This document discusses using a complex adaptive systems approach and fitness landscape theory to understand technology management. It presents a model called the strategy configuration chain that illustrates how organizations classify, select, adopt and exploit technologies through four linked tasks to develop capabilities and a technological configuration. The document also introduces an NK model that can represent an organization's technology strategy as a string of bits and calculate a fitness value for different strategic configurations based on their potential for survival and competitiveness. This provides a way to visualize different strategic options and their relationships on a fitness landscape.
2. 729 Technology management – a complex adaptive systems approach
certainties (operations) [2]. It argues that organisations need to effectively design and
manage configurations (forms and ways of working), capable of achieving and sustaining
the balance between innovation and operational excellence.
With this background, technology management is a practice that entails the
classification, selection, adoption and exploitation of the technologies needed to maintain
an organisation’s current and future survival. It involves a collection of management
disciplines that have responsibility for:
• product technology – the customer purchase
• process technology – the systems utilised to design, manufacture and deliver the
product or service
• management technology – the knowledge and decision-making processes that plan,
manage and control the organisation.
This paper asserts that understanding the factors and events that underpin innovation and
technological change is an evolutionary and systems issue that currently lies at the
junction of information theory and economic history. To recognise the patterns of
evolutionary change, economists have typically studied technological discontinuities as
an aggregate phenomenon to comprehend the nature of competition. Such work has
focused on the policy of innovation [3,4], and the process of change [5,6]. This paper is
influenced by such research, but is not concerned with the teleological theories of
historical change (evolutionism). Rather, it is motivated by the desire to understand
technological evolution as a process of variation, selection, retention and struggle by
developing, integrating and applying fitness landscape theory to technology management
issues.
With this introduction, this paper is organised as follows. Section 2 explains the
notion of complex systems and complex adaptive systems. Section 3 introduces fitness
landscape theory and an accompanying analytical model (the NK model) that represents
the evolutionary properties of complex systems. This section also defines the unit of
analysis (the technology strategy and resulting organisational configuration) by
presenting a conceptual model referred to as the strategy configuration chain. This is
then related to fitness landscape theory to understand the role of evolution in
technological and operational competitiveness. Section 4 discusses the relevance and
implications of these theories and models to the strategic management of technology. It
provides a definition and model of technological fitness, and relates this to four
established technology management theories: the dominant design theory; the technology
S-curve theory; the technology and market trajectories theory and the modularisation of
design theory. Section 5 provides a conclusion.
2 Introduction to complex systems theory
The term system covers a broad spectrum of our social, physical and biological world and
systems thinking is an established body of knowledge that seeks to understand how
entities (social, technical, economic, biological, etc.) function. Khun [7], Capra [8] and
McCarthy et al [9] discuss several eras and movements of systems thinking and despite
3. I.P. McCarthy 730
the different views that each movement has, a common theme is the desire to understand
complicated entities by:
• determining the system boundary, elements, inputs, outputs, relationships and
attributes.
• supporting the integration of views and knowledge to study the total system and how
it interacts with its environment.
In recent years the study of systems has developed with input from various disciplines to
become known as complex systems theory [9–13], or, as some call it, complexity theory
or complexity science. This branch of systems thinking has similar theoretical and applied
motivations to that of other systems concepts (e.g. soft systems methodology [14],
systems dynamics [15] and general systems theory [16]), in that they all seek to model
and understand the behaviour of systems. The distinctive stance taken by complex
systems theory is that it is concerned with systems that exhibit
1 a configuration made up of a large number of elements
2 significant interactions among these elements
3 organisation in the system.
The above systems features generate three highly related characteristics of a complex
system: non-linearity, emergence, and self-organisation. Thus, complex systems theory
acknowledges that certain systems learn and evolve, and cannot be fully described by a
single rule. It is a theory that seeks to understand how the system elements and
interactions self-organise to create new configurations.
Next, it is necessary to introduce the term complex adaptive system. These systems
are complex systems, but the active elements that constitute the system are referred to as
autonomous agents. In organisations, agents would be the decision-making entities
(e.g. operators, control systems, managers, designers, etc.) that receive and process local
information to create the events, outputs and internal dynamics of the system. The
behaviour of an agent is influenced by goal-led operating rules known as schemata. For
example, organisations have schemata (strategies and plans) for issues such as; what
products and services to provide; what technology to use; and how to design and manage
production facilities. Thus, unlike a biological system that blindly changes over time
(animals and plants do not form mission statements and strategic plans to eliminate their
competition), a complex adaptive system has the ability to consciously alter its system
configuration and influence its current and future survival.
