Aarbakke used DataOps provided by Cognite, including contextualized data in Cognite Data Fusion (CDF), to create a dashboard of CNC machine alarms, helping engineers reduce the number of breakdowns and extend the lifetime of the machines.
Aarbakke has dozens of computer numerical control (CNC) machines at its factory in Bryne, Norway. The machines complete complex operations on sometimes rare materials to achieve highly precise product requirements that its customers in the oil and gas industry demands.
Historically, the CNC machines have sometimes been unknowingly operated in a suboptimal way, and there have been no alerts or warnings prior to them breaking down. Issues include high temperatures in coolants or oils, which leads to wear and tear; wrong pH and salinity in the coolant, which can cause corrosion or bacterial or fungal growth; incorrect lube oil consumption; and missed maintenance on the machines.
Aarbakke lacked a master log of these machine alarms, as well as a system to filter out less critical ones. Service managers previously depended on operators to send them a note every time a critical issue occurred. Otherwise the service managers needed to physically go to each individual machine and manually pull a local log to view the alarms.
Aarbakke and Cognite first liberated data about machine alarms from its source system, ingesting it into Cognite Data Fusion (CDF). With all data streaming from one place, the developers then created a dashboard that shows an overview of all alarms but also groups alarms by machine and issue. This helps service engineers pinpoint specific issues and machines and take targeted maintenance actions to address them.
Aarbakke and Cognite plan to add more functionality to the dashboard in the future, including a feature that lets service managers assign levels of criticality to alarms, ensuring that the alarms they deem most important will always be featured at the top of the list.
Improved monitoring of operational parameters and the ability to look at records of alarms and warnings centrally will reduce the number of breakdowns and extend the lifetime of the machines. Beyond that, collecting cleaned, contextualized data about alarms will help drive Aarbakke toward a future in which the company can predict potential failures before they happen.
Aarbakke estimates that the dashboard will cut service costs by 20-30%, reduce downtime, and avoid unplanned stops due to mechanical reasons.