Exploring the process of whole system design

Page 1

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Page 2

Author's personal copy

Exploring the process of whole systemdesign

Fiona Charnley and Mark Lemon, Institute of Energy and Sustainable

Development, De Montfort University, The Gateway,

Leicester LE1 3BH, UK

Steve Evans, Manufacturing Systems Centre, Cranfield University,

Cranfield, Bedfordshire MK43 OAL, UK

This paper explores the adoption of a whole system approach to a more

sustainable and innovative design. A case study methodology was utilised to gain

improved understanding of whole system design and those factors that

substantially influence its success. The paper presents a framework of those

factors including the requirement for trans-disciplinary skills, the dynamics of

a flattened hierarchy and the need to identify relationships between parts of the

system to ultimately optimise the whole. Knowing the factors that influence the

process of whole system design provides designers with the knowledge necessary

to more effectively work within, manage and facilitate that process. This paper

uses anecdotes taken from operational cases, across design contexts, to

demonstrate those factors.

� 2010 Elsevier Ltd. All rights reserved.

Keywords: whole system design, design process, collaborative design

The emergence of increasingly complex problems, combined with grow-

ing concerns for the environment, is fuelling the demand for more inno-

vative and sustainable products, services and systems. Designers are

commonly adopting more holistic and integrated approaches in an attempt

to meet increasing consumer demands (Coley & Lemon, 2008). Whole system

design is one such approach that aims to integrate social, economic and envi-

ronmental phenomena into a comprehensive design solution. The approach

encourages the development of partnerships between actors from a variety

of different backgrounds, disciplines and sectors to develop an innovative, sus-

tainable and optimised solution at a whole system level (Stasinopoulos, Smith,

Hargroves, & Desha, 2009). However, there is limited research concerning the

integrative process that actors are required to follow in order to reach such

a solution.

Literature in the areas of collaborative, sustainable and system level ap-

proaches to design increasingly identifies techniques such as systems thinking

(Senge, 2006) the development of partnerships (Katzenback & Smith, 1993),

Corresponding author:

Fiona Charnley

[email protected]

www.elsevier.com/locate/destud

0142-694X $ - see front matter Design Studies 32 (2011) 156e179

doi:10.1016/j.destud.2010.08.002 156� 2010 Elsevier Ltd. All rights reserved.

Page 3

Author's personal copy

and the use of trans-disciplinary skills (Gibson, 2001; Postrel, 2002) as being

particularly relevant to this type of design approach. However, designers

have been provided with little guidance as to how these techniques should

be implemented efficiently within an operational and substantially complex de-

sign process. This paper provides evidence of the use of a whole system ap-

proach to design and identifies ways in which designers are moving towards

a more holistic and integrated process when developing more sustainable

and innovative solutions to complex design problems. The study presented uti-

lises a longitudinal case study within the automotive industry to conduct an in-

depth and inductive exploration of the whole system design process. Through

the use of multiple observations of design meetings, the analysis of project doc-

umentation and interviews and discussion with project members, a number of

factors were observed which appeared to be common to whole system design.

These factors were then confirmed, modified and validated through use of

a number of additional cases across multiple design contexts. Unique experi-

ences and accounts are presented and utilised to demonstrate a comprehensive

framework of the factors necessary to facilitate good whole system design. The

study concludes by providing the design community with a more accurate def-

inition of what whole system design is and a more detailed account of how to

undertake a holistic and integrated design process.

1 BackgroundDue to a rapid and profound change in contemporary society, the problems

that we now face are complex, incorporating multiple interconnected aspects,

the most pertinent of which are often social, economic and environmental.

Subsequently, there is a growing responsibility to replace incremental im-

provements to existing products with all-encompassing, sustainable and inno-

vative packages of products, services and systems that will provide solutions to

consumer needs and requirements (Bhamra & Evans, 1997; Brezet, 1997;

Lofthouse, 2004). Mainstream businesses are launching new green initiatives

and eco-friendly products each week in an effort to capitalize on society’s ap-

parent shift toward a more environmental ethic. However, there is a concern

that by focusing on environmental sustainability alone, considerable opportu-

nities for improved efficiency, innovation and functionality are being missed.

Most green business efforts essentially are attempts to improve upon tradi-

tional products by somehow making them more environmentally benign,

such as by reformulating the product or increasing its energy efficiency

(Morson, 2007). However, authors are concerned that environmental consid-

erations are still an add-on option as opposed to being central to the way we do

business (Stasinopoulos et al., 2009). There is often little awareness and

understanding of the wider, environmental, social and economic impacts of

design e in other words, the sustainable development aspects (Howarth &

Hadfield, 2006). Infrastructure, buildings, cars and many appliances all have

long lives, in most cases from 20 to 50 years. The size and duration of infra-

structure and building developments, for instance, demand that they should

Exploring whole system design 157

Page 4

Author's personal copy

now be much more critically evaluated for efficiency and function than ever

before (Stasinopoulos et al., 2009).

Senge (2006) states that the un-healthiness of the world today is in direct pro-

portion to our inability to see it as a whole. Organisations are focusing on sus-

tainability as an objective, but they are largely limiting their efforts to what can

be done within the boundary of the firm (Ehrenfeld, 2003). They overlook the

fact that every worker arrives at the office or plant from a home within a com-

munity within a larger society, and imports the elements of the larger cultural

structure. Subsequently, environmentalists want businesses to change their

products fundamentally in anticipation of shifting consumer values and thus

consumer demand (Morson, 2007).

This fundamental change and required movement towards the development of

more sustainable solutions is thought to lie with the way we think about de-

sign. Anarow et al. (2003) suggest that sustainability cannot be achieved in

the absence of whole systems thinking; addressing the problem at a system

level. To gain a whole systems perspective companies are increasingly entering

into the development of partnerships between multiple organisations, often

across disciplines and industrial sectors. This is challenging as it is often uncer-

tain how actors from different organisations are to integrate successfully and

furthermore the holistic process that they should follow, in order to reach

a more sustainable solution, is currently unclear. There are currently multiple

terms being used to describe holistic and integrated approaches to the design

of more radically innovative and sustainable solutions (Coley & Lemon, 2009).

Although they often adopt a slightly different focus, these approaches have

many attributes in common. Whole system design is one such approach which

is becoming increasingly popular, however, there is limited research detailing

the process that actors are required to follow in order to reach a sustainable,

innovative and ‘system level’ solution.

1.1 What is whole system design?The Rocky Mountain Institute (2006) suggests that whole system design

means

“Optimising not just parts but the entire system. it takes ingenuity, intui-

tion and team work. Everything must be considered simultaneously and ana-

lysed to reveal mutually advantageous interactions (synergies) as well as

undesirable ones”

Increasingly the problems that we face are complex, spanning across environ-

mental, social and economical phenomena, and therefore viewing the entire

system necessarily requires the views of experts from multiple disciplines. Of-

ten the solution to these complex problems cannot be visualised, as it is to be

made up of a multi-faceted mix of products, services and systems, and the

process by which to get there can be messy, unstructured and context specific.

