:
^^^
ALFRED
P.
WORKING PAPER
SLOAN SCHOOL OF MANAGEMENT
MANAGING THE INTRODUCTION OF NEW PROCESS
TECHNOLOGY
INTERNATIONAL DIFFERENCES IN A MULTI -PLANT
NETWORK
Warcie J. Tyre
MIT-Sloan School of Management
Revised August 1989
WP# 3004-89-BPS
MANAGING THE INTRODUCTION OF NEW PROCESS
TECHNOLOGY:
INTERNATIONAL DIFFERENCES IN A MULTI-PLANT
NETWORK
Marcie J. Tyre
MIT-Sloan School of Management
Revised August 1989
WP# 3004-89-BPS
-.oiCC
OCT
4 1989
ABSTRACT
This paper examines the intfoduction of new technologies in the manufacturing
environment, focusing on 48 new process introduction projects carried out in
The paper
eight plants in Italy, West Germany, and the United States.
examines regional variations in performance and identifies managerial factors
differences.
which may account for these
In order to understand the requirements of process change, two relevant
dimensions of technological change are measured ("technical complexity" and
Further, three groups of organizational response
"systemic shift").
mechanisms for dealing with technological change are identified. These are
preparatory (or early) search, joint search with experts external to the
plant, and functional overlap within the project team.
Comparing project performance across regions, results are statistically
However, performance on two separate
indistinguishable in Germany and Italy.
This performance gap is
dimensions is significantly worse in the U.S.
explained by differences in the intensity with which the three response
In order to understand the source of
mechanisms are used across regions.
these differences, the paper examines historical differences among plants in
the three regions.
It is apparent that managerial choices over time have
resulted in distinct sets of organizational capabilities, resources, and
assumptions.
These factors, in turn, have important and long-lived effects
on the way plants approach technological problem solving.
INTRODUCTION: NEW PROCESS TECHNOLOGY
PROMISES AND PROBLEMS
Example: The Case of Factory Q
Q
1982 Factory
In
line
--a large,
would be the factory's
promised
continuous operation;
modern age
into the
of metal
management's view, the technology represented the future
of
manufacturing
of 1986,
but was not yet operating
Efficiency was
working.
of the
The
it
In
company, and
the author walked through the line with one of the
process engineers who had worked on debugging
operation,
line.
general.
in
December
in
the
in
high-precision machining
first integrated,
move the factory
to
liigh-volume production facility
installed a new,
--
Eastern United States
line's quality
because engineers had never found time
a
line
was
in
consistent basis.
Product quality was passable,
below acceptable levels.
still
but no one was sure how the new
The
process.
tiie
capacity on
at full
performance should be assessed
to install
and learn
to
use new
measuring equipment as originally planned.
During the same time period, another plant owned by the same company and
located
Northern
in
machining
Italy also
introduced
computer-integrated precision
a
employing many of the same machine designs and producing
line,
similar product.
The
months
of installation;
plant.
By
1986,
line
its
was operating
at
near
efficiency and quality were
underway
several projects were
among the best
in
Back
in
a
of their
Many
O were quick
significant step forward for the plant,
in
ways that
its
line
to
note that the line
For four years,
the
had drained engineering and maintenance resources from other parts
of the plant;
begun
to
and that the pioject
efficiency ratings did not capture.
Nonetheless, the experience had devastated the plant.
new
of the
own.
the U.S., personnel at Factory
had been successful
the
the successful introduction had been promoted or were
managing major projects
represented
in
develop and introduce
to
next-generation technology to support new product thrusts.
personnel involved
a
capacity within 18
full
it
demanded continuing inputs
pay back the sizable investment
1
it
of time
and capital, and had not
represented.
In
the interim,
corporate management had determined that Factory O was probably not ready,
after
it
become the high-performance, automated producer they had hoped
to
all,
The
could be.
The Challenge
of Technological
This case
many
was that the factory was
reality facing plant personnel
be closed.
likely to
repeated, often
is
factories.
extreme
less
in
those introductions which are quickly rejected by the
but those which ci-eate
per-sistent drain on
a
that the difficulties of introducing
in
human and
capital
One major U.S. study found
resources without yielding competitive benefits.
result
too frequently in too
for-m,
When companies introduce new manufacturing processes, the
real failures are not
organization,
Changes
new manufacturing equipment frequently
productivity losses equal to or exceeding the original cost of the
equipment, and that the disruptive effects can persist for two years or more
(Chew, 1985; Hayes and Clark, 1985).
competitive implications as deliveries,
Persistent problems can have serious
and market share
reputation,
slip.
Further process change becomes impossible, while competitors' process
technologies move steadily ahead.
These concerns are particularly relevant
(1986),
for instance,
flexible automation
--
argues that U.
S.
in
the United States.
managers
but they are using
it
"are
Jaikumar
buying the hardware
of
Rather than
vei-y poorly.
narrowing the competitive trade gap with Japan, the technology of automation
is
widening
it
further" (page 69).
Thurow
(1987) blames "America's poor
productivity, quality, and trade performance" squarely on inferior
capabilities in introducing and using process technology
According
to the
(page 1660).
Manufacturing Studies Board, U.S. companies need
"reemphasize process improvements
--
selecting,
available manufacturing technologies -- as
a
using,
critical
to
and implementing
competitive weapon."
(Manufacturing Studies Board, 1986:17)
yet despite evidence of serious problems,
managers can improve their organizations'
Much
of the existing
little
is
known about how
ability to utilize
new technology.
research on process innovation and diffusion has focused
on the decision to adopt new technology,
rather than on the process of
learning to use
a
new technology once
organization (Kimberly,
it
has been brought into the
Most
Rogers, 1983).
1981;
of the existing
r-esearch
on the "implementation" stage of the diffusion-innovation process focuses on
developing organizational attitudes and receptivity to change'useful first step,
but
fails
it
to illuminate the
This
is
a
behaviors needed to
identify and address the problems and uncertainties involved
in
technological
process change.
Process Change as Technological Innovation at th e Plant Level
The premise
is
not just
paper
of the
is
new process introduction
that the success of a
Successful
function of organizational receptivity.
a
technoligcal change also requires active organizational efforts to adapt the
new technology, the existing manufacturing system, and
itself to a
new
set of
demands (Abernathy and Clark,
Van de Ven, 1976).
1987;
Leonard-Barton,
Consequently, new process introductions often
require considerable problem solving and even innovation
(Kazanjian and Drazin,
the organization
1985;
1986;
Rice and Rogers,
at
the plant level
1980).
This paper examines the management of process change at the level of the
individual project and
in
the larger strategic perspective.
It
identifies
organizational mechanisms for adaptation and analyzes their impact on the
success of new introduction projects.
Further, the paper explores whether
national or regional differences in project performance
exist within
of
a
single multi-plant network,
observed differences.
and seeks
The paper suggests
and project approach
to identify the
causes
that managerial choices that
shape organizational assumptions, technical capabilities, and external
linkages amount to long-term strategic decisions.
performance levels and technological options
well
These choices influence
into the future.
This research tradition is represented by two streams of research.
of the early "implementation" research examined the different
requirements of various stages in the innovation process.
Many authors
concluded that implementation, as distinct from development, calls for
relatively tight forms of managerial control (Duncan, 1976; Sapolsky, 1967;
Zaitman, 1973).
Another research stream focuses on creating organizational
or worker receptivity to the new technology (e.g. Majchrzak, 1988; Nutt, 1986)
First,
much
The balance
paper
of the
on existing literature
ot-ganized
is
to identify
in
five parts.
Part
In
I,
I
draw
three principal organizational response
mechanisms for dealing with technological change.
research methodology and the variables used
in
Part
describes the
II
the analysis.
Part
III
examines the data on project attributes and project success, and compares
results across regions.
Part IV,
In
discuss some of the forces which may
I
explain the differential success of introduction projects
Part
V
I.
two broad categories.
change can
efforts to deal with
before
information and making necessary adjustments early,
it
fall
one
undertakes the
it
faces and increases
its
Second, the organization can increase
ability to plan for or control events.
adapt to unexpected events once change has occurred.
ability to
into
By gathering
the organization can prepare.
First,
change, the organization decreases the uncertainty
its
various regions.
ORGANIZATIONAL RESPONSES TO TECHNOLOGICAL CHANGE
An organization's proactive
of
in
presents conclusions and suggestions for further research.
This
involves developing ways to capture, transmit and use new information through
increased coordination and feedback
Theorists argue that
if
preparatory efforts nor
but to accept
1958;
a
in
"real time"
(during task execution).
the organization facing change increases neither
real
time coordination and feedback,
lower level of performance (Galbraith,
1973;
it
has no choice
March and Simon,
Thompson, 1967).
