Technology Management - A Complex Adaptive Systems Approach

Int. J. Technology Management, Vol. 25, No. 8, 2003 728

Copyright © 2003 Inderscience Enterprises Ltd.

Technology management – a complex adaptive systems approach

Ian P. McCarthy University of Warwick, Organisational Systems and Strategy Unit, Warwick Manufacturing Group, Coventry, CV4 7AL, UK Fax: +44 (0) 247 652 4307 E-mail: [email protected]

Abstract: There are systems methods and evolutionary processes that can help organisations understand the innovative patterns and competitive mechanisms that influence the creation, management and exploitation of technology. This paper presents a specific model based on the evolutionary processes of variation, selection, retention and struggle, coupled with fitness landscape theory. This latter concept is a complex adaptive systems theory that has attained recognition as an approach for visually mapping the strategic options an evolving system could pursue. The relevance and utility of fitness landscape theory to the strategic management of technology is explored, and a definition and model of technological fitness provided. The complex adaptive systems perspective adopted by this paper, views organisations as evolving systems that formulate strategies by classifying, selecting, adopting and exploiting various combinations of technological capabilities. A model called the strategy configuration chain is presented to illustrate this strategic process.

Keywords: Technology strategy; evolution; complex adaptive systems; fitness landscapes.

Reference to this paper should be made as follows: McCarthy, I.P. (2003) ‘Technology management – a complex adaptive systems approach’, Int. J. Technology Management, Vol. 25, No. 8, pp.728-745.

Biographical notes: Dr. Ian McCarthy is head of the Organisational Systems and Strategy Unit at the University of Warwick. This unit focuses on understanding industrial organisations and their processes using systems methods that include evolutionary and population ecology concepts. This basic research underpins an applied research agenda that includes studies on operations and technology strategy, new product development and industrial networks.

1 Introduction

Technology-based organisations contribute to dynamic competition, ensure product variety, close market gaps and create jobs. They are considered ‘high technology’ in terms of products and processes, and are key agents in the generation of regional and national economic wealth [1]. Yet, this wealth creation does not just depend on the discovery of new technology; it also requires the ability to successfully exploit discoveries and inventions. This is the theme in March’s seminal paper that considers the relation between exploration of new possibilities (innovation) and the exploitation of old

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

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


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.

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

Compe

tition

Market needs

& ideas

RESOURCES &ROUTINES

InternalForces

R&D Capacity

Design Facilities

OperationsManagem't

OperationalProcesses

Workforce

ProjectManagem't

KnowledgeManagem't

Structure

CONFIGURATION &CAPABILITIES

The resultantorganisational form and

value propositions in termsof cost, innovation, service;

flexibility; reliability, etc.

TECHNOLOGY STRATEGY

The continuous classificationselection, adoption,and exploitation of

technologies needed toachieve business goals

Tech

nolo

gy E

xplo

itatio

n

Technology Classification

Tech

nolo

gy S

elect

ion

Technology Adoption


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

xfN

xfN

ii∑

=

=

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.

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 amongthe capabilities/parts

Table 2 Technology strategy as a three-bit string

Genotype (Strategy)

Gene 1 (Capability 1)



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


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) 101Local 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].

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


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

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


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 andorganisational

innovations (intentional &blind) create newtechnological

configurations.

SELECTION

Dominant technologicalconfigurations and the

correspondingcapabilities are selected

RETENTION

Organisational andindustry forces retain, or

duplicate dominantconfigurations. Rivals

adapt their resources &routines accordingly

STRUGGLE

Competing for resource,such as knowledge, rawmaterial, energy, labour

capacity, etc.

Technology Selection

Tech

nolo

gy C

lassif

icat

ion

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.

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

Rat

e of

tech

nolo

gica

l cha

nge

Source: adapted from Abernathy and Utterback [41]


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

Perfo

rman

ce

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

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

Prod

uct P

erfo

rman

ce

Performance demandedat the high endof the market

Sustainingtechnologyprogress

Performancedemanded

at the low endof the marketDisruptive

technologicalinnovation

Sustainingtechnologyprogress

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


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.

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.


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.