Design-by-Analogy (DbA) is a design methodology that draws inspiration from a source domain to a target domain to generate new solutions to problems or designs, which can benefit designers in mitigating design fixation and improving design ideation outcomes. Recently, the increasingly available design databases and rapidly advancing data science and artificial intelligence technologies have presented new opportunities for developing data-driven methods and tools for DbA support. Herein, we survey the prior data-driven DbA studies and categorize and analyze individual study according to the data, methods and applications in four categories including analogy encoding, retrieval, mapping, and evaluation. Based on such structured literature analysis, this paper elucidates the state of the art of data-driven DbA research to date and benchmarks it with the frontier of data science and AI research to identify promising research opportunities and directions for the field.
{"title":"Data-Driven Design-by-Analogy: State of the Art","authors":"Shuo Jiang, Jie Hu, Jianxi Luo","doi":"10.1115/detc2021-68669","DOIUrl":"https://doi.org/10.1115/detc2021-68669","url":null,"abstract":"\u0000 Design-by-Analogy (DbA) is a design methodology that draws inspiration from a source domain to a target domain to generate new solutions to problems or designs, which can benefit designers in mitigating design fixation and improving design ideation outcomes. Recently, the increasingly available design databases and rapidly advancing data science and artificial intelligence technologies have presented new opportunities for developing data-driven methods and tools for DbA support. Herein, we survey the prior data-driven DbA studies and categorize and analyze individual study according to the data, methods and applications in four categories including analogy encoding, retrieval, mapping, and evaluation. Based on such structured literature analysis, this paper elucidates the state of the art of data-driven DbA research to date and benchmarks it with the frontier of data science and AI research to identify promising research opportunities and directions for the field.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74542054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The goal of the study is to examine the effectiveness of using an incentivized “test/retest” scenario to improve participants’ performance during stressful situations. The study makes use of an electroencephalography (EEG) machine to detect participants’ stress levels during a one-minute typing test. The typing test administered was a standard, “story-typing” test. A total of 23 student participants were randomly divided into two cohorts: the control cohort and the experimental cohort. Participants were asked to complete a preliminary questionnaire self-assessing their ability to handle stressful situations. Both cohorts were then asked to complete the typing test (hereafter referred to as T1) and fill out an Emotional Stress Reaction Questionnaire (ESRQ), indicating their emotions during the typing test. The participants were then asked to complete the typing test and accompanying ESRQ a second time (hereafter referred to as T2). However, prior to the second test, the participants in the experimental cohort were told that the participant that shows the most improvement in their typing speed (measured in words per minute) will receive a $100 gift card. This stimulus is used to increase the already stressful situation for the experimental cohort and examine whether participants’ brain activity changes when the “retest” is incentivized. Each participant’s EEG data and heartrate were measured through the duration of the experiment and t-tests and regression analyses were used to determine if a statistically significant difference existed between cohorts (control vs. experimental) or within cohorts (T1 vs. T2). The results show that there were no significant changes in brain activity, emotions, or typing performance for the control group of participants (no reward offered). However, the experimental group showed an increase in EEG sensor activity; specifically, the sensors that control vision and emotion. Interestingly, the participant’s performance was found to be correlated to their emotional responses, rather than their EEG sensor data. Additionally, the experimental groups’ positive emotions were increased for the second typing test, which is incentivized. The findings lay a foundation for design settings scenarios where preparatory practices can be incorporated.
