A technology maturity assessment framework for Industry 5.0 machine vision systems based on systematic literature review in automotive manufacturing

被引:27
作者
Konstantinidis, Fotios K. [1 ]
Myrillas, Nikolaos [1 ]
Tsintotas, Konstantinos A. [1 ]
Mouroutsos, Spyridon G. [2 ]
Gasteratos, Antonios [1 ]
机构
[1] Democritus Univ Thrace, Dept Prod & Management Engn, 12 Vas Sophias, GR-67132 Xanthi, Greece
[2] Democritus Univ Thrace, Dept Elect & Comp Engn, Xanthi, Greece
关键词
Maturity assessment; machine vision; systematic literature; automotive manufacturing; industry; 5.0; zero defect manufacturing; SURFACE; INSPECTION; MODEL; CLASSIFICATION; RECOGNITION; METHODOLOGY;
D O I
10.1080/00207543.2023.2270588
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
When considering how an intelligent factory can 'see,' the answer lies in machine vision technology. To assess the current technological advancements of machine vision systems and propose a technology maturity assessment framework, a nine-phase Systematic Literature Review (SLR) strategy was implemented. As the automotive industry stands at the forefront of autonomous systems, we analysed 85 works across the entire automotive manufacturing life cycle. The findings revealed that machine vision is utilised in each technological pillar of Industry 4.0, encompassing autonomous robots, augmented reality, predictive maintenance, additive manufacturing, and more. In analysing 22 vision-based applications in 47 automotive components, we clustered machine vision systems' architectural components and processing techniques, ranging from threshold-based methods to advanced reinforcement learning techniques suitable for the I5.0 environment. Leveraging the insights gathered, we propose the I5.0 technology maturity assessment framework for machine vision systems, evaluating nine functional components across five scaling technology levels. This framework serves as a valuable tool to identify weaknesses and opportunities for improvement, guiding machine vision integration into an intelligent factory.
引用
收藏
页数:37
相关论文
共 172 条
[61]   Deep Feature Space: A Geometrical Perspective [J].
Kansizoglou, Ioannis ;
Bampis, Loukas ;
Gasteratos, Antonios .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (10) :6823-6838
[62]   Continuous Emotion Recognition for Long-Term Behavior Modeling through Recurrent Neural Networks [J].
Kansizoglou, Ioannis ;
Misirlis, Evangelos ;
Tsintotas, Konstantinos ;
Gasteratos, Antonios .
TECHNOLOGIES, 2022, 10 (03)
[63]   Do Neural Network Weights Account for Classes Centers? [J].
Kansizoglou, Ioannis ;
Bampis, Loukas ;
Gasteratos, Antonios .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 34 (11) :8815-8824
[64]  
Katika Tina, 2022, 2022 IEEE INT C IM S, P1
[65]   Machine vision based automated inspection approach for clutch friction disc (CFD) [J].
Kaushik, Sujata ;
Jain, Anjali ;
Chaudhary, Tina ;
Chauhan, N. R. .
MATERIALS TODAY-PROCEEDINGS, 2022, 62 :151-157
[66]  
Keele S., 2007, Guidelines for performing systematic literature reviews in software engineering
[67]  
Konstantinidis Fotios K., 2023, Procedia Computer Science, P227, DOI [10.1016/j.procs.2022.12.218, 10.1016/j.procs.2022.12.218]
[68]  
Konstantinidis Fotios K., 2023, Procedia Computer Science, P918, DOI [10.1016/j.procs.2022.12.289, 10.1016/j.procs.2022.12.289]
[69]  
Konstantinidis F. K., 2018, 2018 IEEE INT C IM S, P1
[70]  
Konstantinidis F. K., 2021, 2021 IEEE INT C IM S, P1