Skip to main content
Question

Machine Learning - PM Actions - Work Order Management

  • January 23, 2025
  • 8 replies
  • 83 views

Forum|alt.badge.img+7

HAs anyone used the Machine Learning capability in IFS Cloud or can point me to more information about it? I’m interested in how it can be used to manage PM Actions for automated Work Order generation and Work Order management and scheduling.

 

Thank you!

8 replies

Forum|alt.badge.img+4

Hi ​@taenaren ,
 

The PM and Work Order Management Areas do not currently have machine learning capabilities.
In 24R1, there is a "IFS.ai CoPilot" that functions similarly to a chatbot.

 

https://docs.ifs.com/techdocs/24r2/030_administration/090_automation_optimization/200_machine_learning/

The Machine Learning Service is available if the Cloud deployment model is used to install the IFS Cloud installation.
You won't be able to use this service if you installed using the remote deployment approach.

Thank you & Kind regards,

\ Krishantha 


eqbstal
Superhero (Partner)
Forum|alt.badge.img+21
  • Superhero (Partner)
  • 677 replies
  • January 27, 2025

@KrishanthaRutterKey From your text I deduct that

  1. there are no options in AI, for this moment (so up to and including IFS24R2), that allow a machine learning model to ‘learn’ from what is happening with a work order.
  2. learning capabilities can only be turned on during installation
  3. on remote deployments there is no option to include the learning capabilities

Did I conclude the correct thing per item?

 


Forum|alt.badge.img+7
  • Author
  • Do Gooder (Customer)
  • 14 replies
  • January 27, 2025

Thanks Krishanta.I will take this across to the conversations area. I’m looking for any capability that will help with optimizing planning and scheduling for work order execution.


Mathias Dahl
Superhero (Employee)
Forum|alt.badge.img+32
  • Superhero (Employee)
  • 2856 replies
  • January 28, 2025
taenaren wrote:

 

HAs anyone used the Machine Learning capability in IFS Cloud or can point me to more information about it? I’m interested in how it can be used to manage PM Actions for automated Work Order generation and Work Order management and scheduling.

 

Thank you!

Hi, can you share some of your thoughts on this with us - what use cases do you have in mind? IFS is looking for more use cases where ML/AI can help out.


eqbstal
Superhero (Partner)
Forum|alt.badge.img+21
  • Superhero (Partner)
  • 677 replies
  • January 28, 2025

I know of a request to have the system check if a fault report has been raised and is still open and the second or third is entered for the same failure. So two or three persons detecting a problem and AI should be helpful in detecting that a duplicate of the fault report already exists (while that one is still open).
I understand that the user could open up a second tab to check for any duplication, but there is hardly time for that, so engineer still enters fault report.
I understand that this might be handled via a modification, but it does mean that at start the possible duplication detection must be included in the functionality that the modification performs.

Another one might be to check for work orders that will perform more or less the same and hance it might be worth copying it over.

 


Forum|alt.badge.img+4
eqbstal wrote:

@KrishanthaRutterKey From your text I deduct that

  1. there are no options in AI, for this moment (so up to and including IFS24R2), that allow a machine learning model to ‘learn’ from what is happening with a work order.
  2. learning capabilities can only be turned on during installation
  3. on remote deployments there is no option to include the learning capabilities

Did I conclude the correct thing per item?

 

1. Yes. 
2. IFS Cloud can be installed in one of two ways: IFS hosted (deployment model - cloud ) or on-premises (deployment model - remote ). At this time, only the CLOUD deployment model has ML capabilities.
3. As far as I'm aware, not at this time.

@Mathias Dahl  correct me if I’m wrong.

Thank you & Kind regards,

\ Krishantha

 


Forum|alt.badge.img+4
eqbstal wrote:

I know of a request to have the system check if a fault report has been raised and is still open and the second or third is entered for the same failure. So two or three persons detecting a problem and AI should be helpful in detecting that a duplicate of the fault report already exists (while that one is still open).
I understand that the user could open up a second tab to check for any duplication, but there is hardly time for that, so engineer still enters fault report.
I understand that this might be handled via a modification, but it does mean that at start the possible duplication detection must be included in the functionality that the modification performs.

Another one might be to check for work orders that will perform more or less the same and hance it might be worth copying it over.

 

​Hi @eqbstal ,

This requirement can be implimented through a CRIM.

Thank you & Kind regards,
\ Krishantha


Mathias Dahl
Superhero (Employee)
Forum|alt.badge.img+32
  • Superhero (Employee)
  • 2856 replies
  • January 31, 2025

@KrishanthaRutterKey 

Mchan! After a long time :-)

@eqbstal 

> I know of a request to have the system check if a fault report has been raised and is still open and the second or third is entered for the same failure. So two or three persons detecting a problem and AI should be helpful in detecting that a duplicate of the fault report already exists (while that one is still open).

Thanks, it sounds like a good use case. We have one or two that's similar that we are looking into, possibly already for 25R1, we'll see...

While simple on a high level, doing something like that requires a quite a bit of new infrastructure to be in place. Compared to the relatively modest time savings (duplicated efforts) in this use case, it's a quite big investment, actually. Luckily, we can implement other use cases using the same technology 🙂 

> I understand that the user could open up a second tab to check for any duplication, but there is hardly time for that, so engineer still enters fault report.

Got it!

> I understand that this might be handled via a modification

Well, yes, one can do almost everything in a modification, but this is not small one.

> , but it does mean that at start the possible duplication detection must be included in the functionality that the modification performs.

Not sure what this means, so I'll let it pass...

> Another one might be to check for work orders that will perform more or less the same and hance it might be worth copying it over.

It sounds very similar to the other use case you described. What do you mean by "worth copying over"?

As for the other questions:

> there are no options in AI, for this moment (so up to and including IFS24R2), that allow a machine learning model to ‘learn’ from what is happening with a work order.

I think you are right. While IFS can release ML models that can be trained on customer data, we need to provide the model as well as use it in the right places.

> learning capabilities can only be turned on during installation

I don't see why they could not be installed after some time, but I might be wrong. 

> on remote deployments there is no option to include the learning capabilities

@petya.skinner Any comments on 24R2, AI and Remote deployment?

By the way, as for this use case, I don't think we would need any "training," we would need something called a vector database and similarity search, plus data sync from the database to the vector database, very briefly described…

 


Reply


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings