Innovation at the speed of Information
-
Upload
iam-gautam -
Category
Documents
-
view
218 -
download
0
Transcript of Innovation at the speed of Information
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 1/11
TOOL K IT
nnovation
at the Speed of
nformation
by Steven D. Eppinger
Developing a new product involves trial and
error, but beyond a certain pointy redesignbecomes was teful. A practical and proven
tool, the Design Structure Matrix, can help
streamline the way a com pany innovates.
TH E E X C H A N G E OF I N F O R M A T I O N
is the lifeblood of product develop-
ment. When an electronics company's
circuit designers know what the casing
designers are doing, they design a better-
fitting circuit for the casing. And when
the casing designers know what the
circuit designers need, they design a
casing w here it's easier to put in a bette r
circuit. Such flows of information allow
for experimentation and innovation,
and for that reason, many companiesencourage feedback and iteration in
their product development processes.
This practice is known as concurrent
engineering.
But excessive itera tion can have draw-
backs. A continual back-and-forth of
work inevitably consumes time and re-
sources. And many of the iterations may
turn out to be only marginally benefi-
cial or even w asteful. For example, at a
telecommunications company that mycolleagues and I advised, there were as
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 2/11
TO OL K IT • Inno vation at the Speed of Inform ation
many as 20 regular iterations among
the teams working on tw o key technical
specifications. The result was that few
people worked carefully on specifica-
tions in the early stages, because every-
one knew that extensive rework would
occur down the line.
The lesson is clear. Iteration must be
carefully planned and managed. Good
iteration should be encouraged and bad
iteration eliminated. To do that, man-
agers need representational tools to
help them identify and model iteration.
change, the complete organization of
such a process with a conventional
project-management tool. In addition,
by their nature, high-tech product-
development processes usually involve
many interwoven tasks.
There is, however, a tool that man-
agers can use to obtain a simple and
meaningful representation of such com-
plex processes. This tool, the Design
Struc ture Matrix (DSM), differs funda-
mentally from conventional project-
mana gemen t tools in that it focuses on
The Design Structure M atrix differs from conventional project-management tools in that it focuses on representing informationflows rather than work flows. Thus, it is better able to depict the
key dynamic o f innovation processes.
Unfortunately, the standard project-
manag ement to ol kits such as Microsoft
Project don't contain such tools. The
most commonly used project-planning
tools-principally PERT and CPM net-
work diagrams-a re graphic descrip-
tions of task flows. In th em , tasks are usu-
ally repres ented by boxes or circles, and
sequencing is represented by arrows.
In a complex project, a chart can ru n
to tens or even hundreds of pages, andeach page accommodates only so many
readable circles and arrows. A boxes-
and-arrows depiction of the design pro-
cess for a car's suspension, for example,
would run to more than 30 pages. If th e
project is made up of tasks that can be
completed in sequence w ithout fear of
rework, manageable charts for small
chunks of work can be generated.
But the tools become very hard t o use
if what happens on page 60 forces you
to rework a task on page 18, and, thus,
many of the subsequent tasks down to
page 60 again. It is almost impossible
for managers to comprehend, let alone
Steven D. Eppinger is the General Motors
Leaders for Manufacturing Associate
Professor of Managem ent Science and
Engineering Systems at MIT's Sloan
School of Management in Cambridge,
Massachusetts. He is also codirector of
MIT's Center for Innovation in ProductDevelopment.
representing the information flows of
a project rather th an th e work flows.
As a result, it isbetter able to depict the
key dynamic of innovation processes
such as product development. What's
more, it can often provide representa-
tions of complex develop ment processes
on a single page. In this a rticle, we will
explain how the DSM works and show
how a manager can use it to make de-
velopment processes more efficient.First, though , let's explore why p roduct
development needs a fundamentally
different planning tool.
Product DevelopmentIs About Inform ation,No t Tasks
PERT charts, Gantt charts, and other
common scheduling tools were created
to help managers plan large construc-
tion projects such as btiilding ships or
factories. Although these projects canbe
complex, involving hundreds of differ-
ent tasks or mo re, the planning princi-
ples are fairly simp le: you decide w here
and when tasks should be carried out
No matter how complicated, all con-
struction projects can be thou ght of as
a sequence of discrete tasks, many of
which can be conducted simultaneously
but n one of which should need rework-
ing as a result of later information .
Imagine you are building a house.Some tasks have to be completed in
sequence. You can't frame the wall
until you've built the foundation. You
can't put on the roof until you've buil
the w alls. Sequential tasks, by definitio
rely on information generated by ear
lier tasks. Other tasks, called paralle
tasks, can be carried out sim ultaneousl
You can install w indows, run wires, and
lay plumbing at the same time. These
three jobs need walls, but none need
the other two. Neither sequential no
parallel tasks will need to be reworked
as a result of subsequent tasks. You don
change the foundation afte
building the walls. You do n'
rewire as a result of the win
dow glazing.
Product development i
very different. It requires innovation, and innovation re
quires complex learning (feedback
loops. You repeat prior tasks as you
learn from subsequent ones. Interde
pendent tasks that benefit each othe
in this way are known as coupled task
If you want to come up with a bette
found ation for future hou ses, you migh
rework a fotmdation after building the
walls. The infonnation from such an it
eration is precisely what helps you fin
the improvement, and the presence ocoupled tasks is what distinguishes in
novation processes such as product de
velopment from construction projects
Conventional tools answer the ques
tion, "What other tasks must be com
pleted before I begin this one?" But th
planners of a product development pro
cess need a too l tha t answ ers a very dif
ferent question: "What information do
I need from other tasks before 1 ca
complete this one?"This is the question
the Design Structure Matrix addresses
Drawing the DSM
Constructing a DSM of your company'
existing product-development proces
is a relatively straightforward, if some
times time-consu ming, process.The fir
step, identifying the tasks involved, i
easy and is often available as part of th
project-management documentation
Companies with an established devel
opm ent process already know th e taskneeded to develop a new product. Ford
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 3/11
Innovation at the Speed of Informa tion • T O O L K IT
for example, executes largely the same
process each tim e it develops a new car
engine.
What tak es time is correctly identify-
ing the infonnation needs ofthe various
tasks. You cannot rely on what your
company's managers tell you: they are
usually not the people doing the work,
and they may have an interest in justi-
fying existing or outdated processes.
When we draw a DSM for a product de-
velopment process, we go to the grass
roots and ask individual development
teams what they need from other teams
to do the ir jobs. It's important to focus
on input rather than o utput because we
have found that managers, engineers,
and other product-development profes-
sionals are more accurate in identifying
what they need to know than in de-
scribing what others need to know.
Once you have all this information,
you are ready to draw the project's
DSM. First, list all the tasks in the order
in which they are presently carried out.
Arrange them in the same order hori-
zontally and vertically to form a matrix
of rows and columns. Across each row
corresponding to a task, mark off the
other tasks tha t supply necessary infor-
mation. In other words, looking acrossa row shows you all the information in-
puts you need to complete a task, and
looking down a column shows you ail
the information outputs you'll provide
to other tasks. Consider the simplified
DSM shown below. Reading along row
A
B
CD
E
F
C
H
1
J
A
•
X
XX
XX
B
•
X
X
X
c
•
XXX
XX
D E
•
•
XX X
X
F
•
X
X
c
X
X
•
X
H 1
X
• X
•
j
X
XX
B tells you that task B needs informa-
tion from tasks A, G, and J. Reading
down column B tells you tha t task B sup-plies information for tasks E, H, and J.
Wh at the D SM Can Tell You
A DSM of your development process
provides a useful reality check. First, it
clearly reveals which infonnation ex-
changes involve design iteration and
which do not. In the example just pre-
sented, note the diagonal formed by th e
dots that divide our matrix. All the X's
below the diagonal denote feedforward
infonnation exchanges in which infor-
mation from earlier tasks is available
for later tasks. But an X in th e uppe r
half denotes feedback in which infor-
mation from a subsequent task may force
a reworking of a prior task. These are
th e coupled tasks. Task B, for instance,
needs information from task G, which
is carried out long after B. Executing B
requires making a guess about or as-
suming the missing information from
G. When complete and accurate infor-
matio n from task G is finally available,
a rew orking of task B may be necessary.
Then the development process has to
begin again from B onward, with inter-
vening tasks also being repeated to re-
flect the change to B's output.
