Apparatus Condition Search Using Automated Manufacturing Datasets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for optimizing manufacturing apparatus conditions are complex and require specialized knowledge, making it difficult for general engineers to determine optimal apparatus conditions due to the need for data scientists and programming skills, and there is a shortage of engineers with these qualifications.
Innovation Solution
A computer system that simplifies the process of searching for apparatus conditions by allowing general engineers to input data, construct learning datasets, and analyze manufacturing steps, using machine learning and deep learning to predict optimal conditions without requiring script creation or multiple software usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a processing result simulator or machine learning model is used to support apparatus condition search, then prediction accuracy and processing capability are improved, but the system complexity and requirement for specialized knowledge increase
Solution Approach 1:
The patent introduces an intermediary component that automatically constructs learning datasets from process data and manages the machine learning model. This intermediary layer shields users from the complexity of data preparation and model training, allowing them to benefit from accurate predictions without needing to understand the underlying complex processes.
Solution Approach 2:
The system performs self-service by automatically constructing learning datasets from available process data without requiring manual intervention or specialized knowledge. The automated dataset construction and model training processes enable the system to maintain high prediction accuracy while reducing the burden on users.
2Productivity
If scripts and commands are required to specify apparatus conditions and run the simulator, then processing capability is improved, but ease of operation deteriorates
Solution Approach 1:
The system enables self-service by automatically constructing learning datasets from process data and managing the entire machine learning workflow without requiring users to write scripts or commands. This automation maintains high processing capability while dramatically improving ease of operation for general engineers.
3Productivity
If multiple pieces of software and procedures are required from constructing dataset to running calculations, then processing capability is improved, but device complexity increases
Solution Approach 1:
The patent merges multiple separate software components and procedures into an integrated system. The dataset construction, model training, and prediction functions are combined into a unified platform that automatically manages the entire workflow, reducing the number of separate software components while maintaining processing capability.
Solution Approach 2:
The system achieves multi-functionality by incorporating dataset construction, model training, and prediction capabilities within a single universal platform. This universal system can handle various apparatus condition search tasks without requiring separate specialized software for each function.
4Ease of operation
If general engineers without specialized knowledge are to use the system, then user convenience is improved, but the ability to handle highly difficult processing deteriorates
Solution Approach 1:
The system enables general engineers to perform apparatus condition search without specialized knowledge by automatically constructing learning datasets and managing the machine learning process. This self-service capability allows users with basic skills to handle highly difficult processing tasks that previously required data scientists and programming expertise.
Data Source
AI summary
A computer system is configured to: receive input of a plurality of pieces of learning data, a piece of learning data includes a combination of a value of at least one input parameter which is the apparatus condition and a value of at least one output parameter which is a result of processing; analyze a structure of manufacturing steps of the processing in a piece of standard learning data and a structure of manufacturing steps of the processing in a piece of other learning data; execute, based on a result of the analysis, classification processing of identifying a piece of learning data which is aggregable with a same dataset as the piece of standard learning data; generate a dataset; execute processing of training a model; execute search processing of searching for the value of the at least one input parameter; and present a result of the search processing.