{"title":"Neurocognitive Effects of Incentivizing Students to Improve Performance Through Repeat Attempts in Design Settings","authors":"Devanshi Shah, Elisabeth Kames, Beshoy Morkos","doi":"10.1115/detc2021-72058","DOIUrl":"https://doi.org/10.1115/detc2021-72058","url":null,"abstract":"\u0000 The goal of the study is to examine the effectiveness of using an incentivized “test/retest” scenario to improve participants’ performance during stressful situations. The study makes use of an electroencephalography (EEG) machine to detect participants’ stress levels during a one-minute typing test. The typing test administered was a standard, “story-typing” test. A total of 23 student participants were randomly divided into two cohorts: the control cohort and the experimental cohort. Participants were asked to complete a preliminary questionnaire self-assessing their ability to handle stressful situations. Both cohorts were then asked to complete the typing test (hereafter referred to as T1) and fill out an Emotional Stress Reaction Questionnaire (ESRQ), indicating their emotions during the typing test. The participants were then asked to complete the typing test and accompanying ESRQ a second time (hereafter referred to as T2). However, prior to the second test, the participants in the experimental cohort were told that the participant that shows the most improvement in their typing speed (measured in words per minute) will receive a $100 gift card.\u0000 This stimulus is used to increase the already stressful situation for the experimental cohort and examine whether participants’ brain activity changes when the “retest” is incentivized. Each participant’s EEG data and heartrate were measured through the duration of the experiment and t-tests and regression analyses were used to determine if a statistically significant difference existed between cohorts (control vs. experimental) or within cohorts (T1 vs. T2).\u0000 The results show that there were no significant changes in brain activity, emotions, or typing performance for the control group of participants (no reward offered). However, the experimental group showed an increase in EEG sensor activity; specifically, the sensors that control vision and emotion. Interestingly, the participant’s performance was found to be correlated to their emotional responses, rather than their EEG sensor data. Additionally, the experimental groups’ positive emotions were increased for the second typing test, which is incentivized. The findings lay a foundation for design settings scenarios where preparatory practices can be incorporated.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73692137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Emerging complex engineered systems may have unexpected safety issues due to novel operational environments, increasing autonomy, human-machine interaction, and other factors. To prevent failures in operation or testing that necessitate costly redesign, it is desirable to predict likely failure modes early in the design process. Information about past engineering failures in natural language format presents one possible solution by enabling the retrieval of information that can inform new designs. However, identifying documents containing usable information and extracting the required information can be prohibitively time-consuming when implemented at scale. In this research, an automated natural language processing (NLP) framework is proposed to discover relevant knowledge from documents containing failure-related design information. The framework is applied to NASA’s Lessons Learned Information System (LLIS), which is publicly available. Documents containing usable information are filtered using two different NLP-based models. Next, from the identified usable documents, a failure taxonomy is extracted using a partitioned hierarchical topic modeling approach. Partitions of the document describe different sections of the failure taxonomy — i.e., failure, cause of failure, and recommendations — as indicated by the structure of the original document. The extracted failure taxonomy can be leveraged in early design failure assessment methods. Moreover, the framework can be used to identify documents containing usable failure-related design information from other databases and extract relevant information from these documents.
{"title":"Knowledge Discovery for Early Failure Assessment of Complex Engineered Systems Using Natural Language Processing","authors":"Sequoia R. Andrade, Hannah S. Walsh","doi":"10.1115/detc2021-70694","DOIUrl":"https://doi.org/10.1115/detc2021-70694","url":null,"abstract":"\u0000 Emerging complex engineered systems may have unexpected safety issues due to novel operational environments, increasing autonomy, human-machine interaction, and other factors. To prevent failures in operation or testing that necessitate costly redesign, it is desirable to predict likely failure modes early in the design process. Information about past engineering failures in natural language format presents one possible solution by enabling the retrieval of information that can inform new designs. However, identifying documents containing usable information and extracting the required information can be prohibitively time-consuming when implemented at scale. In this research, an automated natural language processing (NLP) framework is proposed to discover relevant knowledge from documents containing failure-related design information. The framework is applied to NASA’s Lessons Learned Information System (LLIS), which is publicly available. Documents containing usable information are filtered using two different NLP-based models. Next, from the identified usable documents, a failure taxonomy is extracted using a partitioned hierarchical topic modeling approach. Partitions of the document describe different sections of the failure taxonomy — i.e., failure, cause of failure, and recommendations — as indicated by the structure of the original document. The extracted failure taxonomy can be leveraged in early design failure assessment methods. Moreover, the framework can be used to identify documents containing usable failure-related design information from other databases and extract relevant information from these documents.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81095150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the last years, museums have begun to apply new technological solutions to manage their exhibits in a more open, inclusive, and creative way, to improve the visitors’ experience to respond to the need to expand the audience. The main goal is to face the increasing competition in an economy referred to as the “Experience Economy”. To this end, Augmented Reality technology seems to represent a good solution for museum guide systems, to improve visitors’ learning and enjoyment. In this context, the present paper proposes a museum guide system based on Spatial Augmented Reality powered by dynamic projection. The paper describes the overall HW and SW system architecture and reports in detail the developed process adopted to design and implement a museum guide and entertainment application, in the context of the “Studiolo of Federico da Montefeltro” in the Ducal Palace of Urbino. A preliminary survey has been carried out, which involved a total of 79 subjects, aimed at investigating the quality of visitor’s experience, aroused by the proposed application, in terms of the “Four Experience Realms” defined by Pine & Gilmore (1998). Results suggest that the proposed application can be used to stage experiences that satisfy the visitors and may help to enable museums into the Experience Economy.
