Page 235 - Orlicky's Material Requirements Planning

214 PART 3 Managing with the MRP System

the stick-shift choice may have some common parts with another subassembly used only
with automatic transmissions. Requirements for such common items will be overstated
because they are included in the safety stock of both options. In order to segregate such
parts, the BOMs would have to be modularized further, that is, torn apart. In some cases,
it might be desirable to do this, but if this technique is carried to its extreme, the planning
BOM might end up containing single parts only and no subassemblies at any level. The
ultimate module of the product is, of course, the individual component part.
BOM modularization may be a complex task if the product itself is highly complex, if
it is engineered on the “integrated design” principle (nonmodular design), and if it entails a
proliferation of optional features. Judgment must be used in deciding what should be mod-
ularized and how far. Particularly in attempting to segregate common or semi-common (an
item used with diesel and gasoline choices but not with LP gas) parts, the approach should
be conservative; that is, excessive modularization should be avoided.

Effects of Modularization
Modularization affects the timing of subassembly completion. A subassembly promoted
to end-item status takes the place of its former parent in the MPS, which means that it
will be finished later than had its lead time been offset from the timing of the parent. Each
time an item is broken down one level, its lead time somehow must be accounted for.
While subassembly lead times are not significant in most cases, end items in the modu-
lar (planning) BOM should, strictly speaking, be advanced in the MPS by the amount of
their lead time if they have originally been on level 2 or lower. This would complicate
master scheduling, and in practice, it is not done when the lead times of the subassem-
blies in question are very short, as they usually are.
The crucial question in modularizing the BOM is how far downward in the product
structure to go. What is really being done is to determine the correct level in the (origi-
nal) BOM at which to forecast, master schedule, and plan material requirements.
Whether a given subassembly should be forecast and planned or just its lower-level com-
ponents is the fundamental question, and it depends on when it needs to be assembled.
There are two alternatives. One is to assemble the subassembly as a function of exe-
cuting the MPS via the MRP system. This means assembling to stock or preassembling
before the end product itself is scheduled to be built via the final assembly schedule fol-
lowing the receipt of a customer (or warehouse) order. The other alternative is to defer
making the subassembly until such time as the end product itself is scheduled to be built.
Making the subassembly then becomes a function of executing the final assembly sched-
ule. The choice between these two alternatives should be dictated by the nature of the
product in question, as well as by the nature of the business. Lead times and the eco-
nomics of subassembly operations (Is it feasible to make the subassemblies one at a
time?) will determine, in each case, whether the item should be preassembled or it can
wait until final product assembly.