319","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 : 2024-11-17DOI: 10.1016/j.compind.2024.104205
Kay Hönemann , Björn Konopka , Michael Prilla , Manuel Wiesche
Augmented Reality (AR) instructions offer companies tremendous savings potential. However, developing these AR instructions has traditionally been challenging due to the need for programming skills and spatial knowledge. To address this complexity, industry and academia are working to simplify AR development. A crucial aspect of this process is the accurate positioning of AR content within the physical environment, which requires effective AR interaction techniques that enable full 3D manipulation of AR elements. In this study, we conducted an experimental comparison of three different AR interaction techniques with 55 participants to empirically assess their performance, workload, and user satisfaction across tasks related to AR instruction development. Our findings contribute to the design of future AR instructions and AR authoring tools, emphasizing the importance of evaluating AR interaction techniques that can be utilized by users without programming experience tailored to the specific needs of the intended application domain.
增强现实（AR）指令为公司提供了巨大的节约潜力。然而，由于需要编程技能和空间知识，开发这些 AR 指令历来具有挑战性。为了解决这一复杂问题，业界和学术界正在努力简化 AR 开发。这一过程的一个关键方面是在物理环境中准确定位 AR 内容，这需要有效的 AR 交互技术，以实现对 AR 元素的全三维操作。在本研究中，我们对三种不同的 AR 交互技术进行了实验比较，共有 55 名参与者参加，目的是对他们在与 AR 教学开发相关的任务中的表现、工作量和用户满意度进行实证评估。我们的研究结果有助于未来 AR 教学和 AR 创作工具的设计，同时强调了评估 AR 交互技术的重要性，这些技术可以根据预期应用领域的特定需求，为没有编程经验的用户量身定制。
{"title":"A Comparative Study of Handheld Augmented Reality Interaction Techniques for Developing AR Instructions using AR Authoring Tools","authors":"Kay Hönemann , Björn Konopka , Michael Prilla , Manuel Wiesche","doi":"10.1016/j.compind.2024.104205","DOIUrl":"10.1016/j.compind.2024.104205","url":null,"abstract":"<div><div>Augmented Reality (AR) instructions offer companies tremendous savings potential. However, developing these AR instructions has traditionally been challenging due to the need for programming skills and spatial knowledge. To address this complexity, industry and academia are working to simplify AR development. A crucial aspect of this process is the accurate positioning of AR content within the physical environment, which requires effective AR interaction techniques that enable full 3D manipulation of AR elements. In this study, we conducted an experimental comparison of three different AR interaction techniques with 55 participants to empirically assess their performance, workload, and user satisfaction across tasks related to AR instruction development. Our findings contribute to the design of future AR instructions and AR authoring tools, emphasizing the importance of evaluating AR interaction techniques that can be utilized by users without programming experience tailored to the specific needs of the intended application domain.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104205"},"PeriodicalIF":8.2,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-15DOI: 10.1016/j.compind.2024.104212
Anna Gieß , Thorsten Schoormann , Frederik Möller , Inan Gür
Technical coordination between organizations and security concerns are among the major barriers to data sharing. Data spaces are an emerging digital infrastructure that helps address these challenges by sovereignly sharing data across institutional boundaries. The data space concept is at the core of many high-profile research initiatives in the European Union and receives great adoption in practice. Despite the great interest, there is, however, a demand for more conceptual clarity and approaches to describe and design them purposefully. We propose a taxonomy of data space design options grounded in a literature review, an analysis of real-world objects, and over nine hours of expert interviews with data space initiatives. The taxonomy advances our understanding of data space designs and gives a framework to practice making informed design decisions. Our work provides a comprehensive solution space for data space designers to (a) (re-)design data spaces more efficiently and (b) acquire a ‘big picture’ of what needs to be considered.
