Copyright
Martin J Read.

Teamwork to develop technology strategy online

. (page 1 of 1)
Online LibraryMartin J ReadTeamwork to develop technology strategy → online text (page 1 of 1)
Font size
QR-code for this ebook


HD28
.M414

^3



DEWEY



WORKING PAPER
ALFRED P. SLOAN SCHOOL OF MANAGEMENT



Teamworking to Develop Technology Strategy



Martin J. Read & Anthony E. Gear
Balliol College, Oxford

SSWP//3560-93



MASSACHUSETTS

INSTITUTE OF TECHNOLOGY

50 MEMORIAL DRIVE

CAMBRIDGE, MASSACHUSETTS 02139



Teamworking to Develop Technology Strategy



Martin J. Read & Anthony E. Gear
Balliol College, Oxford

SSWP#3560-93



"MASSACHUSETTS INSTIlUTE
0^ TECHMOLOGY



NOV 2 1 ZOOO



.



LIBRARIES



Teamworking to Develop Technology Strategy

Martin J. Read & Anthony E. Gear
Balliol College, Oxford

". , . gather all the food of those
good years that come . . . and that
food shall be store to the land
against the seven years of famine

." Joseph, Genesis Ch. 41,
verses 34-36.

1 Introduction

The strategy of an enterprise is embodied in its strategic
decisions, which are basic to it because "they deal with the
direction of the enterprise, its future size and pattern of outputs
and markets, . . . they determine the kind of company which has to
be managed and the kinds of talents which are called for." (Minkes
[1]).

Technology Strategy is concerned with decisions to develop,
exploit, or maintain the totality of organizational know-how in a
defined set of technologies. Relevant activities in this
connection are identifying and evaluating technologies in terms of
their potential importance to overall strategy. The paper is
concerned with assessing the feasibility and significance of
various tabled technology development options in a team setting.
The paper describes a group decision support system (GDSS) known as
TEAMWORKER, designed to facilitate forward planning by a team, and
demonstrates its use by means of a case application involving
technological forecasting for the time-frame 2010 to 2030.



The literature on technology assessment and forecasting is large.
For reviews of r.anagenent techniques in these fields see, for
example, Jantsch [7], Blake [2], Souder [13] and Ford [4]. Some
techniques seek to make use of the collective intellect of an
assembled group of "experts" rather than that of a single "expert",
by means of juxtaposing disciplines and experiences in order to
promote debate, creativity and intuition. The philosophy of this
paper is based on the use of collective expertise to assess future
technological possibilities in a feedback oriented approach, Gear
[6], Computer based technology is utilized to promote group
communication, and is carefully designed to minimize deleterious
effects of conformity, domination, reticence, indiscipline,
deviation, reconditeness or vested interest.

The TEAMWORKER GDSS comprises hardware and software designed for
the particular group context and process, and is described in a
variety of applications by Gear 15,6] and Read ]10, 11]. In the
case described, the software (sometimes known as "Groupware") is
aimed at improving communication for technology assessment.
Mockler [8], and Finlay [3], have recently reviewed various forms
of groupware and GDSS for aiding strategic management processes.
TEAMWORKER software is aimed at facilitating group dialogue
following feedback of judgements from group to individual and from
individual to individual. This aspect of communication is also
discussed in [17].



The software may be designed around alternative frameworks of
decision analysis [6]. For example, [9] adopted a pairwise
comparative framework, based on a two-level hierarchy, in order to
rank a set of alternative technologies to meet a defined need. A
critical survey of recent developments in applied multiple criteria
decision making and their current status has been undertaken [12].
This work provides a useful input to the design of groupware, given
that it is the group communication process which is of central
concern rather than mathematical routines alone.