近年来，博物馆开始运用新的技术解决方案，以更开放、更包容、更有创意的方式管理展品，以改善参观者的体验，以回应扩大观众的需要。主要目标是面对被称为“体验经济”的经济中日益激烈的竞争。为此，增强现实技术似乎是博物馆导览系统的一个很好的解决方案，以提高游客的学习和享受。在此背景下，本文提出了一种基于动态投影的空间增强现实的博物馆导览系统。本文描述了整个硬件和软件系统架构，并详细报告了在乌尔比诺公爵宫的“Federico da Montefeltro工作室”的背景下设计和实现博物馆导游和娱乐应用程序的开发过程。一项初步调查已经进行，涉及总共79个主题，旨在调查提出的应用程序所引起的游客体验质量，根据Pine和Gilmore(1998)定义的“四个体验领域”。结果表明，所提出的应用程序可以用来提供满足游客的体验，并可能有助于使博物馆进入体验经济。
{"title":"A New Paradigm for the Enjoyment and Exploitation of Cultural Heritage Based on Spatial Augmented Reality: The Case of the Ducal Palace of Urbino","authors":"Alma Leopardi, S. Ceccacci, M. Mengoni","doi":"10.1115/detc2021-68896","DOIUrl":"https://doi.org/10.1115/detc2021-68896","url":null,"abstract":"\u0000 In the last years, museums have begun to apply new technological solutions to manage their exhibits in a more open, inclusive, and creative way, to improve the visitors’ experience to respond to the need to expand the audience. The main goal is to face the increasing competition in an economy referred to as the “Experience Economy”. To this end, Augmented Reality technology seems to represent a good solution for museum guide systems, to improve visitors’ learning and enjoyment.\u0000 In this context, the present paper proposes a museum guide system based on Spatial Augmented Reality powered by dynamic projection. The paper describes the overall HW and SW system architecture and reports in detail the developed process adopted to design and implement a museum guide and entertainment application, in the context of the “Studiolo of Federico da Montefeltro” in the Ducal Palace of Urbino. A preliminary survey has been carried out, which involved a total of 79 subjects, aimed at investigating the quality of visitor’s experience, aroused by the proposed application, in terms of the “Four Experience Realms” defined by Pine & Gilmore (1998).\u0000 Results suggest that the proposed application can be used to stage experiences that satisfy the visitors and may help to enable museums into the Experience Economy.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78484680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
N. Ivezic, B. Kulvatunyou, Elena Jelisic, Hakju Oh, S. Frechette, V. Srinivasan
It is generally observed in inter-organizational communication that present-day data exchange standards are too costly, large, slow to respond to industry demands, and complex to develop and use. These problems in data exchange are felt keenly by the manufacturing industry and its vast supply chains. In addressing these challenges, a successful attempt was recently made that involves two major developments. The first is the Core Components Technical Specification (CCTS), which is an ISO-approved meta-model for data exchange standards. CCTS introduces common data types, uniform structure for data models, and data usage semantics. The second is Score, which is a novel open-source software tool. Score was developed by National Institute of Standards and Technology researchers as a platform to take advantage of the CCTS for data exchange standards development and usage. This paper describes the potential of the CCTS and the Score platform, and describes the current status of the Score platform and Score-enabled industry interactions.
