liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Machine Learning In Design Engineering and Manufacturing
Linköping University, Department of Management and Engineering, Product Realisation. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0002-5950-4962
2023 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

Artificial intelligence (AI) has made significant strides in various fields, challenging conventional notions of computer capabilities. However, while data science research primarily concentrates on refining AI models, there are numerous challenges associated with integrating AI into industrial applications.

Knowledge-Based Engineering, with its potential to streamline the production cycle by reusing engineering knowledge and intent, emerges as a promising avenue for AI in the industry. When engineering knowledge is effectively processed and categorized, neural networks naturally emerge as potent tools for automation.

This thesis presents three case studies that demonstrate the practicality of supervised learning, particularly in the domain of neural networks, to address manufacturing automation challenges. These case studies span various stages of the manufacturing system, encompassing engineering design, production planning, and quality control phases. The first application employs supervised learning to automate the generation of engineering drawings, while the third employs optical character recognition to expedite the quality control process for complex engineering drawings. The second application centers on the estimation of fixturing clamps for welding operations in automobile parts.

In summary, this thesis strives to make a meaningful contribution to the field of design engineering and manufacturing by examining the potential of AI in enhancing processes and addressing automation hurdles. By presenting case studies that showcase the utility of machine learning models in production settings, this thesis aims to stimulate further research in this evolving field.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2023. , p. 48
Series
Linköping Studies in Science and Technology. Licentiate Thesis, ISSN 0280-7971 ; 1979
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
URN: urn:nbn:se:liu:diva-199612DOI: 10.3384/9789180754569ISBN: 9789180754552 (print)ISBN: 9789180754569 (electronic)OAI: oai:DiVA.org:liu-199612DiVA, id: diva2:1819147
Presentation
2023-12-15, ACAS, Linköping University, IEI, A-building, Linköping, 10:15 (English)
Opponent
Supervisors
Funder
Vinnova, 2020-02974Vinnova, 2021-02481Available from: 2023-12-15 Created: 2023-12-13 Last updated: 2023-12-15Bibliographically approved
List of papers
1. Automated and Customized CAD Drawings by Utilizing Machine Learning Algorithms: A Case Study
Open this publication in new window or tab >>Automated and Customized CAD Drawings by Utilizing Machine Learning Algorithms: A Case Study
2022 (English)In: ASME 2022 International Design Engineering Technical Conferences and Computers and Information in Engineering ConferenceAugust 14–17, 2022St. Louis, Missouri, USA: Volume 3B: 48th Design Automation Conference (DAC), St. Louis, MO, USA, 2022, Vol. BConference paper, Published paper (Refereed)
Abstract [en]

This paper describes a methodology for automation of measurements in Computer-Aided Design (CAD) software by enabling the use of supervised learning algorithms. The paper presents a proof of concept of how dimensions are placed automatically in the drawing at predicted positions. The framework consists of two trained neural networks and a rule-based system. Four steps compound the methodology. 1. Create a data set of labeled images for training a pre-built convolutional neural network (YOLOv5) using CAD automatic procedures. 2. Train the model to make predictions on 2D drawing imagery, identifying their relevant features. 3. Reuse the information extracted from YOLOv5 in a new neural network to produce measurement data. The output of this model is a matrix containing measurement location and size data. 4. Convert the final data output into actual measurements of an unseen geometry using a rule-based system for automatic dimension generation. Although the rule-based system is highly dependent on the problem and the CAD software, both supervised learning models exhibit high performance and reusability. Future work aims to make the framework suitable for more complex products. The methodology presented is promising and shows potential for minimizing human resources in repetitive CAD work, particularly in the task of creating engineering drawings.

Place, publisher, year, edition, pages
St. Louis, MO, USA: , 2022
National Category
Engineering and Technology
Identifiers
urn:nbn:se:liu:diva-196468 (URN)10.1115/DETC2022-88971 (DOI)978-0-7918-8623-6 (ISBN)
Conference
ASME 2022 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference
Projects
iProd
Available from: 2023-08-07 Created: 2023-08-07 Last updated: 2023-12-13
2. Application of optimized convolutional neural network to fixture layout in automotive parts
Open this publication in new window or tab >>Application of optimized convolutional neural network to fixture layout in automotive parts
2023 (English)In: The International Journal of Advanced Manufacturing Technology, ISSN 0268-3768, E-ISSN 1433-3015Article in journal (Refereed) Epub ahead of print
Abstract [en]

Fixture layout is a complex task that significantly impacts manufacturing costs and requires the expertise of well-trained engineers. While most research approaches to automating the fixture layout process use optimization or rule-based frameworks, this paper presents a novel approach using supervised learning. The proposed framework replicates the 3-2-1 locating principle to layout fixtures for sheet metal designs. This principle ensures the correct fixing of an object by restricting its degrees of freedom. One main novelty of the proposed framework is the use of topographic maps generated from sheet metal design data as input for a convolutional neural network (CNN). These maps are created by projecting the geometry onto a plane and converting the Z coordinate into gray-scale pixel values. The framework is also novel in its ability to reuse knowledge about fixturing to lay out new workpieces and in its integration with a CAD environment as an add-in. The results of the hyperparameter-tuned CNN for regression show high accuracy and fast convergence, demonstrating the usability of the model for industrial applications. The framework was first tested using automotive b-pillar designs and was found to have high accuracy (approximate to 100%) in classifying these designs. The proposed framework offers a promising approach for automating the complex task of fixture layout in sheet metal design.

