PDF Solutions: Data Analytics for Semiconductors

The human brain’s efficiency is massively more than that of today’s computers. The semiconductor ecosystem is forecasted to reach $1 trillion in revenue by 2030, which means more software, more analytics and new products that need to be tested and analyzed. PDF Solutions, a Lanza techVentures Portfolio Company, is setting a vision for the future of the semiconductor ecosystem to make computing and information processing as efficient and as productive as the human brain at scale.

PDF Solutions is the backbone of data analytics, providing advanced analytics and unique data to the semiconductor ecosystem. Offerings include tools and the insight to create and collect data for AI-driven analysis that allows users to bring new products and process to market at a more rapid clip.

PDF Solutions started in 1990s when the U.S. was concerned about manufacturing and funded university research. Today, it has close to 350 customers representing every segment of the chip industry. Those companies include equipment manufacturers that make equipment for the factories, the facilities producing, testing and packaging wafers and fabless chip companies or system companies designing the chips that go into products.

From manufacturing analytics, PDF Solutions expanded into test operations and assembly operations. Its Exensio platform combines data from across the semiconductor ecosystem into a single scalable database. PDF anticipated the transition to contract manufacturing and how that affected software used to bring design to manufacturing. Since then, it anticipated 3D processing in the wafer fab and assembly facilities, and what that means for analytics and data collection.

John Kibarian, CEO, PDF Solutions

The semiconductor ecosystem is dealing with new architectures and integration schemes and is challenged in needing “to see the invisible.” PDF Solutions developed a design for inspection system using a unique eBeam machine called eProbe to understand the interaction between advanced processes and advanced products to accelerate learning cycles leading to faster time to market.

Kimon Michaels, Co-founder and SVP, PDF Solutions

Designs are more susceptible to systematic defects. Defects, instead of appearing randomly over a chip, appear at specific locations or hot spots. PDF Solutions design analysis software works with AI and pattern recognition to find precise locations for failure.

Companies specializing in software, hardware and analytics tailored for the semiconductor ecosystem like PDF Solutions are critical in providing capabilities for technological advancement and high-volume manufacturing.

It has cycles of learning that lead to an understanding of the technology, an ability to integrate all available sources of data and system-level thinking. That’s why they are perfectly poised to help make computing and information processing as efficient and as productive as the human brain at scale.