Data analysis

• CX-W Quick outlier detection software

In order to generate a program that determines the state of the device, it is necessary to analyze the time series data from the vibration and sound of the device, sensors, etc. and create a discrimination model, but as long as you have specialized knowledge such as a data scientist. There is a shortage of human resources, and there are major challenges in terms of technology and cost which needs to be solved.

• CX-W Quick outlier detection software

  CX-W is an AI-powered embedded software for anomaly detection and prediction. CX-W runs on edge equipment with Windows/Linux OS, and provides ability to train data, inference, and output diagnostics with contexts. Abnormal data sets are not required to start with. CX-W automatically trains correlation between every monitored variables and criteria, then quantifies criticality of each alerts. This enables OT operators to detect anomalies which could not be detected by traditional conventional threshold detection methods.

  

  Features

  ・ Automatic change monitoring by learning operating data
  ・ Threshold generation from both numerical distribution and correlation
  ・ Device can be incorporated and integrated.