158 Design Studies Vol 32 No. 2 March 2011

Page 5

Author's personal copy

Incremental improvements to existing systems rarely meet rising consumer ex-

pectations for solutions to be both effective and environmentally benign

(Blanchard & Fabrycky, 2006). It is therefore suggested that to tackle increas-

ingly complex and disorganized problems; to provide holistic and integrative

solutions, we need to adopt a change in design mentality and to start thinking

differently (Hawken, Lovins, & Lovins, 1999).

Adopting a change in design mentality, however, is not straight forward. De-

signers and engineers have become more and more specialised as scientific and

technological knowledge has increased exponentially. For generations, engi-

neers, scientists and managers prepared themselves to solve complex problems

by becoming increasingly specialised and reducing problems to their constitu-

ent parts and focusing their attention on each part. Designers and engineers

followed highly structured and ‘over the wall’ approaches to design such as

those prescribed by Forsberg and Mooz (1998) and Pahl and Beitz (1996).

As a result engineers and designers are no longer trained across fields and

thus no longer keep up with the latest breakthroughs in every field

(Stasinopoulos et al., 2009). A separation of design functions and processes

means that opportunities are often missed to optimise the whole system, which

can lead to inefficient design, construction delays, oversized heating systems,

higher costs and unnecessary environmental impacts (Anarow et al., 2003).

Stasinopoulos et al. (2009) suggest that this is largely due to the fact that

the engineer only knows their field in detail and has little interaction with other

designers on the project.

The need to address complex problems more systemically, in a systematic way,

and from a multitude of perspectives is highlighting the importance of cross-

disciplinary collaborations and partnerships within industry (Hebel, 2007;

Senge, 2006). However, partnerships are accompanied by numerous expecta-

tions and requirements, and also a more extensive network of actors. Some ac-

tors, who were never previously regarded as designers, are becoming heavily

involved with the actual process of designing. High levels of multi-disciplinary

working not only increase levels of complexity (Mankin, Cohen, & Fitzgerald,

2004) but also create many more issues and concerns to consider and often

they can be conflicting (Howarth & Hadfield, 2006).

1.2 What is a system?Awhole systemdesign approach encourages those involved to regard a problem

as a whole system and not just to concentrate on one particular component of

that system. Additionally, it recognises that a problem is created by every part

of the system in which the problem is embedded, and that the problem can and

should be addressed at every level. Seiffert and Loch (2005) suggest that the

most important property of systems is that they are made up of several parts

that are not isolated, but closely interlinked forming a complex structure.

Global warming, ozone depletion, the international drugs trade and more

Exploring whole system design 159

Page 6

Author's personal copy

recently the crash within the economy are all examples of what Senge (2006)

terms ‘systemic breakdowns’ e problems that have no simple local cause. He

therefore proposes that a systems approach is needed more than ever to start

to manage the overwhelming complexity that is growing around us.

Systems are conceptual devices that we bound with a purpose; however once

bounded they become real and we can explore, and influence, how they emerge

through internal restructuring and their interactions with their environment.

The environment, in systemic terms, is that which lies outside of the system

boundary. It is the ability to acquire and utilise information about that envi-

ronment that forms the basis for an adaptive, and thereby more sustainable

system (Lemon 1999). Anarow et al. (2003) recognise that a whole system ap-

proach focuses on interactions between the elements of a system as a way to

understand and change the system itself. Without this whole system perspec-

tive crucial impacts between components could be missed, therefore disrupting

the system as a whole and overlooking opportunities for improved efficiency

and environmental sustainability. For the purpose of this paper, systems are

defined as a set of independent but interlinked phenomena, or as Sherwood

(2002) defines them ‘a community of connected entities’ that we bound with

a purpose (e.g. the design process). This connectedness means that systems

have emergent properties and cannot be broken into their component parts;

we must consider them as a whole and therefore need to develop mechanisms

for doing so.

2 Understanding the process of whole system designAs highlighted in Section 1.1, there is a limited amount of literature surround-

ing the process of whole system design, how a holistic and integrated process is

to be facilitated and what factors affect its success. Subsequently it was neces-

sary to undertake a largely exploratory and inductive methodology to gain an

expansive insight into the process of whole system design. A case study ap-

proach was adopted which consisted of one large case focusing on automotive

design, and five smaller cases focusing on a cross section of design contexts.

2.1 The automotive case studyThe initial, and most comprehensive, case study was unique in that it provided

the researchers with unlimited access to the design process, and the stake-

holders involved in that process, for a period of three years. This was also im-

portant as data surrounding the complexity of an operational whole system

design project could be collected in real time, as opposed to retrospectively

which is common to this type of research. The case study aimed to adopt

a whole system approach to the design and manufacture of a zero emission,

sustainable sports car utilising hydrogen fuel cell and regenerative breaking

technologies. A whole system approach was adopted by the design team as

it was thought that, through the identification of relationships between the dif-

ferent components of the systems architecture, the system could be optimised

160 Design Studies Vol 32 No. 2 March 2011

Page 7

Author's personal copy

as a whole. Figure 1 presents the positive feedback loop that was designed and

followed by the design team.

The design team decided that in this case ‘a system’ would be confined to the

six companies who were involved in the design and manufacture of the car and

the structure of the car itself. Subsequently, when studying the case, observa-

tions and interviews were limited to the six organisations accordingly.

Data collection involved undertaking twenty-two observations of design meet-

ings; typically lasting a day at a time, eighteen interviews with project members;

typically lasting betweenoneand twohours and the gatheringof relevant project

documentation such as meeting agendas and minutes, project reports and press

releases. Allmeetings and interviewswere recorded using aDictaphone, detailed

notes were made by the researcher and significant sections of meetings and all

interviews were transcribed and then analysed, along with the documentation,

using content analysis techniques (Krippendorff, 2004). The researcher was in-

dependent and had no role within the project or design process other than

that of an observer. Subsequently, the researcher had no input into the facilita-

tionof themeetings and thefindingsand results of the studywerenot sharedwith

the participants until the very end of the project.

2.2 Applying initial findings to multiple design contextsIn addition to the automotive case five additional case studies were carried out.

The central aim of this part of the study was to utilise the smaller cases to con-

firm, modify and validate the findings that were emerging from the automotive

case. Interviews and discussions were again undertaken with relevant stake-

holders alongside the analysis of documentation. To enable the most valuable

and effective data to be collected cases were chosen which were seen to be un-

dertaking a whole system approach to design; interview participants were in-

dividually selected based on their expertise and knowledge surrounding the

whole system design process. As the term ‘whole system design’ is not univer-

sally recognised each participant was chosen against a set of predetermined

Figure 1 Automotive Case

Feedback Loop

Exploring whole system design 161

Page 8

Author's personal copy

criteria, based on literature and the findings from the first phase of research.

Interview participants were therefore chosen who:

- Had engaged in a partnership between two or more organisations in

which there was a democratic governance;

- Had been involved in the utilisation of multiple perspectives to develop

a more holistic and sustainable solution;

- Had participated in the identification of relationships between compo-

nents of a system to develop a solution that would ultimately optimise

the whole system;

- Appreciate the benefit of thinking in a systemic or joined-up way.

A total of eleven interviews were carried out with relevant stakeholders from

five different case studies. Table 1 provides detail surrounding each case and

which design context the case came from.

The aim of carrying out the interviews within the second phase of the research

was twofold:

1) To gain individual experiences of undertaking a whole system design from

a variety of perspectives,

2) To gain critical feedback concerning the findings from the automotive

case study from professionals across a variety of design disciplines.

The first aim was carried out by asking the participants to describe their expe-

rience of working on a project that had adopted a whole system design ap-

proach. During the description the participants were prompted with open

ended questions to encourage them to provide more detail surrounding the

case they had chosen to describe. It was important that they were uninformed

of any of the aims or results of the research to avoid participant bias. Asking

open ended questions encouraged participants to speak freely about their own

experiences and allowed the researchers to identify any similar or additional

themes to those identified within the automotive case study.

Table 1 Case Study Descriptions

Case Description of Case Study Design Discipline

1 A project utilising the integration of multiple stakeholders acrossa community for the regeneration of East and South East Leeds

Community Development

2 A furniture design company who have adopted a highly integratedapproach to their design process and overarching business strategy

Product Design

3 An architecture firm who have adopted a holistic and integrated approachto the design and development of the Olympic Village; Stratford City

Master Planning

4 An environmental consultancy who have adopted a whole system approachto a housing development in Milton Keynes

Built Environment

5 A independent company utilising a whole system approach to the design ofenvironmentally sustainable personal mobility solutions

Business and Transport

162 Design Studies Vol 32 No. 2 March 2011

Page 9

Author's personal copy

The second part of the interview was structured around the themes that re-

sulted from the automotive case study, presented in Section 3. Participants

were asked questions directly related to the findings of the research but asked

to provide answers based upon their own experience. This encouraged them to

comment critically upon the key findings of the research and contribute to the

validation of the themes and study as a whole. The research methodology was

iterative; therefore as the automotive case study was ongoing, findings from

the additional five cases could be fed back into it to form an ultimate compo-

sition of themes and findings.

2.3 Analysing the dataOnce data had been collected and transcribed where appropriate, thematic

analysis was used to identify, analyse and report patterns (themes) within

the data, as prescribed by Braun and Clarke (2006) in their six step process.

This technique was decided upon as it was appropriate to an inductive ap-

proach in which patterns and themes can be identified from different sources

of raw data. Additionally, as the process being observed was complex; consist-

ing of phenomena from multiple disciplines, a thematic approach enabled the

data to be analysed without being simplified; allowing the underlying complex-

ity to remain accessible.

Initially, as the researcher had no previous experience of a whole system design

process, data resulting from the observation of design meetings was messy and

complex. Potentially any information could be relevant to how whole system

design should be carried out and so the notes taken by the researcher were long

and detailed. As more meetings were observed patterns became recognisable

within the data and it was possible to assign themes and sub-themes to groups

of similar data. This contributed to more efficient record taking as the research

developed.

A theme captures something important about the data in relation to the re-

search question, and represents some level of patterned response or meaning

within the data set (Braun & Clarke, 2006). Therefore, within the context of

the study, a theme was defined as a set of behaviours, actions or thoughts

that were displayed by those participants being observed and interviewed

and were perceived by the researchers as significantly influencing the process

of whole system design. Braun and Clarke (2006) suggest that ideally there

will be a number of instances of the theme across the data set, however

more instances do not necessarily mean the theme itself is more crucial. Re-

searcher judgement was required to determine what a theme was; however,

the data set was coded by more than one researcher to ensure validity and

reliability (Krippendorff, 2004).

After approximately 18 months of data collection and analysis within the au-

tomotive case it was realised that data now being collected from this case was

Exploring whole system design 163

Page 10

Author's personal copy

adding to and confirming existing themes as opposed to contributing to any

new ones. Subsequently, the focus of the research moved to the additional

five case studies for confirmation, modification and validation of initial

findings.

3 Factors influencing the success of a whole systemdesignThe initial phase of the research resulted in ten themes that were identified

through consistent engagement with the automotive case study. These themes

were then modified, confirmed and validated, through the use of the smaller

cases, which resulted in eight overall themes that were perceived to signifi-

cantly influence the process of whole system design. This section of the paper

utilises anecdotes taken from the data to communicate the significance of these

themes. The themes presented in the following sections are intended to provide

a provisional framework to guide designers through the process of whole sys-

tem design; they are not defining characteristics of whole system design but are

factors that contribute to good design practise.

3.1 Forming and sustaining a partnershipThe development of partnerships between organisations, as opposed to the use

of sub-contracting or internal collaboration, was recognised by all study par-

ticipants, as a significant enabler of the whole system design process. Partici-

pants particularly noted; the access to multiple perspectives and expertise,

the opportunity to identify linkages between components of a design solution

and the opportunity for improved innovation, as significant advantages of

forming partnerships. However, the majority of participants from all six

case studies also mentioned the challenges and difficulties associated with

developing effective partnerships and maintaining those relationships.

Within four out of the six case studies it was observed that partnerships had

been formed through the use of existing social and professional networks:

“In my experience, the design team are known to each other, it is not

always the same people but it is often the same companies involved in

a project” [Architect, Case Three].

The architect went onto suggest that this can be an advantage as utilising ex-

isting relationships saves time, effort and additionally, due to working on pre-

vious projects, trust and confidence has already been established. Accessing

and forming partnerships with new organisations was described as difficult

and time consuming and often stakeholders have little choice of who to

work with. Within the automotive case, several participants suggested that

the design team consisted of some stakeholders who were not entirely suitable

to the design context and had been chosen due to convenience as opposed to

relevant expertise. This was said to cause inefficiency and slow progress whilst

the partnership learned to make the best of the expertise available.

164 Design Studies Vol 32 No. 2 March 2011

Page 11

Author's personal copy

It was identified, particularly within the larger projects studied such as cases

one, three and four, that meetings were being attended by a multitude of dif-

ferent participants each time; the most common reasons for this were that par-

ticipants had busy work schedules and other commitments that prevented

them from attending regular meetings.

“I’ve been to a meeting, one of my senior directors has been to a meeting, we

all go whenever one of us is free and I think that is just natural, you can’t

guarantee that the core people will be there” [Architect, Case One]

Several participants suggested that this can have a negative effect on progress

and inhibited the development of a shared understanding of purpose, process

and design intent.

Forming and sustaining a partnership has therefore been identified by the au-

thors as a key factor to the success of a whole system design project. It has been

observed that recruiting, and subsequently nurturing, the most appropriate ex-

perience and expertise for the design context can be overlooked or assumed,

however, is necessary for a cohesive and successful whole system design team.

3.2 Human and non-human interactionThe importance of frequent communication between actors in and between

meetings was highlighted by study participants on numerous occasions and

several proposed that it can substantially influence the process of successful in-

tegration. Participants suggested that many of the delays within the automo-

tive case were due to actors not communicating their design decisions early

enough in the process. It was stressed that design decisions need to be commu-

nicated, however small, as they may have a significant impact upon other com-

ponents and ultimately affect the whole system. Participants in three of the six

cases suggested that other team members were often unaware of the high levels

of interaction required of them, particularly if a component was perceived not

to be influential to their part of the design.