In the case described in this paper, a simple scoring scale was
made the basis of individual judgements throughout the meeting.
This meant that rapid interactions were easy, with the ability to
repeat stages, formulate additional items, and respond to new
requests "on the fly". (see also [14] in the connection). For the
given purpose, technology assessment on long time-frames,
refinement of the technique is not the main issue. As Blake [2]
has pointed out: "Over-refinement of the system makes it possible
to obscure issues with masses of detail, in turn causing error in
forecasting". Complex approaches, while of considerable interest,
may be totally unsuitable for certain business uses.

The underlying management technique behind the on-line system,
TEAMWORKER, in application to technology assessment, is GROUP
DELPHI. Some of the advantages of GROUP DELPHI over and above
standard DELPHI are discussed by [15].



A further important aspect to take into account in the design phase
of preparing process support software is the cultural setting and
history of the situation, together with the skills and backgrounds
of the participants. For example, [16] has described the "Emerging
Technology Roadmap" framework of thinking, which is updated by a
small committee of individuals within Motorola.

Each of the above issues needs to be born in mind in order to
design a framework for the group process which can form the basis
of group support software. In Section 3, a case study is described
in order to illustrate design considerations. The group decision
support system (GDSS) used in this application is described in
Section 2, and some conclusions are identified in Section 4.



2. TEAMWORKER SYSTEM

The system comprises a set of hand-held devices, one for each group
member. Digital signals from each unit are transmitted to a remote
receiver, and fed into a micro-processor. The signals are analyzed
in accordance with the software which is being used. Processed
information is displayed on a large screen visible to the group.
The screen is used to:

request information from each group member during the
meeting.



display processed feedback information at a
number of stages.

provide expert advice and menu options during
the meeting.

The arrangement of the system of feedback is shown in Figure 1.

Certain judgmental processes are especially useful, depending on
the issues involved. Some important ones are itemized below:

(i) Voting

Uses include selecting an item or option from a set,
and multiple choice questions.

(ii) Scoring

Uses include evaluating each of a set of items or
options on a pre-defined scale, scoring as a means of
producing lists in rank order, scoring as a means of
quickly filtering long lists to short-lists for closer
attention.



(iii) Comparing

Uses include weighting the relative importance of a set
of criteria by means of pairwise comparison, evaluating
subjective factors and parameters, assessing subjective
probabilities and risks.

( iv) Direct Assessment

Uses include the estimation of parameters,
probabilities, risks and other factors by direct and
subjectively based input values.

The system design facilitates communication at a meeting in support
of group work on four related activities:

(i) pooling of information, expertise, opinions and
beliefs.

(ii) comparison and debate of areas of
agreement /disagreement.

(iii) identification of key elements of the decision task for
further group attention.



(iv) revision and redefinition of the task, options,
criteria, etc. as the group progresses.

3. Case Study; Technology Assessment

Background: This case study concerns the Research and Development
department of a large multinational oil company. It is known that
the future market for petroleum products is uncertain, owing to the
rising costs of production and discovery, and the increasing
environmental concerns associated with the use of petroleum
products.

These and other issues led the R&D department to consider the
science and technologies it should be involved with by the year
2010. This is part of a continuing process led by a "Core Team"
involving the study of two key aspects:

(a) "Business Drivers", i.e. Trends culminating in
opportunities or threats to the organization

(b) "Business Options", i.e. A set of projected
capabilities which would enable implementation
of a specific technology for a defined
business purpose at some future time period.



The organization decided that part of this continuing process
should be a three-day workshop involving senior executives from the
R&D function and from a cross-section of other business functions.
A total of 42 senior executives took part in the workshop, 30 from
the R&D function and 12 from other business areas.

The purpose of the workshop was to exchange views and pool
judgements relating to the business options being considered by the
organization, and to provide guidance to the Core Team on the high
grade options, the options to drop, and the options requiring
further work. It was stressed that this workshop was not so much
a decision-making forum, bur rather an opportunity to exchange
views, evaluate and grade the options.

Workshop Framework

The workshop mission was to define a statement that the Core Team
could use to help frame a Strategic Plan. In this respect, the
Core Team was seen as the "customer" of the workshop.