A DSM can also help you to see how
well your development process is an-
ticipating the need for rework. Here'show. On the DSM, you simply draw
boxes around the tasks that your com-
pany performs concurrently, that is,
interdependently. These are your com-
pany's planned iterations, the tasks
your company recognizes as repeating
and therefore organizes so as to facili-
tate and speed the flow of infonnation
among them. If all the X's above the
diagonal are captured w ithin the boxes,
your organization has planned for these
iterations and has arranged its process
to accommodate them as efficiently as
possible. Our second example, below,
shows that tasks E through I are done
concurrently. However, the company
has failed to prepare for a fair number
of potential iterations: there are still
four feedback marks (now represented
by O's) above the diagonal and outside
ofthe boxed tasks. These are unplanned
iterations.
In practice , of course , successful prod-uct developers are good at recognizing
A
B
cDLLJ
FcH
1
J
A
•
X
XX
X
X
B
•
X
X
X
c
•XX
X
X
X
D
•
XX
X
LLJ
•
X
F
•
X
X
c
o
X
•
X
H 1
X
• X
•
J
o
oo
•X j reforma tion flows
• planned iterations
O unplanned iterations
potential iterations, and their DSMs will
usually reveal a fairly efficient flow of
information. But some companies findthat this exercise reveals muddled pro-
cesses. The telecommunications com-
pany whose development process for
data services is depicted in the chart
"Chaotic Development in Telecommu-
nications" is a case in point. The com-
pany's engineers had long been frus-
trated by the time and energy it took t o
deveiop new services, but it wasn't un til
they saw a DSM that they realized just
how many feedback loops were huilt
into their process.
OptimizingInformation Flows
The DSM is a powerful resource for or-
ganizations like the telecom company
because it helps managers not only
identify problems but also see how to fix
them. Below, we examine four ways to
improve a company's information flows.
Rearrange the sequence of tasks.
The first step in streamlining a product
development process is to determine
whether a different sequence of tasks
will reduce the number of feedback
marks. This involves rearranging the
rows oft he DSM, a process tha t Boeing
executives call "eliminating out-of-
sequence rework." The objective is to
move as many X's as possible from
above the diagonal to below it.
You begin by identifying candidates
for the earliest and the latest tasks. Ide-
ally, the first task would require no in-puts at all, indicating that it would never
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 4/11
Experiencing a Merger?
Acquiring a Company?Beina Acauired?
Ensure a smooth and lsuccessful transitionfor your workforce...
the future depends on it!
I i_Making Mergers Work/
is a powerful two-day interactive
semjnar offenng step-by-step
formulas for managing
large-scale c hange aftera merger or acquisition.
Making Mergers Work!
will arm you with proven
methodology and practical tools
to help you minimize difficulties,maximize results ond ach ievereturn on investment quickly.
Select from these ovoilobte dotes;
January 2Sth and 26th
February 8th and 9thFebruary 22nd and 23rd
CsJxUz of
Co~i.uL
sequentialtasks
paralleltasks
coupledtasks
coupledtasks
need t o be reworked. Refer-
ring to the two previous
DSMs, task A, therefore,
stays first. Task C comes
next because it nee ds infor-
mation only from A; these
are sequential tasks. Then,
we select tasks D and F to
come next because they
require information only
from A and C. Note tha t
D and F require no infor-
mation from each other, so
they can be carried out in
parallel, which we denote by a dashed-
line box in our third example.
Similarly, the ideal last task would not
produce any information required by
other tasks (in other words, its columnwould be blank). Here, that's task H,
which becomes the last task in the
project.
When you can no longer find any
tasks to schedu le early or late, you must
then group the remaining tasks into
blocks, bringing th e X's closer to the di-
agonal. In our exam pie, the most effec-
tive sequence shows two blocks of cou-
pled tasks: B, J, and G must be executed
together, as must tasks E and I. This
DSM plans for ail iterations.
A
cDLl
BJ
E
J_H
A
.
XX
XX
X
c
•XX
X
XX
D
'• "
X
XX
F
•
XX
B
•
X
X
X
J
X•X
X
c
X
•
XX
E
•
X
1 H
X•X •
For a simple DSM like the one here
it's fairly easy to identify by trial and
error the task ordering and couplin
that minimize the number of informa
tion feedbacks above the diagonal. Ithe case of more complicated DSMs
though, you will need to apply a sys
tematic approach involving the use o
computer-based algorithms.
