Procurement Strategy

Carrier Corp.

Student John Perry prepared this case for class discussion rather than to illustrate either
effective or ineffective handling of an administrative situation. Data has been disguised.

It's December 22, 2000, the last day before the holidays. John Perry, a graduate student from
MIT, is about to wrap up his seven-month internship at Carrier Corp. in Syracuse, New York.
Mr. Perry had been assigned to help develop an e-business strategy for procurement of direct
material.
John has spent the past few months working with Ms. Kathy Reardon, project manager of global
supplier web-enablement initiatives at Carrier, and Mrs. Lisa Sadlik. After graduating from
Darden in 1998, Ms. Reardon has spent the past two years in Carrier's exclusive rotational
program. In her first post-rotational job, Kathy has accepted responsibility to become Project
manager for supplier web-enablement. She works as a one-person army both deploying and
supporting project installations. Kathy has gone home to Wisconsin for the holidays and will
conference in for the final wrap-up.
Kathy's supervisor is Lisa Sadlik. Mrs. Sadlik was appointed an officer in the company earlier in
the year as part of her new responsibilities leading global information integration. In the past
nine months, she has quickly assembled a team dedicated to improving operations on a global
scale at Carrier. Lisa spends the majority of her time on the road. In fact, she just returned from a
trip to France the night before.
When John arrives, Lisa is checking e-mail and is already on the phone with Kathy when another
call comes in. It's Dr. Larry Sweet, Vice-President of Operations at Carrier Corp. Dr. Sweet has
been studying budget proposals for next year's projects. He is calling to report that in light of the
slowing economy, Corporate is prioritizing its growing investments in all the various aspects of
e-business. Consequently, the supplier web-enablement initiative may be scaled back or even
canceled. Kathy and John need to follow up on this. What should the priority of e-procurement
be within Carrier? How does it fit with Carrier's overall business strategy? What is the right thing
to do?
Company Background
Carrier Corp. was founded by Willis Haviland Carrier, the father of air-conditioning. In 1915
Carrier and six friends scraped together $32,600 and formed the Carrier Engineering Company.
However, the company had its beginnings in 1902 when, as an engineer with the Buffalo Forge
Company, Carrier designed the first system to control temperature and humidity. His first
customer was a frustrated Brooklyn, N.Y. printer who couldn't print a decent color image
because changes in heat and humidity kept changing the paper's dimensions and misaligning the
colored inks. Initially Carrier's invention was intended for the comfort of machines or industrial
processes rather than people. But ultimately, the larger market has been for the convenience of
people.
Carrier, the company, began manufacturing products in 1922 after the company's namesake
developed one of the most significant achievements in the industry's history - the centrifugal
refrigeration machine. The centrifugal chiller was the first practical method of air-conditioning
large spaces. This achievement paved the way for tall buildings as well as made hospitals,
schools, office buildings, airports, hotels, and department stores more comfortable.
Later, Carrier developed smaller "unit air conditioners". This led to the development of a
residential "Weathermaker" that heated, cooled, humidified, cleaned, and circulated air in homes.
But the Great Depression quickly put an end to residential air-conditioning.
Today, Carrier manufactures more than just air conditioners. They build furnaces, blowers, room
air-conditioners, window units, central air, commercial units, large chillers, racks for grocery
stores, and truck-trailer units. This is industry is known by the acronym, HVAC/R (heating,
ventilating, air-conditioning and refrigeration). Carrier is the largest player in the HVAC/R
market with 1999 revenues topping $7.6 billion. The company is the second largest in the United
Technologies Corporation (UTC) portfolio, a Fortune 100 conglomerate.
Carrier's competitive advantage is the quality of its products. Its quality is a product of both
excellent engineering design and manufacturing. Willis Carrier was an engineer and insisted that
the company be engineering-driven. He was known to disdain non-engineers and was rumored to
have said to one of his business managers, "You are not an engineer. You will never understand
air-conditioning." In contrast, Carrier's largest competitor, Trane, has tried to establish its
competitive advantage as sales and marketing.
Company Organization
Carrier is a global company with almost 90 manufacturing sites. The company is structured
geographically and then by business unit. The global divisions are Latin America (LAO), Europe
(ETO), Asia/Pacific (APO), and North America.
The company is further divided along business units. Commercial Systems and Services (CSS)
sells HVAC systems for large office buildings. Residential and Light Commercial Systems
(RLCS) sells systems to homes and small office buildings. Commercial Refrigeration (CRO)
sells commercial systems to refrigerate food (the 'R' in HVAC/R).
Carrier also has business units that extend both up and down the supply chain. Upstream are the
Carlyle Compressor Division (CCD) and Carrier Electronics (CE) who provide manufactured
components to the main business units. Downstream are Replacement Components Division
(RCD) and Carrier Enterprises, which is Carrier's distribution arm. All told, there are over 200
P&Ls (Profit and Loss) at Carrier.
The global divisions have their own presidents as well as the North American divisions. Each
president operates separately and independently of the others, although all presidents report to
the president of Carrier. A new president of Carrier, Mr. Jon Ayers, was appointed in 1999. The
president describes Carrier as a "federation," stressing the notion that the business units are free
to act as their own "state" as long as they adhere to corporate-mandated strategies.
E-business Initiative
Under the leadership of Mr. Ayers, Carrier established a corporate e-business strategy. This
strategy consisted of eight points and was subsequently titled the “Big Eight.” Each point
addressed different channels of e-business such as business-to-consumer (B2C), business-to-
business (B2B), and business-to-employee (B2E) channels. For example, the B2C initiative
aimed to sell 75,000 window room air conditioners direct to consumers over the internet in the
year 2000 ("75K in Y2K").
One of the many B2B strategies addressed procurement. With the increased tendency to
outsource, procurement has become more and more critical to the operations of the company. For
instance, outsourcing was a contributing factor to why the number of employees working at the
Syracuse campus dropped from 10,000 people to 3,300 over the last ten years. To manage
outsourcing, several people at each plant became dedicated to procurement, or ensuring that raw
material was always in stock.
A new procurement strategy was needed because the company was changing its business
strategy and was moving in the direction of JIT production. Carrier was reducing its own lead
time from 14 days to five days for high volume products and wanted its suppliers to do the same.
It was also decided that a new strategy was needed because the procurement process was very
paper-intensive. Thus, Carrier was looking to reduce costs.
The e-business strategy for procurement was called SupplierLink. The goal of SupplierLink was
to web-enable 50% of the supplier spend by January 1, 2001, meaning that half of the dollars
spent on direct material would be released via the web instead of the FAX. By December 2000,
Carrier had already achieved its goal with some plants achieving as much as 94% on-line. The
goal was met by targeting the largest spend suppliers (see Exhibit 1).
The procurement process at Carrier consists of four steps: (1) forecast, (2) release, (3) supplier
ships, and (4) receive. All but step (3) are very paper-intensive.

