Anomaly Detection of Manufacturing
SmallTrain supports various data. This section describes anomaly detection as a time series forecasting project using SmallTrain.
It can detect equipment abnormalities and promptly notify of deterioration and defects. This will allow you to order the parts needed for repairs earlier and make repairs easier.
Advantages of using SmallTrain for anomaly detection
1. Cost Reduction
There is no need for additional data collection with sensors as there are technologies that allow the user to detect with less data. *
Therefore, it is possible to reduce the large cost of collecting data and expect cost-effectiveness.
*This can be difficult if there is no time series data available (data with time-varying information).
2. Data Acquisition Automation
Data acquisition is also automatically acquired by API connection, and the result is also automatically returned.
3. On-Premises
SmallTrina basically provides results on a cloud basis, but if you cannot upload data to the cloud due to the company’s information management system, Also available on-premise version
Project examples:
PoC for anomaly detection with small data on manufacturing system