Revisit task organization. Havin
rearranged the order of tasks, yo
should now reconsider their organiza
tion. Look again at the two blocks o
coupled tasks shown in the third DSM
above. In principle, the tasks within eac
set should be carried out at the sam
time and in the same place. But loo
Change Management Systems, INC.Call 'Ji at 305 iA? 9707 or visil
www.aboutcms.net
Chaotic Development in Telecommunication
Drawing a DSM ofthe development process or this company's
data services immediately revealed tbe degree of unplanned
iteration. Iterations w ithin the phases were built into the process
(the shaded boxes), but unplanned iterations across the phases
(the O's) were the reality.
planningphase
implementationphase
concept
business requirements
system requirements
network plan
techinical specifications
engineering design
billing implementation
operations engineering
customer service
launch
A
B
cD
ELl
G
H
1
J
A
•
X
X
X
X
H
•
X
X
X
X
c
•X
X
X
X
X
X
•
X
X
X
X
F
X
•
X
X
L
0
oo•
X
0
0
0
•
X
X
H
o
o
•X
X
1 J
0
X
X
•
X •
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 5/11
back to the second DSM, which shows
th e tasks that the hypothetical company
actually carries out concurrently. Task E
is grouped with 1, but tasks B, J, and G
do not fall within the initial group ing of
concurrent tasks; as a result, iterations
involving B , J, and G require many more
tasks in the original process. Clearly, this
company needs to rethink which tasks
are put together with which others.
Keeping interdependent tasks sepa-
rate can cause considerable waste, and
this is where grouping tasks differently
can really help to speed along the pro-
cess. One electronics company we ad-
vised found that it was performing a
set of tightly coupled tasks in two dif-
ferent coun tries. When th e teams were
briefly located together, they were ahleto complete the coupled tasks in just
two w eeks, thereby allowing other work
to proceed using their results.
Reduce information ex chang es. Re-
sequencing the matrix, though , will only
get you partway t o an efficient dev elop-
men t process. The next step is to reduce
the number of information exchanges
by changing the content of some ofthe
tasks. Because the importance and na-
ture of infonnation exchanges between
A DSM clearly reveals which
information exchanges involve
design iteration and which
do not It can also help you to
see how well your development
process is anticipating the
need for rework.
tasks can differ considerably, it is usu-
ally possible to break down coupledtasks into sm aller sets by chang ing task
specifications. Although this can mean
increasing the nu mb er of tasks and peo-
ple, the reduction in the number of
information flows-and thus in poten-
tial iterations - more than compensates
for these investments. The executives
at an aerospace firm we have worked
vtfith, for instance, believe that this ex-
ercise helped them to reduce the num-
ber of potential iterations in the com-
pany's development processes by asmuch as 50%.
I T P A Y S
T o B E T H O U G H T F U L
Join us at The Fuqua School of Business, wh ere th e th ou gh tfu l business leaders
of tomorro w are being taught today.
U.S. Te lephone: 919.660.7700
Europe Telep hon e: 4-49.69.972.6990
w w w . b u s i n e s s . d u k e . e d u
Thoughtful Business Leaders
Worldwide
D U K ETHE FIXJLJA
S C H O O LO F BUSINESS
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 6/11
T O O L KIT • Innovation at the Speed of In formation
Here are three ways to reduce the
need to exchange information. First, you
can transfer key knowiedge between
teams. In some cases, a company can
decouple one task from another simply
by adding to each team someone with
expertise in the other task. These people
must be sufficiently knowledgeable to
suppiy information that would other-
wise have been exchanged in one or
more iterations between teams.
A simiiar effect can be achieved by
using information technoiogy. Com-
puter-aided engineering software, for
exampie, allows design teams to predict
the implications of their designs on each
other, again obviating the need for a di-
rect exchange of infonnation. A piastic-
parts designer at a manufacturer of cellphones can use moid-flow simulation
software to foresee production prob-
lems arising from her design. In this
way, she can anticipate feedback from
the mold-tooling experts that would
otherwise force her to go back to the
drawing board weeks iater.
Second, you can introduce a new task
eariier in the process so as to simpiify
subsequent, time-Intensive iterations
performed by interdependent teams.
