Cognite Data Fusion for Domain Experts

  • 4 October 2022
  • 8 replies

Userlevel 4
Badge +3

This discussion is linked to the course Cognite Data Fusion for Domain Experts.

A data-driven approach to maintenance is more efficient than relying on a fixed schedule or fixing something once it’s already broken. There are plenty of opportunities around us to take advantage of smart maintenance. Look around in your home, your office, or think of your work tasks. Where could data-driven maintenance make a difference?

Share your thoughts! For example, explain what data you have available or would need in order to do smart maintenance, and how this would be an improvement of current methods. 

8 replies

Userlevel 3

We agree and 100% support this, and would like Cognite to build out functionality and capability that can help us do smart maintenance at Celanese.



Userlevel 2

Awesome resource! I just took the course. 
One question, in the last videos it is mentioned that there was a shutdown in July hence the mileage of the valve is counted from that point, also it is suggested to do the same from other maintenance events to measure the mileage between events. Where can I see which were maintenance events? from the time series there are some signals that indicate if the compressor is working or not, but I have not seen one related with a maintenance procedure. 

Userlevel 4

Hi @Luis Ramon Ramirez Rodriguez , glad you liked the course!
You are right that events in Open Industrial Data are harder to use as they are anonymised. We are planning to look into removing some of the abstraction and making more information available about events in this project. Stay tuned in the community for more updates on OID and new courses that work with the data.

Userlevel 2

Great! Thanks Eniko 

Yes, condition monitoring and predictive maintenance is the best way to run operations in asset intensive industries.  


Preston, (Vibration Cat III, CMRP)

Condition based maintenance is a natural evolution that has already presented itself in consumers daily lives.  Newer automobiles, for instance, are much more likely to suggest maintenance based on measured conditions instead of the time-based approach conventionally used.  Proper tire inflation and oil effectiveness are crucial for vehicle performance, and simple sensors provide insights that simplify maintenance tasks and prolong the life of a vehicle.  Home heating/cooling systems have long used current air temperature to control when a thermostat activates climate control - much more effective than trying to guess how many minutes per day to run your heater!

Userlevel 1

I would like to implement smart maintenance for the central cooling and heating units at my house. Now, I get the units serviced on a schedule usually once or twice a year.  Or, I call a repairman out when something goes wrong e.i. not cooling or heating as expected. I’d like the units to communicate to me about maintenance needed based on actually usage/runtime, or predictively notify me that freon is running low or some other part of the system has become unreliable or damaged so that I can get these things addressed before there is a failure, and we are stuck in the Texas summer with no AC. 


I'm looking forward to using Cognite to help me monitor some alarms used in my company's installation. There are many sensors in a refinery. It's so big that I don't know what the best area to start would be. Creating tags and organizing knowledge will help us a lot. I have never used this type of solution before and I believe it will be very useful.