TCAD-enabled Machine Learning -  An Efficient Framework to Build Highly Accurate and Reliable Models for Semiconductor Technology Development and Fabrication

Publikation: Beitrag in FachzeitschriftForschungsartikelBeigetragenBegutachtung

Beitragende

  • Paul Jungmann - , Global Foundries, Inc. (Autor:in)
  • Jeffrey B. Johnson - , Global Foundries Dresden (Autor:in)
  • Eduardo C. Silva - , Global Foundries Dresden (Autor:in)
  • William Taylor - , Global Foundries Dresden (Autor:in)
  • Abdul Hanan Khan - , Global Foundries Dresden (Autor:in)
  • Akash Kumar - , Professur für Prozessorentwurf (Prozessor Design) (cfaed) (Autor:in)

Abstract

The requirements on data-driven Machine Learning models for industrial applications are often stricter, compared to those used for academic purposes, as model reliability is critical in industrial environments. Herein is introduced a framework which enables automated data generation with the goal of efficiently providing a data set sufficient to build a reliable and actionable model. Essential to this framework is the placement of the model training/testing data points, which need to be well distributed across the defined input parameter space. The framework is applied to semiconductor fabrication, wherein TCAD, a set of simulation tools that reproduce the physical processing and the final electrical performance of semiconductor devices, is a well-established capability. Transistor-level processing data is reproduced with TCAD simulations, from which the Machine Learning model is built. The framework described here assures that the resulting Machine Learning model fulfills the accuracy requirements across the parameter space. As an example application, the final Machine Learning model is then used to modify the process for a transistor, to obtain both better electrical performance and reduced variability.

Details

OriginalspracheEnglisch
Seiten (von - bis)268-278
Seitenumfang11
FachzeitschriftIEEE transactions on semiconductor manufacturing
Jahrgang36
Ausgabenummer2
PublikationsstatusVeröffentlicht - Mai 2023
Peer-Review-StatusJa

Externe IDs

WOS 000982419000015
Mendeley ae5137b9-3d72-3ae7-ad02-e0e9be041873

Schlagworte

Forschungsprofillinien der TU Dresden

Schlagwörter

  • Data models, Fabrication, Fabrication Capabilities, Integrated circuit modeling, Machine learning, Reliability, Semiconductor process modeling, TCAD, TCAD-eML, TCAD-enabled Machine Learning, Technology Computer-Aided Design, Training data, TCAD-enabled machine learning, reliability, technology computer-aided design, fabrication capabilities, Fabrication capabilities, Technology computer-aided design, Tcad

Bibliotheksschlagworte