HasSpace Return Value Changes Based on Stacking Order??

Question

question

Stan Davis asked Oct 10, '22 Matthew Gillespie commented Dec 2, '22

HasSpace Return Value Changes Based on Stacking Order??

I am using the HasSpace method in a Find Slot activity. With a X+ > Y+ > Z+ stacking order, the HasSpace is failing to find space for 4 items when there is still plenty of space in the rack (1 slot per bay).

It appears it is returning 0 because the items' dimension would exceed the boundary at the back of the rack...

However, if I change the stacking order to X+ > Z+ > Y+, HasSpace finds space for all items...

With the X+ > Y+ > Z+ stacking order, if HasSpace can't find space on the first row, why doesn't it simply find space by starting a second row??

It seems you'll eventually you'll run into the same problem described above regardless of the stacking method. Thanks - Stan

Software Version:

FlexSim 22.2.2

find slot hasspace

1665409548225.png (12.6 KiB)

1665410862376.png (149.6 KiB)

1665411124696.png (300.8 KiB)

1665411493036.png (296.4 KiB)

· 6

5 |100000

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

Jason Lightfoot ♦♦ commented · Oct 10, 2022 at 03:14 PM

In my quick test it observes the stacking order - can you post your test model?

0 ·

1665414889800.png (155.7 KiB)

Stan Davis Jason Lightfoot ♦♦ commented · Oct 10, 2022 at 07:07 PM

Jason, pls see attached model. I sanitized & scaled it down for submission, but it retains its issue.

The Find Slot's query also uses PartNo labels which are cleared and applied via 'Clear-Rack_Paint' and 'Update_Paint_Rack' respectively

The size of the items is controlled by the DimTable.

The items will be stored for certain changes in size.

I also discovered if the 4 remaining items (Grapes) are placed in the front of the arrival list, they are stored in the rack.

It's a mystery.

Thanks - Stan

Find Slot HasSpace Issue - 10OCT22.fsm

0 ·

1665428789882.png (125.2 KiB)

find-slot-hasspace-issue-10oct22.fsm (107.6 KiB)

Jeanette F ♦♦ Stan Davis commented · Oct 10, 2022 at 08:36 PM

Hello @Stan Davis

Something went wrong in your upload. It shows the text in the body of your reply for a model but there is no attachment. Can you please upload again?

0 ·

1665434179945.png (46.6 KiB)

Show more comments

Joerg Vogel commented · Oct 10, 2022 at 06:29 PM

@Stan Davis , method hasSpace has got a really simple character. It keeps item orientation and stack each item accordingly the order. Optimizing space and produce a compact block is not its metric. You see it in Jasons picture. There are many holes and some items look like if they are floating.

0 ·

Stan Davis Joerg Vogel commented · Oct 10, 2022 at 07:15 PM

Understood. I'm not expecting optimization just placement when there is truly space. Thanks

0 ·

Answer 1 · 2022-10-11T08:22:10Z

Felix Möhlmann answered Oct 11, '22 Matthew Gillespie commented Dec 2, '22

I can corroborate the observation. This looks like a bug/oversight. It appears that if there is already at least one item in the current "row" and the next item would fit beside it going by its x-size but not by its y-size, the hasSpace method returns a false instead of checking if the item could be placed in a new layer.

This makes sense when the "row" is the first one, because the item not fitting in the y-dimension would mean that it is longer than the rack is deep. But for further rows the rack depth will be different from the available depth.

Blue would fit beside green in the x-direction but not y. -> Blue stays in the queue.

If blue is the first item in the "row", it is placed in a new layer.

1665476146863.png (113.7 KiB)

1665476218684.png (123.2 KiB)

· 4

5 |100000

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

Stan Davis commented · Oct 11, 2022 at 04:51 PM

Thanks Felix.

0 ·

Jeanette F ♦♦ commented · Oct 11, 2022 at 05:41 PM

I will send this in as a bug to the dev team.

0 ·

Stan Davis Jeanette F ♦♦ commented · Oct 11, 2022 at 07:39 PM

Thanks Jeanette.

0 ·

Matthew Gillespie ♦♦ commented · Dec 02, 2022 at 09:06 PM

@Stan Davis @Felix Möhlmann This bug has been fixed for 22.0.10 which comes out next week.

0 ·

question