DEMAND PLANNING:

OPTIMIZING OPERATIONS ACROSS THE SUPPLY CHAIN

DEMAND PLANNING TECHNOLOGIES REDUCE INVENTORIES, SPUR COLLABORATION

John Brandt

Compliments of

DEMAND PL ANNING

Demand planning software packages can help

manufacturers to:
Establish baseline sales forecasts, incorporating
multiple inputs;
Perform sophisticated analysis that improves their
use of human and capital resources;
Optimize pricing capabilities; and
Better understand their markets and customers.
Yet to get the most out of these function-rich tools,
manufacturers must be willing to reorganize their
planning processesand to share the resulting
insights with supply-chain partners and customers.

PREDICTING THE FUTURE

How does BMW know how many Mini Coopers it
will sell in North America? How many Priuses should
Toyota build to meet demand in the U.S. this year?
Worldwide? When is it time to tweak production
upward or downwardto reflect a change in
the market?
The short answer to all these questions is: No one
knows for sure. The long answer is: Everyone needs
and wants to know.
Thats why demand planning exists.
Demand planning hasnt gotten the attention that
enterprise resource planning systems (ERP) or customer
relationship management (CRM) have received in
recent years. But thats not to say it isnt a sore spot
for many in the manufacturing and distribution
communities. When I speak in front of industry
groups and ask how many people are happy with their
sales forecasts, I dont see any hands at all, says John
Pavain, president of Norwalk, Connecticutbased MaxQ
Technology, a developer of demand management
systems and a Microsoft partner. Demand planning
is an area that most companies need help in.

POOR FORECASTING = POOR PROFITS

Whether its for consumer packaged goods, industrial
manufacturing, high tech and electronics, or automotive
demand, sales forecasts often lack the sophistication
needed to optimize both operations and profits.
Weve been pretty wrong on forecastingin both

directionsfor our suppliers and for our customers,

says the CEO of a multibillion-dollar high tech
manufacturer. In fact, weve been wrong pretty
massively. With sales forecasting, its sort of, why
bother? The people who do the forecast dont know
because the customers dont know.
Manufacturers and distributors routinely accept the
idea that sales forecasts will err by as much as 10%, and
often 20% or more. In some cases, companies have not
only accepted the fact that forecasts are soft, theyve
begun using software tools to enable them to track the
level of variance over time, adjusting their materials
procurement, operations, and distribution plans
accordingly. We track our forecast accuracy, so that
we can see a forecast wandering out of the norm, says
the director of production planning and forecasting
at a large packaged foods manufacturer. We want
to know exactly the number of weeks we are above
or below a certain accuracy level in our forecast.
That way, we build up a consistent trend of error.
Granted, this food manufacturer has a special business
need to minimize the amount of product lost due to
expiration at different warehouses. Adds this executive,
The software we are using gives us the ability to look
at the freshness level and inventory for each product
at different warehouse locations.
His company is able to adjust product levels at its
warehouses as needed, while avoiding losses due to
product expiration. This feedback capability allows the
business to tweak both production and inventories
for maximum profitability.
These problems are similar to those experienced in
other industries, including high tech and electronics,
where chips, subassemblies, and even finished products
such as computers have limited effective shelf lives.
Here, too, inventories must be managed in such a way
as to maximize sales and minimize spoilage as
products become obsolete or unwanted.
In industries such as steel, excess finished-goods
inventory ends up as expensive scrap later melted
down to fill new orders. Meanwhile, steel producers
lose production time and waste labor producing
and storing the unwanted inventory, some of which
was rejected because its production was too late for
customer demand. Indeed, an entire industrythe
steel service centerhas arisen to cater to the needs
of steel customers who couldnt wait for their orders to

