Cost Contamination -- BPCS-L

Feedback I hoping to see here is * perhaps there is some coding we should check to make sure that some items are setup right * perhaps there is a known BPCS bug and a known work around * perhaps BPCS does not support what we are trying to do, and there are suggested solutions We are on 405 CD modified ... we are using small portions of DRP and JIT and have modified them. I told my boss that for this scenario, I thought we should consult with our tech support, but he is reluctant to call tech support every time something goes wrong ... he wants us first to check what we can check in case we do in fact have some incorrect coding or inappropriate transactions, that we ourselves can identify.

Historically our items fell into several categories * manufactured * purchased * neither ... such as engineering change history comment in BOM and item#s used for purposes other than items Recently we have some items that are BOTH manufactured and purchased, and it is with them that we are having a problem.

Without me going into the business particulars, there are some parts that we can manufacture on extremely short notice (like a few days) or we can source more cheaply in which there is a long lead time (many weeks), so customers have choice of getting their parts in a few days after order, at a higher price, or waiting several weeks and getting them at lower price.

A related topic is making the two different prices easily accessible to customer service, for when they keying in the orders. I suggested having a different customer # for the same customer name address, one for facility 60 and the original for the manufacturing facilities, then place the facility 60 price in the special price file which has differentiation by item and customer, but not by facility.

We use one facility for each of our factories, 20 40 etc. In manufacturing facilities, the cost buckets of items include material, overhead, human labor, machine, outside operations Historically we consider cost variances in purchased materials to be a problem area, but expect variations in the actual cost of manufacturing.

We are using facility 60 for the distribution of the parts we buy on behalf of customers. Parts in this facility are setup using material cost ONLY There should be almost no cost variance because we contract for pricing, receive the parts, receive vendor invoice, do the 3 way match. There are not supposed to be any BOM or routings in facility 60.

Things get setup so they are correct, then later we discover that various items in facility 60 have had added to them a copy of the cost buckets that facility 20 or some other manufacturing facility has, with the same actual costs as that other facility, while standard zero for those non-material buckets. This in turn causes a cost variance ... we have not figured out what damage downstream like to G_L, but basically this is something we not want happening and need to figure out what causes it ... I found out for the first time about this near the end of last week ... some co-workers have known about it longer than me.

I ran a query/400 listing buckets of facility 60 other than material and total, which had got populated with non-zero, then for those items I checked what recent transaction there had been in any facility, since the last time the costs had been believed to be correct. * They had been manufactured and shipped in facilities 20 40 * They had been received and shipped in facility 60 * They had been transferred both ways in and out of facility 60 vs. the other facilities. One co-worker speculates that what is happening is there is a bug in the labor reporting such that when labor updates these items in facility 20 or another manufacturing facility, it also updates those actual costs in all facilities ... I find this hard to believe since the different manufacturing facilities have never contaminated each other in the past.

My speculation is that when the item gets transferred between facilities, for whatever purpose such as doing a consolidated shipment, for consolidated inspection before shipment, or by mistake, the method of transfer causes the TO facility cost buckets to get populated based on which buckets existed in the FROM facility that did not pre-exist in the TO facility.

Currently our cost basis is LAST COST, not any averaging, and our Accounting is at Standard Cost, with some monthly adjustments to compensate for what is in Production Work in Process.

I suggested that perhaps we need an Al fixit program to be run regularly to unpopulate facility 60 of the unwanted cost buckets. Al's boss says no, we first figure out what is causing facility 60 to get the unwanted stuff, perhaps we doing something wrong and it is better to not be doing that.

I plan to do some experiments in which I pick a weekend time when no one else updating BPCS, find one of these items that has on-hand in both facility 20 and 60, is manufactured in 20 and purchased in 60, make sure the costs are correct, then do one of the actions that is suspected of contaminating facility 60 costs, then check the costs again. I have several different things I want to try.

But right at the moment I am covered up in more urgent work I need to do, and next weekend is EOM, so it will probably be a couple weeks until I do my round of tests, by which time I may have added to the list of things I plan to check.

This weekend I spent a large chunk of time with cost repairs. We had over 1,000 items in which the actual cost buckets did not add up right, a handful of items with negative costs, and other problems. The reasons for this are problems previously discussed in BPCS_L and stuff I had not looked at for some time, but now were contributing to some reports not being correct.

PS ... thanks for the XL/400 tips

- Al Macintyre http://www.ryze.com/go/Al9Mac Find BPCS Documentation Suppliers http://radio.weblogs.com/0107846/stories/2002/11/08/bpcsDocSources.html BPCS/400 Computer Janitor