Checking and Validating an Insight
How to check and validate an Insight on the platform
On the platform's main menu, select the "Insights" tab. Then, choose the insight that will be checked.
You can change the status of the Insight depending on the potential issue of the verification:
In Inspection: choose this when the cause of the alert is being verified on the machine;
In Intervention: choose this after the alert has been inspected and confirmed, and when an action is already being taken to correct it.
If a failure has been identified, click on "Check" only when the intervention has already been completed and the failure, therefore, has been avoided.
In cases where no failure has been identified, it is also crucial to click on "Check" and fill in the required information in the opened window.
Always check and validate the platform's insights, regardless of whether a failure has been identified or not.
The purpose of checking and changing an insight status is to evaluate its accuracy –that is, whether a failure has actually been identified or if another factor triggered the alert, such as a change in operation.
Failure identified: If the problem has been corrected (Yes), the platform learns the failure pattern of the asset and becomes more accurate. If the problem has not been corrected (No), it is understood that a failure is still occurring in the asset. If the platform detects an increase in vibration, a new alert will be generated.
Change of Operation: The platform will learn the impact of each type of operational change on the asset and generate more accurate insights accordingly.
No failure: The platform understands that the behavioral change that prompted that insight may be normal for the asset and will take it into consideration when generating future insights.
Sensor issues: Inappropriate positions and other situations involving the sensor can impair the platform's models and insights generation, so it is important to report in this field.
Inspection not performed: The platform understands that that type of alert is not considered concerning by the team, making it unnecessary to perform an asset inspection.
If a failure in the asset has been identified, some information regarding the failure should be filled in, such as:
Failure Criticality: High, Moderate, or Low
Cause of Failure: Lubrication problem, Bent shaft, Wear, Clearance, etc.
Failure Component: Gears, Rotor, Bearing, Coupling, Structure, etc.
Intervention time: Hours or Days
Expected savings: Cost of parts, Cost of Downtime, etc.
Observation: Any relevant information you deem necessary about the failure
Through these pieces of information filled in by you and your team, the platform learns about the behavior of your assets and improves the accuracy of future alerts. Therefore, make sure to fill them correctly.
Once done, simply click on "Check" and the insight will be checked and validated on the platform.
Updated on: 11/27/2023