Managing uncertainty and ambiguity in frontier
R&D projects: A Korean case study
Yong-Il Song *, Dae-Hee Lee, Yong-Gil Lee, Yun-Chul Chung
Korea Institute of Science and Technology, Research Planning Division, 39-1 Hawolgok-dong,
Seongbuk-gu, Seoul, Republic of Korea
Available online 28 June 2007
Abstract
One of the important tasks in planning large, frontier R&D projects is to minimize innate uncertainties
and ambiguities in the early stages of the project. This case study is an attempt to provide a framework to
handle such problems in R&D planning. In it we analyze various elements that define planning conditions,
classify them into basic constructs, and suggest tools and methods to deal with uncertainty and ambiguity.
We utilize two case studies to approach the research questions. Our findings suggest that both initial
planning conditions and the effectiveness of front-end planning management affect the performance of
R&D planning and the later R&D process.
# 2007 Elsevier B.V. All rights reserved.
JEL classification : O32
Keywords: Technology management; R&D planning; Fuzzy front-end; Knowledge management
1. Introduction
As the national R&D strategy of Korea shifts from an ‘‘imitation’’ to an ‘‘innovation’’ based
approach, Korea is facing a set of new challenges in technology development. The prevailing
R&D model in Korea, which has proven effective time and time again, is the introduction and
subsequent modification of foreign technologies. Some of the recent examples in this vein
include the development of technology-intensive products such as DRAM, CDMA, and TFT-
LCD. The reason that Korea has been so successful in developing these products is that major
participants in the Korean R&D system have been able to systematically absorb and internalize
advanced foreign technologies that offer clear technological development paths.
www.elsevier.com/locate/jengtecman
J. Eng. Technol. Manage. 24 (2007) 231–250
* Corresponding author. Tel.: +82 2 958 6005; fax: +82 2 958 6020.
E-mail address: [email protected] (Y.-I. Song).
0923-4748/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jengtecman.2007.05.001
In the process of moving to a knowledge-based, technology-intensive model of economic
expansion, special attention is now being paid to developing a relevant R&D model which the
key players involved in Korea’s national innovation system can utilize for successfully
organizing and managing path-breaking, innovative R&D projects. One of the critical tasks in
this process is to reduce the technological and market uncertainties and ambiguities which are
inherent in such frontier R&D projects. When technologies and markets are ill defined and still in
their emerging stages, the issue of how to handle such uncertainties, as well as the related
development risks, becomes of critical importance. This is especially true in nations like Korea
where technological know-how and creative knowledge stock have not accumulated sufficiently
enough to compete with world technology leaders. Challenges also come from the current R&D
environment in which the technological life cycle gets shorter, and technological breakthrough is
more of an artistic fusion of various technologies over a period of cumulative research by a group
of competent researchers.
This case study is an attempt to deal with some of these issues. We focus particularly on the
issues related to the uncertainty and ambiguity in planning large, cutting-edge technology
development projects supported by the Korean government, and provide a framework to manage
such issues. Researchers have used the term ‘‘fuzzy front-end’’ (FFE) to describe the uncertainties
and ambiguities embedded in the front-end portion of planning activities, or pre-planning activities,
of the product development process (Gupta et al., 1986; Bacon et al., 1994; Khurana and Rosenthal,
1997; Onno Omta and de Leeuw, 1997; Khurana and Rosenthal, 1998). In this study we applied this
concept to investigate how the embedded uncertainties and ambiguities in such R&D projects can
be managed. In particular, we have chosen as case studies two large, forefront research projects,
both of which were composed of many sub-projects with multiple project managers and
participants from many Korean research institutes and universities. Both projects had a highly
complex technological structure, encompassing various technological fields. Both were awarded
similar funding of around US$ 10 million per annum over a 10-year period.
This paper is organized as follows. We first document previous research in R&D planning,
especially in front-end planning activities. Next, historical background is provided of government
initiatives for such large, forefront technological projects in Korea, followed by case studies of the
two selected projects. Then an in-depth analysis of the case studies is performed with a detailed
discussion of the implications of this research. We conclude the paper by providing guidelines for
dealing with front-end fuzziness in such large, highly experimental R&D projects.
2. Research background
2.1. Previous research in R&D planning
Front-end R&D planning can be defined as ‘‘the period between when an opportunity is first
considered and when an idea is judged ready for development’’ (Kim and Wilemon, 1999). It is in
this time period that the organization formulates a concept of final product/derivatives of R&D
and determines whether or not it will invest resources to initiate the formal R&D process
(Moenaert et al., 1995). Planning activities typically include R&D strategy formulation,
opportunity identification and assessment, a technological feasibility study, R&D project
planning, and internal reviews (Cooper, 1997; Khurana and Rosenthal, 1998). Because of
embedded uncertainties, ambiguities, or ‘‘fuzziness’’ with respect to market, technology, R&D
process, funding, etc., this stage is characterized as knowledge seeking, learning, communication
and study, experimenting, and creating.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250232
The issues of uncertainty and ambiguity in the planning period have received particular
attention in the literature (e.g., Bacon et al., 1994; Kim and Wilemon, 1999; Smith, 1998; Doll
and Zhang, 2001). Uncertainty is defined as the absence of relevant information (Galbraith, 1977)
and is a measure of an organization’s ignorance of the value of defining constructs in the planning
process (Doll and Zhang, 2001). Previous research in planning has focused on environmental
uncertainty as the prime cause of front-end fuzziness (Gupta et al., 1986; Bacon et al., 1994;
Moenaert et al., 1995). During this period, decisions are often made without a full understanding
of technological obstacles or opportunities, resource capabilities of the organization, prospective
market needs and responses, or the hidden factors occurring behind the scenes, such as
interactions among involved parties. Due to so many unknown factors and ambiguities, decision
makers often find themselves in a position of being forced to make decisions with inadequate
knowledge and information.
