Machine Learning Methods for Quality Prediction in Production

Abstract

The rising popularity of smart factories and Industry 4.0 has made it possible to collect large amounts of data from production stages. Thus, supervised machine learning methods such as classification can viably predict product compliance quality using manufacturing data collected during production. Elimination of uncertainty via accurate prediction provides significant benefits at any stage in a supply chain. Thus, early knowledge of product batch quality can save costs associated with recalls, packaging, and transportation. While there has been thorough research on predicting the quality of specific manufacturing processes, the adoption of classification methods to predict the overall compliance of production batches has not been extensively investigated. This paper aims to design machine learning based classification methods for quality compliance and validate the models via case study of a multi-model appliance production line. The proposed classification model could achieve an accuracy of 0.99 and Cohen’s Kappa of 0.91 for the compliance quality of unit batches. Thus, the proposed method would enable implementation of a predictive model for compliance quality. The case study also highlights the importance of feature construction and dataset knowledge in training classification models.

On the hybridization of pre-trained deep learning and differential evolution algorithms for semantic crack detection and recognition in ensemble of infrastructures

“…In order to conduct a comprehensive quantitative comparison between the prediction models, OV_ACC and MCC are selected because they are well-known performance measures that succeeded in evaluating the prediction accuracies of classifiers in different engineering disciplines (Sugimori, 2018; Rahim et al , 2018; Trajanov et al , 2018). F−measure, Kappa coefficient, area under the curve, balanced accuracy and Youden's index are chosen to address the imbalanced dataset since OV_ACC and MCC can be misleading due to the over optimistic and highly inflated classification scores (Sankhye and Hu, 2020; Tharwat, 2020; Patel and Thakur, 2017).…”

A method for yarn quality fluctuation prediction based on multi-correlation parameter feature subspace mechanism in spinning process

“…9,10 In order to identify the abnormal fluctuation of quality in the process and give early warning, the change point identification model is used to determine the quality fluctuation stability. 11,12 With the development of production process complexity and the advancement of machine learning techniques, 13 many researches have focused on intelligent quality prediction, including neural networks to build nonlinear relationships for prediction, 14,15 production process quality data mining analysis . 16,17 In order to improve the quality prediction accuracy, Bai et al 18 proposed to combine data dimensionality reduction and support vector machine to build a quality prediction model.…”

“…When the state of the neurons in the visible layer is given, the probability that the j-th neuron in the hidden layer is activated can be calculated. Similarly, when the state of neurons in the hidden layer is given, the probability of the activation of the i-th neuron can be calculated by equation (13).…”

Product Processing Quality Classification Model for Small-Sample and Imbalanced Data Environment

“…Manimala [ 39 ] proposed a data selection method based on fuzzy c-means clustering for power quality event classification. Sankhye et al [ 40 ] used supervised machine learning methods such as classification to predict product compliance quality using manufacturing data collected during production.…”