The Ability of AI to Drive Advances in Semiconductor Design and Manufacturing

AI holds great promise for our industry. This will help bridge the gap between design, manufacturing and testing needed to efficiently produce today’s most advanced hybrid devices.

One way to think about the potential impact of AI in the semiconductor industry is to realize that it is increasingly driven by software engineering.

Designing a chip is essentially like writing code. AI, with tools like Microsoft’s Copilot, is having a huge impact on the way code is now produced. An engineer can now synthesize code, allowing the computer to generate routine, mundane parts. Semiconductor testing is also a form of programming. On the manufacturing side, factories generate gigabytes or even terabytes of data daily with valuable information hidden inside. Humans can’t sift through all this data to find anomalies, but AI can. Using AI to identify these “needles in a haystack” will revolutionize every aspect of the chip industry over the next five years.

AI and ML have already been deployed in advanced testing for complex systems and packages to track the number of test insertion points. Engineers are then able to decide which chipsets belong to which other chiplets in a specific package to optimize testing, and add and subtract tests to adjust production flow.

Today, these techniques are only deployed by a small number of experts within large companies. Anecdotally, a large Japanese semiconductor company with tens of thousands of employees has 30 employees who know how to use AI/ML. For AI to have an impact, it must be ubiquitous.

Not all use cases will benefit equally from the application of AI. When the semiconductor industry applies AI to a specific problem where the width and scope of information is narrow enough for a human to follow, it is difficult for algorithms to do better than a human . An example is where an engineer wants to predict when equipment will require preventive maintenance. It’s hard to beat human understanding of equipment.

A better place to apply AI is with a broad scope of information, such as consuming all product lifecycle management (PLM) data entered into a design, all generated vectors and those that do not. were not, as well as all the equipment. monitor variability data and online inspection data. The space of this data is vast and encompasses several different specialist areas. While different types of engineers focus on each part, integration across these dimensions can be accomplished much more effectively in software and algorithms, and much better than individuals could. I call this human-limited yield improvement because humans cannot understand and discover all the underlying relationships in this huge data space.

The power of AI to transform semiconductor design and manufacturing will be showcased at the PDF Solutions AI Executive Conference on Thursday, December 12 in San Francisco. PDF Solutions and other industry experts will be on hand to lead the way to the industry of the future and tout the democratization of AI for everyone in the chip industry, not just the few specialists in Today. Speakers will explore how AI is revolutionizing semiconductor design and manufacturing and discuss real-world applications of AI in the semiconductor industry.

The conference will provide insight into the power of AI to transform semiconductor design and manufacturing with examples of successful AI applications. Presentations will highlight how industry experts are deploying AI and ML to make a difference in their business. Others will detail how they enable infrastructure to lower the effort barrier, eliminate the expertise barrier required to apply these techniques, and democratize AI.

PDF Solutions AI Executive Conference
Date: December 1‌2, 2‌0‌2‌4
Location: St. Regis Hotel, San Francisco, California.
Agenda and registration