Analyzing the distinctive features and structure of R&D planning, researchers in R&D
planning have attributed the causes of front-end fuzziness to technological risks and uncertainties
(Rosenthal and March, 1988; Wheelwright and Clark, 1995), market uncertainties (Smith and
Reinertsen, 1991; Bacon et al., 1994; Cooper, 1997), regulatory constraints (Gupta et al., 1986),
and/or environmental uncertainties in general (Montoya-Weiss and O’Driscoll, 2000). In a
similar vein, Smith and Reinertsen (1991) point out that turbulent market conditions, as well as
uncertainty with respect to the timing of product introduction, determine the dynamics of
planning. Beck et al. (1999) provide methods to reduce technological uncertainty, using software
tools that facilitate knowledge mapping and technology roadmapping. Clark and Fujimoto
(1991) suggest micro-level structural mechanisms that can be utilized to reduce uncertainty,
including formal information systems, special reports, shared vision and roles of an integrator,
etc. They argue that these structural elements can drastically reduce the lead-time in planning and
provide decision makers with the necessary knowledge and information to make decisions.
Ambiguity is defined as ‘‘a measure of the organization’s ignorance of whether a variable
exists’’ (Doll and Zhang, 2001) and has been much less investigated in an R&D planning context.
It represents multiple and conflicting interpretations about a situation and ignorance about the
existence of problems or issues (Weick, 1979). Doll and Zhang (2001) emphasize ambiguity,
rather than uncertainty, as the primary source of fuzziness in planning. They argue that unclear
cause and effects, departmental differences, and an unanalyzable work process are the reasons for
equivocating in planning, and that enhanced roles of integrators or champions, shared vision and
interpretation, and introduction of interdepartmental, concurrent work processes are ways in
which front-end integrative mechanisms can succeed. We summarize the above discussion in
Table 1.
Research in discontinuous innovation also reveals that front-end planning activities are
influenced by the continuity of R&D projects (DeTinne and Koberg, 2002; Reid and de Bretani,
2004). Researchers argue that discontinuous innovation results in unstructured front-end planning
activities, largely led by capable individuals in both boundary spanning and gate-keeping activities
(Reid and de Bretani, 2004). This finding suggests that uncertainties and ambiguities in front-end
planning activities lead to an unstructured, individual-driven decision making process.
In this paper, we investigate the planning activities of two R&D projects that aim at the
development of futuristic, forefront technologies in bio and electronic technology areas. We
argue that the planning activities of these projects are particularly problematic since the project
technologies are ill defined and the derivatives and outcomes of the projects are fairly difficult to
predict. For these on-going projects, a large sum of investment has already been made; but the
final results are yet to be discovered. We first ask how the project managers perceive the planning
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 233
issues, especially the issues of uncertainty and ambiguity, and how they deal with such problems
in terms of technology planning, organizational arrangements, the R&D process design, and tools
and measures to reduce planning characteristics.
2.2. Backgrounds of large R&D projects in Korea
The National R&D Program of Korea was shaped and launched systematically in 1982 by the
Ministry of Science and Technology (MOST) and aimed at centralizing the development of core
industrial technologies. Reflecting the absence of technological accumulation in Korea at that
time, the program focused initially on narrowing the technological gap between Korea and more
advanced nations as well as fostering the capacity for the technological independence of Korea
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250234
Table 1
Causes and suggested solutions of front-end R&D planning
Factors involved in the front-end process Causes of fuzziness Suggested solutions
Timing of front-end process Lack of organized efforts for more
than half of the front-end time
Reduction of front-end time:
key stake holders should be
involved early in the front-end
process (Smith and
Reinertsen, 1991)
Decision making process Critical decisions are made
before uncertainties and
ambiguities are identified
More integrated and disciplined
decision-making process
(Ettlie, 1997); clear focus and
shared goals among the
development team (Ettlie, 1997;
Clark and Fujimoto, 1991)
Lack of top management attention Delegation of decision making
power to the R&D managers
(Khurana and Rosenthal, 1997)
Non-linear R&D process Concurrent development required Shared knowledge and
interpretation of the issues across
the involved parties (i.e., market
potential, competitive situations,
development goals, etc.)
(Doll and Zhang, 2001)
Technological uncertainties Unresolved technological
uncertainties
Articulated process of R&D
strategy formulation;
introduction of structured R&D
portfolio planning (Khurana and
Rosenthal, 1997)
Technological risk Establishment of program
structure and development
phases; compressing and
overlapping development phases
(Rosenthal and March, 1988)
Uncertainties related to market
and derivatives
Unreliable estimation of market size,
growth, and expected market impacts
Stage-Gate Review Process or
improved market/economic
impact analysis process (Beck
et al., 1999; Cooper, 1997)
by adapting foreign technologies. Therefore, during the 1980’s the program largely supported
small-scale, short-term research projects which involved imitative technological development.
A structural change occurred in the programs during the 1990’s when a number of
governmental bodies besides MOST (e.g., the Ministry of Commerce, Industry, and Energy and
the Ministry of Environment) initiated R&D programs of their own. In a nutshell, Korea pushed
forward government-led technological development projects in certain targeted areas, knowing
that it would be difficult for Korea to be equally competitive with advanced nations in all areas. A
coordinated effort in this regard across different governmental bodies was known as ‘‘the Highly
Advanced Research Project’’ (HAN project). The project was also called the G-7 program
because it aimed at catching up to G-7 nations in technological power. The primary objective of
the program was to establish world-class technological competitiveness in core industries such as
HD TV, broadband ISDN, artificial intelligence, flexible manufacturing systems, medical
technology, substitute energy, etc. This program provided intensive support over more than 5
years for strategic technologies with strong industrial applications. Although the program set
priorities for strategic support and selectively invested R&D funds, it was still designed after the
‘catch-up’ model of technology development.
Moving into the 21st century, national R&D programs in Korea have been experiencing a
paradigm shift. This shift can best be described as a re-orientation of R&D investment from the
strategy of ‘‘catch-up’’ to that of ‘‘technology leadership.’’ This shift emerged from the realization
that it was now much more difficult, if not impossible, for economic growth to continue under the
old model. This realization was driven home after the so-called ‘‘IMF crisis’’ in Korea, a period of
economic downturn which drove the entire Korean economy to near bankruptcy in the late 1990s.
The new thinking is that if Korea ever wants to be an advanced nation in the world, it is imperative to
attain dramatic and creative technological breakthroughs and abandon old ways of obtaining and
copying standardized technologies originating in other nations.
Some examples of such efforts include the Frontier 21 program funded by MOST and the
Vision 21 program pushed forward by the Korea Institute of Science and Technology, a leading
R&D institute in Korea. The two programs have the common aim of attaining world-class
leadership in emerging technological fields. These R&D programs are in sharp contrast to the old
R&D programs in many ways. First of all, the old programs were generally separate, single
projects usually performed over a short-term time horizon, but the new programs are
characterized by a large investment systematically distributed over a long-term period, frequently
over 10 years, and organized by a consortia of research institutes, universities, and/or the research
labs of private firms.
Secondly, the new programs have introduced a responsible management system in which a
project leader has the authority and responsibility to oversee all the activities of the entire R&D
cycle, including R&D planning, technology development, testing and evaluation, and
management of final outcomes. Such an R&D management system was foreign to Korea
prior to the Frontier 21 program. Under the old paradigm, program coordinators were simply
expected to produce anticipated outcomes and were subjected to fairly loosely organized
managerial structures. Thirdly, the new programs pursue future frontier technologies by bringing
different technological fields together to work toward a common end result. Consequently, unlike
the old ways where ‘‘compressed R&D’’1 was possible, frontier technologies are characterized
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 235
1 Compressed R&D refers to developmental activities where a set of technologies is utilized as a unit without
investigating the inherent in-depth technological principles.
by uncertainty and ambiguity with respect to innovation contour and direction, technological
content, and emerging technological trends. Thus, it is difficult for researchers to have a full set of
defined objectives or to understand the utilities of the target technologies under consideration.
Therefore, although technological uncertainty and ambiguity are considered to be innate
characteristics in any planning process, these characteristics are amplified and strengthened in
cases involving cutting-edge technology R&D. Because of this, R&D activities for leading
technologies require special attention as regards research planning, and a much different
approach from previous rounds of government-funded research programs. Most importantly, we
need to acknowledge the embedded technological and market uncertainties in such R&D
activities and devise a continuous improvement mechanism which can allow for corrections to
the elements of the process in response to technological adaptations and changing environmental
conditions.
3. Two case studies of large R&D projects
3.1. Methodology: case study method
The case study method is frequently applied to the theory building stage of research, where
key constructs are not fully understood. It provides a rich context from which we can identify
isolating mechanisms and influencing factors (Soderlund, 2002; Eisenhardt, 1989). In this study,
the idea of reflexive interpretation was employed as follows (Alvesson and Skoldberg, 2000).
Since the case study was performed years after the project was first conceived, it was subject to
several possible interpretative errors. Factors affecting the sources of case studies include ex-post
recollection bias, contamination of memories by ex-post events, emergence of unknown facts,
and arrivals of new issues, to name a few. In order to deal with such possible problems, we
constructed the case study model as follows: (1) multiple interview sources, (2) thorough reviews
of official documents published at the time of planning (i.e., research proposals, evaluation
records, related government documents, etc.), (3) reviews of unofficial documents (i.e., meeting
minutes, annual reports of the Spintronics Study Group, etc.), and (4) testimonies of third parties,
including planning experts who were a part of the planning process. The model is depicted in
Fig. 1.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250236
Fig. 1. Case study model.
For the current study, we chose two cutting-edge technology development projects, the
‘‘Intelligent Microsystem Project’’ (IMP), chosen from the Frontier program of MOST, and the
‘‘Spintronics Project,’’ supported by the Vision 21 program of KIST. Both projects fit exactly into
the ‘‘new’’ breed of government-funded research programs, characterized by a large project fund,
long development time, a highly complex developmental process, high technological and market
uncertainties, etc.
The investigation was performed in the following order. We first reviewed substantial numbers
of official and unofficial documents released from the project managers to gain an overall view of
each project. The specific project process for each case was then investigated in detail in a
chronological order to obtain rich stories about the context of the planning process, both
organizational and technological. For this purpose, we performed a series of structured
interviews for each project over the period of December 2003–July 2004. In order to overcome
possible biases, we carefully selected interviewees.
The interview sources for each project were composed of two to three project leaders/
coordinators, three to four planning team members, at least one planning specialist who took part
in the actual planning process, and top managers and other related third parties at KISTat the time
of planning and program funding. The third parties were primarily used to verify statements
made by the project leaders and planning team members. All the interviews were documented
and sometimes reconfirmed in order to validate the statements. ‘‘Within-case analysis’’ and
cross-case analysis’’ (Eisenhardt, 1989) were subsequently applied to identify, first, the
differences and similarities between the two cases, and second, important emerging patterns and
key constructs shaping the patterns. By carefully looking into each case from the inside, we
identified several issues to which we paid special attention in the construct building process.
While following the idea of reflexive interpretation (Alvesson and Skoldberg, 2000), our
investigation centered around key constructs and constituent elements of the planning process. We
first looked into the key conceptual constructs which define the determinants, or factors, of the
planning process. The factors already identified and suggested in the prior literature include
technological/market uncertainties and ambiguities, the complexity of the planning process,
planning time, and top management involvement, to name a few. Our goal through this exercise was
to find out how the organizations dealt with various issues associated with these factors, and what
might have been the consequences of the lack of proper responses to the issues raised in the process.
The two projects shared many similarities, especially in the project planning stages. Both were
publicly funded, long-term R&D projects with inherent technological uncertainties and risks.
While both projects had high strategic importance to Korea as well as KIST, there were many
factors yet to uncover. In some respects, technological feasibility was even in question. The
projects were also organized in a similar manner. They were, however, different in many respects,
especially in their technological contexts, managerial requirements, and development objectives.
One of the fundamental differences between them was the development goal, and consequently,
technological applicability: While IMP was a product-oriented, demand-driven project; the
Spintronics Project was to develop source technologies that could be applied to next-generation
electronics products/industries.
3.2. The Intelligent Microsystem Project case
3.2.1. Project description
The IMP program was first announced in 1999. As a part of the effort to develop ‘‘source’’
technologies which would create new, leading industrial clusters in 5–10 years, the Korean
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 237
government initiated the Frontier R&D Program in the same year. The IMP was one of the first
projects selected by the government. The objective of this 10-year project was to develop an
endoscopic microcapsule and micro biomedical diagnostic system (PDA) in the field of
biomedical appliances, both of which utilize micro electronic mechanical systems (MEMS)
technologies. The markets for both were expected to grow fast to reach about US$ 100 million for
endoscopic microcapsule technology and US$ 2 billion for micro-PDA technology within 10
years. In the development process, a host of products and parts would be developed along with a
network of new enterprises. The government planned to fund the project for the ensuing 10 years,
starting from 2000, in the amount of US$ 9–10 million per annum (Tables 2 and 3).
3.2.2. Development process
The IMP aimed at developing forefront high-tech products utilizing microsystem technology.
When the RFP was issued by the Ministry of Science and Technology in early 1999, a planning
task force team (TFT) of 20 scientists from KIST was organized almost right away, and after
spending about 6 months of planning time, laid out a 10-year development plan. The 20 scientists
involved in the planning process later became the core of the development team. One of the most
important elements in the planning process turns out to have been the existence of a project
‘‘champion.’’ One of the scientists who was involved in the project from the beginning said,
‘‘We were able to win the project because some key people were convinced that we could
do it. We met people in the government, universities, and other research institutes and
persuaded them that we could do it and do it best. Without these people, the project might
have landed in the hands of others, who seemed more competitive at the time.’’ (Interview
with the program coordinator M).
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250238
Table 2
Objectives for each phase
1st phase 2nd phase 3rd phase
Year
1999
Year
2000
Year
2001
Year
2002
Year
2003
Year
2004
Year
2005
Year
2006
Year
2007
Year
2008
Year
2009
Targets Visual endoscopic
microcapsule
Multi-functional endoscopic
microcapsule
Endoscopic microcapsule for
total GI track
Micro-PDA Micro medical diagnostic
system
Micro medical diagnostic &
system embedded drug delivery
system
Table 3
Investment plan
1st phase 2nd phase 3rd phase Total
Year
1999
Year
2000
Year
2001
Year
2002
Year
2003
Year
2004
Year
2005
Year
2006–2009
Government 3 10 10 10 7.5 8.6 10 40 99
Industry 1.2 5 40 50 1.8 30.4 10 37.7 67
Total 4.2 15 140 150 9.3 11.6 20 77.7 165
Unit: US$ million.
There were initially three groups that submitted the RFP to the government: KIST, Samsung
and Korea Electronics Technology Institute (KETI). According to the interviewees, the
government selected KIST, based not just on the competitiveness of each research proposal, but
on many other factors such as track records, probability of success, distribution of related projects
over research institutes, government decision makers’ attitudes, and proponents of the projects.
The planning and initial decision making process reflected both technological and non-
technological processes. The government specified in the RFP what they wanted as the final
products: forefront technological breakthroughs which could generate new industries and jobs on
the one hand, and source technologies on the other. These two objectives were not entirely
consistent and could not be met simultaneously in a meaningful way, but the TFT had no choice
but to satisfy the demand. In retrospect, the program coordinator said (Table 4):
‘‘We were to pursue two rabbits running in two different directions at the same time. We
had to do it, and we did it awfully. In the first stage of the project, we placed a heavy
emphasis on final products: that is, on how to get to the final products, specified in the
proposal, with all the necessary technological elements in it. But, in the second stage of the
project, the project emphasis was reversed and we are now focusing on the source
technologies . . .. It is an irony that we did it this way. Normally you come after the source
technology first and then final products, but [with all other factors involved in the process]
we did it in exactly the opposite way.’’ (Interview with a program coordinator M).
In order to accomplish the objectives of the project, many technological breakthroughs were
necessary because of the technological obstacles in many aspects of the R&D process. There were
inherent uncertainties and ambiguities in identifying source technologies and other technological
details. In the case of the endoscopic microcapsule, some aspects of the technology roadmap (TRM)
or technological details were found to be unclear and ambiguous. The TFT knew the difficulties
from the beginning, but could not wrestle with the problems long enough to develop a solution
during the planning stage because of the upcoming deadline for the proposal. The technology
roadmap thus contained critical paths which reflected unclear technological feasibility, but were
important for project success. For example, in the case of the development of the endoscopic
microcapsule, three key technological fields, including bio MEMS, optical MEMS, and
microcapsule technologies, had to merge their efforts to produce one final product. Several
technological obstacles were expected, but a feasibility study was not fully performed at the time.
Compounding these problems was the lack of technological expertise necessary to carry out
project planning. Because the project involved forefront, emerging technologies, no one could
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 239
Table 4
Two main categories of the IMP program
Micro medical appliances Micro information appliances
Final products Endoscopic microcapsule Micro-PDA/micro medical diagnostic system
(Personal Digital Assistant)
Types Capsule type Wrist-watch type
Possible
contributions
- To develop high-valued products,
components, technologies
- To have possibility to create world-class markets
- To promote national medical welfare - To be suitable for large enterprises or
industrial consortiums
- To be suitable for medium-sized
enterprises and venture businesses
claim to have adequate knowledge of the field. This inherent deficiency combines with the lack of
field experts in the development process of many emerging technology projects. For instance, in
the IMP case, about 20 scientists took part in the planning process, of whom only 2–3 had
extensive knowledge of the MEMS technologies under consideration. This lack of necessary
expertise directly resulted in a planning deficiency, complicating the actual development process.
The whole project was composed of numerous sub-projects in which technology-oriented
companies, researchers, professors from other research institutes and universities took part from
the beginning. For instance, the endoscopic microcapsule project had about 20 sub-projects
whose final success was critically dependent upon one another. The management of about 25
participating groups and networks soon became the central duty of the project director (Table 5).
