🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

[Example] Predictive Maintenance Part 2

atamulewiczatamulewicz Member Posts: 21 Contributor II
edited December 2018 in Help

Hello All, 

 

Here is the second of two examples of using predictive maintenance from our data scientists here at RapidMiner.

 

Enjoy!

 

Read part 1 here

Tagged:
lionelderkrikor

Answers

  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,190   Unicorn

    Hi @atamulewicz,

     

    I am interested in predictive maintenance and I studied with lot of attention these different processes.

    However, I ask myself "practical" questions : 

     

    - In practice how is obtained the training set  ? An arbitrary extraction of data from the database of the industrial process at a given moment ? or are they (general) rules to create/build the training set ? How is obtained the information about the machines ("failure" / "no failure"). I suppose that this information is not available directly ? 

      - Are such processes builded for a punctual study/project ?

        or in practice, the process is "feeded" by an updated database (with the sensor datas) and then the process product updated predictions. 

      - In this last case, how is, in practice, implemented / interfaced the RapidMiner process with, on the one hand, the database of the industrial process and the other hand with the monitoring system of the industrial process, in other words, how the predictions of the process go back to maintenance department in order to take into account these predictions ?

     

    To sum up, I would like to understand, in practice, how are such processes implemented in an industrial environment ?

     

    That's a lot of questions, so thank you for giving me a little of your time to answer me.

     

    Regards,

     

    Lionel

      

     

     

     

Sign In or Register to comment.