Local Linear Regression Outlier Correction for Substrate Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing substrate processing technologies face challenges in accurately determining and correcting outliers in measured values, leading to inefficiencies in recipe optimization and increased user workload due to the lack of objective criteria for outlier determination.
Innovation Solution
An information processing apparatus that utilizes a local linear regression method to calculate estimated values and automatically determines outliers based on predefined thresholds, correcting measured values at outlier points, thereby reducing user intervention and enhancing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual outlier determination methods are used, then user flexibility is maintained, but user workload increases and objectivity decreases
Solution Approach 1:
The system performs automatic outlier determination and correction without requiring manual user intervention. The calculation unit automatically calculates estimated values using local linear regression, the determination unit automatically identifies outliers based on threshold comparison, and the correction unit automatically corrects measured values, making the system self-sufficient and eliminating the need for manual operations.
Solution Approach 2:
The patent replaces manual mechanical outlier determination processes with an automated information processing system that uses mathematical calculations (local linear regression) and algorithmic decision-making. This substitution eliminates human labor while maintaining or improving determination objectivity through consistent, reproducible computational methods.
2Ease of operation
If automatic outlier correction is implemented, then user workload is reduced, but system complexity increases
Solution Approach 1:
The system is divided into distinct functional modules: input unit for data acquisition, calculation unit for estimating values using local linear regression, determination unit for identifying outliers based on threshold comparison, and correction unit for correcting measured values. This segmentation allows each component to perform a specific function independently, making the overall complex system manageable and easier to implement.
Solution Approach 2:
The system uses configurable parameters such as the threshold value for outlier determination and the local linear regression method parameters. By allowing parameter adjustment, the system becomes adaptable to different measurement scenarios without requiring fundamental redesign, thus managing complexity through parameterization rather than structural complexity.
3Measurement precision
If local linear regression method is used for estimation, then measurement precision is improved, but calculation time increases
Solution Approach 1:
The system applies local linear regression only to selected measurement points rather than all points, and uses a predetermined threshold for outlier determination rather than exhaustive analysis. This partial application of computational methods maintains sufficient measurement precision for correction purposes while significantly reducing the overall calculation time required.
Data Source
AI summary
An information processing apparatus includes an input unit that inputs positional information regarding a plurality of measurement points on a substrate and a plurality of measured values indicating a substrate processing result at each of the plurality of measurement points, a calculation unit that calculates a plurality of estimated values for the substrate processing result at each of the plurality of measurement points selected from the input positional information regarding the plurality of measurement points using a local linear regression method, a determination unit that determines, as an outlier, the measured value at the measurement point where an absolute value of a difference between the measured value and the estimated value at each of the plurality of selected measurement points deviates from a preset threshold, and a correction unit that corrects the measured value at the measurement point determined as the outlier.