BY JAMES COOKE, EDITOR AT LARGE
techwatch
A network design is never done
IT USED TO BE THAT BUSINESSES TURNED TO SUPPLY
chain network modeling for help answering specific questions.
For instance, if a company wanted to determine how many distribution centers to retain after a corporate merger or whether
it should open a new facility to support expansion into a foreign
market, it would use network analysis software to weigh the
pros and cons of various supply chain network configurations.
This type of software is available from such vendors as
Barloworld, IBM, Infor, JDA, LLamasoft, Profit Point, and
Solvoyo, to name a few.
But today’s business climate has become increasingly volatile,
and companies are struggling to contain distribution costs and
meet new customer demands. That’s why
more and more businesses are starting to rec-
ognize that their distribution network
requires periodic modeling. In fact, a recent
study of 60 companies conducted by the
Tompkins Supply Chain Consortium found
that the average length of time between net-
work design studies has dropped from 24 to
18 months. “Designing a supply chain is no
longer something you just do once,” says
Toby Brzoznowski, an executive vice presi-
dent at LLamasoft. “We’re seeing design as
being a process as opposed to a project.”
How can companies benefit from regular
distribution network modeling? For one
thing, periodic evaluations can help a busi-
ness assess whether its network design remains optimal despite
changing inventory or customer requirements. For example, if
customer demand is starting to shift, say to another region of
the country or another part of the world, a network analysis can
help the organization evaluate whether its distribution centers
are still situated so as to allow it to serve customers at the low-
est transportation cost. Or if customers are increasingly order-
ing merchandise online rather than buying from stores, the
software can help a company decide whether it should be
adding DCs to fulfill those online orders.
Along with changing customer demands, a company may also
be dealing with changes to its supplier mix. For example, perhaps the organization has recently switched to a new supplier or
expanded its supplier base in a bid to cut procurement costs.
While that may hold down sourcing
expenses, it can have an adverse effect on
landed costs. A network analysis might
identify ways to reconfigure the distribution network to keep inbound transportation costs low.
Volatile oil prices remain a nagging
worry for logistics managers, prompting a
number of them to get serious about contingency planning. Conducting a distribution network modeling exercise can help a
company prepare a plan to modify the distribution network—
perhaps moving DCs closer
to customers—to keep
down transportation
costs in the event of a
spike in fuel costs.
It’s not just oil prices
that are forcing a network
re-examination. To curb
overall costs, many companies are seeking closer
integration between
manufacturing and distribution, which may
require consolidation or
relocation of facilities.
Alan Kosansky, president of Profit Point
Inc., notes that many manufacturers are
asking the question, “Given our manufac-
turing footprint, what’s the best way for us
to deliver our products to our customers?”
Because distribution networks are based
on assumptions about current customers,
markets, inventory needs, and distribution
costs, volatility often changes the game,
forcing companies to reassess their opera-
tions. In a world where assumptions can no
longer be taken for granted, a distribution
network for today might look very different
from one for tomorrow. ;
