Machine learning device, machine learning method, and program
The machine learning device and method address the challenge of predicting plant-scale manufacturing outputs by acquiring and selecting data sets to create predictive models, enhancing efficiency and accuracy in manufacturing predictions.
WO2025069921A1 Publication Date: 2025-04-03NEC CORP +1
4 Cites 0 Cited by
Patent Information
- Application Number
- PCT/JP2024/031147
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-09-29
- Filing Date
- 2024-08-30
- Publication Date
- 2025-04-03
AI Technical Summary
Technical Problem
Predicting the output of manufacturing equipment in a plant-scale environment is challenging due to the difficulty in collecting sufficient samples and the high cost of performing numerous experiments, making it difficult to apply lab-scale machine learning effectively.
Method used
A machine learning device and method that acquires and selects data sets under varying manufacturing conditions, using transfer learning to create a predictive model for manufacturing outputs, enabling efficient prediction of plant-scale manufacturing outcomes.
Benefits of technology
Enables accurate and efficient prediction of manufacturing outputs in a plant-scale environment by leveraging data from similar experimental methods, reducing the need for costly trial and error.
✦ Generated by Eureka AI based on patent content.
Abstract
The purpose of the present disclosure is to provide a machine learning device, a machine learning method, and a program capable of predicting the output of a high-performance manufacturing device. A machine learning device according to the present disclosure, which predicts the output of a manufacturing device, is provided with: an acquisition unit that acquires a target data set including input information serving as an objective, and a plurality of data sets including input information different from the objective and under a plurality of manufacturing conditions; a selection unit that selects a data set serving as source data from among the plurality of data sets on the basis of prescribed conditions; and a training unit that trains a prediction model for predicting the output of a manufacturing device using the selected data set and the target data set.
Need to check novelty before this filing date? Find Prior Art