To read the full version of this content please select one of the options below:

Machine learning framework for predicting reliability of solder joints

Sung Yi (Mechanical and Materials Engineering Department, Portland State University, Portland, Oregon, USA)
Robert Jones (Portland State University, Portland, Oregon, USA)

Soldering & Surface Mount Technology

ISSN: 0954-0911

Article publication date: 13 August 2019

Issue publication date: 18 March 2020

Abstract

Purpose

This paper aims to present a machine learning framework for using big data analytics to predict the reliability of solder joints. The purpose of this study is to accurately predict the reliability of solder joints by using big data analytics.

Design/methodology/approach

A machine learning framework for using big data analytics is proposed to predict the reliability of solder joints accurately.

Findings

A machine learning framework for predicting the life of solder joints accurately has been developed in this study. To validate its accuracy and efficiency, it is applied to predict the long-term reliability of lead-free Sn96.5Ag3.0Cu0.5 (SAC305) for three commonly used surface finishes such OSP, ENIG and IAg. The obtained results show that the predicted failure based on the machine learning method is much more accurate than the Weibull method. In addition, solder ball/bump joint failure modes are identified based on various solder joint failures reported in the literature.

Originality/value

The ability to predict thermal fatigue life accurately is extremely valuable to the industry because it saves time and cost for product development and optimization.

Keywords

Citation

Yi, S. and Jones, R. (2020), "Machine learning framework for predicting reliability of solder joints", Soldering & Surface Mount Technology, Vol. 32 No. 2, pp. 82-92. https://doi.org/10.1108/SSMT-04-2019-0013

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited