question

sharan-nitin avatar image
0 Likes"
sharan-nitin asked Nil Ns commented

How pull strategy works in RL module?

When is the action space triggered and how does pull strategy works in RL module (referring to tutorial example)?

its confusing!!

FlexSim 23.0.0
reinforcement learning
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

1 Answer

Nil Ns avatar image
0 Likes"
Nil Ns answered Nil Ns commented

Hey Sharan,

1711620484808.png

In the processor, we use a Pull Strategy. Then, we try to maximize the value of the expression. The expression returns 1 if the type of the item is equal to the ItemType parameter; in other cases, it returns 0. This means that it will try to select an item with the same type as ItemType.


We use a trigger in OnPullStrategy that calls the RL before the Pull. The RL will change the value of the ItemType parameter, therefore the processor will try to process an item of the type that the RL decides.


Please note that there is a bug in the RL panel and the triggers may not be working. (Reinforcement learning - FlexSim Community)


Thanks!!


1711620484808.png (10.6 KiB)
· 2
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

sharan-nitin avatar image sharan-nitin commented ·
could you please elaborate on the bug?
0 Likes 0 ·
Nil Ns avatar image Nil Ns sharan-nitin commented ·

The bug is related to how the event is configured. If it is done through the RL tab, the event doesn’t trigger, and consequently, the RL is not called. I’m not clear on which versions this issue has been resolved in and in which ones an alternative method should be sought to make the event execute via RL.

0 Likes 0 ·