{"title":"Discovering data spaces: A classification of design options","authors":"Anna Gieß , Thorsten Schoormann , Frederik Möller , Inan Gür","doi":"10.1016/j.compind.2024.104212","DOIUrl":"10.1016/j.compind.2024.104212","url":null,"abstract":"<div><div>Technical coordination between organizations and security concerns are among the major barriers to data sharing. Data spaces are an emerging digital infrastructure that helps address these challenges by sovereignly sharing data across institutional boundaries. The data space concept is at the core of many high-profile research initiatives in the European Union and receives great adoption in practice. Despite the great interest, there is, however, a demand for more conceptual clarity and approaches to describe and design them purposefully. We propose a taxonomy of data space design options grounded in a literature review, an analysis of real-world objects, and over nine hours of expert interviews with data space initiatives. The taxonomy advances our understanding of data space designs and gives a framework to practice making informed design decisions. Our work provides a comprehensive solution space for data space designers to (a) (re-)design data spaces more efficiently and (b) acquire a ‘big picture’ of what needs to be considered.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104212"},"PeriodicalIF":8.2,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-14DOI: 10.1016/j.compind.2024.104208
Jin-Su Shin , Min-Joo Kim , Beom-Seok Kim , Dong-Hee Lee
It is crucial to detect and classify defect patterns on wafers in semiconductor-manufacturing processes for wafer-quality management and prompt analysis of defect causes. In recent years, continuous technological innovation and advancements in semiconductor-industry processes have led to an increase in unknown defect patterns, which must be detected and classified. However, detection of unknown defect patterns is difficult due to complex reasons, such as training on non-existent defect classes, closed datasets owing to industrial security, and labeling large volumes of manufacturing data. Owing to these challenges, methods for detecting unknown defect patterns in an actual semiconductor-manufacturing environment primarily rely on qualitative indicators, such as intuition and experience of engineers. To overcome these problems, this study proposes a methodology based on open-set recognition to accurately detect unknown defect patterns. This methodology begins with two preprocessing steps: constrained mean filtering (C-mean filtering); and Radon transform to diminish noise and efficiently extract features from wafer-bin maps. This study then develops an entropy-estimation one-class support vector machine (EEOC-SVM), which accounts for the uncertainty in the one-class SVM classification results. EEOC-SVM computes entropy-uncertainty scores based on the distance between decision boundaries and samples and then reclassifies uncertain samples using a weighted sum of uncertainties for each class. This method can effectively detect unknown defect patterns. The proposed method achieves a detection performance of over 98 % for various defect classes based on experiments conducted with new defect patterns occurring in actual semiconductor-manufacturing environments. These results confirm that the proposed method is an effective tool for detecting and addressing unknown defect patterns.
{"title":"Enhanced detection of unknown defect patterns on wafer bin maps based on an open-set recognition approach","authors":"Jin-Su Shin , Min-Joo Kim , Beom-Seok Kim , Dong-Hee Lee","doi":"10.1016/j.compind.2024.104208","DOIUrl":"10.1016/j.compind.2024.104208","url":null,"abstract":"<div><div>It is crucial to detect and classify defect patterns on wafers in semiconductor-manufacturing processes for wafer-quality management and prompt analysis of defect causes. In recent years, continuous technological innovation and advancements in semiconductor-industry processes have led to an increase in unknown defect patterns, which must be detected and classified. However, detection of unknown defect patterns is difficult due to complex reasons, such as training on non-existent defect classes, closed datasets owing to industrial security, and labeling large volumes of manufacturing data. Owing to these challenges, methods for detecting unknown defect patterns in an actual semiconductor-manufacturing environment primarily rely on qualitative indicators, such as intuition and experience of engineers. To overcome these problems, this study proposes a methodology based on open-set recognition to accurately detect unknown defect patterns. This methodology begins with two preprocessing steps: constrained mean filtering (C-mean filtering); and Radon transform to diminish noise and efficiently extract features from wafer-bin maps. This study then develops an entropy-estimation one-class support vector machine (EEOC-SVM), which accounts for the uncertainty in the one-class SVM classification results. EEOC-SVM computes entropy-uncertainty scores based on the distance between decision boundaries and samples and then reclassifies uncertain samples using a weighted sum of uncertainties for each class. This method can effectively detect unknown defect patterns. The proposed method achieves a detection performance of over 98 % for various defect classes based on experiments conducted with new defect patterns occurring in actual semiconductor-manufacturing environments. These results confirm that the proposed method is an effective tool for detecting and addressing unknown defect patterns.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104208"},"PeriodicalIF":8.2,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654839","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 : 2024-11-14DOI: 10.1016/j.compind.2024.104211