Based on the literature on change
fieldwork,
I
identified three "response
organizations to adapt, either
task execution.
These are:
before the new technology
is
in
1)
organizations and on
in
initial
mechanisms" which enable
advance
of technological
change or during
preparatory, or early, search undertaken
put into use;
2)
joint search
during the startup
process with technical experts outside the factory, and 3) functional overlap
between engineering and manufacturing groups
1.
at the plant
level.
Preparatory Search involves the investigation, modification, or
"reinvention" of the new technology and relevant aspects of the receiving
organization before the technology
Rogers,
1980;
is
installed in the factor-y
Rogers, 1983; Van de Ven,
1986)
(Rice and
This may involve adapting
existing manufacturing systems,
1985).
routines and procedures (Bright,
Chew,
1958;
Coordination with (internal or external) developers of process
equipment
an important aspect of preparatory search, allowing the mutual
is
adaptation of source and user during the early phase of the project (Leonard-
Barton,
1988).
The second and third response categories both involve
real-time mechanisn
for adapting to problems and opportunities which develop as the organization
gains experience with the new technology:
2. Joint
facilities or
Search involves coordination with knowledgeable individuals from
organizations external to the manufacturing plant.
of joint organizational search stems from the
set"
concept
(Evan, 1966; Thompson, 1967) or, more specifically,
"technological organization set".
of suppliers of technology,
The
The
notion
"organization
of an
a
unit's
latter can be defined as a coalition
equipment, components, or information.
Research
suggests that joint problem solving among members of the relevant
technological organization set can account for "a
major part of the
company's problem solving capability with respect
to the
(Lynn, 1982:p.
8;
see also Ettlie and Rubenstein,
1980;
new technology"
Imai,
Nonaka and
Takeuchi, 1985)
3.
Functional Overlap involves linking r-elevant functions within the
organization to create "overlapping" subsystems or multifunctional teams for
dealing with change (Galbraith,
1973;
Gerwin, 1981; Landau, 1969).
the
In
manufacturing environment, key functions include the plant technical or
engineering activities and direct management of production output.
Tighter
linkage between these areas moves the locus of decision-making closer to the
source of relevant information, and therefore increases the organization's
ability to
respond to uncertainty (Beckman, 1986; Perrow, 1967).
II.
RESEARCH METHODOLOGY
Methodology and Data Gathering
The introduction
of
new process technology
5
into existing plants
is
a
long
.
and complex process.
Progress may be affected by multiple, often subtle
factors related to the project and
its
Understanding these factors
context.
typically requires in-depth clinical analysis of individual cases.
in
order
to identify patterns
However,
and systematic variations among projects, this
complexity must be compressed into
a
few standardized measures.
To meet these competing demands,
I
collected three kinds of data:
descriptive information about projects, their history and contexts through
open-ended and semi-structured interviews; specific data on project
characteristics and outcomes through
a
written questionnaire; and documentary
evidence about plant operations and projects undertaken from company
archives
Clinical field
work began with open-ended interviews
technical staff at the corporate and division levels.
at
of
managers and
Exploratory interviews
each plant were used to examine salient dimensions of the technological
change process, and
to identify
new process introduction projects and
Data on each project were gathered through
a
series of semi-structured
interviews with principal infor-mants and other project participants.
Personal interviews were also used to introduce
was
filled
follow-up comments four to six weeks later.
response rate.
written questionnaire which
This process resulted
was corroborated by managers
method
in
lOCo
Interviews lasted from one to four hours, and each respondent
was interviewed several times over the course
iterative
a
out be each principal informant, and to collect questionnaires and
of data
at the plant
of a
year or more.
and division
levels.
Information
This
gathering allowed new insights derived during the
course of the research to help guide analysis and interpretation
of the data
collected.
Site Selection
^ In each case the principal informant (also referred to as the
"project manager") was identified during early field work as the person who
had the "most direct, day-to-day responsibility for bringing the new
technology up to speed in the factory".
Other project participants were
identified both by plant management and by principal informants.
The Company operates
components).
It
is
in
a
manufacturing company, and
at a single leading
Research was undertaken
involved eight plants located
in
the United States, West Germany,
single,
and
Italy.
well-defined industry (precision metal
the world market leader
in
industry
its
market share and reputation for product quality.
in
Competition
terms of
the industry
in
focuses on consistent precision quality as well as cost, and senior managers
view the Company's superior manufacturing capabilities as their primary
These
competitive tool.
by extensive in-house
capabilities are supported
New machine
process development activities.
and manufacturing
tools
processes are developed at the central Process Development Laboratory and
Process technology
several divisional technology centers.
is
also
in
purchased
from outside vendors.
The Company
is
organized geographically, and operations
countries are run as separate divisions with
was carried out
in
--
and involved two
These eight plants represent
division.
operating
facilities.
manufacturing
They include one
a
in
management.
Company
three major divisions within the
and the United States
Italy,
local
to
--
different
This study
West Germany,
three plants
in
each
cross-section of the Company's
large factory
in
each division
core part of the product line and often located close to the
a
division offices and technology center.
In
addition,
include more remote and specialized facilities.
studied include one operation judged by
local
In
the sites chosen
each division, the plants
management
to
be
"very
capable", and one judged to be less able technologically.
Limiting the research to one global,
several benefits,
market variations.
single-product corporation had
including controlling for most industry,
The design
product, and
also facilitated access to detailed
confidential information about projects and their historical,
and competitive contexts (see Rogers, 1983;
1986).
On the other hand,
managers
in
I
p.
361;
amount
Graham and Rosenthal,
anticipated that, even within the same company,
different countries might take different approaches to the
introduction of new process technology (Jaikumar,
1988).
and
technological,
1986;
Lynn, 1982; Clark,
Individual divisions had developed along very different lines.
of local
autonomy and the
level of participation
7
by divisions
in
The
Corporate-level decision making
varied significantly among divisions and
lias
over time.
Given this background, the Company offered an excellent oppor-tunity
examine national or regional performance variations
and
to
of the sample,
On the other hand, given the
is
it
to
multi-plant network
it
small size
and restricted nature
not possible to generalize from this study to more
general national or regional differences
is
a
examine some of the managerial differences underlying those
differences.
Nor
in
in
the management of technology.
possible to determine from the data presented here the influence of
national or regional contexts on the managerial factors observed.
The cross-cultural aspect
of the research
struct validity given cultural differences.
raised important issues of con-
This was dealt with through ex-
tensive field work aimed at developing measures and employing vocabulary appropriate to the different contexts studied.
Sample Selection
The sample
of projects studied includes
all
of the
introductions identified where the technology was "new"
new process
in
some way
to a
particular factory and which:
1)
were undertaken during the
last
four years and completed or nearing
completion at the time of the study;
2)
represented
a
total
capital
investment of greater than $50,000
(in
constant 1986 U.S. dollars); and
3)
involved participants
The sample includes
improved versions
a
who were
available for interview.
spectrum
of technological process
of existing
equipment
technologies and production systems.
change, from
to introductions of novel
Production technologies include metal
turning and precision machining equipment, assembly and inspection systems,
thermal treatment and metal forming equipment, and handling systems; the
range of technologies introduced
comparable.
in
each of the three divisions
Four to eight projects were studied
in
is
A
each plant.
total of
48 introduction projects comprise the sample.
Variable Definition
Quantitative measures were developed for each of the following:
Project Attributes
(1)
The
:
size
relative to existing technology
and nature
change,
of the technological
to past practice in the specific user
and
environment;
Response
(2)
The organization's response
:
three mechanisms identified
(3)
Outcome
:
in
The success
Part
change,
to the
in
terms of the
I;
of the introduction project,
in
terms of the
startup time required and the operating benefits achieved.
Further details on variable definition and
Variables are described below.
measurement are presented
Tyre and Hauptman (1989).
in
Project Attributes
Because projects varied widely
involved,
it
attributes.
on
total
(stated
was necessary
Measurement
investment
in
in
to
in
the amount and nature of the change
measure and control for these project
of project scale
new equipment,
was straight-forward and was based
tooling,
constant 1986 U.S. dollars).
and other capitalized items
On the other hand,
the technological challenge involved was more complex.
technological change were derived,
based on the literature on innovation and
These
diffusion combined with early field work.
the technology and the nature of the change
environment (see Appendix
1).
controlling for
Six aspects of
it
relate to the the
represents
in
a
newness
of
specific plant
Data were gathered through twenty-four
separate questionnaire and interview items; this was reduced to six aggregate
scales relating to the
this
way increased
further analyses.
its
amount and nature
statistical
of
reliability
change.