The new task typicaiiy requires eariy
Keeping interdependent tasks
separate can cause considerabie
waste, and this is where grouping
tasks djfferentiy can speed along
the deveiopment process.
agreement about aspects common to
the coupled tasks. The addition of a new
task, carried out by representatives ofthe coupled teams, can break down
some ofthe iterations. For example, two
teams designing coupled parts-say, an
electronic control circuit and its user-
interface keypad-can agree in advance
on the locations of the attachment
points and microswitches. This interface
specification aiiows the circuit-design
team and the keypad team to proceed
in parailei.
Third, you can redefine tasks within
coupled groups. Another way to reduceiterations is to eliminate an informa-
Improving Communication atGeneral Motors'Powertrain Division
W I T H I N THE PO W ER T R AI N D I V I S I O N
of General Moto rs, we used an inter-
esting variant ofthe DSM to im -
prove communication during en-
gine development. The company
had organized its 22 engine product-
development teams (PDTs) into four
system teams, each dealing wit h one
ofthe four engine subsystems: the
short block, the valve t ra in, the in-
ductio n system, and the emissions
and e lectrical system. We suspected,
however, that th is org anization did
not adequately accommodate the
communication needs of all PDTs.
To check ou t our hypothesis, we
surveyed each of th e 22 PDTs about
it5 communication with other
teams, asking each to describe how
often it needed to meet with other
teams. From these data, we drew a
DSM to describe those communica-
tions. Instead of marking each inter-
action w ith an X, we used three sym-
bols to reflect the varying frequency
of team interactions: a large circle
for daily communication, a midsize
circle forweekly m eetings, and a
small dot for monthly conferences.
The results are shown in the chart
"Organizing Communication atGM
PT: Before."The boxes show which
tasks were coupled in CM's originalstructure of system team s.
The DSM revealed that CM 's exist-
ing organization was indeed flawed.
In addition to the regular interac-
tions by the PDTs with in the four
system teams, PDTs also had exten-
sive and frequent contacts outside
their designated groups. The engine
block PDT,for example, which be-
longed to the sh ort block system
team, also had daily meetings with
me mb ersof all three teams in the
valve train system te am , one team in
the induction system team , and two
teams in the emissions and electri-
cal system team.These communica-
tions were entirely outside of any
form al process, and their orga niza-
tion was left to the people involved.
In many cases, the interactions sim-
ply didn't occur and, thus, informa-tio n was not transferred.
To see how GM could improve the
organization of its teams, we applied
a special kind of clustering analysis
to the DSM . First, we identified the
PDTs with c omm unication needs
across the e ntire developm ent orga-
nization and moved their rows to
the bottom of the matrix. We then
grouped the other teams into four
new system teams so as to capture
most of their daily and weekly
meeting needs. The results are
shown in the chart"Organizing
Communication atGM PT: After."
The new DSM demonstrated to
CM that it needed to introduce a
different type of organization. For
a start, while many PDTs could
be assigned to one system tea m
encompassing their principal inter-
actions, a certain amount of cross-
mem bership was required. SomePDTs-pistons, for example-were
assigned to tw o system teams. Two
PDTs, cylinder heads (G) and intake
manifold (J), were assigned to three
system teams. (For visual c larity in
the new m atrix, these teams appear
as Gl and G2 and Ji and J2, respec-
tively.) The boxed groups on the re-
configured m atrix denote the four
new overlapping system teams that
CM created.
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 7/11
Innovation at the Speed of Information • TOOL KIT
But the greatest chalienge facing
GM was how to organize the five
PDTs atthe bottom ofthe matrix,
whose systemwide c omm unication
needs could not be met within the
structure of a system team. The solu-
tion we worked out with GM was to
group these teams into a special
"system integration team " whose
respon sibility was to ensure theoverall integration of work by the
four system teams so that the engine
being developed would meet the
required performance standards.
The system integration team met
its communication needs by leading
monthly program meetings at which
everyone discussed issues related
to overall product performance.
Organizing Com mun ication at GM PT
Before...
short blocksystem team
valve train
system team
inductionsystem team
emissions andelectrical
system team
engine block Acrankshaft Bf lywheel cpistons Dconne cting rods E
„ . lubricat ion Fp cylinde r beads c] camshaft/valve train H
l_ water pump/cooling ippintake manifoldfuel systemaccessory driveair cleanerAIR
_ t b r o t t l e b o d yexhaustEGREVAPigni t ionECM
_ electrical systemengine assembly
j
A B C D E F C H I J K L M N O P Q R S T U V
A * • • • •• B • • • •
. , c• a • Q a a
• • • E •• a . a . F• . a
• • • a
• a a
a .