Juliano Barbosa , Baldoino Fonseca , Márcio Ribeiro , João Correia , Leandro Dias da Silva , Rohit Gheyi , Davy Baia
Natural Language Processing (NLP) has revolutionized industries, streamlining customer service through applications in healthcare, finance, legal, and human resources domains, and simplifying tasks like medical research, financial analysis, and sentiment analysis. To avoid the high costs of building and maintaining NLP infrastructure, companies turn to Cloud NLP services offered by major cloud providers like Amazon, Google, and Microsoft. However, there is little knowledge about how tolerant these services are when subjected to noise. This paper presents a study that analyzes the effectiveness of Cloud NLP services by evaluating the noise tolerance of sentiment analysis services provided by Amazon, Google, and Microsoft when subjected to 12 types of noise, including syntactic and semantic noises. The findings indicate that Google is the most tolerant to syntactic noises, and Microsoft is the most tolerant to semantic noises. These findings may help developers and companies in selecting the most suitable service provider and shed light towards improving state-of-the-art techniques for effective cloud NLP services.
{"title":"Evaluating the noise tolerance of Cloud NLP services across Amazon, Microsoft, and Google","authors":"Juliano Barbosa , Baldoino Fonseca , Márcio Ribeiro , João Correia , Leandro Dias da Silva , Rohit Gheyi , Davy Baia","doi":"10.1016/j.compind.2024.104211","DOIUrl":"10.1016/j.compind.2024.104211","url":null,"abstract":"<div><div>Natural Language Processing (NLP) has revolutionized industries, streamlining customer service through applications in healthcare, finance, legal, and human resources domains, and simplifying tasks like medical research, financial analysis, and sentiment analysis. To avoid the high costs of building and maintaining NLP infrastructure, companies turn to Cloud NLP services offered by major cloud providers like Amazon, Google, and Microsoft. However, there is little knowledge about how tolerant these services are when subjected to noise. This paper presents a study that analyzes the effectiveness of Cloud NLP services by evaluating the noise tolerance of sentiment analysis services provided by Amazon, Google, and Microsoft when subjected to 12 types of noise, including syntactic and semantic noises. The findings indicate that Google is the most tolerant to syntactic noises, and Microsoft is the most tolerant to semantic noises. These findings may help developers and companies in selecting the most suitable service provider and shed light towards improving state-of-the-art techniques for effective cloud NLP services.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104211"},"PeriodicalIF":8.2,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654743","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 : 2024-11-12DOI: 10.1016/j.compind.2024.104206
Boyeun Lee, Saeema Ahmed-Kristensen
Despite growing interest in the use of data for product and service development, a comprehensive understanding of how data is employed in the context of new product, service and product–service system development is lacking. With the aim of deepening understanding of data as a critical resource for generating value through new products and services, we conducted a systematic literature review, conceptualised through a framework and evaluated with a questionnaire survey. This study (1) identifies the relationships between methodologies and various data-x design concepts, together with their contributions; (2) investigates the types of data captured and utilised across the product/service development process; (3) identifies data-driven design (DDD) activities and the types of data for each activity and (4) develops and validates an evidence-based framework of DDD for new product/service development processes. This study is distinct from previous work as our theoretical foundation identifies seven DDD activities alongside the types of data captured and utilised throughout the new product, service or product–service system development. The key findings highlight the relationship between commonly used concepts for using data in product/service development (i.e., data-driven, -enabled, -centric, -aware, -informed, and design analytics) and their methodological differences. The findings show that whereas data is currently captured predominantly from the in-use phase of a product/service, it is mainly used to support concept development. This paper contributes by developing a DDD framework, which helps practitioners understand how data and machine learning approaches can be used for product/service development. The evidence-based framework also contributes to the body of knowledge on data-x design and the understanding of the role of data in product/service development.
