Need help returning only 1min

Question

I'm working on a custom Discrete Job Picklist and am trying to return the locator with the least amount of the item that is required to pick. This is working fine except when the minimum is not distinct.
Here's the bit of code that I'm working with:
and d.total_qoh = (select min(f.total_qoh)
from apps.mtl_onhand_locator_v f
where d.item = f.item)
This returns the mins but give me duplicates when the mins are the same. I'd like to return only 1 min locator per item. I've tried using a select distinct, but this does not work.
Ideally, I'd like to use the min(locator_id) to determine which locator to use should the min(total_qoh) be non-distinct.
I've tried using:
and d.total_qoh = (select *(select min(f.total_qoh)
from apps.mtl_onhand_locator_v f
where d.item = f.item
order by d.locator_id)
where rownum = 1)
But I don't think that order by is allowed in 8.1.6.1 and so it is not working. Likewise, I can't seem to get TOP N to work.
Any suggestions?
-Tracy
Here's the full sql:
select b.wip_entity_name "Job #",
       d.padded_concatenated_segments "Item",
       d.item_description "Description",
       c.REQUIRED_QUANTITY "Qty",
       e.segment1 "Row",e.segment2 "Section", e.segment3 "Shelf", e.segment4 "Bin",
       d.total_qoh "Qty at Loc",
       d.locator_id
from wip.wip_discrete_jobs a, wip.wip_entities b, wip.wip_requirement_operations c, apps.mtl_onhand_locator_v d,  inv.mtl_item_locations e
where a.wip_entity_id = b.wip_entity_id
and   a.wip_entity_id = c.wip_entity_id
and   c.inventory_item_id =  d.inventory_item_id
and   d.locator_id = e.inventory_location_id
and   d.ORGANIZATION_ID = '5'
and   b.wip_entity_name >= '70319'--:job_low
and   b.wip_entity_name <= '70320'--:job_high
and   e.segment1 is not null
and   a.status_type in  ('3', '1') --released and unreleased jobs
and   d.total_qoh = (select min(f.total_qoh)
                    from apps.mtl_onhand_locator_v f
                    where d.padded_concatenated_segments = f.padded_concatenated_segments)
order by b.wip_entity_name, e.segment1, e.segment2, e.segment3, e.segment4; Thanks for any help you can offer.
-Tracy

Answer 1

Richard - this is yielding ora-00904: invalid column. Although I have verified that the column names are correct. I think this may have something to do with not allowing an order by in a subquery for 8.1.6.1 but am not positive. It seems that I either get an invalid column name or missing right parenthesis when using order by in the subquery.
and d.locator_id = (select locator_id
                       from (select locator_id
                                from apps.mtl_onhand_locator_v
                                where d.padded_concatenated_segments = f.padded_concatenated_segments
                                order by total_qoh desc, locator_id desc)
                       where rownum = 1)
-Tracy