Lynn,
     I know the issue of the automatic backflush not accurately backflushing BOM material has been a long standing topic.  I am looking for a one-stop comprehensive explanation of the problem and solution.
      We routinely have work orders that have over issues, under issues, and mixed errors (that is both over issues and under issues on the same work order.)
       We have 2004.1 with IST 7/21/04 version.   Since loading this version in August, we have continued to see the same level of work order issue errors.  The fix that relates to a temporary file for locked records only turns up a very few of the issue errors.
        I therefore think that there is something going on that is beyond record locking.  In many cases our BOM's are very long and I wonder if the length of the BOM may increase the likelihood of error.
        Currrent example. Work has been completed. All units show as closed in inventory transactions as Work Order receipts. Nonetheless the BOM material shows as only issue to 2.38 % up to 21.43%.  When I look at inventory transactions on a component on the BOM the work order issues all show correctly.  The errors in Work Order material issues, wreaks havoc on the allocated quantities for the BOM items.
        This problem may well be caused by something we are doing but we have not been able to isolate any operator error.  On the other hand, the significance of the problem would seem to have raised a much higher level of complaints from the field if others saw this problem as often as we do.
            Help

The backflush default is set to "A"

Lynn,  In my previous message I incorrectly said the date of transaction was 11/2/02.  I meant yesterday 11/2/04.

I am not sure what you mean by errors.  No error message came up at the time of the finished goods unit closes.
 
On the specific incident I mentioned in the previous message, I am told that the material issue screen which is  at WO-A originallly reflected that all but one BOM item was correctly issued at 100%. Iin an attempt to correct the one item a work order rebuild was run.  After rebuild all BOM items were in error by the components associated with 12 units of finished goods.  When we compared WOMAT to INVTXN we can see that a 12 unit close did not hit WOMAT.
       This type of problem has been a long standing and very recurrent problem for us. But I get the sense that the rest of the user community isn't seeing this problem. So I wonder if our software could be contaminated or something peculiar to our use.

No, that might cause WOBOM allocations to be incorrect but it would not cause the WOMAT records to not exist at all.

We have a  issue today similar to what is being discussed in this thread.  An issue was made to a work order with 100 % of this component already issued.  All the fiels contain the4 data, I have looked in WOMAT, INVTXN, the transactions print out in the job cost JC-E and show up in all the inventory reports,  but they do not show in WO-A screen for materials.  WE actually use that screen a lot to check to the issue status quickly.  The transaction was done in Evo.  I don't know if the same would have happened in Classic.  Of course my people are blaming Evo.  
I don't know where the WO-A screen gets the data for the materials screen or I would fix it.  Is that possible?

What program generated the issue?  WO-G, WO-I or backflush by sequence using WO-F or DC-H?  Was there any scrap processing involved?

So what exactly is the problem (or perceived problem)?  WO-G will certainly allow issues over 100% - that in and of itself is not a bug but the % issued in WO-A should reflect the overissue.