The challenges facing technology managers are complex adaptive systems issues that
revolve around emerging and non-linear trends such as the explosion of information and
knowledge, the globalisation of technology, and the accelerating rate of diffusion [17].
Underpinning these trends, are factors such as system uncertainty and risk and the
discontinuities resulting from them. Technology management often involves trade-offs,
such as managing the balance between modularisation and customisation, or balancing
the need for high reliability (e.g. systems such as nuclear power plants that are safety
critical) versus operational leanness (e.g. systems such as automotive assembly plants
that strive for efficiency). With all of these systems, inappropriate decisions and policies
can result in accumulative, disproportional and unpredictable outcomes. For example, the
UK rail transportation company, Rail Track, reduced its rail maintenance program in an
4. 731 Technology management – a complex adaptive systems approach
effort to trim down costs. The consequence was that on the 17 October 2000, a London to
Leeds passenger express train was derailed, causing several fatalities. Early indications
from the accident investigation [18], suggest that the likely cause was a crack in the rail,
which propagated, broke the rail and then derailed the train. This was then followed by a
significant loss of confidence in UK rail travel, which in turn led the government to place
the company into insolvency. This is an example of how a decision or policy can lead to
an irregular chain of events with accumulative and disproportional results.
3 Fitness landscape theory
Fitness landscape theory was developed in the 1930s [19] and is now a key concept for
understanding the structure and interactions in complex adaptive systems. It is based on
the biological view that organisms evolve over time to survive and that this evolution can
be visualised as a journey over a three-dimensional landscape of valleys and peaks. More
recently, fitness landscape theory has been used to investigate a number of physical
science problems including the structure of molecular sequences [20] and mathematical
models of genome evolution [21]. One specific technique that has emerged to model the
diversity of configurations represented by a fitness landscape, is Kauffman’s NK model
[22–24], and it has attracted attention from management and organisational researchers
interested in the diversity of organisational forms and strategies [25–30].
With fitness landscape theory and the NK model, the notion of fitness is a measure of
the ability to survive and produce offspring. As technological and organisational systems
do not sexually reproduce, fitness in such systems is associated with adaptability,
robustness, durability, popularity and survival. For instance, competitive technologies
usually survive longer by internally enduring and adapting to external demands. Thus,
successful technologies often inspire other competing and non-competing technologies to
imitate their configuration, capabilities and mode of working. Ultimately, the term fitness
is a tautology, because what exists must be fit by definition, but for technology
management the objective is to ensure that configurations of product, process and
management technologies are suitable for competing in the future as well as the present.
To understand how fitness landscape theory and NK models relate to technology
management, it is first necessary to define the system that is the subject of this
paper i.e. the technology strategy and the resulting organisational configuration. The term
configuration refers to the make-up of an organisation, its form or defining
characteristics.
Figure 1 illustrates the relationship between technology strategy, resources, routines
and the resulting configuration and capabilities. External forces such as competition,
market needs, and innovation drive the strategy configuration chain, which consists of
four links that represent the management tasks that develop and sustain the resources,
routines and capabilities of an organisation. These tasks are:
Technology Classification – this is the process of searching, identifying and grouping
information about technologies (product, process and management) that may influence an
organisation’s competitive position. This will typically involve tasks such as forecasting
[31], roadmapping [32] and benchmarking [33]. The classification task provides a
comparative framework to store and retrieve information that will facilitate the next
task - technology selection.
5. I.P. McCarthy 732
Technology Selection – organisations have resource and time constraints and are
therefore limited to the number of technologies that can be chosen from those identified
by the classification task. This results in an evaluation and justification process that uses
technological, financial and market criteria to justify, select and invest in technologies.
Technology Adoption – once selected, organisations can adopt technology by various
routes. These include licensing, purchasing (e.g. acquire technology rights or acquire the
firm which possesses the technology), in-house development, seconded change agents
and R&D alliances.
Technology Exploitation – this task seeks to effectively and efficiently exploit the
resulting technological configuration by producing and delivering system outputs that are
competitive in terms of service, cost, quality, etc. This task also involves protecting the
resources and technology that provide the value propositions by patenting, copyrighting
and trademarking.
Figure 1 The strategy configuration chain
Com
petition
M
arket needs
&
ideas
RESOURCES &
ROUTINES
Internal
Forces
R&D Capacity
Design Facilities
Operations
Managem't
Operational
Processes
Workforce
Project
Managem't
Knowledge
Managem't
Structure
CONFIGURATION &
CAPABILITIES
The resultant
organisational form and
value propositions in terms
of cost, innovation, service;
flexibility; reliability, etc.