Place, publisher, year, edition, pages
SPRINGER LONDON LTD, 2023
Keywords
Design automation; Machine learning; Fixtures; CNN; Hyperparameter tuning; EfficientNet
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-192681 (URN)10.1007/s00170-023-10995-0 (DOI)000938262100003 ()
Note

Funding Agencies|Linkping University; Vinnova-FFI (Fordonsstrategisk forskning ochinnovation) [2020-02974]

Available from: 2023-03-29 Created: 2023-03-29 Last updated: 2023-12-13
3. Optical character recognition on engineering drawings to achieve automation in production quality control
Open this publication in new window or tab >>Optical character recognition on engineering drawings to achieve automation in production quality control
2023 (English)In: Frontiers in Manufacturing Technology, E-ISSN 2813-0359, Vol. 3Article in journal (Refereed) Published
Abstract [en]

Introduction: Digitization is a crucial step towards achieving automation in production quality control for mechanical products. Engineering drawings are essential carriers of information for production, but their complexity poses a challenge for computer vision. To enable automated quality control, seamless data transfer between analog drawings and CAD/CAM software is necessary.

Methods: This paper focuses on autonomous text detection and recognition in engineering drawings. The methodology is divided into five stages. First, image processing techniques are used to classify and identify key elements in the drawing. The output is divided into three elements: information blocks and tables, feature control frames, and the rest of the image. For each element, an OCR pipeline is proposed. The last stage is output generation of the information in table format.

Results: The proposed tool, called eDOCr, achieved a precision and recall of 90% in detection, an F1-score of 94% in recognition, and a character error rate of 8%. The tool enables seamless integration between engineering drawings and quality control.

Discussion: Most OCR algorithms have limitations when applied to mechanical drawings due to their inherent complexity, including measurements, orientation, tolerances, and special symbols such as geometric dimensioning and tolerancing (GD&T). The eDOCr tool overcomes these limitations and provides a solution for automated quality control.

Conclusion: The eDOCr tool provides an effective solution for automated text detection and recognition in engineering drawings. The tool's success demonstrates that automated quality control for mechanical products can be achieved through digitization. The tool is shared with the research community through Github.

Place, publisher, year, edition, pages
Frontiers Media S.A., 2023
Keywords
optical character recognition, image segmentation, object detection, engineering drawings, quality control, keras-ocr
National Category
Engineering and Technology Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:liu:diva-195416 (URN)10.3389/fmtec.2023.1154132 (DOI)
Funder
Vinnova, 2021-02481
Available from: 2023-06-20 Created: 2023-06-20 Last updated: 2023-12-13Bibliographically approved
4. Model Architecture Exploration Using Chatgpt for Specific Manufacturing Applications
Open this publication in new window or tab >>Model Architecture Exploration Using Chatgpt for Specific Manufacturing Applications
2023 (English)In: ASME IDETC-CIE, 2023, Vol. 2Conference paper, Published paper (Refereed)
Abstract [en]

Selecting an appropriate machine learning model architecture for manufacturing tasks requires expertise in both computer science and manufacturing. However, integrating state-of-the-art machine learning models and manufacturing processes is often challenging due to the distance between these fields. OpenAI’s popular language model, ChatGPT, has the potential to bridge this gap.

This paper proposes guidelines and questions to explore model architecture options and extract valuable information from ChatGPT’s natural language processing capabilities. While ChatGPT is a powerful tool, it is important to verify any answers obtained against reliable sources before making any decisions. The guidelines compose a flowchart with four queries to give ChatGPT enough context and exisiting input data information. ChatGPT suggestions will be directed towards input processing, output, and architecture proposals. The last query produces keywords based on the chat for a background study on the topic.

A manufacturing case study was conducted to demonstrate the effectiveness of these guidelines. The study involved creating a model to forecast fixturing locations for welding processes in the automotive sector. After conducting four separate interviews with ChatGPT, the authors discuss the selection of architecture based on ChatGPT suggestions and contrast it with previous literature.

The proposed guidelines are expected to be useful in a variety of manufacturing contexts, as they offer a structured approach to exploring model architecture options using ChatGPT’s capabilities, ultimately leading to new and innovative applications of machine learning in this field.

Keywords
fixture layout, machine learning, ChatGPT, manufacturing planning, model exploration
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:liu:diva-199611 (URN)10.1115/DETC2023-116483 (DOI)978-0-7918-8729-5 (ISBN)
Conference
IDETC-CIE 43rd Computers and Information in Engineering
Funder
Vinnova, 2020-02974
Available from: 2023-12-13 Created: 2023-12-13 Last updated: 2023-12-13

Open Access in DiVA

fulltext(6658 kB)272 downloads
File information
File name FULLTEXT02.pdfFile size 6658 kBChecksum SHA-512
e88d5ae4264c16cdcd093a28d910a8c0966a24c16bbebd5ce7f44a3584f06dd8889bf35147b9e2f5c76ee9f64d2979f7f4dff932a45c10452ae8a82d9989ca5a
Type fulltextMimetype application/pdf
Order online >>

Other links

Publisher's full text

Authority records

Villena Toro, Javier

Search in DiVA

By author/editor
Villena Toro, Javier
By organisation
Product RealisationFaculty of Science & Engineering
Production Engineering, Human Work Science and Ergonomics

Search outside of DiVA

GoogleGoogle Scholar
Total: 272 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 600 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf