Semiconductor Parameter Setting Using Correlation Axes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In semiconductor processes, incomplete data sets due to missing values and complex correlations among electrical measurement parameters hinder efficient parameter tuning and sensitivity analysis, leading to increased time and cost in developing semiconductor devices.

Innovation Solution

A method involving classification of electrical measurement parameters by correlation, extraction of correlation axes using artificial neural networks, and prediction of semiconductor device merit based on these axes, allowing for independent analysis and optimization without altering the process recipe.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If electrical measurement parameters are used for sensitivity analysis and parameter tuning, then device development information can be obtained, but the statistical correlation between parameters makes independent analysis impossible and requires complex computation

Engineering Contradiction:

Improveprocess informationVSAvoidcomputation complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts the correlation relationship between electrical measurement parameters by calculating variance inflation factors (VIF). Parameters with VIF values exceeding a threshold are identified as correlated and grouped together. This extraction allows the system to separate correlated parameters from independent ones, enabling independent sensitivity analysis for uncorrelated parameters while identifying multi-parameter interactions for correlated groups, thereby reducing computational complexity.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments electrical measurement parameters into multiple groups based on their correlation relationships. By calculating VIF values and comparing them against thresholds, the system divides parameters into independent groups and correlated groups. This segmentation allows sensitivity analysis to be performed independently on uncorrelated parameters while handling correlated parameters as integrated units, significantly reducing the overall computational burden.

Inventive Principle:

Principle #1Segmentation

2Manufacturing precision

If process recipe is changed to obtain device characteristic data, then parameter optimization can be performed, but a lot of time and money is consumed

Engineering Contradiction:

Improvedevice characteristicVSAvoiddevelopment time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary sensitivity analysis using electrical measurement parameters obtained from normal production processes without changing the process recipe. By analyzing the correlation between electrical measurement parameters and device characteristics in advance, the system identifies which parameters most influence device performance. This preliminary analysis guides subsequent process optimization, reducing the need for extensive trial-and-error process changes and accelerating development time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces electrical measurement parameters as intermediary indicators that reflect process quality without requiring actual process changes. These electrical parameters serve as proxies for device characteristics, allowing the system to predict device performance and guide optimization based on electrical measurements rather than requiring physical process modifications and device fabrication for each analysis iteration.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Ease of manufacture

If sampling is performed to reduce process cost, then measurement cost decreases, but missing values occur creating incomplete data sets

Engineering Contradiction:

Improveprocess costVSAvoiddata completeness

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent implements a feedback mechanism that detects missing values in electrical measurement parameter data and triggers imputation procedures. When sampling produces incomplete datasets, the system identifies missing entries and applies imputation algorithms that leverage correlations with other parameters to estimate and fill in the missing values. This feedback loop maintains data completeness and reliability while preserving the cost benefits of sampling-based measurement approaches.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20230274985A1Apparatus and method for setting semiconductor parameter

Publication Date: 2023.08.31 POSTECH ACADEMY INDUSTRY FOUNDATION

AI summary

Disclosed are a method and apparatus for setting a semiconductor parameter. The method for setting a semiconductor parameter according to an embodiment of the present disclosure is a method performed on a computing apparatus including one or more processors and a memory storing one or more programs executed by the one or more processors, the method including acquiring electrical measurement parameters corresponding to preset semiconductor manufacturing parameters, classifying the electrical measurement parameters into a plurality of groups according to a degree of correlation, extracting a correlation axis reflecting a correlation between electrical measurement parameters belonging to a corresponding group for each classified group, and predicting a figure of merit of a semiconductor device by using data values of electrical measurement parameters belonging to the corresponding group as input based on the correlation axis of each group.