TECHNOLOGY STRATEGY
The continuous classification
selection, adoption,
and exploitation of
technologies needed to
achieve business goals
Technology
Exploitation
Technology Classification
Technology
Selection
Technology Adoption
6. 733 Technology management – a complex adaptive systems approach
In summary, these four technology management tasks create capabilities to provide a
competitive advantage in terms of precision, cost, sustainability, innovation, flexibility,
etc. The capabilities are a combination of resources and routines that form a
technological configuration. The resources (people, knowledge, raw material, etc.) are the
basic inputs to any new technological configuration, but they only provide value when
they become connected. Routines achieve this connection by managing and coordinating
the resources in a particular fashion. They are the “norms, rules, procedures, conventions,
and technologies around which organisations are constructed and through which they
operate” [34].
To understand how fitness landscape theory and the NK model relate to the strategy
configuration chain, Table 1 presents the NK model notation and outlines its relevance to
technology strategy and configurations. This is then followed by a simple and
hypothetical example based on Figure 2 and the data in Table 2. The example has the
following characteristics: N = 3, and K = N-1, and A = 2. This indicates a binary
representation where 1 equals the presence of a certain capability and 0 equals the
absence of a certain capability. An NK model represents the possible solutions or options,
in this case technology strategies and the resulting technological configurations. In the
example, Table 2 lists eight technology strategies (AN
= 23
= 8). For each strategy, a
fitness function f(x) calculates a fitness value of between 0 and 1 that indicates the
potential for survival now and in the future. In this hypothetical example, the fitness
values have been randomly generated, and a value close to 0 indicates poor fitness, whilst
a value close to 1 indicates good fitness. In principle, fitness values can then be plotted as
heights on a multidimensional landscape, where the hills represent high fitness and the
valleys represent low fitness. In Kauffman’s model, the fitness f(x), is the average of the
fitness contributions, fi(x), from each locus i, and is represented as:
)(
1
)(
1
xf
N
xf
N
i
i∑=
=
Table 1 NK model notation
Notations Evolutionary Biology Technology Strategy
N
The number of elements or genes of the
evolving genotype. A gene can exist in
different forms or states.
The number of system elements or
variables that constitute the strategy and
the resulting configuration of capabilities
as determined by the resources and
routines.
K
The amount of epistatic interactions
(interconnectedness) among the
elements or genes. It can range from K
= 0 to a maximum of K = N-1. In this
case the maximum is assumed.
The amount of interconnectedness among
the system elements. This creates trade-
offs or accumulative dependencies
between capabilities such as speed,
precision and cost.
A
The number of alleles (the alternative
forms or states) that a gene may have.
Number of possible states an element
might have. For instance, the quality
capability could have four states:
inspection, quality control, quality
assurance, and total quality management.
C Coupledness of the genotype with other
genotypes.
The co-evolution of one configuration and
its technology strategy with its
competitors.
7. I.P. McCarthy 734
Figure 2 NK model notation
Genome
(technology strategy)
0 01
Gene Loci
(capabilities/elements)
0 and 1 = gene alleles or states
= "K" the interconnectedness among
the capabilities/parts
Table 2 Technology strategy as a three-bit string
Genotype
(Strategy)
Gene 1
(Capability 1)
Gene 2
(Capability 2)
Gene 3
(Capability 3)
Assigned Random
Fitness Value
000 Absent Absent Absent 0.0
001 Absent Absent Present 0.1
010 Absent Present Absent 0.3
011 Absent Present Present 0.5
100 Present Absent Absent 0.4
101 Present Absent Present 0.7
110 Present Present Absent 0.8
111 Present Present Present 0.6
As N=3, a three-dimensional cube can be used to represent the possible strategies and
their relationship to each other (see Figure 3). Each corner point of the cube is a strategy,
labelled with its three-bit string and its hypothetical fitness value. Fitness landscape
theory views evolution as a process of moving from one strategy to another in search of
improved fitness. This process is called the adaptive walk. If a strategy (e.g. point 011) is
chosen from Figure 3, there are three possible one-mutation neighbour strategies (points
010, 111 and 001). If an organisation has strategy 011 and it has a fitter neighbour
strategy (i.e. a higher fitness value) then to evolve and continue surviving the
organisation should consider adopting this strategy (i.e. point 111). In terms of the
landscape metaphor, any movement to a fitter position is equivalent to walking uphill and
conversely any movement to a position of reduced fitness is equivalent to walking
downhill. The directions of the arrowed lines in Figure 3 represent uphill walks towards
global and local peaks. The global peak is the highest point (fittest strategy) on the
landscape (point 110), whilst a local peak is a point on the landscape where there is no
8. 735 Technology management – a complex adaptive systems approach
immediate neighbourhood strategy with a higher fitness (e.g. point 101). Thus, the path
from a local peak to a global peak will always involve at least one downhill journey.
