Automated Design Space Estimation for Complex Manufacturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack automation in determining acceptable regions of variability for input factors and predicting output response regions in complex manufacturing processes, such as semiconductor and pharmaceutical industries, making it difficult to meet multiple operational criteria simultaneously.
Innovation Solution
A computerized method for estimating a design space by creating a multidimensional grid, assessing risk of failure levels at each grid point, and successively eliminating points closest to the boundary to determine optimal input factor values, using techniques like Monte Carlo simulations to account for errors and disturbances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual evaluation methods are used to explore design space, then flexibility in analyzing design options is maintained, but the process becomes time-consuming and inefficient
Solution Approach 1:
The patent replaces manual mechanical evaluation processes with automated computer-based systems. The system automatically generates design options, evaluates them against constraints, and identifies optimal solutions without human intervention in the evaluation process, thereby dramatically improving efficiency while reducing time loss.
Solution Approach 2:
The system performs self-evaluation of design options by automatically comparing generated designs against specified constraints and criteria. The computer-based system serves itself by autonomously identifying feasible design spaces and optimal solutions without requiring continuous human assessment, thus accelerating the design exploration process.
2Reliability
If comprehensive constraints and multiple output responses are considered, then design quality and reliability improve, but the complexity of analysis increases
Solution Approach 1:
The patent segments the complex analysis process into distinct automated steps: generating design options, evaluating each option against multiple constraints, assessing output responses, and identifying optimal solutions. This segmentation allows the system to handle comprehensive constraints systematically without overwhelming complexity, as each segment is processed independently by the computer system.
Solution Approach 2:
The system automatically adjusts and evaluates multiple parameters simultaneously, changing input factors to explore their effects on output responses. By programmatically managing parameter variations and interactions, the system maintains reliability through comprehensive constraint checking while avoiding manual analysis complexity.
3Productivity
If automated methods are implemented for design space estimation, then productivity and efficiency improve, but the device and method complexity increases
Solution Approach 1:
The patent implements a universal computer-based system that performs multiple functions: generating design options, evaluating constraints, analyzing output responses, and identifying optimal designs. This multi-functional approach consolidates what would otherwise require multiple separate tools or manual processes into a single automated system, improving productivity while managing complexity through integration.
Solution Approach 2:
The computer-based system acts as an intermediary between design inputs and evaluation outputs. It mediates the complex analysis by automatically processing design options through multiple constraint checks and criteria assessments, then presenting results to users. This intermediary role handles the computational complexity internally while presenting simplified information externally.
Data Source
AI summary
Described are computer-implemented methods and apparatuses, including computer program products, for estimating an optimal value for each input factor of a design space. The design space is defined by the input factors and output responses for a physical process. The optimal values for the input factors represent a starting point for estimating the design space. Data is received for the input factors, the output responses and criteria. An initial design space is estimated based on the received data. The optimal values for the input factors are determined from the initial design space.