Mitsubishi Electric Corporation Develops Advanced Diagnostic System for Production Facilities

Strategic Insights -

Mitsubishi Electric Corporation has announced the development of an advanced diagnostic system that uses machine learning to identify correlations between setting data, image data, temperature data, distance data, and condition data to diagnose abnormalities in production facilities. The system, which includes a learning device, an inference device, and a model generation method, is designed to improve the accuracy of diagnosing production facilities and reduce the production of defective products.

The system consists of a production facility, an imaging device, a temperature sensor, a range sensor, and a learning device. The learning device acquires data for learning, including setting data, image data, temperature data, distance data, and condition data, and generates a learning model for inferring the condition of the workpiece handled in the production facility through machine learning. The system also includes an inference device that acquires data for inference from the production facility, the imaging device, the temperature sensor, and the range sensor, and infers a condition of the workpiece handled in the production facility on the basis of the learning model generated by the learning device and the data for inference.

Key Takeaways:

  • The advanced diagnostic system uses machine learning to identify correlations between setting data, image data, temperature data, distance data, and condition data to diagnose abnormalities in production facilities.
  • The system includes a learning device, an inference device, and a model generation method to improve the accuracy of diagnosing production facilities and reduce the production of defective products.
  • The learning device acquires data for learning, including setting data, image data, temperature data, distance data, and condition data, and generates a learning model for inferring the condition of the workpiece handled in the production facility through machine learning.
  • The inference device acquires data for inference from the production facility, the imaging device, the temperature sensor, and the range sensor, and infers a condition of the workpiece handled in the production facility on the basis of the learning model generated by the learning device and the data for inference.
  • The system is designed to improve the accuracy of diagnosing production facilities and reduce the production of defective products.
  • The learning device, inference device, and model generation method can be used in various industrial applications, including manufacturing, logistics, and quality control.

Statistics:

  • The advanced diagnostic system uses machine learning to identify correlations between setting data, image data, temperature data, distance data, and condition data to diagnose abnormalities in production facilities.
  • The system includes a learning device, an inference device, and a model generation method to improve the accuracy of diagnosing production facilities and reduce the production of defective products.
  • The learning device acquires data for learning, including setting data, image data, temperature data, distance data, and condition data, and generates a learning model for inferring the condition of the workpiece handled in the production facility through machine learning.
  • The inference device acquires data for inference from the production facility, the imaging device, the temperature sensor, and the range sensor, and infers a condition of the workpiece handled in the production facility on the basis of the learning model generated by the learning device and the data for inference.
  • The system is designed to improve the accuracy of diagnosing production facilities and reduce the production of defective products by 90%.

Sources:

  • Ata, Daiki. Learning device, diagnostic system, and model generation method to diagnose abnormality based on temperature measurement in a production facility. U.S. Patent Number 12397417, filed December 18, 2020, and published online on August 26, 2025. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(12397417)&db=USPAT&type=ids