Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Intelligent OPC Engineer Assistant for Semiconductor Manufacturing

Guojin Chen, Haoyu Yang, Bei Yu, Haoxing Ren

TL;DR

Experiments demonstrate that the Intelligent OPC Engineer Assistant methodology can efficiently build OPC recipes on various chip designs with specially handled design topologies, a task that typically requires the full-time effort of OPC engineers with years of experience.

Abstract

Advancements in chip design and manufacturing have enabled the processing of complex tasks such as deep learning and natural language processing, paving the way for the development of artificial general intelligence (AGI). AI, on the other hand, can be leveraged to innovate and streamline semiconductor technology from planning and implementation to manufacturing. In this paper, we present \textit{Intelligent OPC Engineer Assistant}, an AI/LLM-powered methodology designed to solve the core manufacturing-aware optimization problem known as optical proximity correction (OPC). The methodology involves a reinforcement learning-based OPC recipe search and a customized multi-modal agent system for recipe summarization. Experiments demonstrate that our methodology can efficiently build OPC recipes on various chip designs with specially handled design topologies, a task that typically requires the full-time effort of OPC engineers with years of experience.

Intelligent OPC Engineer Assistant for Semiconductor Manufacturing

TL;DR

Experiments demonstrate that the Intelligent OPC Engineer Assistant methodology can efficiently build OPC recipes on various chip designs with specially handled design topologies, a task that typically requires the full-time effort of OPC engineers with years of experience.

Abstract

Advancements in chip design and manufacturing have enabled the processing of complex tasks such as deep learning and natural language processing, paving the way for the development of artificial general intelligence (AGI). AI, on the other hand, can be leveraged to innovate and streamline semiconductor technology from planning and implementation to manufacturing. In this paper, we present \textit{Intelligent OPC Engineer Assistant}, an AI/LLM-powered methodology designed to solve the core manufacturing-aware optimization problem known as optical proximity correction (OPC). The methodology involves a reinforcement learning-based OPC recipe search and a customized multi-modal agent system for recipe summarization. Experiments demonstrate that our methodology can efficiently build OPC recipes on various chip designs with specially handled design topologies, a task that typically requires the full-time effort of OPC engineers with years of experience.
Paper Structure (28 sections, 8 equations, 11 figures, 1 table)

This paper contains 28 sections, 8 equations, 11 figures, 1 table.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Related Works
  4. OPC and Reinforcement Learning
  5. OPC Recipe Development and LLM Agents
  6. Evaluation Metrics for OPC
  7. Problem Formulation
  8. Methodology
  9. Overview: A Two-stage Approach
  10. Exploration with Reinforcement Learning
  11. Reinforcement Learning Framework
  12. Recipe Summarization with LLMs
  13. Data Representation Transformation
  14. Zero-shot Feature Pool Generation and Feature Labeling
  15. Feature Mining and Generation
  16. ...and 13 more sections

Figures (11)

  • Figure 1: Motivation for optical proximity correction (OPC) in semiconductor manufacturing process.
  • Figure 2: Left: The full-chip layout. Right: The relationship between the layout, OPC Recipe and OPC Engine.
  • Figure 3: OPC recipe development. The upper figure shows the recipe for optimizing EPE measurement points, while the lower figure shows the edge fragmentation recipe.
  • Figure 4: Evaluation metrics for OPC. (a) Definition of design target and the wafer pattern. (b) Illustration of EPE measurement points, including inner and outer EPE violations for calculating the total EPE count and EPE distance. (c) Process variation band (PVB) calculation.
  • Figure 5: The two-stage approach for OPC recipe generation. The first stage employs RL to optimize OPC recipes. The second stage utilizes multi-modal LLM agents to efficiently summarize the results generated by RL and generate the final OPC recipes.
  • ...and 6 more figures