在组织间通信中，通常可以观察到，当前的数据交换标准过于昂贵、庞大、对行业需求的响应太慢，而且开发和使用起来很复杂。制造业及其庞大的供应链敏锐地感受到了数据交换中的这些问题。在应对这些挑战方面，最近进行了一次成功的尝试，其中涉及两项重大发展。第一个是核心组件技术规范(CCTS)，它是iso批准的用于数据交换标准的元模型。CCTS引入了通用数据类型、数据模型的统一结构和数据使用语义。第二个是Score，这是一个新颖的开源软件工具。Score是由美国国家标准与技术研究所(National Institute of Standards and Technology)的研究人员开发的，作为一个利用CCTS开发和使用数据交换标准的平台。本文描述了CCTS和Score平台的潜力，并描述了Score平台和支持Score的行业交互的现状。
{"title":"A Novel Data Standards Platform Using the ISO Core Components Technical Specification","authors":"N. Ivezic, B. Kulvatunyou, Elena Jelisic, Hakju Oh, S. Frechette, V. Srinivasan","doi":"10.1115/detc2021-68067","DOIUrl":"https://doi.org/10.1115/detc2021-68067","url":null,"abstract":"\u0000 It is generally observed in inter-organizational communication that present-day data exchange standards are too costly, large, slow to respond to industry demands, and complex to develop and use. These problems in data exchange are felt keenly by the manufacturing industry and its vast supply chains. In addressing these challenges, a successful attempt was recently made that involves two major developments. The first is the Core Components Technical Specification (CCTS), which is an ISO-approved meta-model for data exchange standards. CCTS introduces common data types, uniform structure for data models, and data usage semantics. The second is Score, which is a novel open-source software tool. Score was developed by National Institute of Standards and Technology researchers as a platform to take advantage of the CCTS for data exchange standards development and usage. This paper describes the potential of the CCTS and the Score platform, and describes the current status of the Score platform and Score-enabled industry interactions.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86629957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Apurva Patel, J. Summers, A. Patel, James L. Mathieson, Michael P. Sbarra, Joshua Ortiz
While fundamentals of DFMA are widely accepted and used in the engineering design community, many CAD environments lack tools that address manufacturing concerns and provide rapid feedback to designers about manufacturing impacts of their design choices. This paper presents an experiment-based testing and validation of a rapid feedback tool that provides users a history-based prediction of manufacturing time based on the current state of the design. A between-subjects experiment is designed to evaluate the impact of the tool on design outcomes based on modeling time, part mass, and manufacturing time. Participants in the study included mechanical engineering graduate and undergraduate students with at least one semester of experience using SolidWorks. The experiment included three different design activities and three different conditions of the design tool. Participants completed up to three sessions with different experimental conditions. Analysis of the data collected shows that use of the design tool results in a small but nonsignificant increase in modeling time. Moreover, use of the tool results in reduced part mass on average, as well as in a within-subject comparison. Tool use reduced manufacturing time in open ended activities, but increased manufacturing time when activities focus more on mass-reduction. Participant feedback suggests that the tool helped guide their material removal actions by showing the impact on manufacturing time. Finally, potential improvements and future expansions of the tool are discussed.
