HCSL Publications

Neural Networks based Quality Control

1. Chatzimichail RA, Hatjimihail AT. Quality Control Using Convolutional Neural Networks Applied to Samples of Very Small Size. Technical Report XXI. Drama: Hellenic Complex Systems Laboratory, 2022.

Abstract

Although there is extensive literature on the application of artificial neural networks in quality control, to monitor the conformity of a process to quality specifications, at least five quality control measurements are required, increasing the related cost. To explore the application of neural networks to samples of quality control measurements of very small size, four one-dimensional convolutional neural networks were designed, trained, and tested with datasets of n-tuples of simulated standardized normally distributed quality control measurements, for 1 ≤ n ≤ 4. The designed neural networks were compared to statistical quality control functions with equal probabilities for false rejection, applied to samples of the same size. When the n-tuples included at least two quality control measurements distributed as Ν(μ,σ2 ), where 0.2 < |μ| ≤ 6.0, and 1.0 < σ ≤ 7.0, the designed neural networks outperformed the respective statistical quality control functions. Therefore, one-dimensional convolutional neural networks applied to samples of 2-4 quality control measurements can be used, to increase the probability of detection of the nonconformity of a process to the quality specifications, with lower cost.

Technical Report

Supplement 1

Supplement 2