Natural Resources Canada operates an Automated Process Fault Diagnosis Tool that uses machine learning and causality analysis to detect and diagnose equipment failures, energy inefficiencies, and quality issues in industrial processes and buildings. This system automates a task that is traditionally time-consuming and expensive for the process industry.
The tool connects to online databases to collect operational data, prepares this information, builds machine learning performance models, and automatically identifies problems across industrial and building systems. It affects both government employees and the public by improving operational efficiency and reducing costs in facilities management and industrial operations.
This system is currently in production and uses data from both public and private sources. The system does not involve personal information. Users should know that the tool makes automated diagnostic recommendations based on pattern analysis, and that these recommendations should be reviewed by qualified personnel before implementation.
