GE Appliances: Reshoring Manufacturing
UV8141
Rev. Feb. 17, 2021
GE Appliances: Reshoring Manufacturing
We Can’t Make the Refrigerators
Chip Blankenship, CEO of GE Appliances, came right out and said it to Jeff Immelt, chairman and CEO
of the General Electric Company (GE), and Keith Sherin, CFO and Vice Chairman: “We have to delay the
launch of the French-door bottom-freezer (FDBF) refrigerator until after Red, White, and Blue.”1
The company had invested $300 million in renovating Appliance Park Building 5 (AP5), a one-million-
square-foot factory built in the 1950s, reshoring manufacturing, and hiring thousands of people in Louisville,
Kentucky. The grand opening of the factory had been two months earlier, with Kentucky’s governor,
Louisville’s mayor, and Immelt in attendance. It was a beautiful factory. The problem was that two months
later, the team at the factory could not make the products.
AP5 was in a combination of stagnation and chaos. The team was unable to thermoform2 the ABS plastic
liners for the inside of the refrigerator cases, and the automatic case-fa
ication line itself was temperamental—
it was producing parts with too much dimensional variation for smooth assembly operations. Each refrigerator
factory had a mini chemical plant inside it to produce foam for cases and doors. Foam served as insulation as
well as structure, so getting the exact amount delivered with no leaks and a successful cure were key
characteristics of success. The AP5 wet system was not delivering foam to the cases with the right chemistry,
and all the liner penetrations for LED lights and many other features were allowing a myriad of foam leaks. To
make matters worse, the workforce was not yet competent on
azing, and the yield of good
aze3 joints in
the sealed system was very low. Foam and
aze yield problems were extra painful because many operations—
and thus significant value—had been expended right up to the moment that a unit had to be scrapped.
These upstream battles turned the final assembly line into a chaotic effort to put together a conforming
product and meet the demand of initial floor loads4 (more than 5,000 refrigerators) with the required mix of
1 This is a field-based case. All information and quotations, unless otherwise noted, derive from the author’s own experiences.
2 Thermoforming was the process of heating a plastic sheet to a pliable condition and stretching it over a mold (with vacuum assist), cooling the
shape, and trimming the flash. It was often used to make inner liners for refrigerator doors and cases.
3 Brazing was a metal-joining process that required a filler metal of a lower melting point than the base metal work pieces so that the base metal did
not melt in the process. It was similar to soldering, but required higher heat and, typically, a tighter fit for capillary flow of the
aze alloy. Brazing in
appliance manufacturing was a key operation to produce a sealed system for refrigerators, which cooled the freezer and fresh-food compartments. It
consisted of a compressor, condenser, evaporator, and expansion valve with associated copper tubing and cooling fans. The term
azing yield refe
ed
to the number of good units coming out of all the
azing operations divided by the total number of units that entered the process.
4 Floor loads were all the appliances necessary to put on floor display at all retail stores and in builder model-home displays. The process of shipping
these in bulk could be triggered when a new model was released to replace an existing floored appliance, or when a new floor spot was won.
This field-based case was prepared by Charles P. Blankenship Jr, Montgomery Distinguished Professor of Practice, University of Virginia Department
of Materials Science and Engineering. It was written as a basis for class discussion rather than to illustrate effective or ineffective handling of an
administrative situation. The author gratefully acknowledges the assistance of Deborah Wexler, Daniel Rowley, and Michael McDermott in preparing
this case. Copyright 2020 by the University of Virginia Darden School Foundation, Charlottesville, VA. All rights reserved. To order copies, send an email
to XXXXXXXXXX. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by
any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of the Darden School Foundation. Our goal is to publish materials of the
highest quality, so please submit any e
ata to XXXXXXXXXX. Do
N
ot
C
op
y
o
P
os
t
This document is authorized for educator review use only by VIK KORTIAN, Macquarie University until Jan 2023. Copying or posting is an infringement of copyright.