Forecast Release Supplier Ships Receive

First, forecasts are faxed or e-mailed either manually or automatically to suppliers after
production has been scheduled. At most plants, scheduling is run once every week and,
consequently, forecasts are sent once every week.
Second, at the same time, order release requirements are generated by the production schedule.
However, not all suppliers ship off the production schedule’s planning system. If the part is
kanban, a crib attendant reorders the part when the crib hits the trigger point. Kanbans account
for 10% of supplier spend. Otherwise, the release is based on the material requirements planning
(MRP) system, which is an integral part of the scheduling system. In some instances, suppliers
release directly off the forecast, and not off a separate release. About 20% of supplier spend falls
into this category. In other cases, suppliers communicate how much of the forecast they will be
able to fill, which, hopefully, is the entire quantity. Carrier confirms the quantity with the
supplier before the supplier ships. Twenty-five percent of supplier spend falls into this category.
In the remaining—and most complex—case, Carrier faxes or e-mails an order release to the
supplier. Parts are released against a blanket purchase order. (Blanket purchase orders generally
specify either a quantity or dollar limit.)
Third, the supplier ships the part.
Last, Carrier receives the part. A shipment is “admitted” to Carrier property. If it is not
immediately received at the dock, the shipment is "dropped." Once at the dock, the receiving
clerk receives the part by matching the bill of lading, the packing slip, and the purchase order.
The receiving process is entirely manual and prone to errors.
SupplierLink
Carrier launched SupplierLink as a tool, free for its suppliers, to simultaneously reduce lead-time
and inventory. Many suppliers were excited by SupplierLink because they knew it would help
them meet their lead-time goals, which were often stipulated in contracts, while keeping their
inventories low.
Carrier's internal consulting group, PDS, defines total cycle time as the time between order and
delivery, beginning with a factory that is empty or where all work-in-process (WIP) has already
been sold. Lead time may or may not be equal to total cycle time. It depends on the business
strategy. Some suppliers strategy is build-to-order. A build-to-order strategy minimizes inventory
at the cost of increasing lead time. In true build-to-order factories,
Lead Time = Order Entry + Supply + Manufacturing + Warehouse + Deliver # build-to-order
Other suppliers strategy is build-to-stock. In this strategy, lead time is minimized at the cost of
increasing inventory. Mathematically,
Lead Time = Order Entry + Deliver # build-to-stock
E-procurement aims to reduce the Order Entry term. Some suppliers spend very little time
processing orders. In contrast, large suppliers who receive 100-200 pages of forecasts and
releases every week may be able to significantly reduce Order Entry term by downloading or
uploading data directly over the internet. Faxed orders are sent at a rate of six pages per minute
over long-distance phone lines ($0.05/min.). In the worst case scenario, it can take the supplier
one day to look over and digest the hundreds of pages it receives. It then can take the supplier
another day or two to enter the order into their computer system. Suppliers estimated that it costs
$30/hr to hire data entry clerks.
In addition to reducing lead time, SupplierLink aimed at reducing costs in purchasing.
SupplierLink was not the only initative with this goal. Other initiatives, such as E-source, aimed
at aggregating supplier spend across plants with the intent to aggregate contracts and purchases.
The ‘Drive for Five' initiative aimed to reduce purchasing costs by 5% by sourcing from low-
cost suppliers. These low-cost suppliers were usually found through internet auctions.
(FreeMarkets, an internet auction company, is partially owned by UTC.) There were many
lessons learned from these projects. Some projects were highly successful in terms of driving
real productivity. Some were not. The major differentiator was the readiness of the internal
Information Systems (IS) to provide linkage of data and processes to enable benefits from being
web-enabled.
The SupplierLink initiative faced its own IS barrier in that it planned to centralize data
management. Carrier hoped to be able to negotiate better contracts with its suppliers by