Reducing the data
and interpretability
in
in
Aggregate scales then served
as
input for exploratory factor analysis to
The
identify underlying dimensions of process change.
factor procedure
employed principal analysis with orthogonal factors (Jackson, 1983;
Analysis revealed convergence around two factors, accounting for
variance
in
The
the aggregate scales of technological change
first factor,
labeled technical complexitv
,
p.
147).
the
537, of
.
measures the number,
novelty, and technological sophistication of new features and improved
The
concepts introduced (such as tooling, measurement and control systems).
second factor, called systemic shift
,
is
the degree to which the new
equipment or system introduces fundamentally changed manufacturing tasks or
operating principles to the plant.
Projects rated high on systemic shift
represent departures from accepted manufacturing approaches, such as moving
from traditional metal removal processes to near-net-shape forming
technology, or moving from reliance on buffer stocks between operations to an
integrated just-in-time flow of manterials
.
When both
technical complexity
and systemic shift are high, the introduction represents
the technology of the factory.
illustrated
The
practical
meaning
a
"radical" shift
of this distinction
in
is
by managers' comments about projects which were rated
particularly high or low on these variables,
FIGURE
1
as
shown
in
Figure
1.
ABOUT HERE
Organizational Response Mechanisms
•^
Oblique roations were also
Factor results are shown in Appendix 2.
Similarly,
tried, however they did not improve results significantly.
While the
inclusion of a third factor does not improve explanatory power.
satisfactory for
considered
percent of variance explained is not large, it is
exploratory work and appears consistent with findings based on
clinical
data.
This dichotomy can be traced back to Perrow (1967); similar
distinctions have been suggested by Abernathy and Clark [2] and Tushman and
Anderson (1986). The measurement and implications of the distinction
described here are discussed in Tyre and Hauptman (1989).
10
1.
Preparatory search:
The intensity
of searcli
and adjustment undertaken
before installation of the new process was measured by four interview and
questionnaire items detailing the involvement of factory personnel
in
proposing and developing the technical features of the new equipment or
system, including early development and testing activities carried out by
factory personnel.
2.
Items and scale reliabilities are shown
Joint Search:
Measures
in
Appendix
of the intensity of joint search are
3.
based on
four interview and questionnaire items describing the role of personnel from
outside of the factory and the importance of their contributions during the
start-up process.
3.
(See Appendix 3.)
Functional overlap was measured by the number of lateral linkages
between plant engineering and production personnel used
Eight possible linkaging mechanisms were
adopted cumulatively (Galbraith, 1973),
a
intensity of integration between functions.
(1)
identified.
in
the project.
Since linkages are
higher score represents
a
greater
Primary information came from
participants' sketches of the "people and project structure" on the
written questionnaire and (2) participants' descriptions of the involvement
and contribution
Project
of various players in the
problem solving process.
Outcome
Multiple measures of project outcome were investigated;
measures are discussed
in
this paper.
two principal
1.
of
startup Time:
The elapsed time between
installation
and productive use
new process technologies has important economic and competitive
implications for the firm (Gunn,
1987;
Thurow, 1987).
Startup time
influences factory efficiency and delivery performance, the ability to carry
The
out other projects, and career growth of project participants.
definition of startup time
is
used
to reflect the idea that the initial
startup period, during which disruption to ongoing operations
significant,
is
more costly
(in
sum
not yet fully
debugged (Chew,
is
operating
1985).
Startup time
is
the
of:
1)
Initial
equipment
2)
is
often
is
terms of resources used and opportunities
foregone) than the later period, when the technology
productively but
following
startup period: the elapsed time
until
parts are being made
in
Introduction period: the elapsed time
new process operates
at
in
months from delivery
of the
production mode; plus
in
months from delivery
acceptable levels or until the the project
until the
is
considered complete.^
Information came from project schedules and key dates reported on the written
questionnaire, with corroboration from project documentation and plant
management.
2.
Operating Improvement: Managerial choices are expected
to affect the
effectiveness of the new technology as well as the efficiency of the
introduction process (Leonard- Barton,
improvements
in
1988;
Rogers,
1983).
Data on
operating performance comes from three interview items
^ This does not indicate when (or whether) the technology reaches fully
satisfactory levels of performance.
Performance at project completion may
differ significantly from original expectations or from current requirements.
12
concerning the usefulness of the technical solutions implemented, the degree
to
which technical objectives of the project were met, and the
operating reliability achieved.
This approach to measuring project performance
.86)
level of
(Cronbach's alpha test of scale reliabilty
is
=
consistent with
studies of R&-D projects, where judgmental assessments of operating
improvement have proven
reliable
(e.g. Allen,
1977;
Allen,
Tushman
& Lee,
1979).
Variables can be organized into an operational model of process change,
depicted
in
Figure
2.
FIGURE
2
ABOUT HERE
13
RESULTS
III.
Project-Level Differences:
The
first
question explored
the relationship among project attributes,
is
organizational responses, and project outcomes for the sample as
Table
whole.
a
shows the correlation matrix for the variables introduced above, and
1
Table 2 shows regression results.
Equations
[I]
and
[III]
in
Table
the effect of project attributes on outcomes; they provide some
support for the validity of the measures used here.
themselves explain
times;
and
Project Attributes, Organizational Response,
Outcome
Project
Project attributes
significant amount (33%) of the variation
a
examine
2
initial
however they do not explain observed variations
in
in
by
startup
the level of
operating improvement achieved.
TABLE
Considerable insight
is
1
ABOUT HERE
gained by taking into accoint variations
use of organizational response mechanisms.
time (Table 2, equation
[II]),
times.
all
is
associated with shorter startup
Moreover, the inclusion of response mechanisms
appears to give
a
the
negative and significant (at p=.05).
greater use of these mechanisms
is,
in
the regression for startup
the coefficients of preparatory search, joint
search, and functional overlap are
That
In
in
the analysis
clearer picture of the effects of systemic shift and
project scale.
Results of the
full
regression for operating improvement (equation [IV]
in
Table
in
successful introductions.
2)
also suggest that organizational
Higher
level of
responses play an important role
preparatory search and joint
search are both associated with higher ratings of operating performance.
major difference
is
expected direction,
" This
that the coefficient of functional overlap,
is
while
in
the
not statistically significant".
result was not expected and
and Hauptman.
14
is
explored
in
more
detail
in
Tyre
The
TABLE
ABOUT HERE
2
These results suggest that preparatory search,
joint search,
and
functional overlap are important mechanisms for coping with technological
process change.
manufacturing
They appear
contribute both to the speed with which
to
introduce new process technology, and (with the
facilities
possible exception of functional overlap) to the degree of operating
improvement achieved.
should also be noted that fast startups are,
(It
in
general, associated with high levels of operating improvement (r=-.63;
p<.05).
identified
Therefore,
appear
to
Regional Differences:
The next step
in
not surprising that the response mechanisms
is
it
enhance performance on both measures.)
U.S. versus Europe
in
the analysis
is
to
compare project performance
systematic differences between projects undertaken
in
Germany and
were
in
a
in
U.S. plants and those
On the other hand, the German and
Italy.
statistically indistinguishable on the
the following analysis
I
subgroups are discussed
Italian
dimensions examined.
treat projects undertaken
in
in
subsamples
Therefore,
Germany and
Underlying similarities and differences
single European sample.
of plants
As shown below, analysis revealed
the U.S., Germany, and Italy.
in
Italy as
the two
Section IV.
The U.S. plants took longer, on average,
to introduce
new process
technology than did European factories, and they reported lower levels of
Table 3 shows summary statistics for the two regions.
operating improvement.
As noted, no consistent differences were discovered between divisions
in
Europe.
TABLE
These results are examined
analysis.
Eruope)
In
in
Table
4,
a
dummy
in
3
ABOUT HERE
greater detail by means of regression
variable indicating the region (U.S. or
which the project took place
15
is
entered into the regressions for
startup time and operating improvement.
projects undertaken
The new variable
project outcomes not explained by other variables
TABLE
Equations
[I]
is
coded
1
for
the U.S. and controls for any "regional effect" on
in
and
[III]
in
4
the analysis.
in
ABOUT HERE
Table 4 show regional differences after
controlling for measured project attributes but not controlling for
variations
in
project locus
Projects undertaken
almost 45 percent.
the mean,
the U.S.
this basis,
associated with significantly poorer project outcomes.
is
in
the U.S. took longer to complete than did projects
European plants by an average
at
On
organizations' responses to change.
