Our internal foundry operation had been closed down a few years before — a victim of tougher EPA regulations and the inability of our ancient equipment and modern union contracts to keep us competitive. We had not yet adjusted to purchasing castings outside. When we had operated our own foundry, we enjoyed lead times so short we could replace scrap the next day. But when we began buying castings outside, the lead times increased dramatically. Since every order required several unique castings that had to be engineered to order, the engineering lead time (combined with the long procurement lead time) strained our ability to stay competitive in the area of delivery lead time to the customer.

As you can imagine, the quality of work life deteriorated as pressure increased on engineering, purchasing and manufacturing to "get the product out somehow." This was aggravated by a factory scrap rate that added even more lead time by requiring us to reorder castings, pay expedite premiums and internal overtime each time we ruined one.

We middle managers were delighted when a vice president from the home office showed up with the solution to this problem. He asked the MIS guy to program a percentage safety stock field into the part master to cause us to overbuy castings, at least the common ones. We certainly did not wish to overbuy the unique ones and knew their low volumes would never yield a percentage of even one piece.

From a late '90s view, having to program a safety stock function sounds bizarre. But, way back then, companies often chose to program MRP in house. A few good programmers could pretty much replicate the simple MRP packages for sale at the time, and we were led by top managers who thought we were unique in the world of manufacturing and that no "off the shelf" software could possibly help us. But, gratefully, they wanted no part of the detail, so we middle managers were allowed to spec the system. As long as we told the guys on top that the BOM processor would only process our BOMs and that inventory calculations only worked for our inventory, we could do anything we wanted. So we had a primitive MRP system up and running but had not yet addressed safety stocks when the vice president carried out his preemptive strike. Of course, we should have programmed a scrap factor, but that would require explaining something new to the vice president. And, if geared to current demand instead of some statistical average, a percentage safety stock would function like a scrap factory anyway. So we let the vice president and MIS gurus go ahead with the safety stock and decided to tidy things up later.

Well, the safety stock worked. Scrap remained high, but we always had a few extra castings to eliminate the replacement lead times. The vice president had ordered a 20 percent safety stock on all castings. This would provide extras for the high-volume castings but not for the unique ones — assuming the calculations rounded down. We certainly wanted no extra unique castings; they would be obsolete if not consumed.

But soon we noticed extras of these unique ones. These extras were for orders shipped weeks ago and were thus destined for the landfill. The vice president returned and bounced around like a superball demanding to know who ordered extra unique castings. A quick review of past MRP output clearly showed these extras were derived from safety stock, so the MIS guy was dragged into the inquisition to explain how a percentage safety stock could generate demand for unique castings. These castings were almost always needed in quantities of one. A 20 percent safety stock would yield a requirement of .20 — how did that get turned into 1.0?

"We rounded up, of course," the MIS guy explained defensively.

"Who told you to round up?" The vice president's anger turned to fear — sensing for the first time the magnitude of the problem. This was not human error limited to a few part numbers, this was systemic; and every part number created after we had programmed for the safety stock would be affected.

"Nobody told us to round up. But it just made sense. If we rounded down you wouldn't generate any safety stock at all," the MIS guy argued

"That was the whole point of using a percentage safety stocks instead of fixed quantity safety stocks," the vice president shouted.

We had generated $250,000 in obsolescence in approximately 60 days because the vice president and the MIS guy failed to define the project in sufficient detail. We stopped generating obsolescence, but the costs of this fiasco were just beginning. The vice president felt it was not in the best interest of his career to scrap a quarter of a million dollars so he pressured engineering to design this stuff into future products. We should have scrapped it all. Dumping $250,000 in non-salvageable obsolescence only cost $250,000. Selling product containing components that won't quite perform to customer specs — components designed into the product because they were available instead of right for the job — that can cost much more.

Copyright © 2020 by APICS — The Educational Society for Resource Management. All rights reserved.

Web Site © Copyright 2020 by Lionheart Publishing, Inc.
All rights reserved.


Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: +44 23 8110 3411 |
E-mail:
Web: www.lionheartpub.com


Web Design by Premier Web Designs
E-mail: [email protected]