presenting one face to the supplier. Suppliers would benefit from a single point of contact as they
would be able to lower their cost of doing business by dealing with just one company, instead of
many plants.
However, each of the plants had their own MRP system. The plants’ independence is very much
a product of Carrier’s business strategy to grow through acquisition. Acquired companies are
offered the financial resources of the Carrier family, but, for the most part, the acquired
companies continue to operate in much the same manner as before. This includes their back-end
system. Consequently, the way in which plants interact with the same supplier differs from plant
to plant. For instance, some plants forecast 'gross' requirements and others forecast 'net'
requirements. Ms. Reardon, after spearheading several meetings to reach consensus on data and
data presentation, explained, "The biggest barrier to a successful implementation is the back-end
systems. The different plants at Carrier all run different MRP systems. However, depending on
the plant, the data in the MRP is not always trustworthy. At some plants, after MRP had been
run, the orders are manually changed. If nothing else, SupplierLink is helping the companies
realize where some of their systems need improving."
While there were certainly possibilities for cost reduction, SupplierLink required financial
outlays, all of which were funded at the Corporate level. Four database extracts had to be written
for each plant’s computer system. Since each plant had its own MRP system, no work could be
leveraged across the company. Each extract cost about $5000. Each supplier would have to write
their own database extracts.
Besides investing in software, Carrier invested in hardware. The company had not done much
business over the internet in the past and needed to upgrade its existing internet infrastructure.
For example, in June 2000, response time, the time from when a web page is requested to when
the web page appears in the browser, was several seconds. The supplier web-enablement team
had a goal of having two second response time so that users would not become frustrated with
the system. The upgrade included re-wiring some of the local area network (LAN). The upgrade
also included redesigning the entire Carrier wide area network (WAN). Carrier had a single
connection point to the internet backbone at their headquarters in Farmington, CT. If a plant’s
WAN line was down, that plant could not access the internet. Or, if the internet connection in
Farmington was down, the entire company could not access the internet. Consequently, each
plant added its own T1 line connection to the internet to the tune of $200/month. SupplierLink
was certainly not the only e-business initiative requiring additional bandwidth and they were
"billed" proportionately to assist with the upgrade.
Other hardware investments included a mission-critical, application, web server. In the web
environment, the server would essentially replace the FAX machine. Since Carrier's IT security
policy stated that servers requiring access from outside the Carrier domain must not be hosted
within the Carrier domain, SupplierLink was hosted at Genuity, a web hosting company. The
cost of hosting, $4000 per month, was divided among several e-business initiatives which shared
the server. It took over eights months for the server to be fully configured.
The final barrier to the success of SupplierLink was selling the solution to Carrier's own
employees. SupplierLink represented a process change. Material planners and buyers,
SupplierLink’s end users, needed to feel comfortable with the system and know how to use it for
SupplierLink to be successful. They were the ones responsible for keeping parts in stock. If a line
went down because there are no parts, the material planners would be held accountable. A line
that was down for half a day (four hours) waiting for parts to be rushed in might cost the
company more than $100,000. Ms. Reardon personally trained the buyers and planners at Carrier
and their counterparts at the suppliers. Adoption by the material planners had been mixed so far
—some were strongly in favor of SupplierLink while others were lukewarm to the application.
Likewise, adoption by the suppliers had been mixed. While many suppliers were enthusiastic
about SupplierLink, some suppliers had been reluctant to participate, preferring to use the FAX
machine as before.