The "U.S.
of 8.8
effect" on operating
new equipment and systems introduced
percent below those introduced
in
in
aggregate months, or an increment of
Europe
improvement
in
is
comparable;
the U.S. were rated 48
terms of operating improvements
in
achieved.
Once organizational responses are taken
into account,
however, the
association between regional locus and projet success becomes statistically
insignificant.
The
addition of preparatory search, joint search,
functional overlap reduces the effect of the U.S.
time by approximately two-thirds (Equation
[II]),
dummy
and
variable on startup
and decreases
its
effect on
operating improvement by approximately one half (Equation [IV]).
Organizational Responses to Change
in
Europe and the U.S.
This analysis suggests that while the U.S. plants studied did tend to
perform poorly on process introductions relative to their European
counterparts,
a
large part of the gap
project teams
in
the U.S.
technological change,
new equipment.
to
is
accounted for by the tendency of
respond somewhat
less
vigorously to
both before and after the actual installation of the
This argument
is
further supported by comparing the
relationship between project attributes and organizational response variables
across the two regions (Table 5).
U.S. -based project teams were less likely
16
than their European counterparts to respond to particularly challenging
projects with intensified preparatory search or joint search.
is
a
marked tendency
undertake
to
large-scale projects than
of systemic shift
the U.S. are
still
Europe,
In
or six on functional overlap (out of
such project
in
of functional
overlap
in
projects rated either five
six
Conversely, five projects
(the lowest possible score),
as
in
opposed
the U.S.
to
one
Europe.
V.
ANALYSIS OF REGIONAL DIFFERENCES
Regional differences
underlying differences
in
in
project performance cannot easily be attributed to
the products produced (which are closely comparable
across regions) or the nature of the projects undertaken: as shown
4,
there
possible high score of 8), while no
a
U.S. project rated higher than four.
had an overlap rating of
fact,
the face of
strongly associated with the
is
undertaken, absolute levels
relatively low.
In
in
Further, while the use
smaller introductions'.
U.S. -based projects
of functional overlap in
degree
in
preparatory search
less
Table
in
significant unexplained performance differences persist even after
controlling for the size and nature of the change.
results suggest
Rather,
that there exist managerial differences across regions which influence teams'
use of the response mechanisms examined here.
This finding raises important
questions about the source of regional differences
in
turn,
in
in
search activities and,
the ability of the plants studied to introduce new manufacturing
technologies.
The following analysis seeks an answer
the manufacturing organization and
its
in
the historical evolution of
process technology
in
each division.
As discussed below, managerial choices made over time dictated that regional
operations evolved along different lines, acquiring different capabilities
and operating assumptions.
Further, these attitudes and capabilities, once
created, proved to be long-lived and resistant to change:
starting
in
1985
One reason for this surprising finding is that, in many cases, largedollar projects involved the introduction of new equipment from the Company's
central Process Development Lab.
U.S. project managers frequently expressed
the belief that it was the Lab's responsibility to fully test their equipment
to deliver working production machines.
the paper.
and
17
This issue
is
discussed later
in
.
management
(the year before initiation of this study),
division levels began taking vigorous action to address
weaknesses
historical
corporate and
what they viewed
as
However the evidence suggests that
U.S. operations.
in
at the
embedded organizational constraints continued
to affect efforts to introduce
new process technology.
As background,
is
it
important to note that these differences developed
within particular social, political, and economic contexts at national and
regional levels.
Both specific managerial decisions, and the impact they had
However,
on operating units, can be examined within that larger environment.
for the purpose of this paper
comparisons;
I
shall focus
I
on specific intrafirm
do not examine their relationship with national or cultural
variables.
Organization Structure and Technology Strategy
1.
European Operations
have
Local
:
managers
in
the Company's European divisions
high degree of autonomy over operating decisions.
a
However, strategic
decisions relating to manufacturing operations and technology are
made by
centralized management committees composed of senior managers from division
and central staffs.
Product
line
rationalization,
as well as process
and
product development, have been aggressively pursued and centrally coordinated
through this vehicle
As
a
result,
in
all
European operations since the mid 1960's.
there has been
a
coherent strategy for process development
the European divisions for almost 20 years.
In
in
the early 1970's, the
Company's central Process Development Laboratory^ introduced the
first
generation of proprietary "ABC" machine tools, which incorporated new
German and
concepts for precision metal finishing.
Italian
plants soon
adopted the technology, completing hundreds of introduction projects.
Describing the process, one senior manufacturing manager observed,
long time to
learned
°
a
debug those early machines and we had many problems.
huge amount about the equipment, and about how
The Process Development Lab
country.
studied
It
is
romote
in
location
is
located
in
a
took
But we
bring new
different European
of the divisions
and language from any
18
to
"It
a
technology into the plants."
the late 1970's, the Development Lab introduced the second generation
In
of
"ABC" machine
The new
tools.
concepts but incorporated
was based on existing machining
line
This
proprietary microprocessor control system.
a
development enabled European plants
As
technology from the start.
a
to introduce
plant
one standardized control
manager explained, "We made sure that
not one machine was introduced that did not have the possibility to
communicate with other machines on the floor."
the early 1980's, the Company's development thrust turned to linking
In
The
separate machines on the shop floor.
German and
Italian
plants of
result was the introduction in
centrally-controlled,
a
fully integrated
machining and assembly system capable of 24-hour low-manned operation.
Finally,
the latter 1980's the objective has been to develop more flexible
in
integrated systems, where appropriate, for greater market responsiveness
in
low-volume product types.
2.
United States Operations
:
Recent history
differs sharply from the above.
enjoyed
full
autonomy over
local
policies
arm's length relationship with the parent
management committees.
The
in
the Company's U.S. division
Until the mid-1980's the U.S.
result
division
and operations, maintaining an
Company and
centralized
its
was the development
of a different
environment for technological change.
Until
1985,
the U.S.
division treated
independent profit centers.
plants as decentralized,
its
There was very
little
product
rationalization or coordination of process development.
line
According
to the
engineering manager at one U.S. plant, "Until recently, there was no
real
long-term planning on the subject of manufacturing processes and equipment.
Every
capital
While
tools
a
request was reviewed
number
of early
were introduced
years.
When
in
in
isolation."
examples of internally-developed "ABC" machine
the early 1970's, process investment slowed
factories did
buy new
capital
equipment they tended
the second-generation, microprocessor-based
19
"ABC" machines,
to
in
later
eschew
relying instead
There were
on external equipment vendors and traditional manual technology.
some localized efforts
to integrate separate
machining oper-ations, but plants
ended up with multiple, incompatible machine control systems.
Further, while
the German and Italian divisions maintained active equipment development
efforts to complement
work
at the central
development center had been closed
Development Lab, the U.S. equipment
1972.
in
With that move,
many
of the
functions of the division-level technology group had been discontinued.
U.S. factory operations during the 1970's and early 1980's were
characterized by fragmented product assortments and low quality levels
relative to the
Company's standards
its
in
period, as one division manager described
making money, but
in
a
short-term mode.
During this
other divisions.
it,
"The U.S. division had been
They were
not investing
in
technology or productivity."
Organizational Assumptions and Technical Capabilities
Partly as a result of the pattern of decisions described above,
the
German and
Italian
both
divisions boasted strong installed bases of process
technology, strong manufacturing capabilities, and technically capable
personnel which were largely lacking
Productivity growth
in
whereas the comparable figure
the Company's United States division.
the U.S. was
in
variations prevent exact comparisons,
management
in
Europe had averaged 7% per year between 1970 and 1985,
to be considerably
defect and quality levels, and
1.5°o.
Similarly,
while local
U.S. plants were judged by Company
behind their European counterparts
in
measures
of operating efficiency
in
terms of
such as
cycle time and machine set-up time.
Following the introduction of coordinated
operations
in
1985,
Company management
differences between regions
performance indicators began to narrow.
in
U.S.
However, patterns and assumptions
developed during the earlier period continued strongly
in
of
specific manufacturing
to
influence the
way
which new process technology was introduced into plants.
Most important appears
to
be
a
relatively
weak organizational
infrastructure for continuous process development and refinement.
In
the
U.S., such activities and their results were not tracked consistently before
20
As one senior manager
1986.
to
show
profit besides
a
manufacturing engineer who had worked
to a
"There were other way;
the region expalained,
in
According
worrying about operating efficiencies."
in
both European and North
American plants;
In America, it's easy to get plant engineers to start working on large
projects, with formal project management structures, but it's extremely
People tend to
difficult to keep attention focused on the details over time.
drift away to other problems when the work is only half done, leaving all the
little tasks that actually mean the difference between success and failure in
a
new system.