“I can’t be going to all the meetings because a lot of the stuff isn’t relevant”

[Designer, Automotive Case]

“Unless he, as an architect, perceives that his design can benefit from talking

with the engineers, there’s nothing in the contractual arrangements that ex-

isted. So unfortunately industry is set up to avoid any of this (interaction)”

[Environmental Consultant, Case Four]

During project meetings, within the Automotive Case, it was common for dis-

cussions to digress into areas of detailed design. When asked about this, par-

ticipants suggested that although detailed discussion is necessary, it often

inhibits open discussion and participation during meetings. Several partici-

pants suggested that encouraging discussions to return to a system level dur-

ing design team meetings encourages all to participate and share ideas;

Exploring whole system design 165

Page 12

Author's personal copy

additionally this is when linkages between sub-systems are most likely to be

identified.

“I think partners need to be reminded at the beginning of every meeting that

discussions are to be kept at a whole system level” [Designer, Automotive

Case]

A lack of communication was observed, by the researcher, to inhibit the prog-

ress of integration and occasionally design decisions, which had not been

communicated to the rest of the team, resulted in substantial delay later on

in the process. To prevent this within future whole system design projects,

it is recommended that actors should be made aware of the requirements

and expectations that a whole system design process demands, early on within

the project.

Based on the study, the authors conclude that maintaining system-level con-

versations, during design meetings, supports team progress and efficiency. It

was observed that necessary detailed design discussions were more effective

when held between a select group of participants outside of whole team

meetings.

3.3 Individual characteristicsMany of the participants in the study referred to the process of whole sys-

tem design as being different from a traditional design process and therefore

requiring different skills.

“It is a completely different skill set . you have to be able to view things

from the outside of the object, you have to be able to look down on the object”

[Architect, Case One]

During a discussion with architects from Case Four it was agreed that it was

trans-disciplinary skills, such as the willingness to learn across boundaries and

the ability to think systemically, that were required to significantly enable ac-

tors to appreciate the impact of their design decisions upon other sub-systems

and the final design solution. In accordance with this a designer from case five

identified that in addition to high levels of domain specific expertise, actors

who have that expertise will also have a broad and inquisitive view to

make the process of whole system design easier and more successful.

“I am sure that the role or prestige of the specialist has reached its absolute

zenith. there is an ever increasing role for polymaths and I think the day of

the polymath is returning because in whole system design that is the core

skill” [Designer, Case Five]

Findings suggest that it is necessary for actors participating in a whole system

design process to possess a balance of discipline specific and trans-disciplinary

skills. It was observed by the researcher that participants who were familiar

with traditional design processes tended to find utilising trans-disciplinary

166 Design Studies Vol 32 No. 2 March 2011

Page 13

Author's personal copy

skills difficult. Additionally, skills such as thinking systemically are difficult to

teach. It is suggested that sourcing actors who already possess and under-

stand the benefits of utilising trans-disciplinary skills should be part of the re-

cruitment process within a whole system design project. However, these skills

are difficult to spot and therefore guidelines about how to identify those re-

quired characteristics should be developed early on. An example of this is

the search for actors who display an enthusiasm to further their own learning

and development and who show interest in areas aside from their own area of

disciplinary expertise.

3.4 Understanding of purpose and processFindings early on in the study suggested that participants found a whole sys-

tem approach to be different to that of a more traditional design process. Fur-

thermore, multiple participants within the automotive case in particular

suggested that they were unsure of what a whole system approach was and

what the benefits of undertaking the approach were.

“I don’t know what a whole system design is expected to be” [Designer,

Automotive Case]

It was observed that the benefits of whole system design were often not dis-

covered by participants until later on in the design process. Once these bene-

fits became more evident to these participants then the purpose and process of

the project became more apparent. Participants from four of the six case stud-

ies agreed that if the whole system approach, that they were expected to

adopt, had been comprehensively highlighted to them at the beginning of

the project, along with the reasoning behind that decision, then it would

have made the process easier to adopt. Participants said that it had been

wrongly assumed that all team members were aware of what a whole system

approach was.

One participant within case five assigned the confusion around the approach

to the fact that the process of whole system design is new to everyone and

therefore still needs exploring.

“At the moment we are not very good at it (whole system design) and we

haven’t had much practise; no one has. We haven’t had very long to work

out how to put whole system design teams together at all” [Designer, Case

Five]

One aspect that actors within the study found challenging was the concept of

emerging properties; i.e. qualitatively new situations that arise out of the in-

terconnections within the design process. Parts of a whole system design often

appear counter-intuitive unless the system is regarded as a whole.

“Its completely counter-intuitive. we are asking people to counteract an aw-

ful lot of what they’ve learnt. How can you seriously expect to get a more

Exploring whole system design 167

Page 14

Author's personal copy

efficient, cheaper or lighter car by choosing a component that’s less efficient,

more expensive or heavier; I mean its just inconceivable for most people”

[Designer, Automotive Case]

It is the authors’ opinion that this demonstrates the necessity for actors to de-

velop a shared sense of purpose and process within the context of the whole

system, including its emergent properties. Often the benefits of a design can-

not be seen as emergent properties are not being included; from this view it is

counter-intuitive. Subsequently, the ability of actors to identify linkages be-

tween the components of a design, leading to the identification of emergent

properties, largely influences the process of whole system design.

Itwas observed by the researcher in all case studies that themore cohesive a team

becomes the easier it is to form a shared understanding of purpose. However, it

appears that the principles of whole systemdesign are frequentlymisunderstood

or unknown and therefore it should not be assumed that all actors have a shared

understanding of the process required to reach a whole system solution.

3.5 Alignment of interestsThroughout the study different sets of interests were observed; the reasons why

individual actors and organisations wanted to be part of the process and what

they wanted to achieve from its outcome, the interests of the project team and

the interests of the intended consumers or those who are to benefit from the

project outcomes.

A designer from case five suggested that in traditional design the consumer and

the manufacturer are at “polar opposites” in terms of needs and requirements.

Participants from the automotive case expanded upon this:

“The manufacturer of a motor vehicle wants to make an ongoing profit

through regularly servicing the vehicle once it is sold and relies on compo-

nents needing to be replaced; to an extent failure is built into the design.

On the other hand the consumer requires reliability, efficiency and quality

and does not want to be frequently spending more money, time and effort

on replacing components of the vehicle”

Participants highlighted that establishing an alignment of interest between all

parties involved in the process of a whole system design was “fundamental”

and that “the opposition of interests is a terminal barrier to whole system de-

sign”. In all cases participants agreed that openness and honesty surrounding

the interests, expectations and requirements of all parties was an important

aspect to achieving a more cohesive and holistic solution, which satisfies as

many requirements as possible.

Study participants from case four proposed that currently there is a lack of

alignment between legislation and the aims of whole system design.

168 Design Studies Vol 32 No. 2 March 2011

Page 15

Author's personal copy

“The vast majority of the industry will design to building regulations such as

the Code for Sustainable Homes. But it should be right and it certainly can’t

stop the innovators from innovating which currently it actually, categorically,

definitely, absolutely is doing and it should certainly not be pushing the non-

innovators down a bad road which it is” [Managing Director, Case Four]

Architects in Case One added to the argument by suggesting that targets and

legislation can prevent designers from achieving the most optimum and effi-

cient solutions.

Based on the findings of the research the authors’ conclude that not only

does an alignment of interests need to be identified between a project team

and its intended consumers, but alignment also needs to be sought from the

policy makers that are imposing stringent targets and legislation upon those

projects.