Another problem arose due to the time limitation for proposal submission. Because of the
requirements of the RFP, the researchers had only 6 months for proposal preparation. The TFT
had a meeting almost every day, but even then they did not have enough time to consider all the
elements in the project. Considering that the IMP represented one of the largest government-
funded R&D projects, the TFT had too short a planning period to generate a coherent and
comprehensive plan. A deficient planning process combined with technological uncertainty and
ambiguity can lead to bigger problems in later development stages such that the component
technologies cannot be merged as expected. The development project of the endoscopic
microcapsule was a good example of this phenomenon. Although in the planning process all the
essential technologies such as optical MEMS, bio MEMS, micro battery, etc., were expected to
merge together to produce the endoscopic microcapsule, in actual fact, the researchers found that
there were significant technological obstacles in doing so.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250240
Table 5
Sub-projects of IMP
Categories Sub-projects
Endoscopic microcapsule Development of intro-body motion mechanism
Medical assessment of endoscopic microcapsule
Basic research on intro-body motion techniques
Development of video-image production module using invisible light
Ultrasonic imager for endoscopic microcapsule
Development of source technology for ultra-minute telemetry module
Development of microfluidics source technology and micro tool for
diagnosis and treatment
Development of pre-test treatment and assessment technology
Design and production of detection device using functional membrane
Micro-PDA (phase 1) Development of power-saving micro drive storage device
Development of micro cordless input device
Development of virtual micro-display
Development of front-end module for cordless transmitter and source
technology for integrated circuit
Common technologies Development of high-power micro battery
Development of micro rapid prototyping technology
Development of micro machining center
Development of micro fabrication cell
Development of creative source technology
System integration System design, development, and assessment
3.3. The Spintronics Project case
3.3.1. Project description
The second case study is the ‘‘Spintronics’’ project, first officially introduced in 2002 to achieve
two purposes: to develop the fabrication technology and materials for spin electronic devices on the
one hand, and spin-photonic devices and magnetic SAW technology on the other. Technological
motivation for the project lay in the limitation of the current charge-based electronics with respect to
size reduction, performance, etc. Spin electronics (Spintronics) is an emerging electronic
technology which bypasses the limitations of current electronics by using the spinning
characteristics of electrons and photons. Examples of spintronic technologies are shown in Fig. 2.
Spintronics is known for having some notable advantages over traditional electronics such as
speed, non-volatility, and low susceptibility to environmental influences (reliable performance).
Low power consumption is also an attractive aspect of spintronics. In the first stage, which
encompassed the first three years of the project, KIST planned to infuse US$ 4 million per
annum. Large research facilities and devices, such as those found at the Micro/Nano Fabrication
Center, and facilities for thin film fabrication as well as device processing and measurement,
were already established or in the process of establishment. Depending upon the progress
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 241
Fig. 2. Types of Spintronic Technology.
Table 6
Objectives for each phase
1st phase 2nd phase 3rd phase
Year
2002
Year
2003
Year
2004
Year
2005
Year
2006
Year
2007
Year
2008
Year
2009
Year
2010
Year
2011
Targets � Spin injection technology � Spin-FET device design,
fabrication, testing
technology
� Spin electronic device application
technology
� Spin transport control
technology
� Magnetic semiconductor
critical temperature
improvement technology
achieved and the appraisal of research results after the first 3 years, the funding was scheduled to
continue for about 10 years (Tables 6 and 7).
3.3.2. Development process
The project was developed in a traditional top–down planning framework. In 1998, the top
management of KIST was seeking two to three large projects in the core technological fields of
KIST, i.e., nanotechnology, biotechnology, environmental technology, etc., which would draw a
large sum of research funds to KIST while at the same time, conform to the research direction of
KIST. By then, the Spintronics Study Group had already been active for a while, and had
produced some preliminary documents on the necessity and possibility of performing successful
research in spintronics (Institutional Reports on Vision 21 Projects, 1999). In early 1999, top
management requested the division directors of KIST to prepare research proposals with a
required investment of about 5 billion won (about US$ 4 million) per annum. Research proposals
were then reviewed by the proposal assessment team which was composed of top management
and outside specialists in the respective research areas. Ultimately, two projects were selected:
the spintronics project from the future technology division, and the chemo-informatics project
from the biotechnology division.
Since the project was the outcome of an institution-wide search and deliberation to shape the
core technological fields of the future at KIST, top management support was very strong from the
beginning. When the decision was made at the top, the division head immediately formed a task
force team composed of three scientists who had previously been involved in related research
projects like GMR Head, MRAM, and Photon device. The team was later strengthened to include
other experts from both inside KIST as well as from other institutes. The team was the central
force in the planning process and also in the subsequent development process.
Top management support was a strong motivational factor in the entire process. Spintronics
was an area noted for its technological complexity and difficulty. It had been pursued by
technology leader nations, such as the US, Japan, and Germany, for decades without much
progress. Under such challenging circumstances, top management support was probably the
single most important reason that the Future Technology Research Division decided to pursue the
project. One of the project coordinators recalled:
‘‘The project was initiated at the top. We were all thrilled at being able to pursue such a
daunting task—I mean, such a highly difficult technology development project. You see,
when the bottom-line decision is already made, you don’t really have to be much concerned
about defending yourself, about the necessity (of the project), budget requirement,
infrastructure, etc. You just need to check how you go about doing it, of course to make a
success story.’’ (Interview with a program coordinator H).
In the idea generation and project design stage, the TFT faced numerous problems in
performing technology competence analysis. First of all, there were not enough specialists in the
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250242
Table 7
Investment plan
1st phase 2nd phase 3rd phase Total
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Institutional funding 5.4 6.5 8.6 10 10 12 12 12 15 15 106.5
Unit: US$ million.
field of spintronics at KIST, especially in the areas of quantum theory and manufacturing devices.
At first, the TFT was less concerned about the inadequate presence of experts, as they believed
that internal education and external recruitment of experts would resolve most of the problems.
However, it soon became clear that in such a highly advanced technological field, knowledgeable
experts with adequate experience were scarce even in the most advanced nations. In fact, building
a team of experts with creativity and experience in the field was the most critical element for the
success of the project. Insufficient technological leadership in the project was later magnified and
reflected in the actual development as well as the planning process. The technological feasibility
of each sub-component of the project could not be checked in sufficient detail. Furthermore, the
project as a whole and its key components were inadequately evaluated and prepared.
Compounded with the lack of technological experts, the complexity of the project required the
participation of many outside professionals from universities and other research institutes.
Professors from Seoul National University, Korea University, Yonsei University, etc. got involved
from the early stages of project development, even including the preparation of the proposal.
