Page 230 - Orlicky's Material Requirements Planning

CHAPTER 11 Product Definition 209

300 customer orders in any one month is unlikely to break down exactly this way, and
thus some safety stock would be desirable.
As mentioned previously, the proper way to handle safety stock under the MRP
approach is to plan it at the MPS level. Thus the transmissions would be deliberately
overplanned (the statistical technique of determining standard deviations for a binomial
1
population may be used to establish the correct safety stock), and transmission quanti-
ties such as 275 and 100 would be put into the MPS. This would not be done in every peri-
od because unused safety stock is rolled forward, that is, applied in subsequent periods.
The same approach would be followed for the other optional features. Each of the option-
al choices would have to be covered by an appropriate module of the BOM for use by the
MRP system. Under this approach, the total number of BOMs would be as follows:
Basic tractor 1
Wheel arrangement 3
Fuel and horsepower 6
Transmission 2
Steering 2
Rear platform 2
Axles 2
Hitch 2
Power takeoff 3
Radiator shutters 1
Operator cab 1
Total 25

This total of 25 compares with 6,912 if each tractor configuration had a BOM of its
own. Now, if the engineers add special fenders, it would add only two BOMs to the file
(regular fenders, part of the basic tractor, would become an optional choice and would
have to have a BOM of their own) instead of doubling it.
At this point in the discussion, the reader may be wondering how this type of prob-
lem is being handled in a real-life situation if the manufacturer does not have BOMs set
up in modular fashion. The chances are that there would be several BOM for some of the
6,912 possible configurations covering the (arbitrarily established) models that are being
recognized. These BOMs also would be used for all other configurations by adding and
subtracting optional components.
The add-and-delete technique solves some but not all of the problems. Its principal dis-
advantages are vulnerability to human error, the slowing down of order entry, and awk-
2
wardness in establishing proper historical data for option forecasting purposes. Under

1 In this case, the standard deviation is based on the proportion of the population’s individuals who possess one of
two qualities—male or female, Democrat or Republican, stick shift or automatic transmission.
2 D. Garwood, “Stop: Before You Use the Bill Processor,” Production & Inventory Management 11(2), 1970.