Breadcrumb navigation

Case Studies : Manufacturing Monitoring Solution: Invariant Analysis Technology(SIAT) [2:56]

Movie script

Nowadays, Manufacturing companies are using AI to monitor the work place.

However, due to a lack of human resources who handle AI

and an unpredictable judgment basis because of black box algorithm processes,

Manufacturing companies are facing challenges to realize monitoring by AI on-site.

NEC's invariant analysis technology collects and analyzes a large amount of data,
automatically models the invariant relationship of the normal state of equipment,

compares the predicted value obtained from the model with the actual measured data,
and accurately detects changes in real time, is comprehensive, and continuous,
before any issue occurs.

Current On-site staff can handle this technology even without having analytical expertise such as IT and statistics.

Nippon Steel, Japan's No. 1 steel maker, had many unknown equipment failures even experts had never experienced due to the increasing burden on their facilities caused by advanced customer requests.
They were facing the limits of predictive detection by conventional threshold management.

So, based on the measurement data from more than 2,000 physical sensors,
the behavior of the device is automatically modeled by AI to prevent trouble.

Since it is possible to find the signs of failure that are difficult to resolve and visualize the basis, the cause can be identified and the recurrence of trouble can be prevented.

Invariant analysis technology is also used for quality control and performance evaluation using sound.
Since it learns as a normal state even in a noisy environment, abnormalities can be easily detected.

In the past, it was difficult to distinguish abnormal sounds in a noisy manufacturing site.
However, invariant analysis technology learns as a normal state even in a noisy environment, making it easy to detect anomalies.

Therefore, it is possible to monitor a wide area with a small number of sensors using an ordinary microphone,

so sound monitoring can be applied to a wide range of works.

By utilizing invariant analysis technology that can be used on manufacturing sites,
NEC realizes smart monitoring by AI.

Orchestrating a brighter world


Many manufacturers are using NEC's unique AI technology "Invariant analysis technology" to realize smart monitoring. In this video, we will introduce an example of streamlining plant equipment inspection and operation monitoring in a company that has introduced this solution, and an example of solving problems at a maintenance site by detecting signs of failure by sound.

Related Link

Your Opinion/Comments

Did you like the contents? Please check on a scale of 5.
  • *
    Please send messages in English. Thank you for your cooperation.
  • *
    Response to comments will not be made.
  • *
    All personal information will be handled according to our privacy policy.


(February 14, 2022)