Figure 3 A fitness landscape (N=3 and K=2)
(0.8)
110
(0.6)
111
(0.5)
011
(0.1)
001
(0.00)
000
(0.3)
010
(0.4)
100
(0.7)
101
Local peak
Global peak
As this is a simple example, consisting of only three different capabilities it is relatively
easy to visualise the space or landscape of strategic options using a three-dimensional
cube. If the example was more complex and dealt with several capabilities, then a
Boolean hypercube could be used to map the strategic options. For instance, Figure 4
illustrates the landscape of strategic options generated by four capabilities (innovation,
reliability, flexibility and sustainability). The arrowed lines represent the structure of
landscape. The dotted lines with single arrows represent the path with the greatest gain in
fitness for each step from point 0000 to one of global peaks (1111). The dashed lines with
double arrows specify that two neighbouring strategies have the same fitness. When a
point has all lines directed to it, then this point is said to be a peak (either local or global).
In Figure 4, there are two strategies (1101 and 1111), both with global peak fitness values
of 0.67. The presence of more than one global peak suggests that the capabilities might
have epistatic interactions with each other (a K factor of N-1) that results in
accumulations and trade-off in overall fitness. This notion is consistent with the theory
of capability trade-off [35,36].
9. I.P. McCarthy 736
Figure 4 A Boolean hypercube of four technological capabilities
0000
(0.35)
1000
(0.49)
0100
(0.43)
0010
(0.43)
1100
(0.54)
1010
(0.57)
1001
(0.52)
0001
(0.44)
0110
(0.53)
0101
(0.53)
0011
(0.44)
1110
(0.62)
1101
(0.67)
1011
(0.58)
0111
(0.53)
1111
(0.67)
1111
(0.67)
Innovation Reliability Flexibility Sustainability
Fitness
= Optimal path from 0000 to 1111
= Neighbouring strategies with same
fit
3.1 The K and C factors
As described in Table 1, K represents the interactions or dependency between the
different elements (capabilities) of a technological configuration or strategy. If K=0, this
indicates no interaction and therefore no trade-off or accumulative effect between the
elements. The resultant landscape if visualised as a mountainous terrain, is relatively
simple and smooth, except for one single global technology strategy (Figure 5). As K
increases from 0 toward its maximum of N-1, the landscape changes to an increasingly
rugged, uncorrelated, and multi-peaked landscape (Figure 6). This is because the NK
model assumes that the contribution of an individual capability to the overall fitness of a
strategy depends on the status of that capability, and its effect on the status of other
capabilities in the configuration.
Originally, Kauffman’s NK model did not consider how systems and their strategies
might respond to the actions of other competing systems in the same environment. Yet,
complex adaptive systems do co-evolve and Kauffman therefore introduced a C factor to
include the concept of coupledness and acknowledge that systems rarely exist in
isolation. For example, if the fitness of one strategy is increased, it is almost certain to
affect the fitness of other competing strategies. This is because the fitness of any complex
adaptive system is not only determined by its own performance, but also by the
10. 737 Technology management – a complex adaptive systems approach
performance of other systems in its environment. Fitness, like competitiveness is a
relative concept. Also, the landscape is not static, it is continually changing as
innovations emerge, operational performance alters and competitors enter and leave.
Thus, the fitness of a technology or strategic configuration is a function not only of its
own characteristics and behaviour, but also of the characteristics and behaviour of all of
its rivals. This results in three possible situations: competition, exploitation, and
mutualism [37]. Competition is when a configuration seeks to hinder the fitness of other
configurations. Exploitation is when one configuration (configuration A) stimulates the
fitness of another configuration (configuration B), whilst the presence of configuration B
inhibits the fitness of configuration A. Mutualism is the situation when configurations
stimulate individual and collective fitness. For example, personal computer
manufacturers and software technology providers have mutual dependence.
Figure 5 Fitness landscape for K=0
The dominant global peak
Figure 6 Fitness landscape for K=N-1
Several local peaks
11. I.P. McCarthy 738
4 The strategic management of technology
The evolutionary concepts presented in this paper do not just apply to biological systems.
If an entity (technological, social, or economic) evolves, then a complex adaptive systems
approach provides a framework to study the evolutionary and systems processes.
To understand how fitness landscape theory relates to the strategic management of
technology, the model created by Campbell [38] and developed by Pfeffer [39] and
Aldrich [40] to represent the evolution of organisations is used. The model identifies four
processes that underpin the evolution of a population of organisations: variation,
selection, retention and struggle.
