AI-Based Testing Frameworks for Next-Generation Semiconductor Devices

Abstract

The semiconductor industry has been growing at a fast pace for the last decade. The ability to keep up with Moore's law and the ever-increasing chip complexity has become a tough challenge for manufacturers. The cost associated with testing these ICs has almost risen to that of the design. AI/ML is being actively explored as a solution to assist in almost every kind of issues arising in IC design and technology. IC testing plays a critical role in the production of reliable and functional chips. The increased complexity of the chips is leading to the introduction of new approaches and methods for improving existing and outdated methods. The test approach, automation, and tools used for 3D IC and FPGAs differ from the conventional ones used. The increase in chip complexity and the number of cores leads to a large power gouging. Test patterns for cores should be applied in a way that drastic changes to the current testing infrastructure and flows are avoided. Exploring opportunities for involving ML methods in different allied fields, such as screening equipment, failure cost reduction, optical DFT testing, etc. could also lead to improvements.