{"title":"Testing and Validation of a Custom CAD Tool to Support Design for Manufacturing: An Experimental Study","authors":"Apurva Patel, J. Summers, A. Patel, James L. Mathieson, Michael P. Sbarra, Joshua Ortiz","doi":"10.1115/detc2021-69820","DOIUrl":"https://doi.org/10.1115/detc2021-69820","url":null,"abstract":"\u0000 While fundamentals of DFMA are widely accepted and used in the engineering design community, many CAD environments lack tools that address manufacturing concerns and provide rapid feedback to designers about manufacturing impacts of their design choices. This paper presents an experiment-based testing and validation of a rapid feedback tool that provides users a history-based prediction of manufacturing time based on the current state of the design. A between-subjects experiment is designed to evaluate the impact of the tool on design outcomes based on modeling time, part mass, and manufacturing time. Participants in the study included mechanical engineering graduate and undergraduate students with at least one semester of experience using SolidWorks. The experiment included three different design activities and three different conditions of the design tool. Participants completed up to three sessions with different experimental conditions. Analysis of the data collected shows that use of the design tool results in a small but nonsignificant increase in modeling time. Moreover, use of the tool results in reduced part mass on average, as well as in a within-subject comparison. Tool use reduced manufacturing time in open ended activities, but increased manufacturing time when activities focus more on mass-reduction. Participant feedback suggests that the tool helped guide their material removal actions by showing the impact on manufacturing time. Finally, potential improvements and future expansions of the tool are discussed.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91154813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Kato, Ken Yoshitugu, N. Maeda, T. Hirogaki, E. Aoyama, Kenichi Takahashi
Most industrial robots are taught using the teaching playback method; therefore, they are unsuitable for use in variable production systems. Although offline teaching methods have been developed, they have not been practiced because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have attempted to calibrate the position and posture but have not reached a practical level, as such methods consider the joint angle when the robot is stationary rather than the features during robot motion. Currently, it is easy to obtain servo information under numerical control operations owing to the Internet of Things technologies. In this study, we propose a method for obtaining servo information during robot motion and converting it into images to find features using a convolutional neural network (CNN). Herein, a large industrial robot was used. The three-dimensional coordinates of the end-effector were obtained using a laser tracker. The positioning error of the robot was accurately learned by the CNN. We extracted the features of the points where the positioning error was extremely large. By extracting the features of the X-axis positioning error using the CNN, the joint 1 current is a feature. This indicates that the vibration current in joint 1 is a factor in the X-axis positioning error.
{"title":"Finding Features of Positioning Error for Large Industrial Robots Based on Convolutional Neural Network","authors":"D. Kato, Ken Yoshitugu, N. Maeda, T. Hirogaki, E. Aoyama, Kenichi Takahashi","doi":"10.1115/detc2021-68237","DOIUrl":"https://doi.org/10.1115/detc2021-68237","url":null,"abstract":"\u0000 Most industrial robots are taught using the teaching playback method; therefore, they are unsuitable for use in variable production systems. Although offline teaching methods have been developed, they have not been practiced because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have attempted to calibrate the position and posture but have not reached a practical level, as such methods consider the joint angle when the robot is stationary rather than the features during robot motion. Currently, it is easy to obtain servo information under numerical control operations owing to the Internet of Things technologies. In this study, we propose a method for obtaining servo information during robot motion and converting it into images to find features using a convolutional neural network (CNN). Herein, a large industrial robot was used. The three-dimensional coordinates of the end-effector were obtained using a laser tracker. The positioning error of the robot was accurately learned by the CNN. We extracted the features of the points where the positioning error was extremely large. By extracting the features of the X-axis positioning error using the CNN, the joint 1 current is a feature. This indicates that the vibration current in joint 1 is a factor in the X-axis positioning error.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86877305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Material removal rate (MRR) plays a critical role in the operation of chemical mechanical planarization (CMP) process in the semiconductor industry. To date, many physics-based and data-driven approaches have been proposed to predict the MRR. Nevertheless, most of the existing methodologies neglect the potential source of its well-organized and underlying equipment structure containing interaction mechanisms among different components. To address its limitation, this paper proposes a novel hypergraph neural network-based approach for predicting the MRR in CMP. Two main scientific contributions are presented in this work: 1) establishing a generic modeling technique to construct the complex equipment knowledge graph with a hypergraph form base on the comprehensive understanding and analysis of equipment structure and mechanism, and 2) proposing a novel prediction method by combining the Recurrent Neural Network based model and the Hypergraph Neural Network to learn the complex data correlation and high-order representation base on the Spatio-temporal equipment hypergraph. To validate the proposed approach, a case study is conducted based on an open-source dataset. The experimental results prove that the proposed model can capture the hidden data correlation effectively. It is also envisioned that the proposed approach has great potentials to be applied in other similar smart manufacturing scenarios.