• a
• • • •
• a , .
• . a . •
• • a • a •
a - -
m 9 9
. • .
G • •• H •
• • 1• • •
•• • •
a .
• a
•
a
• • •
• • •
• a •
•
a
• a
• a • a .
a . • .
) • • • • •
• • L • • •• • AM • •
a * N *
• > • • • Oa a . . •
• • •
• • a
a • . » •
• • a a a
• •
a .
•
• • a
• • a •
a • • •
p • • • •• Q . • .
R • •• • S * *• • • • T *• • • • • Ua a . • a •
a
•
a
•
a
•
•
a
•
•
a
a
a
a
a
a
•
a
•
V
frequency of team interactions
• daily • weekly • monthly
These DSMs use symbols of varying size to
reflect the requency of team interactions .
The boxes show which product develop-
ment teams (PDTs) are grouped together
into system teams. In the "before"DSM,
we see that m any PDTs had frequent
contacts outside their designated groups.The reorganization first isolated
the PDTs with communication needs
across the whole organization and then
regrouped the remaining PDTs into new
system teams so that most daily and
weekly meeting needs would be met.
Th e "after"DSM shows the four new
overlapping teams as well as the system
integration team that communicated
with PDTs throughout the organization.
..,and After
team i
team 2
teams
team 4
integrationteam
crankshaftf lywheelconnecting rodspistonslubricat ionengine blockcamshaft/valve traincylinder headsintake m anifoldwater pump/coolingfuel systemair cleanerthrott le bodyEVAPcylinder beadsintake ma nifoldAIRexhaustECRaccessory driveigni t ionECMelectrical system
_ engine assembly
B
rEDF
AHr,1|1
1
KMO
RC?n
NP0L
TI IV
B C E D E A
B • • • • •
• c - •• E • • '
a • •
• • •
• F •
• • A
H G l Jl 1 K M 0 R G2 J2 N
•
a • •
• • • •
• . W H • • •
• • • •G I • •• • • ^ j ! •
a a • a • a 1
• -
• . .
• K • • • • • •
• M • • • •
a a • O • • • •
• • R
a
• * 4 *t a •
. . . a .
. .
G 2 » "
L
a
•
•
•
•
. a
• : •
a a
• •
• 12 * • '
• • NU •• • • p •• • • • Q
•
•
•
•
•
• • • • • • • • • • • • • • • • L
a . . . . . a « « a a . a a . a .
• a . a « a . . . a . . . . . . .
• a a * « « a » « a « • . 9 9 9 • * *
S 1
• •
••
•• •
• •
a
•
a a
a
a a
• •
• •
• T• •• a
U
a
a
•
•
•
•
•
•
•
•
*
•
•
V
•a
a
a
a
•
•
•
•
a
•
•
a
a
•
a
a
•
V
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 8/11
TO OL K IT • Innovation at the Speed of Inform ation
tion exchange within a group by add-
ing an extra task or two to intervene
between existing tasks in the group. To
see how this can be done, it's instruc-
tive to compare how two suppliers de-
sign the same dashboard component,
a speedometer cluster for General Mo-
tors. (See the chart "Reducing Informa-
tion Exchange by Decoupling Tasks.")
up to four steps of the process can be
affected, and much of the process may
have to be repeated (see O's in the left
side of the exhibit).
Supplier B takes a different approach ,
reducing the coupled tasks to just two:
casing design and lighting details. A
wiring circuit is worked out hased on
output from the first two tasks, and
A coupled process encourages iterations and th e
search for creative solu tions. But som etimes speed is
more imp or tant than innovat ion.
Supplier A adopts a concurrent engi-
neering process, carrying out th ree tasks
(casing design, wiring layo ut, and light-
ing details) at once, with three taskteams working in close proximity. The
three teams go through a number of
iterations and take a relatively long time
to finalize th eir plans, but the end result
is a fairly advanced prototy pe. However,
if the testing phase reveals problems,
the speedometer is then crudely proto-
typed. Once the prototype has been
tested, wiring revisions are made to
produce the final design. Although itinvolves an extra task-re visin g th e wir-
ing-in fact, this process is faster than
the first because the re is mu ch less iter-
ation between two task teams than
among three. The extra step also antici-
pates and allows for changes to th e pro-
Reducing Information Exchange
by Decoup ling Tasks
Supplier A has three teams that carry out their tasks concurrentfy. After a number
of iterations, a fairly advanced prototype is developed an d tested. But if problems
arise during the testing phase, up to four steps of the engineering process ca n be
affected (see O's).