In
both Italy and Germany, plant-level engineers spent
specified
a
great deal of
These were formally
time developing and achieving annual improvement plans.
terms of productivity, quality, and other measures of operating
in
effectiveness, and monitored at the division (and even corporate) level.
While improvement activities were managed differently
plant-level personnel
divisions,
the two European
in
both countries generally had several
in
process improvement projects ongoing
These "miniprojects"
any time.
at
frequently involved iterative development and testing of new tooling or
devices
conjunction with the division's
in
(especially
Italy)
in
local
technical center or
According
with outside suppliers.
to
one senior
technical manager;
specialized by product line, we
concentrate not just on making those parts but are also
Each division is a
responsible for both product and process development.
Underlying this is a strong
development center for a specific technology.
While we
requirement -- and will -- to improve on what now exists.
rationalize product lines, we integrate technical ideas across plants and
across divisions.
When we say that European plants are
mean
they
that
As
this
quote suggests, and important difference between the two regions
concerns the way
both
potential
to
solutions.
expect
in
and outside
solutions
suppliers, whereas project teams
a
resource
apparent
in
for
continued
the
it,
to
explore technical
While counterexamples exist,
purchase
to
of expertise,
which project teams used external sources
the company
inside
in
U.S.
equipment
from
project teams tended
vendors
Europe were more apt
development
way project teams
of
in
21
a
the
given
item.
different
problems and
to
or
component
use outsiders as
This
regions
contrast
viewed
is
the
development
new
of
tooling
suppoft new process technology.
to
viewed
was
development typically
tooling
as
a
Europe,
In
requiring
task
critical
the
combined knowledge of both plant-level engineers and tooling experts from
and
internal
external
For
suppliers.
instance,
one
leader
project
an
in
plant explained that;
Italian
We
identified very early what would be the most critical problem to
developing tooling of sufficient precision to allow us to take
advantage of the new CNC technology. And we devoted a great deal of effort
to a thorough tooling study.
An important part of this process was working
with the tooling experts at the equipment developer -- the real workers who
use those machines -- to understand how to utilize this new technology.
solve:
Developing
were
which
systems
tooling
not
instances,
"project"
call
continued
In
after
the
the
exact
effort
many U.S.
contrast,
development.
vendor,
introduction
optimization
tool
all
me,
part
several
In
'You
of
was
focus of
a
itself.
boundaries of the
leader told
project
optimal
and was
introduction
really
the
can't
project--
our equipment."
project managers expressed the view that the most
approach was simply
the equipment
at
we are always doing, on
that's something
efficient
and
As one German
itself.
the
all
emphasis
this
blurred
during,
before,
attention
but
usable
just
considered one of the factory's regular responsibilities,
to
purchase
a
satisfactory tooling package from
thereby avoiding the need
to
invest
time
One manager attributed the problems which plagued
in
tooling
particular
a
introduction to the fact that the machine had not been purchased along with
full
complement
tooling
of
from
Another
vendor.
the
explained that the external supplier had delivered
right off the bat.
more
trials
or
It
ask
never presented
for
a
tool
set
extremely
a
manager
"which worked
problem, so we never had
He was
changes."
a
project
satisfied
to
run any
with
the
arrangement.
The
notion that technology can be purchased "off the shelf" extended to
new manufacturing equipment from the Company's Process Development Lab. As
senior U.S.
a
manufacturing engineer explained;
The new machines from Central Lab require a shift in skills and
operating procedures -- but that is all provided.
You don't need to have all
the knowledge in place because the machines come complete with hardware and
22
software, feference manuals, and service engineers from Central to do the
Once you learn how to push the buttons, these
training and initial set-up.
machines are quite simple to use.
was not borne out
however,
expectation,
This
practice.
in
In
almost
every instance where equipment from the central Lab was introduced into
U.S.
project participants complained that the machine did not
plant,
according
operate
performed
argued
engineers
their
which U.S.
in
with equipment vendors,
of
who
engineers
service
debugging the equipment
Engineers from the Process Development Lab,
managers
U.S.
that
run
to
instances
the
that
up did not complete the task
or training local operators.
turn,
and
specifications,
to
set
initial
and
equipment for them.
expected
were
there
While
invested heavily
project participants
all
personnel
technical
a
fact
in
in
joint
in
Lab
notable
search
but one such case involved vendors external to
the company and technical features which were relatively well-developed (low
to
medium technical complexity).
European
While
part
project
support from the
technical
development
the
of
divisional
participants
simultaneously
they
must
responsibility
about the
complained
also
Lab,
centr-al
the
with
rest
lack
stressed
factory.
of
that
As
a
manager explained;
the Labs are experts in the machine technology, but we are
understand the manufacturing process. Therefore, to take
machine which meets the basic specs and make it respond to all our
a
requirements --that's the job of the factory.
Engineers
who
the ones
in
really
Engineering Infrastructure:
The different
United
"infrastructures"
the
organization
attitudes
appeared
States
in
and
capabilities
the two regions -- that
and
development
of
Although Germany and
environment.
displayed
rooted
be
to
is,
technical
Italy
in
Europe and
different
in
the
engineering
contrasting approaches to
in
talent
differed
in
in
the
manufacturing
many aspects
of their
engineering infrastructures, both divisions shared common themes which were
absent
1
in
Italy
:
the U.S.
In
Italy,
according
to
the
23
division's
director
of
manufacturing
technology;
There are at least two
The word "engineer" is hard for us to define.
First and foremost, the
possibilities this can refer to within the plant.
He is in the
backbone of this organization is the "shop floor engineer".
He is responsible for
generally a foreman or assistant foreman.
line:
He
production and quality, as well as for cost and quality improvements.
Second, there are factory
also has hands-on responsibility for new projects.
These are the future-oriented engineers; they support
technical offices.
production people in ongoing operations but they also develop new kinds of
But they never have direct project responsibility for new process
solutions.
introductions.
It's important that line managers have responsibility for new
technology from the start.
Indeed,
many instances
in
overlap
functional
Italian
in
was
it
difficult
because
pr-ojects
to
measure the degree
respondents
of
had
sometimes
trouble distinguishing engineering and production as separate functions.
several
but also maintaining
for manufacturing,
development
Personnel
rotation
between
office.
All
through
a
levels
rotation"
of
experience
Promising
center and
supervision
in
begins
variety
a
considerable job
includes
and
the
the
with
machine
experts
attendance
manufacturing
of
technical
plant
and
project
Italian
plants
technical
to
center
itself
very
is
understand their needs,
with outside machine shops or component suppliers to design and realize
ideas
for
machines or devices
use
for
move between plants over the course
the plants.
in
of
their career,
sharing knowledge and experience among plants
division
was noted for
through
extensive
fertilization
its
ability
in-house
to
utilize
training,
in
addition,
the division.
a
people
basis
The
for
Italian
assignments,
and
cross-
among areas and plants.
by
this
system
at
the
mechanics and maintenance people.
described
In
creating
and
new
and build on existing knowledge
team
Many observers within the Company pointed
developed
go
and
move between the division technical
also
works closely with
In
responsibility
direct
management and technical
which
While
division
Italian
individuals
the factories.
staff
its
the
in
having
as
seat in the plant technical office.
a
manufacturing
manufacturing
of
includes
situations.
small,
direct
"spiral
generally
was described
individual
an
cases,
level
of
to
production
the
deep
setters,
capabilities
supervisors,
As the Company's director of manufacturing
it;
24
Our Italian plants put less emphasis on the pure engineering excellence
people than do, say, the Germans; but their strength is in the very
They have created a very
careful, conscious management of their technology.
high level of interest and ability among all their people.
of their
The outcome,
are wonderful at making all kinds of process ref inements-of the management there, not the product line or anything
But it's not a game-They love to "fiddle" with their equipment.
are very conscientious about applying their knowledge to their
The
and
it's
else.
they
one U.S. engineering manager remarked, was that;
as
Italians
because
operations.
You take a machine which we consider great, but they make lots
of little innovations and end up running twice as much product on it.
2.
Germanv
quite
In
:
German
the
division,
Engineering
differently.
is
the plant-level engineer
organizationally
management and generally reports directly
undertaken
apprenticeship
important
distinctions
by operating
within
Many
the plant management.
to
have formal engineering degrees,
engineers
plant-level
conceived
is
from production
distinct
as
engineering
the
plant
there
Indeed,
personnel.
the
based
office
from the
distinct
on
are
individuals'
formal technical training.