These experts became the sub-project leaders for the subsequent stages of development. For
example, in 2002, the first year of the actual project, 14 outside research teams were involved in
the project. As strategic cooperation was an essential part of the project, issues surrounding
project management sprang up even in the planning stage.
More than half of the project funds were to be spent by the extramural research partners who
were geographically dispersed and thus difficult to monitor. The sheer complexity of the project
required many sub-projects to be performed simultaneously. However, geographically dispersed
sub-project leaders (more than 15 scientists at the first stage) made such a tight project
management scheme highly problematic from the beginning. Managing the network of
participating scientists became one of the most difficult tasks of the entire project. This problem
continuously haunted the management team of the project as each group focused mainly on their
portion of the project without looking at the entire picture (Fig. 3).
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 243
Fig. 3. Technology roadmap of Spintronics Project.
Another problem was related to the large scope of equipment and facilities required for the
successful completion of the project. A Micro/Nano Fabrication Center with ‘‘Clean Room’’ had
already been established before the project started, but spintronics needed much more specialized
facilities for thin film fabrication and device processing and measurement. In the first three years
alone, more than 5 billion won (about US$ 4 million) was used to purchase and develop facilities
and devices such as UHV Sputter, MOCVD, nano imaging and patterning equipment, AGM, etc.
The facilities and equipment also needed to be concentrated in one location. Given this situation,
what bothered the TFT the most was how to optimize the use of limited funds so as to purchase all
the necessary equipment for research while at the same time to provide sufficient support to all
the external research groups. The large number of independent groups participating in the project
resulted in managerial havoc in the actual development process.
There are other interesting points highlighted in this case. One of the technological objectives
the TFT initially set was the development of photonic devices. As research continued in this area,
it became obvious that the expected results could not be produced. Two years into the project,
management made the painful decision to discontinue that aspect of the project. This exemplifies
the riskiness associated with R&D projects for emerging technologies and shows again how
critical R&D planning is in such projects. When planning is insufficient, the project suffers from
a waste of time, money and effort at its later stages. Another related issue is the role of planning
specialists in such projects. In the spintronics case, the TFT brought in a planning specialist only
in the later stages of the project, but the specialist’s role turned out to be critical in finalizing the
planning process. As opposed to the original proposal which emphasized the utilization of an
internal workforce, the planning specialist provided fresh input, emphasizing a network of
external research groups as the prime workforce of the project. This idea was accepted and
changes were made accordingly. The planning specialist also suggested making clear in the
proposal that heavy investment in facilities and equipment was needed for the success of the
project, arguing that emerging technologies could be adequately researched only with the support
of proper technological devices and equipment. This suggestion turned out to be valuable as fund
expenditures were now in sync with the plan.
4. Analysis and discussion
4.1. Implications for large-scale development projects
Large-scale, forefront technology R&D projects often end up with fairly complex research
structures and face various types of uncertainty and ambiguity, sometimes beyond the capacity of
participating researchers to handle. Such projects require many sub-projects to be processed
simultaneously. Consequently, the overall size of the project gets larger, and the innovation
process becomes non-linear. Oftentimes in such R&D projects, the boundary between basic
research and applied research becomes blurry; the technology life cycle becomes shorter, and
basic research moves directly to industrial development without going through the stage of
applied research. Therefore, all the R&D efforts at different stages of technological development
interact, forming the networked innovation pattern. As such, patterns of innovation related to
industrial development have a direct impact on patterns of innovation in both basic and applied
research. This situation characterized the two cases dealt with in this paper. We first provide a
summary table of the two projects (Table 8).
As the summary table shows, one notable difference between the two was in the pre-planning
activities. In the IMP case, program planning was already performed by the technocrats at MOST,
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250244
who initially organized the Frontier Program. Basic requirements for the project were already set
forth by the government, eliminating the possible confusion that is often entailed in such a long-
term, large-scale project. By contrast, in the case of the Spintronics Project, a prolonged period of
information searching and the development of a preliminary (technological) feasibility study was
done by the Spintronics Study Group.
4.2. Implications for management of front-end planning activities
The two cases share similarities and differences with respect to the foci of planning and factors
influencing the process. Both suffered from technological uncertainties and ambiguities so that
researchers were not sure of the technological feasibility of the various R&D activities and
development goals. The missing links were not thoroughly reviewed or their trends investigated. For
example, the endoscopic microcapsule project had some big missing links for which the researchers
hoped to find solutions later while performing the actual project; however, the project had to be
significantly modified later on. These weaknesses were compounded by the lack of planning know-
how and/or proper experts to provide guidance and assistance in such a situation. Adequate
investigation of enabling technologies would have reduced the waste of resources later in the actual
R&D process. It is inevitable that research focusing on emerging technologies experiences similar
problems. R&D objectives for government-funded projects are often imposed by the government,
which distorts the overall research orientation significantly. Our research found that clear and
consistent objectives and goal-oriented planning are important for a successful project outcome.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 245
Table 9
Key factors of R&D planning and their influence: IMP case
Key factors of planning Impacts Outcome/influence in planning and beyond
Project champion + Strong motivation factor
Participation of core scientists + Congruence of planning with actual R&D
Limited planning coverage � Unrealistic business development planning
Participation of experts from other institutes + Creation of R&D network
Insufficient technological expertise � Key problem factor in both R&D and planning
Inadequate use of planning tools � Uncertainties/ambiguities undiscovered
Time limitation � Inadequate planning
Table 8
A summary of the two case studies
Comparison items IMP case Spintronics case
Funding sources Frontier program of MOST Institutional funding by KIST
Planning process
characteristics
� Program planning done by MOST � Preliminary info. Search and planning
done by Spintronics Study Group
� Tight schedule management due to
submission deadline
� Loosely organized planning team
� Planning done by structured task force team � Final decision made in the context
of organizational choice
� Planning driven by project champion
Key factors of
winning the projects
� Project champions � KIST-university research network
� Cordial KIST-technocrat relationship � Long-term institutional support
� Core group of competitive scientists � Top management support
We also found that depending on the project characteristics planning activities might be
geared in different directions. For example, for the IMP project, the existence of a project
champion and government support played significant roles in the decision making process while,
for the Spintronics project, emphasis was placed upon top management support and early
participation of core researchers. It was interesting to note that not many planning tools were
used in either project besides a technology roadmap. This was primarily because of the time
limitation and lack of a planning budget. For both projects, planning was thought to be highly
risky, but budgetary support was not adequately secured.
