Substrate Processing Data Management by Recipe Step Range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor manufacturing, substrate processing apparatuses generate large volumes of data that often exceed the memory capacity of data managers, leading to unstable data acquisition and processing due to the high number of sensors and complex processes.
Innovation Solution
A substrate processing system with a controller and memory that allows for data acquisition and storage of processed data, along with a data manager that acquires data on a step-by-step basis, specifying the number of steps to ensure the data falls within a predefined acquirable range, thereby stabilizing data acquisition and preventing memory overflow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data acquisition is performed from all sensors in substrate processing apparatus, then complete process monitoring is achieved, but data size exceeds memory capacity of data manager
Solution Approach 1:
The patent segments the acquisition process by dividing process data into individual step units. The data manager specifies a number of steps to acquire, obtaining data for only those specific steps rather than all steps simultaneously. This segmentation allows complete monitoring capability while controlling data volume by acquiring data in manageable portions.
Solution Approach 2:
The patent implements partial action by acquiring process data for a selected subset of steps rather than all steps. The data manager specifies a number of steps (e.g., 1-10 steps) to acquire out of the total process steps, obtaining sufficient process monitoring data while keeping the data size within memory capacity limits.
2Reliability
If all process data is acquired and stored, then comprehensive analysis is enabled, but processing stability deteriorates due to memory overload
Solution Approach 1:
The patent segments data acquisition into discrete step-based units with controlled quantities. By specifying a number of steps to acquire and converting this to a data size threshold, the system processes data in manageable segments rather than handling all data at once, improving processing stability and reducing memory management complexity.
Solution Approach 2:
The patent dynamically adjusts the acquisition parameter (number of steps) based on memory capacity constraints. The data manager calculates an appropriate number of steps to acquire by considering the data size threshold derived from available memory capacity, thereby adapting the data acquisition volume to maintain processing stability.
3Loss of information
If data acquisition range is increased to capture more process steps, then better process coverage is achieved, but data acquisition time increases
Solution Approach 1:
The patent implements partial action by acquiring data for a specific number of steps that provides adequate process coverage without excessive acquisition. The data manager determines an optimal number of steps based on memory capacity, achieving sufficient process monitoring coverage while limiting acquisition time by not acquiring all possible step data.
Solution Approach 2:
The patent adjusts the acquisition parameter (number of steps) dynamically based on system constraints including time and memory capacity. By converting memory capacity into a data size threshold and determining the corresponding number of steps, the system optimizes the balance between process coverage and acquisition time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a technique, which includes: at least one substrate processing apparatus including a controller configured to be capable of controlling processing of a substrate according to a recipe including at least one step and a first memory configured to be capable of storing data of the at least one substrate processing apparatus that is reported during the processing of the substrate; and a data manager connected to the at least one substrate processing apparatus, configured to acquire the data, and configured to be capable of specifying the number of the at least one step that is included in the data to be acquired such that the data falls within a predefined data acquirable range.