Presentation Information
[8p-S202-7]Accelerating Chemical Resistance Evaluation of Resins Using Machine Learning for Semiconductor Production Equipment
〇Takeru Nakamura1, Mitsuru Yambe1, Shogo Kunieda1, Yosuke Hanawa1, Hitoshi Kamijima2, Toshiaki Shintani2, Shunya Sugiyama2, Yoshihiro Hayashi3, Ryo Yoshida3 (1.SCREEN Holdings, 2.ISP, 3.ISM)
Keywords:
Semiconductor Production Equipment,Machine Learning
In semiconductor production equipment, chemical resistance testing of the resin used as equipment material is required when adopting a new process.
Such testing took a lot of time and cost, and was a challenge for equipment development.
To address this issue, we have been developing a machine learning model to predict the chemical resistance of resins in order to reduce testing costs.
As previously, a binary classification model has been developed to predict the presence or absence of chemical resistance, but since chemical resistance is judged based on multiple items, the basis for the judgment result is not clear with only a binary classification result.
In this presentation, we report on the development of a model to predict the temporal change in weight of resins in a chemical immersion test as one of our efforts to clarify the judgment results.
Such testing took a lot of time and cost, and was a challenge for equipment development.
To address this issue, we have been developing a machine learning model to predict the chemical resistance of resins in order to reduce testing costs.
As previously, a binary classification model has been developed to predict the presence or absence of chemical resistance, but since chemical resistance is judged based on multiple items, the basis for the judgment result is not clear with only a binary classification result.
In this presentation, we report on the development of a model to predict the temporal change in weight of resins in a chemical immersion test as one of our efforts to clarify the judgment results.