Variation – this is the evolutionary process that generates technological variety. For a
market and consumers to select differentially among products and services there must be
some variation. This difference can be any factor (operational performance, product
design, innovative materials, systems integration, etc.) that creates a competitive
demarcation. Aldrich [40] states that variations may be intentional (planned) or blind
(unplanned). Variation is intentional when an organisation deliberately attempts to
resolve problems, improve performance or exploit opportunities. Planned variation
involves using resources such as design teams, internal change agents, R&D departments
and external consultants to undertake and realise strategic and tactical programs of
improvement and change. Such programs seek to promote innovative activities and
enhance the fitness of an organisation. Blind variation is an innovation that was not
planned and often involves an element of trial and error learning, serendipity, accidents
and so forth. It can also take the form of new knowledge or experience introduced into an
organisation by newly recruited employees or acquired businesses.
Selection – once there are a variety of technological configurations and resulting
products, the evolutionary process that differentially chooses or eliminates these systems
is known as selection. This process can be either internal or external [40]. Internal
selection occurs when the resources and routines within an organisation determine
whether a variation is adopted or not. For example, certain organisational cultures can
create a positive reinforcement of old innovations. The result is that organisations carry
on doing what they know best, rather than exploring new technologies. External
selection takes place when factors external to an organisation determine the selection.
Customers who request a certain technical benchmark, functionality or performance, and
industry regulations that govern product standards, are examples of external selection.
Retention – this process preserves, copies or imitates technological configurations and
products that are perceived to be successful. It can take place at two levels, the
organisational level and the population level. Organisational retention occurs through the
industrialisation and documentation of successful knowledge and routines (e.g. industry
best practice reports) and by existing personnel transferring knowledge to new personnel
and vice versa. Population level retention arises when new knowledge is spread from one
organisation to another. This can occur through personal contacts, marketing publicity, or
through observers, such as academics, who publish case studies about new technologies
and operational practices. Retention is the process that promotes technologies (product,
process and management) that are perceived to be beneficial.
Struggle – this is the process of competing for resources that are limited. During the
industrial revolution, raw material and energy were key resources, whilst the present need
12. 739 Technology management – a complex adaptive systems approach
is for knowledge-based resource such as skilled workers, research partners and value
adding suppliers. With new industries, the pioneering organisations responsible for
creating the industry often benefit from relatively unconstrained access to the necessary
resources, but as competition and members of the industry grow, the resources become
limited, the process of struggle intensifies and failure rates begin to increase.
To understand how these four evolutionary processes affect the strategic management
of technology and the tasks defined by the strategy configuration chain, Figure 7 is
presented.
Figure 7 Model of technological fitness
VARIATION
Technological and
organisational
innovations (intentional &
blind) create new
technological
configurations.
SELECTION
Dominant technological
configurations and the
corresponding
capabilities are selected
RETENTION
Organisational and
industry forces retain, or
duplicate dominant
configurations. Rivals
adapt their resources &
routines accordingly
STRUGGLE
Competing for resource,
such as knowledge, raw
material, energy, labour
capacity, etc.
TechnologySelection
TechnologyClassification
Technology Adoption and Exploitation
Innovation &
Adaptation
Durability &
Robustness
Reproduction
The four technology management tasks are mapped onto the evolutionary processes as
three driving links, with technology adoption and technology exploitation occupying the
same link. All four management tasks act as driving and consuming forces that underpin
technological evolution. This conceptual model is the basis for the following definition of
technological fitness:
Fitness is a fundamental characteristic of technology management that involves
innovating and replicating appropriate resources and routines to successfully attain
technological capabilities that satisfy strategic objectives and market needs. It is an
evolutionary indicator of an organisation’s ability to ensure current and future survival
in one or more markets by effectively balancing exploration (classification, selection and
adoption) and exploitation.
13. I.P. McCarthy 740
Thus, if a technological configuration currently exists it is fit. The potential for future
fitness depends on how effectively an organisation understands and conducts its
technology management tasks to achieve future configurations that can effectively
exploit and explore according to how competition, market needs and invention shape the
evolutionary processes.
4.1 Relevance and implications for technology management
This section examines the implications that fitness landscape theory and the model
presented in Figure 7 could have for the strategic management of technology. Four
classic technology management theories are presented and the concepts of fitness,
connectivity, coupledness, variation, selection, retention and struggle are considered.
