TY - JOUR
T1 - Investigations on Explainable Artificial Intelligence methods for the deep learning classification of fibre layup defect in the automated composite manufacturing
AU - Meister, Sebastian
AU - Wermes, Mahdieu
AU - Stüve, Jan
AU - Groves, Roger M.
PY - 2021
Y1 - 2021
N2 - Automated fibre layup techniques are widely used in the aviation sector for the efficient production of composite components. However, the required manual inspection can take up to 50 % of the manufacturing time. The automated classification of fibre layup defects with Neural Networks potentially increases the inspection efficiency. However, the machine decision-making processes of such classifiers are difficult to verify. Hence, we present an approach for analysing the classification procedure of fibre layup defects. Therefore, we comprehensively evaluate 20 Explainable Artificial Intelligence methods from the literature. Accordingly, the techniques Smoothed Integrated Gradients, Guided Gradient Class Activation Mapping and DeepSHAP are applied to a Convolutional Neural Network classifier. These methods analyse the neural activations and robustness of a classifier for an unknown and manipulated input data. Our investigations show that especially Smoothed Integrated Gradients and DeepSHAP are well suited for the visualisation of such classifications. Additionally, maximum-sensitivity and infidelity calculations confirm this behaviour. In future, customers and developers could apply the presented methods for the certification of their inspection systems.
AB - Automated fibre layup techniques are widely used in the aviation sector for the efficient production of composite components. However, the required manual inspection can take up to 50 % of the manufacturing time. The automated classification of fibre layup defects with Neural Networks potentially increases the inspection efficiency. However, the machine decision-making processes of such classifiers are difficult to verify. Hence, we present an approach for analysing the classification procedure of fibre layup defects. Therefore, we comprehensively evaluate 20 Explainable Artificial Intelligence methods from the literature. Accordingly, the techniques Smoothed Integrated Gradients, Guided Gradient Class Activation Mapping and DeepSHAP are applied to a Convolutional Neural Network classifier. These methods analyse the neural activations and robustness of a classifier for an unknown and manipulated input data. Our investigations show that especially Smoothed Integrated Gradients and DeepSHAP are well suited for the visualisation of such classifications. Additionally, maximum-sensitivity and infidelity calculations confirm this behaviour. In future, customers and developers could apply the presented methods for the certification of their inspection systems.
KW - Automation
KW - Defects
KW - Non-destructive testing
KW - Process monitoring
UR - http://www.scopus.com/inward/record.url?scp=85111322401&partnerID=8YFLogxK
U2 - 10.1016/j.compositesb.2021.109160
DO - 10.1016/j.compositesb.2021.109160
M3 - Article
AN - SCOPUS:85111322401
SN - 1359-8368
VL - 224
JO - Composites Part B: Engineering
JF - Composites Part B: Engineering
M1 - 109160
ER -