Substrate Thickness Pre-Check for Reliable ML Processing Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The reliability of output data from trained models in substrate processing apparatuses is not guaranteed, leading to potential non-optimal processing conditions that may result in deviations from the target state of the substrate.
Innovation Solution
A substrate processing apparatus that includes a thickness gauge, a controller, and storage, where the controller uses input data and reference data to determine whether to perform substrate processing based on comparison results and thresholds, ensuring reliable processing conditions by measuring the substrate's thickness before processing and comparing it with target values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a trained model is used to output processing conditions, then automation and productivity are improved, but the reliability of the output data cannot be guaranteed
Solution Approach 1:
The patent applies preliminary action by measuring the substrate thickness before processing and comparing it with the target thickness in advance. This pre-check ensures that only substrates within the acceptable thickness range proceed to processing, preventing unreliable or non-optimal processing conditions from being applied. The thickness measurement and comparison are performed prior to the actual processing step, allowing the system to proactively identify and exclude substrates that would result in poor processing outcomes.
2Productivity
If processing conditions from a trained model are applied without verification, then productivity is improved, but manufacturing precision deteriorates
Solution Approach 1:
The patent implements feedback by measuring the actual substrate thickness and comparing it with the target thickness to determine whether to proceed with processing. This feedback mechanism ensures that only substrates meeting the thickness specifications are processed, maintaining manufacturing precision. The comparison result directly influences the decision to apply processing conditions, creating a closed-loop control system that prevents defective substrates from undergoing processing.
3Reliability
If a thickness check and comparison system is added, then reliability of processing conditions is improved, but device complexity increases
Solution Approach 1:
The patent applies universality by integrating the thickness measurement and comparison functionality into the existing substrate processing system. The control unit that already manages processing conditions is extended to also perform thickness verification, allowing a single component to serve multiple functions. This multi-functionality approach avoids adding separate dedicated verification equipment, thereby limiting the increase in device complexity while still improving reliability.
Data Source
AI summary
A substrate processing apparatus, a substrate processing system, and a data processing method are provided. A substrate processing apparatus (100) includes a thickness gauge (8), a controller (102), and storage (103). The thickness gauge (8) measures the thickness of an object (TG) included in a substrate (W). The controller (102) enters input data in a trained model (LM), thereby causing the trained model (LM) to output processing conditions. The storage (103) stores reference data (RE) acquired based on pieces of training data (LD) used to construct the trained model (LM). The controller (102) acquires pre-processing measurement data indicating the thickness of the object (TG) before the substrate processing is performed, and creates input data based on target data indicating a target value for the thickness of the object (TG) and the pre-processing measurement data. The controller (102) compares the input data and the reference data (RE) to determine whether or not to perform the substrate processing.