{"title":"Predicting the Material Removal Rate in Chemical Mechanical Planarization Process: A Hypergraph Neural Network-Based Approach","authors":"Liqiao Xia, Pai Zheng, Chao Liu","doi":"10.1115/detc2021-68250","DOIUrl":"https://doi.org/10.1115/detc2021-68250","url":null,"abstract":"\u0000 Material removal rate (MRR) plays a critical role in the operation of chemical mechanical planarization (CMP) process in the semiconductor industry. To date, many physics-based and data-driven approaches have been proposed to predict the MRR. Nevertheless, most of the existing methodologies neglect the potential source of its well-organized and underlying equipment structure containing interaction mechanisms among different components. To address its limitation, this paper proposes a novel hypergraph neural network-based approach for predicting the MRR in CMP. Two main scientific contributions are presented in this work: 1) establishing a generic modeling technique to construct the complex equipment knowledge graph with a hypergraph form base on the comprehensive understanding and analysis of equipment structure and mechanism, and 2) proposing a novel prediction method by combining the Recurrent Neural Network based model and the Hypergraph Neural Network to learn the complex data correlation and high-order representation base on the Spatio-temporal equipment hypergraph. To validate the proposed approach, a case study is conducted based on an open-source dataset. The experimental results prove that the proposed model can capture the hidden data correlation effectively. It is also envisioned that the proposed approach has great potentials to be applied in other similar smart manufacturing scenarios.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90110794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a machine learning approach for building an object detector for interactive simulation of planar linkages from handmade sketches and drawings found in patents and texts. Touch- and pen-input devices and interfaces have made sketching a more natural way for designers to express their ideas, especially during early design stages, but sketching existing complex mechanisms can be tedious and error-prone. While there are software applications available to help users make drawings, including that of a linkage mechanism, it is both educational and instructive to see existing sketches come to life via automated simulation. However, texts and patents present rich and diverse styles of mechanism drawings, which makes automated recognition difficult. Modern machine learning algorithms for object recognition require an extensive number of training images. However, there are no data sets of planar linkages available online. Therefore, our first goal was to generate images of sketches similar to hand-drawn ones and use state-of-the-art deep generation models, such as β-VAE, to produce more training data from a limited set of images. The latent space of β-VAE was explored by linear and spherical interpolations between sub-spaces and by varying latent space’s dimensions. This served two-fold objectives — 1) examine the possibility of generating new synthesized images via interpolation and 2) develop insights in the dependence of latent space dimension on bar linkage parameters. t-SNE dimensionality reduction technique was implemented to visualize the latent space of a β-VAE in a 2D space. Training images produced by animation rendering were used for fine-tuning a real-time object detection system — YOLOv3.
