Semiconductor Smart Manufacturing Evaluation Using Depth Levels

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing semiconductor factories face challenges in evaluating the smart manufacturing implementation degree quantitatively and qualitatively, making it difficult to compare and monitor the implementation across multiple factories.

Innovation Solution

An autonomous evaluation method and device that analyzes task completion ratios of automation functions, maps them to an evaluation matrix to determine depth levels, and calculates comprehensive indicators using expansion ratios to quantify the smart manufacturing implement degree.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If smart manufacturing technology is introduced into semiconductor processes, then process stability is improved and human intervention errors are reduced, but the complexity of evaluating implementation degree across multiple factories increases

Engineering Contradiction:
Improveprocess stabilityVSAvoidevaluation complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The evaluation system segments the complex evaluation into multiple dimensions: automation depth levels (L0-L5) for individual apparatuses, factory-wide implementation degrees, and expansion ratios. This segmentation allows systematic assessment of smart manufacturing implementation across diverse apparatuses and factories, transforming the evaluation complexity into manageable components.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces quantitative parameters including task completion ratios, depth levels (L0-L5), expansion ratios, and comprehensive implementation degrees. These parameter changes enable objective comparison and evaluation of smart manufacturing implementation across multiple factories, converting qualitative assessment into measurable metrics.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If multiple types and models of manufacturing apparatuses are used, then manufacturing versatility is improved, but the difficulty of evaluating smart manufacturing implementation degree increases

Engineering Contradiction:
Improvemanufacturing versatilityVSAvoidevaluation difficulty
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The evaluation system applies universality by creating a standardized framework that can assess diverse apparatuses (lithography, etching, deposition, etc.) using common criteria: automation function completion ratios and depth levels. This universal evaluation matrix enables consistent measurement across different apparatus types and models, resolving the evaluation difficulty despite manufacturing versatility.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If quantitative evaluation metrics are implemented, then comparison accuracy between factories is improved, but the data collection and processing requirements increase

Engineering Contradiction:
Improvecomparison accuracyVSAvoiddata quantity
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The system extracts key evaluation indicators from complex manufacturing data: task completion ratios for automation functions, depth levels (L0-L5), and expansion ratios. By extracting these essential metrics, the system achieves accurate quantitative comparison between factories without requiring processing of all raw manufacturing data, thus balancing measurement precision with data quantity management.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS20250216833A1Autonomous evaluation method and autonomous evaluation device for semiconductor smart manufacturing
Publication Date: 2025.07.03 UNITED MICROELECTRONICS CORP
  • US20250216833A1 patent drawing
  • US20250216833A1 patent drawing
  • US20250216833A1 patent drawing

AI summary

An autonomous evaluation method and an autonomous evaluation device for semiconductor smart manufacturing are provided. The autonomous evaluation method for the semiconductor smart manufacturing includes the following steps. For each of a plurality of apparatuses, a task completion ratio of each of a plurality of automation functions implemented in a plurality process tasks is analyzed. The task completion ratios of the automation functions are mapped to an evaluation matrix to obtain a depth level corresponding to each of the apparatuses. A plurality of expansion ratios corresponding to the depth levels of the apparatuses are analyzed. A comprehensive indicator is analyzed according to the depth levels and the expansion ratios.