Formal
are
responsibilities
responsible
directly
engineer
is
process
in
are
and
cost
for
clearly
The
chief
the formal project leader for planning and administration of new
and
introductions,
responsibility
of
a
process
implementation
engineer
in
his
generally
is
or
her
and
quantity.
They are
also
expected
to
the
group.
managers, meanwhile, are responsible for meeting output targets
quality
Engineers
bifurcated.
improvements.
quality
direct
Production
terms of
in
support the engineers'
improvement efforts, principally by allocating people and development time
to
improvement and introduction projects.
However,
broken down
studied
in
structure.
these clear distinctions often
--
in
actual
operating
Germany presented exceptions
One
break down
procedures.
to
plant was composed of several
Two
the
small
or are actively
-
of
the
standard
three
shops or subfactories;
within them, engineers reported to manufacturing shop managers.
at
least
one
of
these
shops,
the
production
25
plants
organization
manager was
also
In
fact,
the
in
chief
.
engineer.
working
another plant,
In
addition
in
production floor"
at
managers
Pi'oduction
by degreed engineers on
respect
considerable
who
engineers
on the
"sat
among production
expertise
technical
considerable
levels.
all
(degreed)
"feal"
each department and reported to manufacturing managers.
in
there was
Further,
personnel
the
to
there were also engineers
the technical office,
in
the
basis
technical
their
of
shown
were
typically
As one degreed engineer explained, "The manufacturing supervisor
know-how.
--
and the department manager are really engineers too
engineers."
While
managers
at
were the
first
their
level
this
ones
manufacturing
received some formal technical training,
all
respond
to
they are the floor
varied,
capabilities
technical
problems on
technical
to
the
and they
Indeed
line.
production department managers frequently came from the plant's engineering
office.
many cases, junior engineers
In
reported
in
sort
a
of
charge
in
projects
introduction
of
arrangement
matrix
informal
the
to
production
department manager.
Relative
to
personnel
technical
Italy,
in
Germany were
less
among plants or between plants and the division technical
rotate
division
at
On
considerable efforts were made to link technical development
the other hand,
activities
to
likely
office.
various
five
First,
facilities.
(including two of the three plants
seven
the
of
plants
in
were located
here)
studied
the
in
a
geographically centralized "complex" along with the division headquarters and
size
of
The
center.
technical
group
the
division
Italy,
in
technical
was
which was several times the
staff,
responsible
designing,
for
helping plants to implement new equipment and systems.
located
plant
in
the
central
engineers
personnel
with
and
complex,
engineers
particular
consulted or even
there
the
at
areas
of
was
expertise
exper-ience
borrowed for projects undertaken
in
plants
between
factory
Fui-ther,
center.
or
in
interplay
considerable
technical
and
building,
Especially
were
other plants
often
in
the
penchant
for
complex
Whereas
famous
for
was
Italy
"fiddling" with
known within
the
production equipment once
doing
a
great
deal
of
it
technical
26
Company
for
was on the
its
floor,
preparation
of
Germany was
both
the
new
equipment
and
before
factory
existing
the
attempting
new introduction.
a
While this did not always occur (as measured by levels of preparatory search
in
studied)
projects
the
One
importance.
consistently was
it
recognized
utmost
being of
as
project leader explained;
This was one of our most successful introductions because we worked
explicitly to identify all the most important unknowns, and to develop
solutions, before the equipment was shipped or, in many items, even before it
really think about it, you can identify and address 95
If you
was ordered.
percent of the hard issues beforehand.
United
3.
States
The
:
structures
organization
formal
U.S.
of
factories
resembled the German pattern, with separate engineering departments reporting
to
the plant manager.
plant
was
requirements
supervisor
Manufacturing
integration
more
blamed
of
difficult
delays
manager-
training
managers,
or
on-line
for
testing
frequently
meanwhile,
In
set
to
and
production
several
instances,
priorities
attain.
to
subcomponent
of the small
technical
new equipment on
introducing
in
the production
getting
of
difficulty
and
often
engineers
plant-level
However, with the exception
informal
studied,
aside
time
or
tooling
of
attributed
the
for operator
devices.
disruption
and
startup delays to the absence of engineering support during startup and later
debugging
of the
Further,
capabilities
technical
much
in
new equipment.
many individuals
training
capabilities,
the plants,
or
attention
less
was
true
the U.S.,
through
Europe.
in
Unlike the situation
"Many
They do
from our industry.
In
In
that
engineering
formal and in-house
rotation
job
Plant-level
received
engineering
and the relationship between engineers and production people
varied widely.
said with dismay,
involved.
company argued
the
development
personnel
than
turnover among engineers was
manager
in
plants were too thin.
U.S.
in
frequently-cited
a
of the
engineers
in
problem.
in
high
European divisions,
As one senior
these plants
ai-e
not even
not really understand the products and processes
"
addition,
managers and technicians on the plant
According
necessary
technical
worked
various U.S. operations;
in
skills.
27
to
one
European
floor
often
lacked
engineer who
had
The production supervisor is the place where there is very often a big
gap.
In
too many cases, the supervisor does not have sufficient
understanding of modern production technologies, so he ends up relying of
Sometimes the operators have long
operating people for process expertise.
So it
experience and special aptitudes, but sometimes that is not the case.
can be hard to carry out big projects successfully.
Another engineer, who had worked
forty
years,
pointed
serious
to
"engineering backup"
in
in
factories.
local
the Company's U.S. division for some
erosion
process
of
among the
capabilities
his view.
In
The operator level is not, or should not be, what is critical.
Productivity of both old and new machines comes from the engineering backup
That
in the plants -- the engineers, maintenance people, and supervisors.
small group has got to be the organizational intelligence between management
and operators. They have to be able to teach operating people, to guide them
The
to focus on
key problems, not just rely on operators' expertise.
competence of that group and how it is cultivated is the key to the ability
to bring in new machine technology.
V.
CONCLUSIONS
This paper has analyzed the performance of new process introductions
undertaken by
the
a
single technology-based
United States.
differences
differences
response
company operating
The analysis has suggested
in
performance can be traced,
in
the way project teams
mechanisms
identified
--
the short
in
in
both Europe and
that the observed
run,
regional
underlying
to
The
respond to technological change.
preparatory
search,
joint
and
search,
functional overlap -- were shown to support improved project performance
teams
in
Europe and
mechanisms
in
including
I
differences
the
U.S.
differed
propensity
their
in
to
use
these
dealing with challenging introductions.
More important,
historical
the
in
As demonstrated, project
terms of startup time and operating improvement.
have argued that these response patterns are rooted
in
organization
the
of
management
technical
of
tasks
the corporation, and between the firm and
its
manufacturing
at
the
factory
in
technology,
level,
within
outside technical partners.
This study identified three related areas of managerial choice that determine
28
development
the
of
charged with the guidance
body
served
continuous
improvement
showed
plainly
up
performance
of process
was
levels
project participants brought to
by
Europe, this
and
across
emphasis
on
This
processes.
the
effort
improvement and quality
the attitudes and assumptions which
in
new process introductions.
organization and outside suppliers of technology.
It
was supported
competence
the
in
Finally,
the third area of
regions was the system for building
managerial choice which differed across
technical
In
time
cross-boundary problem solving between the manufacturing
tradition of
a
over
top-to-bottom
productivity
of
European divisions, and
in
activities.
efforts
manufacturing
existing
high
in
development
organization.
management body
strategy-level
development
consistent,
a
of
the manufacturing
in
multi-faceted,
a
coordinate
to
Second
facilities.
capabilities
technical
of
was the existence
First
manufacturing
organization.
Italy
In
in
continuous managerial emphasis on personnel development, cross-
particular,
and
training,
appeared
cross-fertilization
which brought unusual capabilities
have
to
resulted
in
a
workforce
both ongoing process improvement and
to
new process introductions.
This paper leaves many questions unanswered, and perhaps suggests some
new ones.
single
First,
company.
choices
the
in
and
economic,
impossible
to
it
must be remembered that
No attempt has
Company's
cultural
in
these
countries or regions involved.
this
made
study was confined
explicitly
operating
different
forces
generalize
been
Europe
and
findings
to
units
the
to
to
An important next step
a
political,
Therefore
organizations
in
to
managerial
broader
U.S.
other
link
in
it
is
the
the research would
be to compare these findings to work by researchers examining more general
patterns
would be
in
to
industries.
broadly
these and other geographic regions.
expand the existing study
On the other hand,
consistent
with
compared the development
contexts,
al.
the
a
An interesting second step
larger sample of companies and
findings
reached
conclusions
of
to
this
paper are
by other authors
who have
reviewed
manufacturing technology
in
in
different
national
including Piore and Sabel (1984), Jaikumar (1985) and Dertouzos et
(1989).