To achieve this alignment, it is recommended that requirements, needs, expec-

tations and concerns of all partners should be discussed openly early on in the

design process. Partners should be encouraged to be honest about what moti-

vates and drives them. It was observed by the researcher that it is common at

the start of a project for actors to ‘keep the peace’ by agreeing with shared

goals; however this could lead to conflict later on in the process.

3.6 Sense making and system boundariesAs the process of whole system design is frequently unclear it was assumed that

sense making (Weick, 1995) would play a large part in that process. However,

sense making activities often occur sub-consciously and it was therefore diffi-

cult for participants within the study to recognise and relay specific accounts of

when they had occurred. However, through observations and prompting ques-

tions it was uncovered that sense making activities took place within each case

study and that such activities have a substantial impact upon the whole system

design process.

In most of the case studies participants proposed that it was necessary to spend

time making sense of what a system was and where the system boundary

should be drawn. Participants within Cases three and four suggested that often

team members differ in what they perceive to be the most important aspect of

a design. They recommended that to enable the team to make sense of the

whole system, each team member should be asked to draw and detail the de-

sign from their perspective. This may enable a comprehensive architecture of

the system to be developed which includes multiple perspectives and require-

ments. Additionally, based on the research, the authors’ propose that the de-

velopment of a shared architecture would enable actors to identify linkages

between different components of the design more easily.

Exploring whole system design 169

Page 16

Author's personal copy

The use of system boundaries was re-occurring within the cases studied and

was identified as a sense making mechanism. Several participants suggested

that system boundaries were used to provide actors with security and cer-

tainty regarding the problem context. However not all participants were com-

fortable with the use of system boundaries: “by identifying a system boundary

we are trying to impose a limit on something that normally grows organically”

(Designer, Case Two). It was observed in cases three and four that occasion-

ally system boundaries could inhibit interaction and the development of rela-

tionships with the external environment. One architect explained that

physical limits may be decided upon and referred to as the boundary of the

system; for example, the edge of a city. However, within a whole system de-

sign process he suggested that it was inevitable that parts of that boundary

are going to be blurred and crossed and emergent properties in time will ex-

pand that limit.

The process of whole system design is complex due to the integration of mul-

tiple stakeholders and perspectives. Based on the findings of the research the

authors propose that sense making techniques such as forming a common lan-

guage and sense of purpose can assist with creating a project view and archi-

tecture. Additionally, defining a system boundary is a way of simplifying the

complexity of a whole system design. However, it is recommended that this

should not be used as an enduring structure as eventually the complexity of

the system needs to be acknowledged.

3.7 Facilitating whole system designFindings from the automotive case revealed a substantial feeling of uncer-

tainty, amongst the design team, surrounding the process of whole system de-

sign and study participants suggested that this uncertainty was inhibiting

progress. Following discussions surrounding uncertainty and ambiguity with

participants from all case studies the feeling of uncertainty was proposed to

be closely related to the absence of a leader or manager.

“So the house builders build their houses, the architects design them, the

school workers deal with the school issues, the social workers deal with the

youth issues but who on earth is supposed to manage the whole system? At

the moment in East Leeds I have no idea who is overseeing that system

approach” [Architect, Case One]

The role of ‘facilitator’ was observed to be missing within each of the case in

the study. It is suggested that it is important, within a whole system design, to

have an individual who is able to regard the system from above and to iden-

tify gaps or potentially overlooked relationships between sub-systems.

The role of the facilitator should not be confused with a leader or manager. It

was observed within the study that the most successful whole system design

projects were managed through a flattened hierarchy. Participants from

170 Design Studies Vol 32 No. 2 March 2011

Page 17

Author's personal copy

multiple cases agreed that a flattened hierarchy, consisting of stakeholders with

equal shares in the project, encourages joint ownership and democratic gover-

nance. Participants from cases three and five suggested that the encouragement

of shared ownership amongst a design team led to a feeling of empowerment

and allowed decisions to be made more efficiently. Additionally, participants

thought that where a strong sense of ownership did exist, members of a team

were more likely to tie their identity to a project’s outcome, thus injecting extra

effort to ensure its success. This was observed throughout the study and it is the

authors’ opinion that this went a long way to supporting the process of whole

system design. Cases in which a flattened hierarchy was successful (the automo-

tive case and cases two and five) also appeared to positively influence job sat-

isfaction as actors said that they felt valued and their ideas were being

recognised without having to get every single aspect signed off. The concept

of shared responsibility and a flattened hierarchy was observed to work at its

best when accompanied by a substantially integrated team.

3.8 IntegrationThe successful integration of actors and disciplines was identified across all six

case studies as being central to the identification of advantageous relationships

within the system and therefore key to the success of a whole system design.

Participants proposed that the blurring of individual roles and disciplinary

boundaries enables cross-disciplinary learning to be achieved and subse-

quently the impact of design decisions are more readily appreciated. However,

it was identified early on in the automotive case that the blurring of roles can

mean that responsibility is not accounted for.

“Tasks are ignored and no one takes responsibility until eventually someone

is forced to. Usually that task is not that person’s role or responsibility”

[Engineer, Automotive Case]

Subsequently this can result in components being missed out all together.

“I always worry that we’re missing something; that the consortium is missing

something. Obviously you can do the best you can but I always have this hor-

rible feeling that there’s going to be a gap between two bulk heads where

a wire should be” [Designer, Automotive Case]

Towards the endof theautomotive case study,participants suggested that this chal-

lenge could have been more effectively addressed through the development of

a more cohesive team and cross-disciplinary learning. This was supported by par-

ticipants fromother case studies and it was suggested that although not every actor

need understand the details of each and every component it is beneficial to under-

stand how each component impacts upon the rest of the system:

“You don’t have to understand every single detail of how they work, its

much more important to have a feel for what they do and how they fit

into the system” [Designer, Case Five].

Exploring whole system design 171

Page 18

Author's personal copy

Additionally, all six cases highlighted that for successful integration to take

place, each actor needed to possess the skills to ‘monitor’ ‘keep an eye on’

or ‘be aware of’ the whole system:

“You’d need the blurring of roles and you need, either you’d need someone

who is on top looking down or you need a great deal of curiosity from every

body involved” [Designer, Automotive Case].

Ultimately the integration of a design team was seen to have a substantial im-

pact upon the success of a whole system design project. Successful integration

has been observed to positively influence the other factors necessary for good

whole system design; particularly ‘forming and sustaining a partnership’, ‘hu-

man and non-human interaction’, ‘understanding of purpose and process’

and the ‘alignment of interests’. Additionally, developing an integrated team

is significantly assisted by the role of the facilitator as discussed in Section 3.7.

4 Discussing whole system designSection 1.1 highlighted that due to a lack of literature surrounding the process

of a whole system design; it was not possible to develop a precise definition,

this is now possible following the longitudinal observation of one case study

and engagement with five additional cases.

“Whole system design is an integrated and emergent approach to the design

of more radically innovative and sustainable solutions. It encourages those

involved to look at a problem as a whole; take multiple factors into account

and utilise relationships between different parts of the problem as opposed to

addressing one aspect at a time”

This paper has highlighted a number of factors that have been observed to

substantially influence the success of a whole system design process and in do-

ing so has created a framework to guide designers who will undertake such

a process. Figure 2 presents this framework and highlights some of the key

findings from the study that enable and inhibit the process.