{"title":"D3 framework: An evidence-based data-driven design framework for new product service development","authors":"Boyeun Lee, Saeema Ahmed-Kristensen","doi":"10.1016/j.compind.2024.104206","DOIUrl":"10.1016/j.compind.2024.104206","url":null,"abstract":"<div><div>Despite growing interest in the use of data for product and service development, a comprehensive understanding of how data is employed in the context of new product, service and product–service system development is lacking. With the aim of deepening understanding of data as a critical resource for generating value through new products and services, we conducted a systematic literature review, conceptualised through a framework and evaluated with a questionnaire survey. This study (1) identifies the relationships between methodologies and various data-x design concepts, together with their contributions; (2) investigates the types of data captured and utilised across the product/service development process; (3) identifies data-driven design (DDD) activities and the types of data for each activity and (4) develops and validates an evidence-based framework of DDD for new product/service development processes. This study is distinct from previous work as our theoretical foundation identifies seven DDD activities alongside the types of data captured and utilised throughout the new product, service or product–service system development. The key findings highlight the relationship between commonly used concepts for using data in product/service development (i.e., data-driven, -enabled, -centric, -aware, -informed, and design analytics) and their methodological differences. The findings show that whereas data is currently captured predominantly from the in-use phase of a product/service, it is mainly used to support concept development. This paper contributes by developing a DDD framework, which helps practitioners understand how data and machine learning approaches can be used for product/service development. The evidence-based framework also contributes to the body of knowledge on data-x design and the understanding of the role of data in product/service development.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104206"},"PeriodicalIF":8.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-12DOI: 10.1016/j.compind.2024.104210
Zheng Gao , Danfeng Sun , Kai Wang , Huifeng Wu
The connectivity of devices and systems in the Industrial Internet of Things (IIoT) enables interoperability and collaboration between industrial systems. Device access is the pathway to achieve connectivity, while protocol matching is the basis for device access. Protocol matching is a complex task due to the diverse range of device types, numerous protocols, the issues related to protocol privatization, and the reliance on domain knowledge. These complexities result in inefficient device access. To improve the device access efficiency, a Device Protocol Matching Model (DPMM) is proposed in this paper, which uses only the basic device information to find the best-matched protocol, including protocol identification and basic data. The DPMM adopts a two-stage strategy, consisting of an ontology creation stage and a protocol matching stage. In the ontology creation stage, a simplified device ontology is built to enable the uniform expression of device information and the representation of domain knowledge. In the protocol matching stage, a protocol matcher based on the Two-layer Cooperative Iteration (TCI) algorithm is designed to find the best-matched protocol. In the TCI, to achieve the global optimization of protocol matching efficiency, a penalty mechanism-based weight update method and learning-based matcher evolution are designed. Experiments in two scenarios: a communication base station and a copper smelting production line, are conducted to validate the effectiveness of the DPMM. The experimental results demonstrate that the DPMM can achieve automatic protocol matching with an average matching index of 80.3% and an average hit rate of 35.1%. Moreover, it significantly reduces network resource consumption by up to 96.7%, and increases the hit rate by up to 12.1 times compared with the existing methods.