Dominant Design Theory – the premise of this theory is that innovation is altered
significantly once a dominant design of technology emerges [41]. This is a selection and
retention issue that determines which technologies become standard or successful. A
current example would include the battle between Bluetooth and Wi-Fi to be the
wireless-data standard for mobile phones, laptops and hand held PCs. Previous examples
include the supremacy of the VHS video system over its rivals Betamax and Philips, and
the dominance of automobiles powered by gasoline engines, rather than by steam engines
or electric powered engines. Each of these dominant technologies evolved to create a
fitness landscape with an overriding single peak, similar to that shown in Figure 5. The
dominant design is not necessarily the optimal or best design. It is often selected and
retained, because it is a technological system that has superior integration with other
technological systems (products, processes, suppliers, customers, etc.) in its environment.
Technologies, capabilities and configurations compete, co-evolve (C factor) and
demonstrate coupledness (K factor). For example, once a dominant product technology
emerges, it often reduces and even eliminates the innovation (variation) activity and
fitness of competing product technologies, whilst increasing the innovation activity of the
associated process technology required to manufacture or deliver the dominant product
(Figure 8).
Figure 8 The impact of dominant design on the rate of technological change in processes and
products
Product Technology
Process Technology
Time
Rateoftechnologicalchange
Source: adapted from Abernathy and Utterback [41]
14. 741 Technology management – a complex adaptive systems approach
Technology S-Curve Theory – this theory asserts that the rate of performance
improvement generated by any variation has certain limits. This performance rate can be
represented using one or more S-curves. For managers, this theory advocates that they
should monitor any decline in performance and then respond appropriately, before the top
of the S-curve is reached (Figure 9). Thus, the ideal management practice is viewed as a
series of S-curves that balance and smooth the processes of variation and struggle
(investment in innovation). The ability to continually create new S-curves is an indicator
of an organisation’s ability to understand and manage the diversity of routines and
resources that it can accommodate. This can only be achieved if an organisation has
excess diversity (additional resource) to explore and innovate. This means that an
organisation should have the resources to effectively produce and deliver products
(exploitation), and excess resource to explore and experiment (planned and unplanned
variation), and thus create new knowledge and routines. Those organisations that have
most excess diversity tend to be large multinational corporations who have the market
reputation and financial assets to successfully compete for resources (struggle).
Figure 9 Understanding technology strategy as a series of S-curves
Time and Effort
Performance
1st Technology
2nd Technology
3rd Technology
The path of the ideal strategy
Source: adapted from Christensen [42]
Technology and Market Trajectories Theory – this theory states that the patterns of
technological change and innovation are influenced by the co-evolving and intersecting
trajectories of:
1 the performance demanded by the market need (or problem) versus
2 the performance delivered by the technology (or solution) [43].
The main observations of the theory are that the performance delivered trajectory of a
technology is often steeper than the performance demanded trajectory of the market.
Thus, a technology frequently has the fitness to satisfy future needs in addition to current
needs. Also, the concept of sustaining technologies and disruptive technologies [42] is
represented in terms of the impact on performance (Figure 10). Sustaining technologies
improve the performance of established products according to the performance criteria
that mainstream customers have historically valued. This process of retention often leads
15. I.P. McCarthy 742
to architectural innovations that link or combine existing technologies. Disruptive
technologies create new value propositions that establish new markets. Levinthal [44]
compares the emergence of such technologies to the speciation process in biology.
Figure 10 Intersecting trajectories
Time
ProductPerformance
Performance demanded
at the high end
of the market
Sustaining
technology
progress
Performance
demanded
at the low end
of the marketDisruptive
technological
innovation
Sustaining
technology
progress
Source: adapted from Christensen [45]
In summary, this theory is consistent with the notion that managers should view their
organisations as complex adaptive systems consisting of temporary repositories of
routines and capabilities. The distribution of these capabilities and routines is subject to
the struggle for knowledge, technology, raw material and skills in the population. Thus,
the dynamic of survival and competitiveness are determined by the relative connectivity
of the different technology strategies and configurations in an industry. The different
growth rates that exist in industries create the structural changes that define a fitness
landscape.
Modularisation of Design – this area of technology management is concerned with
understanding how technological designs create capabilities for the future and how they
might influence competitiveness. For designers of sub-assembly based products this is an
interconnectedness issue (K factor). If one component of a particular
sub-assembly should change, then how will this affect the overall performance of the
product? If a full understanding of the possible interactions exists, then the product is
modularised. This allows designers to amend and develop product sub-assemblies
quickly and with minimal cost. Today, the personal computer (PC) is an example of a
modularised product. The components that make up a PC can be purchased and
assembled with relative ease. Much of the hardware and software functionality is based
on a ‘plug and play’ concept. The result is that companies such as Dell, Compaq and
Time were able to enter and compete in the PC market, without acquiring capabilities in
computer science or software engineering. These organisations exist, because they have
the capability to competitively assemble and deliver PCs (a systems assembler). When
there is low or even no modularisation of design for a product (e.g. military aircraft), an
16. 743 Technology management – a complex adaptive systems approach
organisation competes on the basis of specialist and propriety knowledge, and integration
expertise (a systems design and integration capability). Twenty years ago, PCs were
designed and manufactured solely by organisations that were system designers and
integrators. Hence, industries and products go through cycles of modularisation and de-
modularisation.