Supplier B couples only two tasks and creates a wiring circuit based on
output from them. A soft prototype is built and tested, and an additional task
of wiring revision is introduced. However, this process is faster than Supplier A's
because less iteration occurs between the two initial task teams, and the extra
step of planned iteration virtually eliminates unplanned iteration.
Supplier AHighly innovative, but slow
Supplier BFast, bu t less innovative
A B C D E F A B C D F C
casing designwiring layoutlighting detailstoolinghard prototypetesting
• X X
X - X
X X •
X X X
ooo
oX •
X •
casing design Alighting details Bwiring layout csoft prototype Dtesting Erevision F
hard tooling c
• X
X •
X X
X X
X X
*
X •
X •
X •
X X X '
totype, and planned iterations virtually
eliminate unplanned ones.
In general, any decision to decoupl
a task (in this case, wiring) depends on
how the company views the trade-of
between speed and quality. A coupled
process encourages iterations and the
search for creative solutions and thu
is more likely to produce a significan
improvement in the quality of the prod
uct being developed. But sometime
speed is more important than innova
tion. Then, a faster, less coupled proces
is preferable.
Manage unplannable rework. So
far, most of our discussion has focused
on relatively small and easily managed
processes. In our theoretical example
for instance, it was possible to op timizea process so that all of the X's were
moved close to or below the diagonal
All coupled tasks could be carried ou
concurrently, so all iterations could be
planned for.
In real life, however, product devel
opm ent is a large and complicated pro
cess. It is rare that a company will be
able to design a process in which all cou
pled tasks can be carded out together
There will generally be tasks that can
only be conducted later in the procesbut that provide information for task
completed months earlier.
Consider Intel, one of the most con
sistently successful product developer
and a company that regularly leads the
way in developing the next generation o
microprocessors. The p roduct develop
men t process for semiconductor chips a
Intel consists of 60 distinct tasks, and
the DSM is shown in the chart "Semi
condu ctor D evelopment at Intel." Abov
the diagonal, the X's denote planned
information exchanges, and the O's sig
nify unplanned iterations.
As the boxed areas on the DSM re
veal, Intel groups most of the tasks into
15 concurrently m anaged stages. Som
of the stages overlap, that is, some o
the tasks in one grou p are also included
in the next. As you might expect, there
is little inc entive for a highly successfu
company like Intel to improve its tried
trusted process through rearranging andreorganizing its tasks. Equally, there i
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 9/11
Innovation at the Speed of Information • T O O L K I T
Semiconductor Development at Intel
This DSM describes Intel's proven and successful
process for developing and producing semiconductors.
This complex DSM clarifies to Intel where to concentrate
its efforts to improve the process. As the matrix reveals,
the process includes 1 5 planned stages, most of which
involve substantial iteration. There are also a significant
number of unplanned iterations (the O's), v^'hich are the
subject of Intel's process-improvement efforts. Three of the
iterations occur so tate that the company treats feedback
from the later tasks (the • 's ) as "generational learning"
feedback, to be used in the design and development of
subsequent products.