More important,
this
study suggests that
29
a
company's strategic guidance
of
manufacturing
its
development
of
processes,
suggested
by
and
analysts
in
evidence
empirical
to
attention
its
intimately
This
this
a
paper
firm's
show how
to
current
support
through the introduction
More research
global corporation.
is
historical
standing
of
(e.g.
assertions"^
which
in
understanding
of
Further,
the
of the serious questions
process
of
long-term
ability
to
been
has
Schonberger,
and
little
these areas.
also
its
in
1986;
explores
ability
these
technological
these issues are of
clinical
I
detail
have tried
redefine
to
itself
the future.
connections
The questions involved are important
organizations.
its
about technology influence not
choices
but
new technologies
merited
to
Thurow, 1987), however there has been
1986;
the effects of different managerial clioices
only
the
to
linked
interdependence
observers
industry
Manufacturing Studies Board,
little
are
new manufacturing technologies.
utilize
on
and
technology,
people and
vital
within
the
for increasing our
change
practical
in
manufacturing
concern
in
light
which now surround the global competitiveness of the
U.S. manufacturing sector.
An important exception to this is work by Hayes and Clark (1985) and
their colleagues, who showed that plants which were "better managed" in terms
quality performance and inventory levels were also able consistently to
introduce new equipment with less disruption than more "poorly managed" plants.
of
30
REFERENCES
Abernathy,
Innovation
Press.
in
William J.
(1978) The
the Automobile Industry
Productivity
Baltimore:
,
Dilemma: Roadblock to
Johns Hopkins University
J. and Kim B. Clark (1985) "Innovation: Mapping the Winds
Creative Destruction" Research Policy 14, 3-22.
Abernathy, William
of
,
(1977) Managing the flow of technology
Allen, T.J.
,
Cambridge MA: MIT Press.
T.J., D.M.S. Lee and M.L. Tushman (1980) "R&D performance as a
function of communication, project management, and the nature of work" IEEE
Transactions on Engineering Management Vol. EM-27, 2-12.
Allen,
,
Beckman, Sara (1986) "Manufacturing Flexibility and Organization Structure",
unpublished Ph.D. thesis, Stanford University.
James R. (1958) Automation and Management
Business School Division of Research.
Bright,
Boston,
MA:
Harvard
Chew, W. Bruce (1985) "Productivity and Change: Understanding Productivity at
the Factory Level" paper for the 75th Anniversary Colloquium of the Harvard
Business School.
"Project Scope and Project Performance: The Effect of
Clark, Kim B. (1988)
Parts Strategy and Supplier Involvement on Product Development" Harvard
Business School Working Paper 88-069
Richard K. Lester and Robert M. Solow (1989) Made
Dertouzos, Michael L.
America:
Regaining the Productive Edge Cambridge MA: MIT Press.
,
in
Duncan, Robert B. (1976) "The Ambidextrous Organization: Designing Dual
Structures for Innovation", in Ralph H. Kilmann, Louis R. Pondy, and Dennis
Volume I, 167-188.
P. Slevin (eds.) the Management of Organization Design
,
Ettlie,
John and Albert Rubenstein
Implementation
648-668.
of
Production
(1980)
Innovations"
Learning Theory and the
Decision Science 11 (4) (October)
"Social
William M.
(1966) "The Organization-Set: Toward a Theory of
Interorganizational Relationships" in James D. Thompson (ed.) Approaches to
Organization Design Pittsburgh: University of Pittsburgh Press.
Evan,
Galbraith, Jay (1973) Designing Complex Organizations Reading, MA: Addison
Wesley.
Gerwin, Donald (1981) "Control and Evaluation in the Innovation Porcess: the
Case of Flexible Manufacturing Systems" IEEE Transactions on Engineering
Management vEM-28, no 3, 61-70.
.
Graham, Margaret B.W. and Stephen R. Rosenthal (1986) "Institutional Aspects
of Process Procurement for Flexible Machining Systems," Manufacturing
Roundtable Research Report Series, Boston University, Boston, MA.
31
(1987) Manufacturing for Competitive Advantage Cambridge, MA:
Ballinger Publications.
Gunn, Thomas G
.
Robert H. and Kim B. Clark (1985) "Exploring the Sources of
Productivity Differences at the Factory Level," in Clark, K.B., Hayes, and
Lorenz (eds.) The Uneasy Alliance Boston, MA: Harvard Business School Press.
Hayes,
Nonaka and H. Takeuchi (1984) "Managing New Product Development:
Imai, K.
How Japanese Companies Learn and Unlearn" in Clark, Hayes, and Lorenz (eds
The Uneasy Alliance Boston, MA: Harvard Business School Press.
,
I
.
.
Jackson, Barbara Bund (1983) Multivariate Data Analysis Homewood,
D.
IL:
)
Richard
Irwin.
Jaikumar, Ramchandran (1986) "Postindustrial Manufacturing." Harvard Business
Review 64 (6) (November-December) 69-76.
Robert K. and Robert Drazin (1986) "Implementing Manufacturing
Human
Innovations: Critical Choices of Structure and Staffing Roles"
Resources Management 25:30, 285-403.
Kazanjian,
Kimberly, John R. (1981) "Managerial Innovation" Paul C. Nystrom and William
Starbuck, (eds.) Handbook of Organizationa D esign Vol. 1, Oxford, England:
Oxford University Press Vol. 1, 84-104.
l
Landau, M. (1969) "Redundancy, rationality, and the problem
and 'overlap " Public Administration Review Vol. 29, 346-358.
of
duplication
,
Leonard- Barton, Dorothy (1987) "Implementation as Mutual Adaptation of
Technology and Organization" working paper 88-016, Harvard University
Graduate School of Business Administration.
Lynn, Leonard H, How Japan innovates - A Comparison with the U.S.
Oxygen Steelmaking Boulder Co, Westview Press, 1982.
in
the Case
of
Majchrzak, Ann (1988) The Human Side of Factory Automation
Jossey-Bass.
San Francisco,
Manufacturing Studies Board (1986) Toward a New Era in U.S. Manufacturing:
The Need for a National Vision Washington, D.C.: National Research Council
March, James G. and Herbert Simon (1958) Organizations New York: John Wiley
and Sons.
Nutt, Paul C. (1986) "Tactics
Journal 29, no. 2, 230-261.
of
Implementation"
Academy
of
Management
Perrow, Charles (1967) "A Framework for the Comparative Analysis
Organizations American Sociological Review 32 (2) (April) 194-208.
of
'
Piore, Michael and Charles
York, Basic Books.
F.
Sabel (1984)
32
The Second
Industrial Divide
New
Rice, Ronald E. and
Process," Knowledge
Everett Rogers
(4) 499-514.
(1980)
"Reinvention
the
Innovation
(1982) Diffusion of Innovations (Third Edition)
New York:
Sapolsky, Harvey (1967) "Organizational Structure and Innovation"
Business 40 (7) 497-510.
Journal of
Rogers, Everett M.
1
in
The Free Press.
Schonberger,
Richard (1986) World Class Manufacturing New York: The Free
Press.
Thompson, James
D.
(1967) Organizations
Thurow, Lester C. (1987) "A Weakness
December, 1659-1663.
Tushman,
M.L.
and
P.
in
in
Action
New York: McGraw
Hill.
Process Technology" Science v 238,
Anderson
organizational environments"
(1986) "Technological discontinuities and
Administrative Science Quarterly Vol. 31, 439,
465.
Tyre, Marcie J. and Oscar Hauptman (1989) "Technological Change in the
Production Process: Organizational Implications and Responses" Working paper
#3050-89-BPS Cambridge MA: MIT.
Van
de
Ven, Andrew, (1986)
"Central Problems
Management Science 32 (5) 590-607.
in
the
Management
of
Innovation"
Zaitman, Gerald, Robert B. Duncan, and Jonny Holbeck (1973)
Organizations New York: John Wiley and Sons.
33
Innovation
in
Figure
Two Aspects
1
of Technological
Process Change
SYSTEMIC SHIFT
CHANGE IN PRINCIPLES OF PRODUCTION
low
high
THE NEW MACHINE REPRESENTED
A QUANRJM LEAP
IN
high
-EVERYTHING ABOUT THE
PRECISION
WAY YOU PRODUCE WITH
AND CONTROULABILrrY. THE
THIS
RESULT WAS ONE OF THE MOST
COMPLEX FINISHING PROCESSES
THERE
IS.