Over the last decade there has been a significant increase in public awareness

surrounding the issues we are facing regarding environmental sustainability.

Consequently this has also had an impact upon the enhanced understanding

of ways in which improved sustainability can be achieved. Although not

guaranteeing a more sustainable outcome, whole system design is one ap-

proach that has been suggested by authors as providing more optimised and

innovative solutions that can achieve higher levels of sustainability at a whole

system level. Involving multiple stakeholders and perspectives within the de-

sign process and creating an integrated holistic view of the system should build

adaptability and flexibility into the design solution.

The process of integrating multiple perspectives, needs and requirements is not

without its challenges. Those involved within a whole system design process

172 Design Studies Vol 32 No. 2 March 2011

Page 19

Author's personal copy

Fo

rm

in

g a

nd

S

us

ta

in

in

g a

P

artn

ers

hip

Who

le S

yste

m D

esig

n re

quire

s th

e in

tegr

atio

n of

m

ultip

le p

ersp

ectiv

es c

ombi

ned

with

co

mpl

emen

tary

exp

ertis

e, e

xper

ienc

e, a

bilit

y an

d co

mpe

tenc

e

Influencing Factors

A wide spanning social network enables the formation of partnerships

Maintaining a core design team throughout the project enables the development of a shared

understanding of purpose, process and design intent

Utilising existing contacts based on familiarity or convenience may inhibit access to relevant

expertise

Spending time and effort on recruiting the ideal actors enables the development of a successful

and cohesive team

Hu

man

a

nd

N

on

-H

um

an

In

te

ra

ctio

n

Who

le s

yste

m d

esig

n re

quire

s fre

quen

t co

mm

unic

atio

n be

twee

n al

l par

ts o

f the

sys

tem

Influencing Factors

Frequent communication between actors enablesintegration

Interconnectivity between sub-systems, systems and the external environment enables advantageous relationships to be discovered and enables the

elimination of components

System level discussion during project meetings promotes equal participation

Reluctance to interact with external systems inhibitsthe implementation of the final solution

In

div

id

ua

l C

ha

ra

cte

ris

tic

s

Who

le s

yste

m d

esig

n re

quire

s ac

tors

to

have

a b

alan

ce o

f dis

cipl

ine

spec

ific

expe

rtise

and

tran

s-di

scip

linar

y sk

ills Influencing Factors

Possessing a balance of discipline-specific expertise and trans-disciplinary skills allows actors to appreciate

the impact of design decisions

Familiarity with traditional design processes can hinder the ability to think systemically

A lack of knowledge and education can restrict understanding of the benefits surrounding trans-

disciplinary skills

A lack of incentive inhibits the development and utilisation of trans-disciplinary skills

Un

derstan

din

g o

f P

urp

ose an

d P

ro

cess

It is

nec

essa

ry to

dev

elop

a s

hare

d un

ders

tand

ing

of th

e en

d go

al a

s w

ell a

s th

e w

hole

sys

tem

s ap

proa

ch b

eing

ado

pted

to g

et th

ere Influencing Factors

Regarding the system as a whole enables the identification of emergent properties and the understanding of counterintuitive aspects

Lack of commitment hinders the process of whole system design

A lack of understanding surrounding the principles of whole systems design significantly delays the process

Overlooking emergent properties restricts the development of a whole system solution

Alig

nm

en

t o

f In

terests

An a

lignm

ent b

etw

een

indi

vidu

al a

nd

proj

ect m

otiv

atio

ns a

nd a

lso

betw

een

the

proj

ect

and

the

inte

nded

con

sum

ers

Influencing Factors

Establishing an alignment of interests between actors, the project team AND the potential consumers enables the development of a more comprehensive

solution

Openness and honesty between actors early on in the design process allows an

alignment to be established

Un-communicated motivations can lead to conflict and ineffective integration

Legislation can inhibit innovative design as opposed to increasing motivation

Sen

se M

akin

g

Sens

e m

akin

g ac

tiviti

es a

re n

eces

sary

to g

ain

a sh

ared

und

erst

andi

ng o

f the

sys

tem

and

the

mul

tiple

per

spec

tives

ass

ocia

ted

with

it Influencing Factors

Gaining an understanding of cross-disciplinary

terminology promotes effective communication

Making sense of perspectives, requirements, needs etc. is required for a shared architecture of the final

solution to be developed

Defining system boundaries enables actors to develop structure and certainty

A system boundary can restrict the development of relationships with the external environment

Facilitatin

g W

ho

le S

ystem

D

esig

n

The

role

of t

he fa

cilit

ator

enc

oura

ges

join

t ow

ners

hip,

a s

hare

d de

moc

racy

and

a fl

atte

ned

hier

arch

y

Influencing Factors

The presence of a facilitator promotes the reduction

of uncertainty and increases integration

The facilitator identifies gaps in the system

A flattened hierarchy enables shared ownership

Shared ownership encourages participation, empowerment to make decisions, and efficiency

A shared democracy assists the role of the facilitator

Feelings of uncertainty hinder decision making and progress

In

te

gra

tio

n

Inte

grat

ion

of a

ctor

s, d

isci

plin

es a

nd s

ub-s

yste

ms

is n

eces

sary

to d

evel

op a

n op

timis

ed a

nd h

olis

tic

solu

tion

Influencing Factors

High levels of collaboration enable more successful integration

The blurring of disciplinary boundaries and roles assists cross-disciplinary learning

Cross-disciplinary learning allows the impacts of design decisions to be appreciated and linkages to

be identified

The blurring of roles can result in the exclusion of components

Familiarity with traditional design processes can inhibit integration

Figure 2 Factors that influence a whole system design process

Exploring whole system design 173

Page 20

Author's personal copy

are recommended to look to research in the disciplines of collaborative, multi-

disciplinary and participatory design and also concurrent engineering. This re-

search outlines some of the key factors that inhibit integrative working such as

the difficulty of maintaining a core design team (Lee, 2008) and the frequent

lack of communication (Sonnenwald, 1996). Literature within these disciplines

has also highlighted methods and techniques for improving successful integra-

tion such as the need for sense making activities (Klein, Moon, & Hoffman,

2006; Weick, 1995), the development of a shared understanding

(Kleinsmann, 2006) and the use of extended social networks for access to

relevant knowledge and expertise (Granovetter, 1973, 1983, Leenders, van

Engelen, & Kratzer, 2003).

One of the key principles of whole system design is the identification and use of

beneficial relationships and linkages between different parts of a system to ulti-

mately optimise the whole. The study has shown that it is important for stake-

holders to have an understanding of the benefits of taking this approach.

Studies within the disciplines of Enterprise Architecture and Enterprise Engi-

neering demonstrate the necessity of gaining a perspective of the whole system

and suggest methods for how designers can better understand and monitor the

relationships between sub-systems, systems and the external environment

(Giachetti, 2010). The research reported in this paper suggests that there are

certain characteristics that assist stakeholders in developing and maintaining

a systems view such as the use of systems thinking, the ability and to learn across

disciplines and the curiosity and incentive to monitor an assembly of sub-sys-

tems. Katzenback and Smith (1993) agree that for a design team to respond

to multi-faceted challenges they need to have a broad range of skills and

knowhow.Literature surrounding thedevelopment andutilisationof trans-dis-

ciplinary skills in design is limited (Cabrera, Colosi, &Lobdell, 2008).However,

stakeholders can learn from literatures that have explored the use of systems

thinking, such as Senge (2006) and Katzenback and Smith (1993) and identify

how these skills can be applied to more innovative and sustainable design.