,10 , 2 0 , , 3 0 , , 4 0 ,
Set customer targetEstimate sales volumesEstablish pricing directionSchedule project time lineDevelopment methodsMacro targets/contraintsFinancial analysisDevelop program mapCreate initiaTQFD matrix
10 Set technical requirenient5
Write customer specificationHigh-level modelingWrite target specificationDevelop test planDevelop validation planBuild base prototypeFunctional modelingDevelop product modulesLayout integration
30 Integration niodelingRandom testingDevejop test parametersFinalize schematicsValidation simulationReliability modelingComplete product layoutContinuity verificationDesign rule checkDesign package
30 Generate masKSVerify masks in lab
Run wafersSort wafersCreate test programsDebug productsPackage pro ductsFunctionality testingSend samples to customersFeedback from customers
40 Verify sample functionalityApprove packaged productsEnvironmental validationComplete product validationDevelop tech. publicationsDevelop service coursesDeterrnine marketing nameLicensing strategyCreate dem onstrationConfirm quality goals
50 Life testingInfant mortality testingM f g . process stabilization
Develop field support planThermal testingConfirm process standardsConfirm package standardsFinal certificationVolume productionPrepare distribution network
60 Deliver product to customers
, 5 0 , ,60
20
0000 0 0
XX
X
X
x x x xX X XX
XX
X XX XX
X XX X X
X X X X X X XX X X X X X XX X XXX
X X
X X
X X
XXXX
x x x x x x x x
X - X XX• X XX
X X X X X ' X X.Xx x x F Jtx
X X X X X ' X Xxxxx
XXX XX
X X X
XXXX
XXX
XXX
0 0 0 0
0 0
XX X XX
• XX •
X
0
Xx -XXX" '
0 0
00 0
XX
X X
0000 0
000 0
000 0
XXX
x x x x xx x x x x
X information flows D planned stages O unplanned iterations generational learning
little to be gained from breaking down
the existing coupled g roups.
But a s the O's show, a significant num-
ber of potential unplanned iterationscan occur when errors are discovered
during th e developm ent process. Results
from Intel's thermal testing (task 54),
for example, could force the company
to rework package design (task 2 9 ) or torework functional modeling (task 17 ) or
both. This rework would also require
the company to redo some intervening
tasks. T h e value ofthe D SM in cases tike
this resides principally in making ex-plicit where information exchanges of
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 10/11
TO OL K IT • Innova tion at the Speed of Inform ation
this kind might occur. The company
then decides what to do about them.
Sometimes, there's little it can do.
The interdep enden t tasks may be so far
apart that a delay caused by incorporat-
ing late information effectively means
erational learning feedback," as Intel
did (see • 's near the top of the chart).
Rather than entirely rework the devel-
oped product and come out behind
competitors in that product cycle, the
company will either abandon the proj-
By strip pin g away the m ystery arou nd the exchange
of inform at ion durin g inn ovat ion, the DSM can give
managers far more contro l over some of their company's
most risky and expensive projects.
starting the whole process again. These
situations usually arise because some
fundamental mistake in assumptions
was made at the beginning of the proj-
ect. In Intel's case, creating a product
demo nstration (task 48) had th e poten-tial to reveal that the company's esti-
mates of sales volumes and pricing levels
were faulty (tasks 2 and 3)- If th e mis-
takes were serious, the product wouid
have to be completely redesigned.
In such cases, we recommend treat-
ing the negative infomiation as "gen-
ect altogethe r if th e information reveals
fatal flaws or launch the product as de-
signed if the flaws are minor. Mean-
while, the information will be fed into
the design and development of a prod-
uct in the next generation.In most cases, though, development
teams prefer to minimize the probabil-
ity that a later task will generate infor-
mation that necessitates rework. Thus,
it makes sense to transfer key knowl-
edge and create prior tasks, as we dis-
cussed above. But individual solutions
can emerge unexpectedly. At Intel, fo
instance, managers found they could
reduce th e likelihood of failures in ther
mal testing by having a thermal-testing
engineer con tribute to package design
Solutions of this kind will never entirely
eliminate unplanned iterations, buthey will certainly reduce the probabil
ity of them .
In our experience, the information gen
erated in a DSM analysis has alway
yielded new insights to improve the
ways companies develop products and
services. By stripping away the myster
around the exchange of information
during innovation, the DSM can give
managers far more control over some o
their company's most risky and expensive proje cts.
For a more in-depth tutorial on how to crDSM models and for links to DSM researand so ftware, go to ht tp-JAveb.mit.edu/d
Reprint ROIOIL
To order reprints, see the last page
of Executive Su mmaries.
"Our idea is to use this logo if Congress won't approve a postal rate increase.
7/28/2019 Innovation at the speed of Information
http://slidepdf.com/reader/full/innovation-at-the-speed-of-information 11/11
Harvard Business Review Notice of Use Restrictions, May 2009
Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for
the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material
in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may
not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any
other means of incorporating the content into course resources. Business licensees may not host this content on
learning management systems or use persistent linking or other means to incorporate the content into learning
management systems. Harvard Business Publishing will be pleased to grant permission to make this content
available through such means. For rates and permission, contact [email protected].