Electronic Packaging Science and Technology

14Artificial Intelligence in Electronic Packaging Reliability

14.1 Introduction

Artificial intelligence (AI) is to mimic human intelligence. In the big data era, many new AI applications have been invented, for example in distance teaching and home office using advanced 5G communication technology and 3D IC devices. Here, we ask what will be the technical innovation in AI apps, which can enhance the study of reliability of electronic packaging technology? After all the analyses presented in the early chapters in this book, we should know the answers on what are the most important issues in reliability and how we can get help from AI?

The most important issue in reliability is “experience.” In any company, an employee who has had experience on reliability is valuable. He/she may know roughly the best design, proper integration of different materials, and critical processing steps for a new device, so that he/she can have design‐for‐reliability (DfR) in packaging technology. The new product can achieve the best compromise among the requirements of reliability, cost‐effectiveness, and short time to market. No doubt, innovation is always important.

Fundamentally, reliability failure in electronic devices is due to microstructure change, which leads to void or hillock formation. Our goal of AI is to build a machine to have the experience of the expert on reliability based on having big data, in which every event of reliability failure due to microstructure change is included.