There is still much discussion between disciplines surrounding what constitutes

a system and how the use of system boundaries can assist and hinder the process

of design. A useful way of thinking of a system is to define it as a ‘system of in-

terest’ (Checkland, 2000; Collins, Blackmore, Morris, &Watson, 2007). Collins

et al., (2007) suggest that systems thinking involves being aware of systems of

interest in their contexts and acknowledging what they are affected by and af-

fect. System boundaries can change over time so what might be contextual at

one time might be within a system of interest at another, therefore people

make at least implicit boundary judgements about what lies within or outside

of them (Collins et al., 2007). The study identified the need for an alignment

of interest to be formed between a system and its external environment or a de-

sign solution and its user/the consumer. Designers can learn from research

which has prescribed methods and techniques surrounding how this alignment

174 Design Studies Vol 32 No. 2 March 2011

Page 21

Author's personal copy

can be achieved. An example of this is the modelling of requirements to enable

the successful communication of goals, targets which also assists the decision-

making processes and makes themmore transparent (Stechert & Franke, 2009).

Additionally, literature in participatory design supports findings that a facilita-

tor plays a principle role within an integrated design team by having the ability

to oversee the relationships between systems (Brown, 2008; Wojanh, Dyke,

Riley, Hensel, & Brown, 2001). As the role of a facilitator within a whole system

design team is not clearly defined then this literature provides stakeholders with

a valuable insight.

Table 2 Learning from previous research

Theme Disciplines that can belearnt from

Specific Literature

Forming and Sustaining a PartnershipWhole System Design requires theintegration of multiple perspectives;complementary expertise, experience,ability and competence

New Product DevelopmentParticipatory DesignSocial Network TheoryEnterprise Architecture

Leenders et al., 2003Lee, 2008Granovetter, 1973, 1983Giachetti (2010)

Human and Non-Human InteractionWhole system design requires frequentcommunication between all parts ofthe system

Team-based designCollaborative DesignConcurrent Engineering

Leenders et al., 2003Sonnenwald, 1996

Individual CharacteristicsWhole system design requires actors tohave a balance of discipline specificexpertise and trans-disciplinary skills

Systems TheorySustainable Design

Katzenback & Smith, 1993Senge, 2006Cabrera et al., 2008

Understanding of Purpose and ProcessIt is necessary to develop a sharedunderstanding of the end goal as wellas the whole systems approach beingadopted to get there

Enterprise ArchitectureCollaborative Design

Giachetti (2010)Kleinsmann & Valkenburg, 2008Kleinsmann, 2006Dong, 2005

Alignment of InterestsAn alignment between individual andproject motivations and also between theproject and the intended consumers

Multi-Disciplinary DesignParticipatoryDesign

Stechert & Franke, 2009

Sense Making and System BoundariesSense making activities are necessary togain a shared understanding of the systemand the multiple perspectives associatedwith it

Organizational StudiesCollaborativeDesignEnvironmental Science

Weick, 1995Klein et al., 2006Kleinsmann, 2006Kalay, 2001Collins et al., 2007

Facilitating Whole System DesignThe role of the facilitator encourages jointownership, a shared democracy and aflattened hierarchy

ParticipatoryCross Functional Teams

Luck, 2007McDonough, 2000

IntegrationIntegration of actors, disciplines andsub-systems is necessary to develop anoptimised and holistic solution

Collaborative DesignHealth and Social CareConcurrent Engineering

Kleinsmann, 2006Brown, 2008Wojanh et al., 2001

Exploring whole system design 175

Page 22

Author's personal copy

This discussion has suggested that whole system design shares many attributes

with other approaches to design and therefore it is important that we can learn

from these disciplines. Table 2 highlights the key themes of the study alongside

some of the relevant literature. This literature has been identified as being able

to contribute to knowledge surrounding the process of whole system design.

As the practise and process of whole system design crosses so many disciplin-

ary boundaries this table is intended to assist design practitioners in sourcing

relevant literature to guide them through that process.

5 ConclusionsThe study presented within this paper aimed to provide insight into the process

of whole system design and to identify the factors that influenced its success.

The automotive case study provided unique access to an operational whole

system design project from beginning to end. This enabled valuable knowledge

to be gained into multiple aspects of the complex design process.

The adoption of a qualitative, exploratory and inductive approach enabled the

collection of a vast amount of primary data without any predetermined judge-

ments as to what factors were most pertinent. As more data was obtained the-

matic analysis was used to identify patterns and relationships within the data.

This resulted in the consolidation of 10 themes that were thought to be com-

mon to a whole system design process.

The second phase of the study allowed initial results to be validated and mod-

ified across a number of additional design contexts. This was important to

ensure that the final framework of factors could be accessible to as wide

a community of designers as possible.

Results of the study indicated that there are multiple factors that influence the

success of a whole system design process. The paper highlights these factors

and uses examples from the cases to demonstrate best practise within whole

system design. The identification of relationships between parts of a system

to ultimately optimise the whole, the need for actors involved in the process

to develop trans-disciplinary skills and the dynamics of a flattened hierarchy

were identified as being some of the key necessities of whole system design.

At the beginning of this study in February 2006, understanding surrounding

whole system design was limited and there was a multitude of terminology sur-

rounding holistic approaches to design. Four years on, it appears that there is

still no consensus as to the terminology being used to describe holistic ap-

proaches to design and, additionally, there has only been a limited increase

in the utilisation of the term ‘whole system design’. There has, however,

been a noticeable increase in the understanding surrounding some of the key

principles that a whole system approach to design promotes. The development

of national and international partnerships across disciplines, thinking

176 Design Studies Vol 32 No. 2 March 2011

Page 23

Author's personal copy

systemically, and involving stakeholders within the design process (Luck,

2007), are increasingly being recognised as necessary components of more sus-

tainable design.

Oneof thebiggest challenges facedby thosewishing topromoteapproaches suchas

whole system design is the ability to encourage designers, developers, engineers,

planners, strategists, and government officials to think holistically and to view

the bigger picture. For decades we have been taught and trained to develop disci-

plinary expertise, and to view theworld fromwithin that discipline, and therefore it

is understandably difficult and counter-intuitive for experts to begin to learn from,

interact and integrate with other disciplines. There is evidence, however, that styles

of teaching are recognising the need to think holistically and to develop trans-dis-

ciplinary skills and understanding. TheNatural Edge Project inAustralia operates

as a partnership for education and develops curriculum supplements for students

from the age of 10 regarding sustainable education one of which is entitled ‘Whole

SystemDesign:An IntegratedApproach toEngineering’ (TNEP, 2008).Addition-

ally the SchumacherCollege in theUKhas recently introduced anMSc in ‘Holistic

Science’whichcalls intoquestion ‘western scientificmethodswhichhavebeendom-

inated by specialisation in disciplines and by reductionism’ and instead ‘explores

new trans-disciplinarymethodologies that go beyond reductionism in understand-

ing whole systems’ (Schumacher College, 2008). This investment into the develop-

ment of skills to support the design of more innovative and holistic design for the

future is encouraging.

