Page 347 -

330 Chapter 9

Box 9.5
An example: Ford

Ford and Firestone suffered the death of a thousand cuts, in part because of a catastrophic
failure to share knowledge. Information that might have alerted the companies to the
calamitous mismatch of Ford Explorers and Firestone tires was scattered in different places
in both companies, each item innocuous in isolation. Yet Ford ’ s knowledge-sharing
scheme is one of the best in the world. The company ’ s Best Practices Replication Process
has produced a billion-dollar benefi t for the automaker. Why did it not help in this case?
The Ford process was started in 1995 when a VP of manufacturing on a trip to Europe
saw that the plant there had ideas Americans could use and vice versa. Back home, he
assembled his operations people and asked them to fi gure out a way to share best practices.
At the same time, another Ford group was addressing reengineering issues through the
Rapid Actions for Process Improvement Deployment (RAPID) program. These were work-
shops aimed to eradicate small ineffi ciencies. They soon turned to the challenge of repli-
cating the solution so the RAPID need not be reinvented again. The two merged to become
Ford ’ s Best Practices Replication Process. In 4.5 years, more than 2,800 proven superior
practices have been shared across Ford ’ s manufacturing operations. The documented value
of this shared knowledge so far is $850 million. Another $400 million stands to be won
from work in progress, bringing the grand total to $1.25 billion. Royal Dutch/Shell and
Nabisco have licensed the process and portions have been patented.
Ford made three key decisions: fi rst, the process would be managed with distinct roles
and responsibilities. Second, no practice would get into the system unless proven. Third,
every improvement would be described in the language of the work group involved: time,
head count, gallons, and quality. These work groups are communities of practice. Each
CoP has a company-wide administrator, picked by the director of manufacturing. The role
takes a half a day a week. At the plant level, each CoP chooses someone as the focal point
and that role takes one to two hours a week. No one is paid extra. The best practices process
has forty-two steps. The focal point looks for a neat new process (or its inventors go to
him or her). He or she makes up a web page that prompts him or her to quantify benefi ts
such as time or material saved. The focal point then e-mails it to the community admin-
istrator, who compares it with other plants, and if it passes muster, designates it as a gem.
It is then immediately posted on the intranet and e-mailed to every focal point in the
community. One way or another, each focal point must report a decision: to adopt or
adapt it, and say when; to investigate it; or to reject it and explain why. The web displays
a scorecard to all users — by community and by plant. It may show, for example, that of
sixty-one gems in painting, the St. Louis plant has done or agreed to forty-two, was inves-
tigating two, had rejected seven as inapplicable and nine as economically not feasible, and
had originated and contributed two.