Modeling Storage Systems Where Items Can Span Adjacent Racks

Question

question

Stan Davis asked Sep 19, '22 Stan Davis commented Sep 20, '22

Modeling Storage Systems Where Items Can Span Adjacent Racks

Since a slot's depth cannot be independently specified and the 'y' value controls the entire rack's depth, what are some possible approaches in modeling systems where large items are stored by spanning adjacent racks?

For example, if there are 2 racks back-to-back with unique items stored respectively in each rack's lower levels, how do you model large items being stored on the top level that span across both racks?...

We also have situations where items are spanned at lower levels.

If a slot's depth could be independently controlled, you could specify the required depth for Rack A's slot and then set Rack B's corresponding slot as not storable.

I'd appreciate any ideas on how to approach this. Thanks - Stan

Software Version:

FlexSim 22.2.2

racks slots

1663623907630.png (10.7 KiB)

· 2

Joerg Vogel commented · Sep 19, 2022 at 11:07 PM

Any slot can hold items larger than rack depth. It is only restricted by method hasSpace. Leave this method from your find slot query and you can place items wherever you like.

0 ·

Stan Davis Joerg Vogel commented · Sep 20, 2022 at 09:40 PM

Thanks Joerg. Understood, but still need to employ hasSpace for non-spanned items (see below). - Stan

0 ·

Answer 1 · 2022-09-20T11:16:56Z

Jason Lightfoot answered Sep 20, '22 Stan Davis edited Sep 20, '22

You can't assign two different slots to an item, so the question is then whether you need to reserve the slot in the other rack. (You may need a map of the counterpart slots if the bay/levels do not match).

If there are dedicated slots for these spanning items, you can mark them as unstorable in one of the racks and just place oversized items in the 'primary' rack by removing the space constraint from the query.

If the slots can be use by other non-overlapping items, then you can set a slot label on the counterpart indicating it's occupied and include that in all the slot queries (testing AND slot.occupied=0).

Another method to consider would be to create a dummy items and assign those to the counterpart slots - and then manage their unassignment when the item is picked from the rack.

· 1

Stan Davis commented · Sep 20, 2022 at 09:36 PM

Jason,

Greatly appreciate the ideas. For our situation, a 'span' slot is dedicated and only holds one item. 'Regular' slots still require the hasSpace method.

Therefore, I've employed a decision activity to test for the depth of the item.

If its depth is greater than that of the racks, the hasSpace is not part of the query and the found slot is then set to occupied.

If the depth is smaller, then the hasSpace method is employed.

Still working out details but headed right direction. Thanks again - Stan

0 ·

1663709427793.png (139.7 KiB)

question