5 Conclusions
A systems approach to technology management provides frameworks, theories and a
universal language to understand and manage the range of technological issues,
relationships and values that are the basis of all business systems. A complex adaptive
systems approach is evolutionary in nature, but acknowledges that variation can be both
planned and unplanned, and that the processes of selection and retention are influenced
by organisational routines and resources. The term adaptive, refers to one of the key
differences between biological systems and organisational systems i.e. a decision-making
capacity to consciously evolve in response to certain goals and objectives. This decision-
making capacity resides in the four technology management tasks (classification,
selection, adoption and exploitation) that drive the strategy configuration chain (Figure
1) and technological fitness (Figure 7).
Organisations are complex adaptive systems that create (variation) and adopt
(selection and retention) innovative technologies, which in turn drive evolutionary
change. Fitness landscape theory and the NK model provide a method for visualising and
communicating the adaptive and imitative reality of the strategy process that
accompanies this evolution. Hence, the management of technology is a process of
understanding the profile of the landscape and how the four management tasks coupled
with the processes of variation, selection, retention and struggle shape the landscape.
An organisation occupies a certain position on the landscape, because it is fit and has
evolved a combination of capabilities and routines that place it on the landscape. The
competitiveness of its position is determined by the interconnectedness of its capabilities
and the relative performance of its competitors. Identifying where an organisation should
be on a landscape is a benchmarking and forecasting exercise that is realised via the
process of planned variation. Yet, history and theories of technological change reveal that
the variation process can often be unplanned and serendipitous.
References
1 De Vol, R., Wong, P., Catapano, J. and Robitshek, G. (1999) ‘America’s high-tech economy,
growth, development, and risks for metropolitan areas’. Research Report, Santa Monica, CA:
Milken Institute.
2 March, J. G. (1991) ‘Exploration and exploitation in organisational learning.’ Organisation
Science, Vol. 2.1, pp.71-87.
3 Schumpeter, J. (1939) Business Cycles, New York: McGraw-Hill Book Company, Inc.
4 Ruttan, V. (1959) ‘Usher and Schumpeter on invention, innovation, and technological
change’, Quarterly Journal of Economics, November.
5 Marquis, D. (1969) ‘The anatomy of successful innovations’, Innovation, Vol. 1, pp.35-48.
17. I.P. McCarthy 744
6 Nelson, R.R. and Winter, S.G. (1982) An Evolutionary Theory of Economic Change,
Cambridge: Harvard University Press.
7 Kuhn, T. S. (1962) The Structure of Scientific Revolutions, Chicago: University of Chicago
Press.
8 Capra F. (1986) ‘The concept of paradigm and paradigm shift’ Re-Vision, Vol. 9, No. 1, p.3.
9 McCarthy, I.P., Frizelle G. and Rakotobe-Joel, T. (2000) ‘Complex systems theory –
implications and promises for manufacturing organisations’, International Journal of
Technology Management, Vol. 2, Nos. 1–7, pp.559-579.
10 Anderson, P. (1999) ‘Complexity theory and organisation science’, Organisation Science,
Vol. 10, No. 3, pp.216-232.
11 Choi, T.Y., Dooley, K.J. and Rungtusanatham, M. (2001) ‘Supply networks and complex
adaptive systems: control versus emergence’, Journal of Operations Management, Vol. 19,
pp.351–366.
12 Dooley, K. and Van de Ven, A. (1999) ‘Explaining complex organisational dynamics’,
Organisation Science, Vol. 10, No. 3, pp.358–372.
13 Morel, B. and Ramanujam, R. (1999) ‘Through the looking glass of complexity: the dynamics
of organisations as adaptive and evolving systems complexity’, Organisation Science, Vol. 10,
No. 3, pp.278-293.
14 Checkland, P. (1981) Systems Thinking, Systems Practice, Wiley & Sons.
15 Forrester, J. W. (1961) Industrial Dynamics, Cambridge MIT Press.
16 Bertalanffy, L.V. (1968) General System Theory, (Ed.) Braziller, New York.
17 Zehner II, W. B. (2000) ‘The Management of Technology (MOT) Degree: a bridge between
technology and strategic management’, Technology Analysis & Strategic Management,
Vol. 12, No. 2.
