Response Surface Modeling for Efficient Experimental Point Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting experimental design parameters in complex processes are inefficient, especially when dealing with a large number of control parameters, leading to increased experimental points and difficulty in obtaining optimal conditions within the experimental range.
Innovation Solution
An information processing method using an experimental design approach that creates multiple tables of experimental conditions, adjusts level values, and calculates response surfaces to efficiently set additional experimental points, ensuring the inclusion of target values and reducing the number of redesigns required.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of control parameters increases, then the comprehensiveness of the experimental design improves, but the number of experimental candidate points increases rapidly
Solution Approach 1:
The patent segments the experimental design process into multiple stages: first creating a base experimental design without the first control factor, then selectively adding experimental points that incorporate the first control factor at different level values. This segmentation allows the system to handle multiple control parameters without generating all possible combinations, thus reducing the total number of experimental points while maintaining comprehensiveness.
Solution Approach 2:
The patent introduces a new dimension by adding the first control factor with multiple level values (one level value and second level value) to the existing experimental design space. Instead of expanding uniformly in all dimensions, the method strategically adds points along the new dimension only where necessary, reducing the combinatorial explosion that would occur with full factorial design.
2Measurement precision
If additional experimental points are selected to compensate for missing points, then the accuracy of characteristic models improves, but the complexity of selecting optimal combinations increases
Solution Approach 1:
The patent performs preliminary action by first creating a base experimental design table without the first control factor, identifying missing points, and then systematically adding experimental points that include the first control factor. This preliminary structuring reduces the complexity of selecting optimal combinations by providing a framework that guides the addition of necessary points rather than searching through all possible combinations.
Solution Approach 2:
Instead of selecting the exact number of additional points needed, the patent adds experimental points with the first control factor at different level values beyond the minimum required. This partial excessive action ensures that accuracy requirements are met while providing flexibility in managing the complexity of combination selection.
3Reliability
If the experimental range is expanded to include target values, then the reliability of obtaining optimal conditions improves, but the number of redesigns required increases
Solution Approach 1:
The patent incorporates feedback by calculating response surfaces based on the experimental data and using this information to determine whether the experimental range includes target values. If the response surface analysis indicates that target values are not included, the system identifies which additional experimental points with the first control factor should be added to expand the range appropriately, rather than blindly expanding or redesigning the entire experiment.
Solution Approach 2:
The patent changes parameters by introducing the first control factor with specific level values (one level value and second level value) to expand the experimental range. This parameter change is done strategically based on response surface analysis, allowing the system to include target values while minimizing the number of additional points needed and avoiding unnecessary redesigns.
Data Source
AI summary
An information processing method includes: creating a first table; calculating a first response surface using the first table in which an object variable is recorded, the object variable acquired when experimental conditions of a first control factor is fixed to a first level value; adding the experimental conditions of the first control factor in which the first level value is set to the first table when the first response surface does not include a target value related to the object variable; creating a second table by adding a plurality of combinations of the experimental conditions for each of the plurality of control factors in which a first plurality of level values are set and the first control factor in which a second plurality of level values are set to the first table; calculating a second response surface including the target value using the second table; and outputting the second response surface.