Jul. 14, 2021

Introducing a High-performance Model that Combines AI with an Inspection Algorithm in the INSPECTRA® Series of Semiconductor Wafer Inspection System

Toray Engineering Co., Ltd. (Head office: Chuo-ku, Tokyo, CEO & COO: Takashi Iwade, hereinafter "Toray Engineering") has developed AI-based image classification technology that has been used in the life science field, and successfully applied in the industrial field for inspecting products. TASMIT Inc. (TASMIT, Head office: Kohoku-ku, Yokohama, CEO: Kenji Sato, hereinafter "TASMIT"), a subsidiary of Toray Engineering, has optimized this technology for the optical semiconductor wafer inspection process, and the semiconductor wafer inspection system INSPECTRA®. It will be equipped on INSPECTRA as "AI-ADC (Automatic Defect Classification)" and will start full-scale sales from August.
The sales target is 5 Billion JPY in FY 2021 and 9 Billion JPY in FY 2025.

In product inspection using AI, it is necessary to link a large number of example images to classification categories and create inspection conditions before operation in order to improve accuracy. In the past, these conditions had to be created manually, and there was the problem that it took time and effort to improve accuracy. The newly equipped "AI-ADC" has a pre-classification AI-based function for condition setting and a "Self-Organizing Map function" that enables visualization of the inspection condition creation flow (hereinafter "SOM" function). For this reason, it is possible to greatly improve efficiency in the preparation stage by reducing the manual condition setting work. As a result, we succeeded in reducing the time required from classification to condition creation under the same conditions to about one-sixth of the conventional time (according to Toray Engineering’s research).
This technology was developed for the life science field by LPIXEL Inc. (Headquarters: Chiyoda-ku, Tokyo, Representative Directors: Yuki Shimahara and Tomihisa Kamada), which performs image analysis. This is the first time in the industry that Toray Engineering has diverted it to industrial use through open innovation with our company.
In addition, the "random forest" method is used for defect inspection using AI-equipped INSPECTRA®, and highly sensitive and highly accurate automatic judgment is realized by using the judgment criteria subdivided into about 300 items. Compared to the "deep learning" method, the basis for judgment is clearer, so by combining it with the SOM function, it is possible to make adjustments to improve accuracy when overkill or incorrect judgment occurs.
We have been proposing this product to some customers since 2020, but now we have decided to proceed with full-scale sales.
The Toray Engineering Group will continue to contribute to society by pursuing cutting-edge AI technology and providing solutions that make full use of our technology, engineering technology, and know-how.

The details of the newly developed "AI-ADC" are as follows.

Product name:
INSPECTRA® AI-ADC equipped type
Product features:
  • When creating defect classification conditions, it is possible to efficiently create inspection conditions by using the automatic classification and the SOM function which visualizes the classification results.
  • Highly sensitive and highly accurate automatic judgment is realized by random forest AI inspection using subdivided judgment criteria.
  • Killer defect detection for process improvement
  • Overkill and oversight suppression
Sales goals:
  • 5 Billion JPY in FY 2021
  • 9 Billion JPY in FY 2025