{"title":"Sketch-Based Mechanism Simulation Using Machine Learning","authors":"Anar Nurizada, A. Purwar","doi":"10.1115/detc2021-72149","DOIUrl":"https://doi.org/10.1115/detc2021-72149","url":null,"abstract":"\u0000 This paper presents a machine learning approach for building an object detector for interactive simulation of planar linkages from handmade sketches and drawings found in patents and texts. Touch- and pen-input devices and interfaces have made sketching a more natural way for designers to express their ideas, especially during early design stages, but sketching existing complex mechanisms can be tedious and error-prone. While there are software applications available to help users make drawings, including that of a linkage mechanism, it is both educational and instructive to see existing sketches come to life via automated simulation. However, texts and patents present rich and diverse styles of mechanism drawings, which makes automated recognition difficult. Modern machine learning algorithms for object recognition require an extensive number of training images. However, there are no data sets of planar linkages available online. Therefore, our first goal was to generate images of sketches similar to hand-drawn ones and use state-of-the-art deep generation models, such as β-VAE, to produce more training data from a limited set of images. The latent space of β-VAE was explored by linear and spherical interpolations between sub-spaces and by varying latent space’s dimensions. This served two-fold objectives — 1) examine the possibility of generating new synthesized images via interpolation and 2) develop insights in the dependence of latent space dimension on bar linkage parameters. t-SNE dimensionality reduction technique was implemented to visualize the latent space of a β-VAE in a 2D space. Training images produced by animation rendering were used for fine-tuning a real-time object detection system — YOLOv3.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79446515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The paper presents a method to develop Virtual Reality (VR) platforms based on serious games for the rehabilitation of severe memory loss. In particular, it is related to retrograde amnesia, a condition affecting patient’s quality of life usually caused by brain stroke. Nowadays, the standard rehabilitation process consists in showing pictures of patient’s familiar environments in order to recover the memory. Past research works have investigated the use of 3D scanners for the virtualization of real environment and virtual reality for the generation of more immersive interaction to design serious games for neurocognitive rehabilitation. Reached results highlighted a time-consuming development process to interface each new environment with the game logic specifically developed for the serious games. Furthermore, a complete VR platform must also consider the medical monitoring and the data management oriented to a more objective medical assessment. The proposed method allows the design of VR platforms based on patient-specific serious games for memory loss starting from the 3D scanning acquisition of familiar environments. The 3D acquisition is performed using the Occipital Structure Sensor and the Skanect application. A modular procedure has been designed to interface the virtual objects of each acquired environment with the modules of the game-logic developed with Unity. The immersive Virtual Reality is based on the use of the HTC Vive Pro head mounted display. Furthermore, the method permits to associate the patient-specific serious game to a set of software modules for the medical monitoring and the data management for the generation of reports useful for the evaluation. The solution has been evaluated by measuring the time needed to develop a whole VR platform for two different familiar environments. Less than 5 hours are required to complete the design process.
{"title":"A Method to Develop Virtual Reality Platforms for the Medical Rehabilitation of Severe Memory Loss After Brain Stroke","authors":"Daniel Lanzoni, A. Vitali, D. Regazzoni, C. Rizzi","doi":"10.1115/detc2021-70319","DOIUrl":"https://doi.org/10.1115/detc2021-70319","url":null,"abstract":"\u0000 The paper presents a method to develop Virtual Reality (VR) platforms based on serious games for the rehabilitation of severe memory loss. In particular, it is related to retrograde amnesia, a condition affecting patient’s quality of life usually caused by brain stroke. Nowadays, the standard rehabilitation process consists in showing pictures of patient’s familiar environments in order to recover the memory. Past research works have investigated the use of 3D scanners for the virtualization of real environment and virtual reality for the generation of more immersive interaction to design serious games for neurocognitive rehabilitation. Reached results highlighted a time-consuming development process to interface each new environment with the game logic specifically developed for the serious games. Furthermore, a complete VR platform must also consider the medical monitoring and the data management oriented to a more objective medical assessment.\u0000 The proposed method allows the design of VR platforms based on patient-specific serious games for memory loss starting from the 3D scanning acquisition of familiar environments. The 3D acquisition is performed using the Occipital Structure Sensor and the Skanect application. A modular procedure has been designed to interface the virtual objects of each acquired environment with the modules of the game-logic developed with Unity. The immersive Virtual Reality is based on the use of the HTC Vive Pro head mounted display. Furthermore, the method permits to associate the patient-specific serious game to a set of software modules for the medical monitoring and the data management for the generation of reports useful for the evaluation. The solution has been evaluated by measuring the time needed to develop a whole VR platform for two different familiar environments. Less than 5 hours are required to complete the design process.","PeriodicalId":23602,"journal":{"name":"Volume 2: 41st Computers and Information in Engineering Conference (CIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82039951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：