PEN: Process Estimator neural Network for root cause analysis using graph convolution. (January 2022)

• Record Type: Journal Article
• Authors: Leonhardt, Viktor; Claus, Felix; Garth, Christoph
• Is Part Of: Journal of manufacturing systems. Volume 62(2022)
• Page Start: 886
• Page End: 902
• Publication Date: 2022-01
• Subjects: Root cause analysis -- Variation source identification -- Graph convolution neural network; Manufacturing processes -- Periodicals; Production engineering -- Data processing -- Periodicals; Robots, Industrial -- Periodicals; Production, Technique de la -- Informatique -- Périodiques; Robots industriels -- Périodiques; Electronic journals; 670.42
• DOI: 10.1016/j.jmsy.2021.11.008
• Languages: English
• ISSNs: 0278-6125

Abstract

Highlights: Real-time root cause analysis using graph convolution neural network. Solving the standard state-space model using a graph convolution neural network. Numerical optimization ensure high performance by using Chebyshev polynomials. Optimizing the architecture to noise afflicted data by filtering only in low frequencies. The convolution, applied on the product's outer hull, neglecting Key Product Characteristics. Abstract: Root cause analysis in modern multistage assembly lines is a challenging, yet widely used technique to increase the product quality. Improvements – due to Industry 4.0 – aim for near-zero-defects manufacturing. Thus, we propose a novel root cause analysis: the Process Estimator neural Network (PEN) to solve the sparse, nonlinear problem of the state-space model empowering a graph convolution neural network. The contributions of this paper are: (1) study a novel problem of utilizing nonlinear deep neural networks to solve the state-space model; (2) elaborating the use of a graph convolution neural network to scope with the current limitations of linear approaches, which cannot process dense 3D point cloud data of the outer skin of the product; (3) how to analyze the trained network for fine tuning. We showed through a realistic experiment how PEN performs on huge 3D point clouds (188.000 points or higher) in form of meshed CAD models of first-order shell elements. These experiments set an example on how to overcome the fundamental performance limitations of current linear approaches.