DEMAND PL ANNING

be coated, slit, shipped, and delivered. The automotive

industry faces an even more complex problem. General
Motors, for instance, began trimming production this
year at certain plants in late summer and early fall to
bring inventories more closely in line with slowing sales
figures. Yet even as dealers around the country found
themselves burdened with unusually high inventories
on their lots, GM kept shipping vehicles because
the company books revenue and profit at the time it
builds and ships vehicles, not when they are purchased
by consumers.
This system has predictable and painful results: As of
July, the average time it took to sell a new GM vehicle
was up to a record 91 days, forcing the company to
offer sales incentives of up to $4,500 in rebates per
vehicle. If, however, an effective demand planning
process had been in place, timely analysis of key
factorsincluding sharply rising gasoline prices and
negative consumer attitudes toward the safety and
environmental friendliness of SUVswould have
prompted GM to scale back production of Hummers
and other large SUVs, whose sales were hit the hardest
in the first half of the year.
Now, however, with the 2005 models ready to hit
showroom floors, inventory-swamped dealers
were understandably reluctant to accept additional
inventories. In turn, the reduction in dealer orders
finally curtailed production. Yet existing inventories
remain high, costly to both carry and unload at
discounted prices. Says one securities analyst
regarding the situations impact on GMs earnings:
Its the companys choice whether they pay it
back now or pay it back later.
Poor demand management is costly for manufacturers
and distributors in other ways as well. One large maker
of data storage products that distributes via consumer
electronics retailers found it was constantly playing
catch-up with its dealers, routinely missing shipments
and order dates. Although we were spending more
than a week each month planning and forecasting,
we were still facing fines by our retailers for missing
shipment dates, reports the director of sales planning.
To help solve the problem, the company purchased
a demand forecasting software package. The software
takes into account each products demand pattern,
while allowing decision makers to adjust forecasts
and inventories to meet the companys goals.

Other downsides to inadequate demand planning

include poor customer satisfaction, out-of-stock items,
and high inventoriesall of which drain the bottom
line. Any time a customer finds an item out of stock
and goes to another store and buys a competitors
product, both the manufacturer and retailer have
lost sales.
Even with demand planning, forecasting remains
difficult; variables ranging from promotions to store
openings to retail channel shifts must be included in
estimating demand. When Dell mails out catalogs
and sends out e-mail blasts, says MaxQs Pavain, they
must use past history as a guide to plan how much of a
sales increase is expected.
Still more difficult to assess is the demand for a new
product. You can do market research and look at
similar products, but for products that are instant hits
or fads, its very much like trying to predict the future,
says Pavain. Who knew Apples iPod was going to
take off, whereas the Newton [an early handheld PDA]
was a dud.

DEMAND PLANNING TECHNOLOGY

Where just five years ago only large enterprise clients
could afford sophisticated demand planning systems,
new software packages are well within the budgets
of most companies. Yet despite the availability of new
technology, many small and midsize manufacturers
and distributors continue to forecast using simple
spreadsheets. Says one technology analyst, [Microsoft]
Excel is the most widely deployed demand planning tool.
Unfortunately, spreadsheets typically lack sophisticated
forecasting algorithms, arent built to track multiple
inputs, and fall short in performing sophisticated
analysis for various user groups within a business.
Companies interested in more effective management
of their supply chains have found that function-rich
demand planning software can be a powerful
tool to help managers take guesswork out of the
planning process.
Why? Because demand planning software recognizes a
wide array of variables that influence sales projections.
For example, while sales representatives typically view
demand in a particular region based on the total

DEMAND PL ANNING

revenue it produced, operations planners in the same

firm view demand by quantities of each product sold.
Demand planning software accounts for both variables.
Demand planning systems typically produce a baseline
forecast based on shipment or order history. Once
established, users can change the baseline to account
for other data, including promotions, advertising
inserts, seasonality, new sales outlets, and other factors.
At the same time, demand planning software factors
in potential constraints such as materials availability
and plant capacity. You can set up the system to switch
from a make to a buy approach for a certain
part or component any time insufficient in-house
capacity exists.
These systems typically yield four or five data streams,
with varying forecasts. At this point, management
plays a role in the planning process, overlaying
broader strategic goals, such as reducing inventory
and improving customer service. In addition, these
systems usually offer notification functions, providing
planners with automatic updates when certain events
occur, such as a change of more than 30% from the
previous months forecast. Most demand planning
packages also offer simulation capabilities to compare
multiple forecasts and their impacts on both revenue
and profitability.
Automotive manufacturer and distributor Hyundai
Motor America uses point-of-sale data to build a
demand forecast for its service parts. The company
uses a demand management system to digest more
than three years of parts sales data to calculate
demand forecasts by part number for four warehouses,
leading to an order plan for some 50,000 parts. The
order is then sent to Korea, where the company is
headquartered and where many of its parts suppliers
are located. As a result, the company has reduced
inventory and boosted its fill rate.

