 |
 |
 |
 |
|
 |
 |
|
 |
precision in demand forecasting becomes certain insofar as large scale orders are made
through buyer consolidation, and forecasting for non-stock items is obviated with the
presence of a CMP broker.
With regard to global supply chain management, and particularly focusing on
the value of postponement, we are interested in evaluating the effects of a CMP
broker along several dimensions: 1) inventory costs in early and late differentiation,
2) lead times in early and late differentiation, and 3) the overall value of postpone-
ment. We are also interested in evaluating whether early or delayed differentiation
is the best strategy to use when a global supply chain is enabled by a CMP broker.
Within this realm of inquiry, we ask three questions. First, we consider the role
of demand variability—the degree to which the quantity ordered varies over time.
Anand and Mendelson (2000) showed that the value of postponement increased
in demand variability and information precision. Further, because a CMP-broker
facilitates greater information precision in demand forecasting, we believe it is likely
that an CMP-broker enabled supply chain will increase the value of postponement
over a non CMP-broker supply chain. Thus, in our first hypothesis, we state:
•
H1: A CMP-broker enabled supply chain will, in the face of increasing demand
variability, produce a greater value of postponement than a non-brokered supply
chain.
Second, we consider the role of demand correlation—the degree to which
the order quantities of different products are similar, or the same. Anand and
Mendelson (2000) showed that the value of postponement falls with demand
correlation, but that it increases with information precision. Thus, we believe it is
likely that a CMP-broker enabled supply chain will have a greater value of
postponement as demand correlation decreases than a non-brokered supply chain.
This is because of the additive effect anticipated by a CMP-broker’s greater
information precision regarding demand forecasting. Thus, our second hypothesis
is as follows:
•
H2: A CMP-broker enabled supply chain will, in the face of decreasing demand
correlation, produce a greater value of postponement than a non-brokered supply
chain.
Third, we consider the role of demand pooling—the degree to which a large
number of small orders becomes a small number of large orders. Lee(1996) and