Supervised Learning Approach for Surface-Mount Device Production

Jabbar, Eva; Besse, Philippe; Loubes, Jean–Michel; Roa, Nathalie Barbosa; Merle, Christophe; Dettai, Rémi

doi:10.1007/978-3-030-13709-0_21

Cited by 11 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work is based on a dual-level defect detection method. In (Jabbar et al, 2018) some treebased machine learning methods are used to predict the defects found in AOI using SPI data. (Gaffet, Ribot, Chanthery, Roa, & Merle, 2021) proposes an unsupervised univariate method for monitoring the In-Circuit Testing machine (located at the end of the Surface Mount Technology lines) and components.…”

Section: Related Workmentioning

confidence: 99%

A Hierarchical XGBoost Early Detection Method for Quality and Productivity Improvement of Electronics Manufacturing Systems

Gaffet¹,

Roa²,

Ribot

et al. 2022

PHME_CONF

Self Cite

View full text Add to dashboard Cite

This paper presents XGBoost classifier-based methods to solve three tasks proposed by the European Prognostics and Health Management Society (PHME) 2022 conference. These tasks are based on real data from a Surface Mount Technologies line. Each of these tasks aims to improve the efficiency of the Printed Circuit Board (PCB) manufacturing process, facilitate the operator’s work and minimize the cases of manual intervention. Due to the structured nature of the problems proposed for each task, an XGBoost method based on encoding and feature engineering is proposed. The proposed methods utilise the fusion of test values and system characteristics extracted from two different testing equipment of the Surface Mount Technologies lines. This work also explores the problems of generalising prediction at the system level using information from the subsystem data. For this particular industrial case: the challenges with the changes in the number of subsystems. For Industry 4.0, the need for interpretability is very important. This is why the results of the models are analysed using Shapley values. With the proposed method, our team took the first place, capable of successfully detecting at an early stage the defective components for tasks 2 and 3.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hierarchical XGBoost Early Detection Method for Quality and Productivity Improvement of Electronics Manufacturing Systems

Gaffet¹,

Roa²,

Ribot

et al. 2022

PHME_CONF

Self Cite

View full text Add to dashboard Cite

show abstract

“…According to the authors, XGBoost models were shown to be capable of capturing complex nonlinear characteristics of the sensor data and dealing with anomalies in the data set. Jabbar et al [18] proposed a decision-making tool based on XGBoost to support operators in the manufacturing of printed circuit boards. XGBoost was shown to yield both high accuracy and high recall in the classification of defects when trained on real-world data.…”

Section: Related Workmentioning

confidence: 99%

Multistage Quality Control Using Machine Learning in the Automotive Industry

et al. 2019

View full text Add to dashboard Cite

Product dimensional variability is a crucial factor in the quality control of complex multistage manufacturing processes, where undetected defects can easily be propagated downstream. The recent advances in information technologies and consequently the increased volume of data that has become readily available provide an excellent opportunity for the development of automated defect detection approaches that are capable of extracting the implicit complex relationships in these multivariate data-rich environments. In this paper, several machine learning classifiers were trained and evaluated on varied metrics to predict dimensional defects in a real automotive multistage assembly line. The line encompasses two automated inspection stages with several human-operated assembly and pre-alignment stages in between. The results show that non-linear models like XGBoost and Random Forests are capable of modelling the complexity of such an environment, achieving a high true positive rate and showing promise for the improvement of existing quality control approaches, enabling defects and deviations to be addressed earlier and thus assist in reducing scrap and repair costs.

show abstract

“…Indirectly supervised learning approaches enhance the outcome of current automatic optical inspections and measured solder joint dimensions. This can be used in the production line to support the operator in assessing defect calls from the automatic optical inspection and identify false positive classifications (Jabbar et al, 2019). Specifically, Jabbar et al compared tree-based machine learning algorithms for this use case and achieved a good performance (Jabbar et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

“…This can be used in the production line to support the operator in assessing defect calls from the automatic optical inspection and identify false positive classifications (Jabbar et al, 2019). Specifically, Jabbar et al compared tree-based machine learning algorithms for this use case and achieved a good performance (Jabbar et al, 2019). Supervised deep learning has also been applied to real production environment data to enhance the outcome of the optical inspections and reduce labor cost (Chang, Wei, Chen, & Hsieh, 2018).…”

Section: Introductionmentioning

confidence: 99%

Application of Machine Learning Methods to Predict the Quality of Electric Circuit Boards of a Production Line

et al. 2022

View full text Add to dashboard Cite

For the data challenge of the 2022 European PHM conference, data from a production line of electric circuit boards is provided to assess the quality of the produced components. The solution presented in this paper was elaborated to fulfill the data challenge objectives of predicting defects found in an automatic inspection at the end of the production line, predicting the result of a following human inspection and predicting the result of the repair of the defect components. Machine learning methods are used to accomplish the different prediction tasks. In order to build a reliable machine learning model, the steps of data preparation, feature engineering and model selection are performed. Finally, different models are chosen and implemented for the different sub-tasks. The prediction of defects in the automatic inspection is modeled with a multi-layer perceptron neural network, the prediction of human inspection is modeled using a random forest algorithm. For the prediction of human repair, a decision tree is implemented.

show abstract

Supervised Learning Approach for Surface-Mount Device Production

Cited by 11 publications

References 10 publications

A Hierarchical XGBoost Early Detection Method for Quality and Productivity Improvement of Electronics Manufacturing Systems

A Hierarchical XGBoost Early Detection Method for Quality and Productivity Improvement of Electronics Manufacturing Systems

Multistage Quality Control Using Machine Learning in the Automotive Industry

Application of Machine Learning Methods to Predict the Quality of Electric Circuit Boards of a Production Line

Contact Info

Product

Resources

About