The Situation
At the same time Carrier was investing in SupplierLink, e-business analysts were pointing to a
new model, the public electronic marketplace. E-marketplaces promised to connect the world on
a single platform. They promised to reduce the price of doing business but required that all e-
market participants follow the same procurement process. After reviewing its processes and its
supply base, Carrier opted not to join an e-marketplace but to further develop its private solution,
SupplierLink.
There were indications that the economy was slowing so all aspects of the business—supply
chain management, customer service, e-business, etc.—needed to be reviewed. Corporate had
now indicated that the SupplierLink initiative may be subject to scaling back as Carrier
prioritized its projects. This was the situation SupplierLink was facing as Mr. Perry left.
Exhibit 1: Profile of Suppliers

Supplier Dollar volume (in $M) Part volume (in 000s)

A 11.8% 13,000

B 8.4% 409

C 5.0% 1001

D 2.7% 12,566

E 2.2% 333

F 2.0% 8,552

G 2.0% 36

H 1.5% 131

I 1.4% 3,375

J 0.8% 10

K 0.8% 15

L 0.8% 73

M 0.8% 60

N 0.7% 35

O 0.7% 51

P 0.6% 58

Q 0.6% 72

R 0.6% 49

S 0.5% 29
Exhibit 2: Product Mix at Different Plants

Indianapolis – furnaces, fan coils

*
central air conditioning
Exhibit 3: Org Chart