Initially, a report on each of the key business drivers was
presented and discussed. This provided a framework for the
subsequent assessment of the business options. After this, the
full set of business options generated by the Core Team over the
preceding months was presented by the Core Team. At this point the
TEAMWORKER system was used as part of a structured process,

8



involving a combination of discussion and polling, to determine a
sub-set of up to 13 options to be studied more intensively at the
workshop, and then to become the subject of post-workshop
activities. This was called the "Screening Process".

The next stage involved a number of small teams each working on one
of the sub-set of options. Each team was organized to have at
least one delegate from a business function. Presentations were
then made by each team in a plenary session, detailing the
advantages and disadvantages of the business option studied by that
team.

Following this, the TEAMWORKER system was used to score the sub-set
of options on each of two criteria; the relative likelihood that
the option would be feasible in the timescale under consideration,
and the relative commercial impact that the option would have on
the organization in this timescale. This was known as the "Mapping
Process". The output from this process consisted of a two
dimensional grid showing the workshop's evaluation of each of the
sub-set of options on the two criteria.

The outputs from these two processes were used to formulate the
final guiding statement to the Core Team. These processes are now
described in more detail below.



The Screening Process

Each of the set of 35 Business Options was presented by selected
participants. The details of these options cannot be listed in
this paper, but included items such as "Chemical Vehicles
Development".

Following the presentations, and some further clarifying
discussions, each workshop delegate used a 1 to 5 scale to score
each of the business options. The business option considered of
most importance (in the opinion of a particular person) was given
a 5, and the option considered of least importance given a 1. All
other options were then given a unitary value of between 1 and 5
relative to the individually defined most and least important
options. This approach served to anchor the end points of the
subjective scoring scale of each participant independently.

TEAMWORKER was used to collect the scores and feedback summarized
information to the workshop for discussion and debate. The first
feedback screen provided to the delegates, following a particular
option vote is shown in Fig. 2. This is a histogram showing the
numbers of scores obtained for each value on the 1-5 scale, the
overall average and standard error, and the level of confidence
associated with statement that the sub-sets of scores from those
R&D Function members were significantly different from those of the
Business Function members. The average used was not a simple

10



average of all 43 delegates, but the mean of the averages of the
two sub-groups. This form of average was subsequently used to aid
determination of the options to be considered at the next stage of
the workshop, and the Core Team did not want the larger number of
R&D delegates to have a more than equal (quantitative) influence in
the process. Histograms of the type shown in Figure 2 clearly
showed how much disagreement there was in the workshop for each
option. In addition, the calculated confidence level indicated
whether there was substantial disagreement between the two groups
of members. A feedback screen was used in order to compare the
sets of R&D and Business scores (Fig. 3) . Each of these feedback
screens frequently led to further discussion and debate,
particularly when significant differences were revealed, often
resulting in rescoring of an option by the group.

When all the options had been considered, a "High-Low" chart
showing the average score for each option, together with the
standard error, was presented to the group (Fig. 4) . Included with
this chart was a ^grey area' which depicted the zone where the cut-
off for the sub-set of options to be studied further was located.
This zone was established in a debate. Further discussion
concentrated on this grey area, with some rescoring of options by
the group.

The outcome of this process was a defined sub-set of 12 options for
further attention at the workshop (the "Mapping Process") . Several

11



other options were considered worthy of further work after the
workshop (sometimes because there was not yet enough information to
make a judgement) , and other options were recommended to be
discarded.

The Mapping Process

The 42 delegates were divided into 12 work teams, each to develop
one of the selected options established in the Screening Process,
by studying the feasibility and marketability of their option.
Each team had a minimum of 3 and a maximum of 4 members, including
one Business Function Member, and spent a day developing their
option, and preparing literature which could be presented to, and
discussed and evaluated by all, the other workshop members. A
series of presentations was given by each team, when the results of
team deliberations were disseminated to the rest of the workshop.