AT
FIRST,
NEW TECHNOLOGY IS
DIFFERENT; IPS LIKE
HAVING A RESEARCH
WE
COULD NEVER REALLY PREDICT
PROJECT, NOT JUST A
THE OUTCOME."
PRODUCTION MACHINE."
"IT
SOME ADJUSTMENT IS
WAS NOT THAT THE
SPECIFIC TECHNICAL
SOLUTIONS WERE SO HARD
ALWAYS NECESSARY AT
THE OPERATOR LEVEL, BUT
TO DEVELOP. BUT THAT
WE HAD TO LEARN A WHOLE
THIS
•A
WAS NOT REALLY
NEW APPROACH TO
HAPPENING'."
MANUFACTURING."
low
34
Figure
2
Model of Process Change
PROJECT ATTRIBUTES
-
Technical Complexity
-
Systemic Shift
-
Project Scale
ORGANIZATIONAL
RESPONSE MECHANISMS
-
Preparatory Search
-
Joint Search
-
Functional Overlap
V
PROJECT OUTCOME
Startup Time
Operating Improvement
35
Table
1
Correlation Matrix of Project Attributes,
(n=48)
Response Mechanisms, and Outcomes
Table
Project Attributes,
A.
Effect on
2
Response Mechanisms and Project Performance
STARTUP TIME
Independent Variables:
PREP-
TECHNICAL
COMPLEXITY
i
.49**
(.12)
SYSTEMIC
PROJECT
ARATORY
SHIFT
SIZE
SEARCH
.32**
(.12)
JOINT
SEARCH
FUNCTIONAL
OVERLAP
.05
(.12)
r2=.33 (df=45); F=8.7 (p<.001)
II
Regional Differences
in
Means: Europe and the U.S.
MEAN
STANDARD DEVIATION
6
Table 4
Regional Differences
Effect on
A.
Startup Time and Operating Improvement
in
STARTUP TIME
Independent Variables:
PREP-
TECHNICAL
C OMPLEXITY
.
50**
(.11)
SYSTEMIC
PROJECT
ARATORY
JOINT
SHIFT
SIZE
SEARCH
SEARCH
.
30**
(.11)
.
FUNCTIONAL
OVERLAP
U.S.
DUMMY
.30**
1
(.12)
(.12)
r2=.40 (df=44); F=8.7 (p<.001)
IL .54**
69**
(.09)
.33**
.39**
(.09)
(.07)
(.14;
-.33**
-.28*
(.10)
.11
(.10)
(.15)
r2=.64 (df=41); F=13.2 (p<.001)
B.
Effect on
OPERATING IMPROVEMENT
PREP-
TECHNICAL
COMPLEXITY
m
-.23
(.14)
SYSTEMIC
PROJECT
ARATORY
JOINT
SHIFT
SIZE
SEARCH
SEARCH
.13
(.14)
FUNCTIONAL
OVERLAP
U.S.
DUMMY
-.10
-.33**
(.15)
(.15)
r2=.10 (df=44); F=2.0 (NS)
JY
-.25*
-.20
-.25
(.13)
(.19)
(.15)
.45**
(.13)
27**
.20
(.20)
(.14)
-.16
(.10)
r2=.31 (df=41); F=4.() (p<.(X)5)
Regression coefficients are standardized to
Standard errors are shown in parentheses.
facilitate
**
=
=
significant at .05 confidence level; *
39
comparison
of effect sizes;
.10 confidence level.
Table 5
Organizational Responses
[1]
Projects Undertaken
in
to Technological
EUROPEAN PLANTS
Change
(n=31)
Organizational Response
PREPARATORY
SEARCH
JOINT
SEARCH
FUNCTIONAL
OVERLAP
Project Attributes
TECHNICAL COMPLEXITY
SYSTEMIC SHIFT
PROJECT SCALE
[2]
Projects Undertaken
in
.51**
.01
.27*
.29*
.39**
.05
.03
.26*
-.08
U.S.
PLANTS
(n = 17)
Organizational Response
PREPARATORY
Project Attributes
TECHNICAL COMPLEXITY
SYSTEMIC SHIFT
PROJECT SCALE
Appendix
Aggregate Scales
1
of Technological
DEFINITION
Newness
of technical features
Change
NUMBER OF
CRONBACH
VARIABLES
ALPHA
86
relative to technology
previously available
Familiarity of technical
features relative to equipment
existing in the plant at the
time of the introduction
.83
The degree
.77
technology
to
is
which the new
based on
technical or operating
principles new to the plant
or the company
The degree to which the new
technology is a prototype
,65
installation
The degree
of
change
in
product
,64
specifications introduced with
the new equipment
The stringency
of performance
requirements (cost and quality)
relative to existing standards
41
,65
Appendix
3
Derivation of Organizational Response Variables
I.
Preparatory Search
What
role did technical support and production personnel play
following activities before the equipment arrived at the factory?
1.
2.
in
the
Propose purchase of the new equipment.
Developing new technological concepts.
(1-5 scale:
support; 5
1
=
involved; 2
responsible.)
not
= fully
=
were informed;
3
=
gave advice;
4 = major
Describe any studies undertaken by plant personnel prior to delivery of
the new equipment. (Interview item, with questionnaire commentary.)
3.
(1 = no studies; 3 = some study, but not unusual amount; engineering personnel
performed some analyses; 5 = intensive study; we made a major tool life
project, and several people were assigned; we made a large effort to resolve
problems in the systems choice before we even ordered the equipment...)
Describe the process of pretesting the equipment at the vendor (or,
in
the plant) prior to delivery (or to startup).
(Interview item, with questionnaire commentary.)
4.
where applicable,
We had a runoff test, but it was perfunctory; the vendor had everything
up just right, and we were not allowed to change the conditions; the people
we sent to the qualifying test did not know what to look for; 5 = We did
extensive runoff testing at the vendor's, and we refused to accept the
equipment until it worked to our standards; once we received the equipment, we
set it up off-line and did extensive testing or routines and tooling.)
(1
=
set
Scale alpha = .67
2.
Joint Search
How helpful
1.
introduction?
a.
were
people
from
the
following
groups
during
this
Equipment vendor or in-house development center?
Personnel from sister plants?
Outside advisors?
d. Technical experts from division technical center?
b.
c.
(1-5 scale:
Made
critical
contributions
no real contribution.)
Where did people get the know-how to accomplish the introduction
new equipment -- how true are the following statements?
2.
43
of the
a.
(1-5
Outside experts provided guidance.
vendor provided essential know-how.
Not at all true.)
introduction
b. Training by the
scale: True for this
Describe the role of outside experts in the startup process.
(Interview item, with questionnaire commentary)
3.
(1 = Not a partner in the problem solving process; sold the equipment and the
normal services (such as initial installation help, replacement parts but was
5 = Part of the problem solving team; the
not really part of the project team;
vendor gave us lots of important ideas, we discussed many of our problems with
about
this
technology
by working with the supplier; the
learned a lot
him;
expert from the head office came to this plant and worked closely with me and
my colleagues, he was an important partner in the introduction project.)
I
Scale alpha
-
.79
Functional Overlap
3.
Primary information came from:
1.
"people and project structure" and their
Participants' sketches of the
of the sketch;
commentary
Participants' descriptions of the problem solving process:
involved and the modes of communication and contribution;
2.
who was
Information was coded for mechanisms listed below:
Mechanisms for Mutual Adiustment and Feedback
(1
Individuals from different functions work together directly
and solve problems in the new process.
A manager who is not directly
linking role between functions.
involved
-
-
The introduction
is
in
organized as
The introduction is one of
multi-functional task team.
a
many
special
participates
operator participates
in
for
both
plays
technical
a
and
interfunctional task team.
activities
One or more individuals
engineer
introduction
identify
the plant or department.
-
-
the
A project manager(2) has ongoing responsibility
production activities
-
with
to
performed by
act as liaisons between
directly on the production
an engineering study).
44
a
multi-level,
functions
line;
a
(e.g.
an
production
-
-
transfer of personnel across functional areas
area with the locus of technological decision-making
Permanent or temporary
links the production
.
(3)
-
The introduction takes
place
within
a
semi-autonomous,
multi-functional
unit within the plant.
Descriptions adapted from Galbraith (1973).
Formal or informal designation; multiple managers possible.
Only transfers directly related to the introduction are considered.
Previous transfers undertaken as a general policy are not included.
(1)
(2)
(3)
45
D
/g
i
/
vj
Date Due
FEB 1^199
MAY 22^
JUL 12
1991
S0V.25
APR.
1
1
1994
3
TDflO
DDS77707
D