Predictive Maintaiance/ Failuer Template

sunnyalsunnyal Member Posts: 44 Contributor II
edited January 2020 in Help

How do we use this template to run for our own data. Is there a step-by-step process to change the sample data set to the real data set in the required format to run this? I hope we do not have to re-create this, rather just change the required dataset in the required format? Please correct me here

 

Also, are there any other community developed Equipment Predictive Failures Models/Templates that we can use?

 

Thx in Advance

 

 

Answers

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn
    Put a breakpoint on the Retrievr operator and inspect how the data is structured. Then take your data and transform it that way. Then you can use it with your own data.

    Search through the Community, there might be some sample processes.
  • sunnyalsunnyal Member Posts: 44 Contributor II

    Thank you for the swift response. Can you please tell me the location of the reference data? where is the location  of below highlighted? If I just replace the sample file with mine in the required format would that do the trick? Also, how do we modify this model (template) to predict the equipment failure over time  (i.e.) degradation curve

     

    reference data.JPG

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    That's the right data file but I wouldn't "over-write" that file. I would load the "Predictive Maintenance" process and save it into your own repository and then copy over the data files. Ideally your data should be the same format as the reference data file. However, if it's not, not all is lost. You'd have to tweak the process to load your own data. 

     

    I've never made a degradation curve but i'm assuming it'll be in some X,Y format where X is the date. You can visualize that in the Charts and you can predict the series using the Series Extension.

Sign In or Register to comment.