Following the presentations, all delegates used a 1 to 5 scale to
score each of the subset of options on each of two criteria; the
relative likelihood that an option would be feasible in the given
timescale and the relative impact the option would have on the
organization if it was feasible. As with the Screening Process,
each delegate first (privately) gave the top option on each
criterion a score of 5, and the bottom option a score of 1. All
other options were then given a score of 1 to 5, relative to the



12



individual's top and bottom options on each of the criteria taken
in turn.

TEAMWORKER was used to collect the scores on each criterion. The
first feedback screen showed the scatter of scores on the two
criteria (Fig. 5) . In this figure, the areas of the dark squares
are proportional to the numbers of scores in each zone of the grid
respectively. This gave a visual indication of the degree of
disagreement existing within the group. It was also easy to view
the histogram of scores for each of the criteria (Fig. 6) . The
Average is the simple mean of the two sub-group averages (R&D and
Business) , and the Standard Error shown is the standard error of
this mean. These feedback screens led to further discussion and
debate, particularly when significant differences were observed,
sometimes resulting in rescoring.

Once all the options had been scored, a two dimensional grid was
shown in order to display the position of each option on the two
criteria, using the averages and standard errors in both directions
(Fig. 7) . The more important options were those with a high
Relative Likelihood score and a high Relative Impact score, placed
towards the top right of the figure. These were identified as the
high-grade options in the advice given by the workshop to the Core
Team.



13



4. Discussion and Conclusions

The TEAMWORKER system was undoubtedly useful in terms of providing
a degree of structure to a complex task carried out by a large
group of experts. It also enabled rapid identification of areas of
strong disagreement, making it easy to prompt relevant debate. A
number of aspects or issues which arise from this case application
are identified as follows:

(i) The Core Team wanted to use a scoring approach which
was easy to understand and implement, so a 1-5 scale
was adopted in order to score each option rather than,
for example, a pairwise or ratio scale approach.

(ii) Each participant was asked to identify independently
their personal lowest and highest rated options, and to
allocate scores of 1 and 5 to these respectively. This
ensured that the full scale was used, and serves as a
basis for producing a ranked list of options from each
delegate.

(iii) The scores themselves were used in order to calculate
average and standard deviation values. The rigour of
this form of aggregation is questionable. An
alternative approach could be aggregate rankings rather
than scores. In discussion, the Core Team preferred to

14



work with the scores in terms of clarity of the
feedback, bearing in mind that this was intended to
generate debate rather than create excessive dependence
on a mathematical routine and associated output.

(iv) The group process achieved a high level of sustained
attention and application, involving all the delegates
over several days.

(v) The feedback screens frequently resulted in verbal
expressions of surprise, typically concerned with
differences of judgement, and easily served to generate
focussed debate.

(vi) Re-scoring after feedback and discussion did not
invariably reduce differences of opinion, but
nonetheless opinions frequently changed at these
stages.

(vii) There was an occasional review of the definitions of
options, especially during debate following feedback.

(viii) The workshop resulted in the group reaching a
collective decision: a defined short-list of
options, to each of which a small group was assigned
for further indepth review and assessment after the

15



Workshop. There appeared to be commitment to this
on-going work, perhaps as a result of the high level
of involvement of the participants.

(ix) The workshop was a collective highlight or episode in
an on-going process of technology assessment and
strategic planning, rather than a start or finish in
its own right.

We conclude this paper with a quotation made by the organizing
officer shortly after the event: "We might have had some other,
more conventional, meeting but I don't think it would have been
nearly as productive".



16



REFERENCES



Minkes, Al (1987)



"The Entrepreneurial Manager",
Penguin , p. 138.



Blake, Stewart P (1978)



"Managing for Responsive

Research & Development". W. N. Freeman

& Co., San Francisco.



3 Finlay, Paul and Marples,
Chris (1992)



"Strategic Group Decision Support Systems
A Guide for the Unwary". Long Range
Planning . Vol. 25, 3, pp. 98-107.