XXXXXXXXXX or XXXXXXXXXX
mailto: XXXXXXXXXX
mailto: XXXXXXXXXX
Page 2 UV8141
models. Right behind this initial salvo was the requirement by retailers to ensure sufficient inventory of the
promotional models to support Red, White, and Blue, the annual Fourth of July promotional extravaganza that
was rivaled only by Black Friday in the appliance industry.
It was clear that Immelt and Sherin were furious. Blankenship and Mark Krakowiak, GE Appliances CFO,
sat on the other end of the videoconference line waiting for their response.
Sherin said, “I knew we shouldn’t have invested in Louisville.”
Immelt said, “Come on, Chip. You used to make jet engines all day long. Are you telling me you can’t make
a refrigerator?”
Blankenship said, “Jeff, we can make a refrigerator. We just can’t make 550 per shift with the right fit, feel,
and finish to deliver to customers.”
In frustration, Immelt said, “Should we just give this product back to MABE?”5
Sherin chimed in and said, “We probably should just get the Mexican or Korean sites back online and stop
further investment in Louisville.”
GE Infrastructure: The Future of GE
Although the Appliances and Lighting business divisions were GE’s most recognizable activities to the
average US citizen in the first decade of the 2000s, they were not key drivers of revenue or operating profit for
the company. GE was investing into growth platforms such as aircraft engines, power generation, medical-
imaging systems, and oil and gas equipment.
All of these growth platforms were in long-cycle industries with high intellectual-property moats6 yielding
significant, profitable service annuities. The Appliance and Lighting businesses were categorized as cash
generators that were used to fund investment activities in other businesses and were thus starved for capital as
a matter of policy.
As a direct result of this corporate landscape, GE Appliances and GE Lighting were left to manage their
own product and business portfolio growth with very limited capital for R&D, factory modernization, or even
equipment maintenance. (See Exhibit 1 for a GE Appliances timeline.) GE Appliances embarked on a design,
source, and sell strategy with a joint venture (JV) in Mexico, MABE, for gas ranges, dryers, and side-by-side
and FDBF refrigerators. It enlisted Korean companies LG and Samsung for microwaves, higher-feature
efrigerators, and high-capacity top-load washing machines. It also formed a JV (called Little Swan) with a
Chinese washing-machine company to produce front-load washers.
Product development required very little capital and just enough engineering resources and industrial design
effort to specify the visual
and language, a functional specification for performance of the appliance and, in
some cases, the control systems. Program management was required to sequence the development process with
test and validation, including field tests, prior to commercial product launch. The financial return on investment
(ROI) of this strategy was very good—in fact, too good.
5 MABE was GE Appliances’ joint venture (JV) in Mexico.
6 Intellectual property moats were used to describe how difficult it would be for competitors to duplicate a company’s capability. Do
N
ot
C
op
y
o
P
os
t
This document is authorized for educator review use only by VIK KORTIAN, Macquarie University until Jan 2023. Copying or posting is an infringement of copyright.
XXXXXXXXXX or XXXXXXXXXX
Page 3 UV8141
The Downside of a Phenomenal ROI
The downside of GE outsourcing the detailed product design and manufacturing of many new products
was twofold. First, of all the players in the US appliance market, GE Appliances owned the leading distribution
and service network—which made for excellent engagement with builders, retailers, and consumers. But since
GE did not directly oversee the manufacturing process, there was a gap in an otherwise tight and timely
feedback loop. The company was losing its ability to incorporate customer and consumer feedback in a
continuous-improvement loop, even though GE repairpeople were in consumer homes addressing problems
one-by-one. Any and all improvements were outsourced, translated, and managed by others.
Second, although initially GE had been pleased with on-site representatives from the sourced appliance
suppliers, it became apparent that they were doing more than just facilitating good communication with their
headquarters team. The suppliers were learning from and preparing to attack the US appliance market
themselves. GE had helped accelerate the launch of some able competitors with attractive
ands and capability.
Fix, Close, or Sell?
In 2008, GE started a sale process, but was not able to attract an offer high enough