We summarize our findings in Tables 9 and 10. The identified major factors are helpful for
reducing the uncertainty and ambiguity in the planning process of a large R&D project and their
influence on the later stages of the R&D process. In the IMP case, the existence of a project
champion and the participation of core outside experts early on in the planning process was found
to have a positive influence on the R&D planning/actual development process, while insufficient
technological expertise, lack of planning tools, and a planning time limitation all affected later
processes negatively. In contrast, the planning process of the spintronics case was influenced
positively by a matrix planning structure, the participation of planning experts, an excellent sub-
project management structure, and adequate planning for the acquisition and utilization of
research facilities. It was negatively influenced by insufficient technological expertise, high
technological complexity, and a loosely organized planning process. Top management support
was found to work in both ways: the impact could be positive as a motivation factor, but negative
because it could lead to a decision making process based on institutional concerns rather than
technological or market demand reasons.
Based on these observations, key constructs and corresponding factors of front-end planning
activities are identified. As shown in Fig. 4, we developed a five-construct model, including
Proper Planning Environment, Organizational Support, Goal Orientation, Knowledge Sharing,
and Technical Expertise. Each construct reflects several key factors of front-end planning, which
in turn stem from major obstacles, or fuzziness, in R&D planning.
Proper Planning Environment refers to the adequacy of planning time and budget, and the
availability of planning tools and know-how. Organizational support measures in-house and
external support for the project. Goal Orientation indicates the degree of goal congruence and
goal-orientation of planning. Similarly, Knowledge Sharing deals with the degree of sharing of
knowledge, vision, and process among the participants. Finally, Technical Expertise is the
construct that asks whether technological expertise and enabling technologies are available
and to what degree they are used. We believe that the model provides major tools to guide the
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250246
Table 10
Key factors of R&D planning and their influence: spintronics case
Key factors of planning Impacts Outcome/influence in planning and beyond
Top management support +/� (+) Key motivation factor
(�) Institutional decision making factor
Matrix/network planning structure + Planning scope expanded
Insufficient technological expertise � Key problem factor in both R&D and planning
Involvement of planning experts + Realistic R&D planning
High technological complexity � High technological uncertainties/ambiguities
Sub-project management/facility planning + Creation of R&D network
Loosely organized process � Inadequate R&D planning
front-end planning process for large, risky R&D projects focusing on emerging technologies, and
thus, to analyze the adequacy of the front-end planning activities.
4.3. Implications for large-scale, forefront technological R&D projects
Albeit problematic in many respects, the IMP and spintronics cases provide valuable lessons for
planning large-scale, frontier technology R&D projects lessons which can influence the success of
the project significantly. First, comprehensive pre-planning and planning activities are imperative
for those R&D projects with high technological uncertainty resulting from aiming for technological
breakthroughs. By performing a technology feasibility study using planning tools such as a
technology tree, R&D portfolio analysis, technology roadmap, and/or TRIZ, project teams would
be able to reduce uncertainty significantly. Even though it would not be possible to eliminate innate
uncertainties and ambiguities completely, the uncertainties could be tested and calculated to see if
the obstacles could be handled and treated properly in the actual development process. The two
cases we assessed in this study provide ample evidence that it is better to be prepared in the
beginning than be sorry later, wondering if the objectives of the project can ever be achieved despite
spending millions of dollars. Second, a related issue is the ability to locate early on the ‘‘enabling
technology’’ which could be used to overcome the technological uncertainties and fill the gaps in
the technological roadmap. Such enabling technologies can act as a catalyst in the process of
technological breakthroughs. TRIZ is known to be useful for locating such technologies.
Third, the objectives must be consistent and coherent so that they can be translated into
comprehensive and detailed project plans. Oftentimes, a government-funded project requires
researchers to achieve a broad set of conflicting objectives, which places a great deal of pressure
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 247
Fig. 4. Key factors and major constructs of uncertainty/ambiguity reduction model.
upon the researchers and the project as a whole. The Frontier program seemed to make just such a
mistake because the government wanted participating scientists to develop source technologies
that could influence many related fields simultaneously, and final cutting-edge products that
would generate industries of the future. This inconsistent message made the entire project so
confused and complicated that investment had to be made in widely dispersed sections of the
R&D chain, with emphasis being placed on source technology development in one stage and on
final products in the other.
Fourth, the TFT must have ‘‘champions’’ for the project. Champions are the people who have
absolute confidence in the project and are ready, at any cost, to support it. When the technological
feasibility of the project is ambiguous, and project success uncertain, project champions can
provide the necessary impetus to get the project launched and to keep it running. In the case of IMP,
although 20 scientists participated in the planning process, only 2 were champions with adequate
expertise and the necessary confidence to be an integral part of the planning process and push the
process forward to completion. Because of their dedication, KIST was, in fact, able to win the
project from the government in the midst of fierce challenges by the other two competitors.
Fifth, in the case of large-scale, frontier technology development projects, any attempts to
save time, money, and effort in the planning process must be opposed vigorously. In such
projects, there are numerous missing, ambiguous, uncertain links and relations in the project/
technology roadmaps, and thus the planning team must spend time and money to identify the
problems, issues, and difficulties beforehand and develop corresponding solutions, or at the very
least, directions. Without this comprehensive preparation process, development teams are likely
to find themselves in a rocky situation deep into the project after wasting millions of dollars.
Sixth, the planning process needs to be translated into the actual R&D process in a seamless
manner. This requires the early participation of all the major project leaders in the planning
process and the sharing of goals and objectives among these major participants. Both the cases
we analyzed in this paper commonly showed this characteristic as reflected in the fact that the
members of the TFT were later appointed as the sub-project leaders. This ensured that the whole
project was consistent and object-oriented, which is especially important for large-scale, frontier
technology development projects.
