Correction Value Computation for Manufacturing Process Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional manufacturing processes face challenges in maintaining process repeatability and product quality due to fluctuations in manufacturing apparatus states, leading to unintended fluctuations in performance indicators and product quality, especially with the increasing complexity and miniaturization of semiconductor integrated circuits.
Innovation Solution
A correction value computation device and method that acquire index values representing fluctuation phenomena, device models, and process models to transform sensor output values into fluctuation amounts for process elements and performance indicators, computing correction amounts to stabilize control parameters and maintain process repeatability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional manufacturing processes are used without correction mechanisms, then the manufacturing process can be executed with simple control, but fluctuations in manufacturing apparatus states cause unintended fluctuations in performance indicators and product quality
Solution Approach 1:
The system performs preliminary actions by acquiring index values representing fluctuation phenomena and transforming them into fluctuation amounts for process elements before the manufacturing process executes. Correction values are computed in advance based on these fluctuation amounts and control ranges, allowing the system to proactively compensate for apparatus state changes rather than reacting to performance deviations after they occur.
Solution Approach 2:
The system implements feedback by acquiring index values from sensors that monitor manufacturing apparatus states, transforming these into fluctuation amounts, computing correction values based on control ranges, and applying these corrections to control parameters. This closed-loop feedback mechanism continuously adjusts control parameters to maintain performance indicators within target ranges despite apparatus fluctuations.
2Reliability
If correction values are computed and applied to control parameters, then performance indicators can be maintained within target ranges, but the computation and transformation processes increase system complexity
Solution Approach 1:
The system performs preliminary transformations by converting sensor index values into fluctuation amounts for process elements before computing correction values. This preliminary processing organizes the data in a structured manner that facilitates subsequent correction computations, making the overall system more manageable despite the multiple transformation steps involved.
Solution Approach 2:
The system introduces intermediary elements including index values that represent apparatus states, fluctuation amounts that quantify deviations, and correction values that bridge the gap between measured states and desired performance. These intermediaries facilitate the complex computation process by breaking it down into manageable transformation steps between different parameter representations.
3Measurement precision
If multiple transformations and computations are performed to compute correction values, then accurate control parameter adjustments can be achieved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary transformations by converting index values to fluctuation amounts before computing correction values. This preliminary organization of data into meaningful fluctuation measurements enables more efficient subsequent computations compared to processing raw sensor data directly, reducing the overall processing time while maintaining accuracy.
Solution Approach 2:
The computation process is segmented into distinct transformation steps: transforming sensor index values to fluctuation amounts, computing correction values from fluctuation amounts and control ranges, and applying corrections to control parameters. This segmentation allows each step to be optimized independently and enables parallel processing where applicable, reducing total computation time while preserving the accuracy benefits of multiple transformations.
Data Source
AI summary
A device for computing correction for control parameter in a manufacturing process executed on a manufacturing apparatus includes circuitry which acquires an index representing fluctuation in a manufacturing apparatus, acquires an apparatus model and a process model, acquires an output from a sensor in the manufacturing apparatus, transforms the output into first fluctuation for a process element, transforms the index into second fluctuation for the process element based on the apparatus model, computes fluctuation for performance indicator from the first and second fluctuation based on the process model, computes correction for the performance indicator from control range for the performance indicator and the fluctuation for the performance indicator, and converts the correction for the performance indicator into correction for each process element based on the process model such that correction for control parameter in process executed on the manufacturing apparatus is computed from the correction converted for each process element.