{"title":"Improving device access efficiency using a device protocol matching model","authors":"Zheng Gao , Danfeng Sun , Kai Wang , Huifeng Wu","doi":"10.1016/j.compind.2024.104210","DOIUrl":"10.1016/j.compind.2024.104210","url":null,"abstract":"<div><div>The connectivity of devices and systems in the Industrial Internet of Things (IIoT) enables interoperability and collaboration between industrial systems. Device access is the pathway to achieve connectivity, while protocol matching is the basis for device access. Protocol matching is a complex task due to the diverse range of device types, numerous protocols, the issues related to protocol privatization, and the reliance on domain knowledge. These complexities result in inefficient device access. To improve the device access efficiency, a Device Protocol Matching Model (DPMM) is proposed in this paper, which uses only the basic device information to find the best-matched protocol, including protocol identification and basic data. The DPMM adopts a two-stage strategy, consisting of an ontology creation stage and a protocol matching stage. In the ontology creation stage, a simplified device ontology is built to enable the uniform expression of device information and the representation of domain knowledge. In the protocol matching stage, a protocol matcher based on the Two-layer Cooperative Iteration (TCI) algorithm is designed to find the best-matched protocol. In the TCI, to achieve the global optimization of protocol matching efficiency, a penalty mechanism-based weight update method and learning-based matcher evolution are designed. Experiments in two scenarios: a communication base station and a copper smelting production line, are conducted to validate the effectiveness of the DPMM. The experimental results demonstrate that the DPMM can achieve automatic protocol matching with an average matching index of 80.3% and an average hit rate of 35.1%. Moreover, it significantly reduces network resource consumption by up to 96.7%, and increases the hit rate by up to 12.1 times compared with the existing methods.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104210"},"PeriodicalIF":8.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654739","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 : 2024-11-11DOI: 10.1016/j.compind.2024.104204
Yuanda Lin , Shuwan Pan , Jie Yu , Yade Hong , Fuming Wang , Jianeng Tang , Lixin Zheng , Songyan Chen
To meet the growing demand for lightweight models and rapid defect detection in mini-light emitting diode (LED) chip manufacturing, we developed a highly efficient and lightweight multi-branch gradient backhaul (MBGB) block. Based on the MBGB block, a mini-LED surface defect detector was designed, which included an MBGB network (MBGB-net) for the backbone and an MBGB feature pyramid network (MBGB-FPN) for the feature fusion networks. MBGB-net was introduced to reduce resource utilisation and achieve efficient information flow while enhancing defect feature extraction from mini-LED wafers. MBGB-FPN optimises the parameter utilisation, thereby reducing the demand for computational resources while maintaining, or even improving, the detection accuracy. Furthermore, a partial convolution module is integrated into the detection head to reduce the computational overhead and improve the detection speed. The experimental results demonstrated that the method achieved optimal performance in terms of both accuracy and speed. On the mini-LED wafer defect dataset, it achieved an mAP50 of 87.2% with only 9.3M parameters and 21.6G FLOPs, reaching an impressive FPS of 345.4. Furthermore, on the NEU-DET dataset, an mAP50 of 77.5% was achieved using the same parameters and FLOPs.