4.4. Implications for planning success
Planning success can be measured from two perspectives: winning project funding and the
successful execution of the project. In normal, incremental innovation cases, the probability of
the latter (successful execution of the project) may constitute the most vital criteria for winning
the project. In the case of complex, forefront technology development cases, however, other
factors are found to interact in the project selection process, which may or may not be related to
actual project success. For example, In the IMP case, it seems that the existence of project
champions and the cordial relationship between responsible technocrats and KIST were
instrumental factors in winning the project. In the spintronics case, it was top management’s
vision for the future contour of KIST that ultimately determined the program selection.
This finding suggests that when market and technological uncertainties and ambiguities are
high and the risks associated with the projects are difficult to calculate, R&D organizations look
to other sources for decision making. Institutional ties, top management support, or the existence
of project champions are a few examples that have proven to be vital for project selection under
such circumstances. This observation is consistent with Reid and de Bretani (2004), who find that
discontinuous innovation tends to be unstructured and depends on capable individuals rather than
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250248
structured planning systems. How such factors are, in fact, related to actual project performance
is yet another research question to be investigated in the future.
5. Conclusion
In this case study, we looked at the structural, technological elements that comprised the
planning stage of large-scale, forefront technology development projects supported by the
Korean government. We provided a five-construct R&D planning model which can be used to
deal with the inherent ambiguity and uncertainty in such research projects. Though the model has
been derived from Korean R&D experiences, we believe that it is still applicable for any large,
forefront technology R&D project. We argue that a comprehensive planning process is
imperative for such R&D projects because the complexity of the project exposes any insufficient
planning, and when ignored, always leads to much higher costs later in the actual development
project. A network of participating researchers and experts with adequate knowledge backed by
top management support, needs to be secured early on in order to minimize the vagaries of such
research. We also found that due to the extensive networking that occurs in actual research, vision
sharing and coordinated structural arrangements are essential, even during the planning process.
Some elements such as political and government support also seem to be important in carrying
out such large-scale projects that have high national visibility. Our findings suggest that in the era
of international cooperation, adequate upfront planning of R&D activities must include ample
planning time, a reasonable budget, qualified experts, and the identification of clear goals in order
to optimize the investment and R&D activities at later stages of the project. As a further avenue of
study, this model should be enriched and tested for its relevance and applicability to large-scale,
forefront technology R&D projects in general.
References
Alvesson, M., Skoldberg, K., 2000. Reflexive Methodology. Sage Publications, London, UK.
Bacon, G., Beckman, S., Mowery, D., Wilson, E., 1994. Managing product definition in high technology industries: a pilot
study. California Management Review (Spring), 32–56.
Beck, D.F., Boyack, K.W., Bray, O.H., Siemens, W.D., 1999. Bringing the fuzzy front end into focus. Working paper,
Sandia National Laboratories.
Clark, K.B., Fujimoto, T., 1991. Product Development Performance: Strategy. Organization, and Management in the
World Auto Industry. Harvard Business School, Boston, MA.
Cooper, R.G., 1997. Fixing the fuzzy front end of the new product process: building the business case. CMA 71 (8), 21–23.
DeTinne, D.R., Koberg, C.S., 2002. The impact of environmental and organizational factors on discontinuous innovation
within high-technology industries. IEEE Transactions on Engineering Management 49-4, 352–364.
Doll, W.J., Zhang, Q. 2001. Clarifying the fuzziness in the concept of front end fuzziness: a dual theoretical rationale.
Working paper, University of Toledo.
Eisenhardt, K.M., 1989. Building theories from case study research. Academy of Management Review 14, 532–550.
Ettlie, J.E., 1997. Integrated design and new product success. Journal of Operations Management 15 (1), 33–55.
Galbraith, J., 1977. Organizational Design. Addison-Wesley, Reading, MA.
Gupta, A., Raj, S., Wilemon, D., 1986. A model for studying R&D-marketing interface in the product innovation process.
Journal of Marketing 50 (2), 7–17.
Khurana, A., Rosenthal, S.R., 1997. Integrating the fuzzy front end of new product development. Sloan Management
Review 38 (2), 103–120.
Khurana, A., Rosenthal, S.R., 1998. Towards holistic ‘front end’ in new product development. Journal of Product
Innovation Management 15, 57–74.
Kim, J., Wilemon, D., 1999. Managing the fuzzy front end of the new product development process. In: Proceedings,
PICMET ’99, vol. 2.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250 249
Moenaert, R.K., De Meyer, A., Souder, W.E., Deschoolmeester, D., 1995. R&D/marketing communication during the
fuzzy front end. IEEE Transactions on Engineering Management 42 (3), 243–258.
Montoya-Weiss, M.M., O’Driscoll, T.M., 2000. From experience: applying performance support technology in the fuzzy
front end. Journal of Product Innovation Management 2 (17), 143–161.
Onno Omta, S.W.F., de Leeuw, A.C.J., 1997. Management control, uncertainty, and performance in biomedical research
in universities, institutes and companies. Journal of Engineering and Technology Management 14 (3), 223–257.
Reid, S.E., de Bretani, U., 2004. The fuzzy front end of new product development for discontinuous innovations: a
theoretical model. Journal of Product Innovation Management 21–23, 170–184.
Rosenthal, S.R., March, A., 1988. Speed-to-market: disciplines for product design ad development. In: Executive
Summary of Research Findings and Conference Proceedings, Boston University School of Management, Manu-
facturing Roundtable.
Smith, P.G., Reinertsen, D.G., 1991. Developing Products in Half the Time. Van Nostrand Reinhold, New York.
Smith, R., 1998. Managing innovation: integrating technological, market and organizational change. Journal of
Engineering and Technology Management 14 (3), 111–113.
Soderlund, J., 2002. Managing complex development projects: arenas, knowledge processes, and time. R&D Manage-
ment 32 (5), 419–430.
Weick, K.E., 1979. Social Psychology of Organizing. Addison-Wesley, Reading, MA.
Wheelwright, S.C., Clark, K.B., 1995. Leading Product Development: The Senior Manager’s Guide to Creating and
Shaping the Enterprise. The Free Press, New York.
Y.-I. Song et al. / J. Eng. Technol. Manage. 24 (2007) 231–250250