Ford, David (1988)



Develop your Technology Strategy"

Long Range Planning Vol 21, 5, pp. 85-95.



Gear, T and Read, M J
(1988)



"On-Line Group Decision Support".
Proceedings of Vlllth International
Conference on Multiple Criteria Decision
Making , Manchester. Published by Springer-
Verlag.



6 Gear, Anthony E and Read,
Martin J (1993)



"On-Line Group Process Support". Accepted
for Publication in OMEGA.



7 Jantsch, E (1972)



Technological Planning and Social Futures .
Wiley, New York.



8 Mockler, Robert J and
Dologite, DG (1991)



"Using Computer Software to Improve Group
Decision-Making". Long-Range Planning . Vol. 24
4, pp. 44-57.



9 Prasad, AVS and

and Somasekhara N. (1990)



"The Analytic Hierarchy Process for Choice
of Technologies".

Technological Forecasting and Social Change ,
38, pp. 151-158.



10 Read, M and Gear, T (1989)



"Interactive Group Decision Support".
Proceedings of MCDM International
Workshop on Multiple Criteria Decision
Support , Helsinki. Published by Springer-
Verlag.



11 Read, Martin & Gear, T
(1993)



"Decision Support for Management Meetings
Accepted for Publication in O R Insight .



12 Stewart, T J (1992)



A Critical Survey on the Status of
Multiple Criteria Decision Making Theory and
Practice. Omega , Vol. 20, No. 5/6, pp
569-586.



13 Souder, William E &

Ziegler, Robert W (1977)



"A Review of Creativity and Problem Solving
Techniques" Research Management , July pp. 34-42.



14 Tozar, Edwin E (1986)



Developing Strategies for Management Information
Systems". Long Range Planning. Vol 19, 4
pp. 31-40.



15 Webler, Thomas; Levine

Debra; Rakel, Horst; and
Renn Ortwin (1991)



"A Novel Approach to Reducing Uncertainty".
Technological Forecasting and Social Change .
Vol 39, pp. 3-263.



16 Willyard, Charles H (1987)



"Motorola's Technology Roadmap Process".
Research Management , Sept-Oct., pp. 13-19.



17 Vetschera, Rudolph (1991)



"Integrating Databases and Preference Evaluations
in Group Decision Support". Decision
Support Systems , 7, pp. 67-77.



HAACSETS



SCREEN
FEEDBACK



1 ^ - ,




• ' 1 1 '>


,'^^=N::'.








^^ .-""




CHAIR Y''-^ >


'



MONITOR




RECEIVER
< / KXXJLE



fN



Figure 1. TEAMWORKER Feedback System




LO
OO



O

o



OO

II

LU
CO



CNJ
CO

II

CD

:>

<



CSJ



CO



— r—

CSJ
CO



-LO












CO




Csl



CO
■ to

<



— r-

CSJ



CSJ

Csl



— r-



— T"

csl



— r—
CSJ



Figure 2. Histogram of Scores for Complete Group



-LO



i



"?5

aat






ti/ifrfrfrnirfftfiffMrftm



— r-
Csl



— J—



— t—
CO



'""^■^



-CM



_r5

CO

<



CO
CM



— I—

Osl



— I—
CO
CM



— T"
I —



LO CM



CO
CO



CM



— I—

CO
CM



— r—
CO
CM



— I—

o

CM



— I 1—

^ oo



LO CM



-LO




Figure 3. Histogram of Scores for R&D and Business Groups




-Lr>





Figure 6. Histogram of Scores for Relative Likelihood and Relative Impact



R
e
I

a
t
i

Y

e

1

m

P
a

c

t



1



B




cHi




{^



e



12 3 4 5

Relative Likelihood



Figure 7. Grid Showing Average Scores and Standard Errors for Each Option



^.7^0



Date Due




3 9080 02239 2200



'i'.')ii.(i!/i;,¥„*t%?;«iis#i






1

Online LibraryMartin J ReadTeamwork to develop technology strategy → online text (page 1 of 1)