{"title":"MBGB-detector: A multi-branch gradient backhaul lightweight model for mini-LED surface defect detection","authors":"Yuanda Lin , Shuwan Pan , Jie Yu , Yade Hong , Fuming Wang , Jianeng Tang , Lixin Zheng , Songyan Chen","doi":"10.1016/j.compind.2024.104204","DOIUrl":"10.1016/j.compind.2024.104204","url":null,"abstract":"<div><div>To meet the growing demand for lightweight models and rapid defect detection in mini-light emitting diode (LED) chip manufacturing, we developed a highly efficient and lightweight multi-branch gradient backhaul (MBGB) block. Based on the MBGB block, a mini-LED surface defect detector was designed, which included an MBGB network (MBGB-net) for the backbone and an MBGB feature pyramid network (MBGB-FPN) for the feature fusion networks. MBGB-net was introduced to reduce resource utilisation and achieve efficient information flow while enhancing defect feature extraction from mini-LED wafers. MBGB-FPN optimises the parameter utilisation, thereby reducing the demand for computational resources while maintaining, or even improving, the detection accuracy. Furthermore, a partial convolution module is integrated into the detection head to reduce the computational overhead and improve the detection speed. The experimental results demonstrated that the method achieved optimal performance in terms of both accuracy and speed. On the mini-LED wafer defect dataset, it achieved an mAP50 of 87.2% with only 9.3M parameters and 21.6G FLOPs, reaching an impressive FPS of 345.4. Furthermore, on the NEU-DET dataset, an mAP50 of 77.5% was achieved using the same parameters and FLOPs.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104204"},"PeriodicalIF":8.2,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654741","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 : 2024-11-08DOI: 10.1016/j.compind.2024.104203
Pierluigi Del Nostro , Gerhard Goldbeck , Ferry Kienberger , Manuel Moertelmaier , Andrea Pozzi , Nawfal Al-Zubaidi-R-Smith , Daniele Toti
The demand for advanced battery management systems (BMSs) and battery test procedures is growing due to the rising importance of electric vehicles (EVs) and energy storage systems. The diversity of battery types, chemistries and application scenarios presents challenges in designing and optimizing BMSs and determining optimal battery test strategies. To address these challenges, semantic web technologies and ontologies offer a structured and common vocabulary for information sharing and reuse in battery management and testing. This work introduces the Battery Testing Ontology (BTO), a standardized, comprehensive, and semantically flexible framework for representing knowledge in electrical battery testing and quality control. BTO models a variety of electrical battery cell tests, specifying required test hardware and calibration procedures, mechanical fixturing of batteries, and referencing electrical measurement data. For example, it supports electrochemical impedance spectroscopy, self-discharge and high-voltage separator tests, the latter specifically demonstrating separator requirements, hardware specifications, and measurement details. Positioned within the ontology ecosystem of materials science, BTO aligns with the Elementary Multiperspective Material Ontology (EMMO) and related domain ontologies such as the Characterization Methodology Ontology (CHAMEO). This work elaborates on BTO’s development, structure, components and applications, highlighting its significant contributions to the field of battery testing.
{"title":"Battery testing ontology: An EMMO-based semantic framework for representing knowledge in battery testing and battery quality control","authors":"Pierluigi Del Nostro , Gerhard Goldbeck , Ferry Kienberger , Manuel Moertelmaier , Andrea Pozzi , Nawfal Al-Zubaidi-R-Smith , Daniele Toti","doi":"10.1016/j.compind.2024.104203","DOIUrl":"10.1016/j.compind.2024.104203","url":null,"abstract":"<div><div>The demand for advanced battery management systems (BMSs) and battery test procedures is growing due to the rising importance of electric vehicles (EVs) and energy storage systems. The diversity of battery types, chemistries and application scenarios presents challenges in designing and optimizing BMSs and determining optimal battery test strategies. To address these challenges, semantic web technologies and ontologies offer a structured and common vocabulary for information sharing and reuse in battery management and testing. This work introduces the Battery Testing Ontology (BTO), a standardized, comprehensive, and semantically flexible framework for representing knowledge in electrical battery testing and quality control. BTO models a variety of electrical battery cell tests, specifying required test hardware and calibration procedures, mechanical fixturing of batteries, and referencing electrical measurement data. For example, it supports electrochemical impedance spectroscopy, self-discharge and high-voltage separator tests, the latter specifically demonstrating separator requirements, hardware specifications, and measurement details. Positioned within the ontology ecosystem of materials science, BTO aligns with the Elementary Multiperspective Material Ontology (EMMO) and related domain ontologies such as the Characterization Methodology Ontology (CHAMEO). This work elaborates on BTO’s development, structure, components and applications, highlighting its significant contributions to the field of battery testing.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104203"},"PeriodicalIF":8.2,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：