18 Health and Safety Executive (2000) ‘Train derailment at Hatfield, 17 October 2000’,
First HSE Interim Report, London, UK.
19 Wright, S. (1932) ‘The roles of mutation, inbreeding, cross-breeding and selection in
evolution’, Proceedings of the Sixth International Congress of Genetics, Vol. 1, pp.356-366.
20 Lewontin, R.C. (1974) The Genetic Basis Of Evolutionary Change, New York; London:
Columbia University Press.
21 Macken, C.A. and Perelson, A.S. (1989) ‘Protein evolution on rugged landscapes’,
Proceedings of the National Academy of Sciences of the United States of America, Vol. 86,
No. 16, pp. 6191-6195.
22 Kauffman, S.A. and Weinberger, E.D. (1989) ‘The NK model of rugged fitness landscapes
and its application to maturation of the immune-response’, Journal of Theoretical Biology,
Vol. 141, No. 2, pp.211-245.
23 Weinberger, E.D. (1991) ‘Local properties of Kauffman N-K model – a tunably rugged energy
landscape’, Physical Review A, Vol. 44, No. 10, pp.6399-6413.
24 Kauffman, S. A. (1993) The Origins Of Order: Self Organisation And Selection In Evolution,
New York: Oxford University Press.
25 Maguire, S. (1999) ‘Strategy as design: a fitness landscape framework’, in M. Lissackand and
H. Gunz (Eds.) Managing Complexity in Organizations: A View in Many Directions, Quorum
Books: Westport, CT, pp.67-104.
26 Merry, U. (1999) ‘Organisational strategy on different landscapes: a new science approach’,
Systemic Practice and Action Research, Vol. 12, No. 3, pp.257-278.
27 Beinhocker, E. D. (1999) ‘Robust adaptive strategies’, Sloan Management Review, Vol. 40,
No. 3, pp.95-106.
28 McKelvey, B. (1999) ‘Self-organisation, complexity, catastrophe and microstate models at the
edge of chaos’ in Variations in Organisation Science – in Honour of Donald T. Campbell,
J. A.C. Baum and B. McKelvey (Eds.), Sage Publications, pp.279-307.
18. 745 Technology management – a complex adaptive systems approach
29 Ruef, M. (1997) ‘Assessing organisational fitness on a dynamic landscape: an empirical test of
the relative inertia thesis’, Strategic Management Journal, Vol. 18, No. 11, pp.837-853.
30 Rivkin, J. (2000) ‘Imitation of complex strategies’, Management Science, Vol. 46,
pp.824-844.
31 Martino, J.P. (1993) Technological Forecasting for Decision Making, 3rd ed., New York:
McGraw-Hill.
32 Groenveld, P. (1997) ‘Roadmapping integrates business and technology’, Research
Technology Management, Vol. 40.5, pp.48-55.
33 Camp, R.C. (1989) Benchmarking. The Search for Industry Best Practices that Lead to
Superior Performance, ASQC Quality Press & Quality Resources, Milwaukee, Winconsin.
34 Levitt, B. and March, J.G. (1988) ‘Organisational learning’, Annual Review of Sociology,
Vol. 14, pp.319-340.
35 Skinner, W. (1974) ‘The focused factory’, Harvard Business Review, Vol. 52, No. 3,
pp.113-121.
36 Filippini, R., Forza, C. and Vinelli, A. (1998) ‘Trade-off and compatibility between
performance: definitions and empirical evidence’, International Journal of Production
Research, Vol. 36, No. 12, pp.3379-3406.
37 Metcalfe, J.S. (1998) Evolutionary Economics And Creative Destruction, London: Routledge.
38 Campbell, D.T. (1969) ‘Variation and selective retention in socio-cultural evolution’, General
Systems, Vol. 14, pp.69-85.
39 Pfeffer, J. (1982) Organisations And Organisation Theory, Boston: Pitman.
40 Aldrich, H. E. (1999) Organisations Evolving, London: Sage Publications.
41 Abernathy, W. and Utterback J. (1978) ‘Patterns of industrial innovation’, Technology Review,
June-July, pp.40–47.
42 Christensen, C.M. (1992) ‘Exploring the limits of the technology S-curve, parts I and II’,
Production and Operations Management, Vol. 1, pp.334-357.
43 Dosi, G. (1982) ‘Technological paradigms and technological trajectories’, Research Policy,
Vol. 11, pp.147-162.
44 Levinthal, D. (1997) ‘Adaptation on rugged landscapes’, Management Science, Vol. 43,
pp.934-950.
45 Christensen C.M. (1997) The Innovator’s Dilemma: When Technologies Cause Great Firms to
Fail, Harvard Business School Press, Boston.