online. Why? Two reasons: First, many companies are

reluctant to share data, and second, even when willing,
few firms are able to commit the managerial and
technological resources necessary to make this kind of
collaborative planning work. Says one manufacturer,
Youre relying on your customer to have the same
expertise that you do, and thats not always the case.
Another obstacle to collaboration is technological
integration (or its lack) among various players in the
supply chain. Sooner or later, companies large and
smallmanufacturers, distributors, and retailers alike
will connect using true application-to-application
integration technologies such as XML and Microsofts
.NET initiative. In the meantime, companies looking
to collaborate regarding supply-chain needs and
demand forecasts generally do so on a customized,
one-on-one basis.
One technology that is proving a boon to demand
planning is the Web portal. Many progressive
manufacturers and distributors are using portals to
link their dealer and retailer networks for both product
ordering and sales forecasting. For example, Fleetwood
Enterprises, a leading maker of recreational vehicles
and manufactured housing, is using a Web portal to
upload dealer sales data to improve forecasting.
For Fleetwood and others, the payoffs derived from a
more effective demand planning system ripple through
the supply chain. But for the individual manufacturer
or distributor, they tend to center around key
financial benefits:
Reduction of inventories,
Reduced warehouse space, and
Improved customer satisfaction, resulting in
increased sales.
For instance, by simply using a demand planning
package, one manufacturer of tire products was able to
cut supply orders by half and reduce peak inventories
by 25% in the first year.

SUPPLY-CHAIN COLLABORATION
Manufacturers and distributors seeking to optimize
their forecasts try to include their supply-chain
partners in the planning process. Yet according to one
estimate, only 10% of manufacturers and distributors
are using collaborative supply-chain planning systems.
The reality is that while many suppliers work with key
customers to build forecasts, few perform this exercise

These benefits can be so significant, in fact, that some

manufacturers and distributors are going the extra mile
for customers, providing vendor-managed inventories.
You do the replenishment for the customers and
manage their inventory of your products for them,
explains MaxQs Pavain. Distributors offer the same
service when they stock retailers shelves. Customers
appreciate the extra attention, while manufacturers
and distributors get direct demand signals from the

DEMAND PL ANNING

marketremoving much of the uncertainty from the

planning process.
One company using vendor-managed inventory is GM
Brazil. Using customer demand data provided through
its dealer network, the companys vendor -managed
inventory system drives sales forecasts for more than
20,000 parts. With the new system, initial forecast
accuracy was more than 90%.

efficient, and more responsive supply chainsand

happier customers. Collaborative forecasting gives
the supplier more information, says Pavain, and
the more information [a supplier] has, the better
the supplier can serve [customers]. The benefits are
significant reductions in inventory and warehouse
space, an improved ability to supply the customer,
and increased sales.

SUMMARY
THE PAYOFF
Effective demand planning can provide significant
benefits, including:
More profitable linking of pricing, promotions,
and other sales incentives;
Better informed supply-chain decisions as a
result of more accurate data regarding customer
demand;
More efficient and predictable use of plant
capacity, equipment, and labor; and
Better understanding of customer needs so that
manufacturers and distributors can improve
products, deliveries, and satisfaction.

Manufacturers in diverse industries have dramatically

improved their forecasting accuracy with demand
planning software packages. But to turn their crystal
balls into improved bottom lines, organizations must:
Develop the products and services based on
customer demand, not guesswork;
Establish pricing based on dynamic
market feedback;
Manage effectively to forecasts, thereby
optimizing the performance of facilities, people,
inventories, and suppliers; and
Extend their forecasting capabilities across the
entire supply chain with collaborative planning.

Manufacturers and distributors who have invested in

new demand planning systems have benefitted not
only through improvements in internal operations,
but also from the greater collaboration fostered by
these technologies. The end results are tighter, more

Microsoft Dynamics is a line of integrated, adaptable business management solutions that enables you and your
people to make business decisions with greater confidence. Microsoft Dynamics works like familiar Microsoft software
such as Microsoft Office, which means less of a learning curve for your people, so they can get up and running quickly
and focus on whats most important. And because it is from Microsoft, it easily works with the systems your company
already has implemented. By automating and streamlining financial, customer relationship and supply chain processes,
Microsoft Dynamics brings together people, processes and technologies, increasing the productivity and effectiveness
of your business, and helping you drive business success.
Worldwide (1) (701) 281-6500
U.S./Canada Toll Free (1) (888) 477-7989
www.microsoft.com/dynamics
2004 Microsoft Corporation. All rights reserved. Microsoft and Windows are either registered trademarks or trademarks of Microsoft Corporation in the United States
and/or other countries. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.