AcknowledgementsThe authors would like to thank the experts who took part in this study for

their kindness and patience whilst frequently being observed and interviewed.

Also thanks to the anonymous reviewers who pointed out weaknesses and

made useful suggestions for improving the paper. All remaining weaknesses

are, of course, our own.

ReferencesAnarow, B., Greener, C., Gupta, V., Kinsley, M., Henderson, J., Page, C., &

Parrot, K. (2003). Whole-Systems framework for sustainable consumption andproduction, Vol. 807. Denmark: Danish Ministry of the Environment. 1e51.

Bhamra, T., & Evans, S. (1997). An ecodesign model based on industry experi-ence. IEE Colloquium (Digest) 37e40.

Blanchard, B., & Fabrycky, W. (2006). Systems engineering and analysis (4th

edn.). USA: Pearson Prentice Hall.Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative

Research in Psychology, 3, 77e101.Brezet, H. (1997). Dynamics in eco design practise. UNEP Industry and Environ-

ment, 20, 1e2.Brown, B. (2008). Blurred roles and permeable boundaries: the experience of mul-

tidisciplinary working in community mental health. Health and Social Care in

the Community, 8, 425e435.

Exploring whole system design 177

Page 24

Author's personal copy

Cabrera, D., Colosi, L., & Lobdell, C. (2008). Systems thinking. Evaluation andProgram Planning, 31, 299e310.

Checkland, P. (2000). Soft systems methodology: a thirty year retrospective. Sys-tems Research and Behavioral Science, 17, 11e58.

Coley, F., & Lemon, M. (2008). Supporting multiple stakeholders in the design ofmore sustainable and systemic solutions: the case of the sustainable sports car.Design Principles and Practises: An International Journal, 2, 27e33.

Coley, F., & Lemon, M. (2009). Exploring the uncertainty surrounding the designand perceived benefit of sustainable solutions. Journal of Engineering Design,20, 543e554.

Collins, K., Blackmore, C., Morris, D., & Watson, D. (2007). A systemic ap-proach to managing multiple perspectives and stakeholding in water catch-ments: some findings from three UK case studies. Environmental Science and

Policy, 10, 564e574.Dong, A. (2005). The latent semantic approach to studying design team commu-

nication. Design Studies, 26, 445e461.Ehrenfeld, J. (2003). Putting the spotlight on metaphors and analogies in indus-

trial ecology. Journal of Industrial Ecology, 7, 1e4.Forsberg, K., & Mooz, H. (1998). System engineering for faster, cheaper, better.

Center of Systems Management.

Giachetti, R. (2010). Design of enterprise systems: Theory, architecture, andmethods. CRC Press.

Gibson, C. (2001). From knowledge accumulation to accommodation: cycles of

collective cognition in work groups. Journal of Organisational Behaviour, 22,121e134.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology,

78, 1360e1380.Granovetter, M. (1983). The strength of weak ties: a network theory revisited.

Sociological Theory, 1, 201e233.Hawken, P., Lovins, A., & Lovins, H. (1999). Natural capitalism creating the next

industrial revolution. USA: Little Brown.Hebel, M. (2007). Light bulbs and change: systems thinking and organisational

learning for new ventures. The Learning Organization, 14, 499e509.

Howarth, G., & Hadfield, M. (2006). A sustainable product design model.Materials and Design, 27, 1128e1133.

Kalay, Y. E. (2001). Enhancing multi-disciplinary collaboration through seman-

tically rich representation. Automation in Construction, 10, 741e755.Katzenback, J., & Smith, D. (1993). The wisdom of teams. Massachusetts:

Harvard Business School Press.Klein, G., Moon, B., & Hoffman, R. F. (2006). Making sense of sensemaking:

alternative perspectives. IEEE Intelligent Systems, 21, 70e73.Kleinsmann, M. (2006). Understanding collaborative design. PhD Thesis. Nether-

lands, Delft University.

Kleinsmann, M., & Valkenburg, R. (2008). Barriers and enablers for creat-ing shared understanding in co-design projects. Design Studies, 29,369e386.

Krippendorff, K. (2004). Content analysis: An introduction to its methodology.USA: Sage Publications.

Lee, Y. (2008). Design participation tactics: the challenges and new roles for

designers in the co-design process. Co-Design, 4, 31e50.Leenders, R., van Engelen, J., & Kratzer, J. (2003). Virtuality, communication

and new product team creativity: a social network perspective. Journal ofEngineering and Technology Management, 20, 69e92.

178 Design Studies Vol 32 No. 2 March 2011

Page 25

Author's personal copy

Lemon, M. (Ed.). (1999). Exploring environmental change using an integrativemethod. Reading: Gordon and Breach. (ISBN 90-5699-193-0).

Lofthouse, V. (2004). Investigation into the role of core industrial designers inecodesign projects. Design Studies, 25, 215e227.

Luck, R. (2007). Learning to talk to users in participatory design situations.Design Studies, 28, 217e242.

Mankin, D., Cohen, S., & Fitzgerald, S. P. (2004). Developing complex collabora-

tions: Basic principles to guide design and implementation. (Beyerlein).McDonough, E. (2000). Investigation of factors contributing to the success of

cross-functional teams. Journal of Product Innovation Management, 17,

221e235.Morson, K. (2007). The push to define ‘green’. Available at: http://www.serve.

com/Lincolnheritage2/articles/address/2004toNow/environment/2007-08-The_

Push_to_Define_Green.htm (Accessed on 23.11.2006).Pahl, G., & Beitz, W. (1996). Engineering design e A systematic approach (2nd

ed.). London: Springer.Postrel, S. (2002). Islands of shared knowledge: specialization and mutual under-

standing in problem solving teams. Organization Science, 13, 303e320.Rocky Mountain Institute. (2006). Hypercar design and technology. Available

from: https://old.rmi.org/sitepages/pid390.php (Accessed on 07.02.2008).

Schumacher College. (2008). Transformative learning for sustainable living.Available at: http://www.schumachercollege.org.uk/ (Accessed on 20.06.2008).

Seiffert, M., & Loch, C. (2005). Systemic thinking in environmental management:

support for sustainable development. Journal of Cleaner Production, 13,1197e1202.

Senge, P. (2006). The Fifth discipline. UK: Century.

Sherwood, D. (2002). Seeing the forest for the trees. London: Brealey.Sonnenwald, D. (1996). Communication roles that support collaboration during

the design process. Design Studies, 17, 277e301.Stasinopoulos, P., Smith, M., Hargroves, K., & Desha, C. (2009). Whole system

design: An integrated approach to sustainable engineering. London: EarthScan.Stechert, C., & Franke, H. J. (2009). Managing requirements as the core of multi-

disciplinary product development. CIRP Journal of Manufacturing Science and

Technology, 1, 153e158.The Natural Edge Project. (2008). The natural edge project. Available at: http://

www.naturaledgeproject.net/ (Accessed on 20.06.2008).

Weick, K. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage.Wojanh, P., Dyke, J., Riley, L., Hensel, E., & Brown, S. (2001). Blurring bound-

aries between technical communication and engineering: challenges of a multi-disciplinary, client-based pedagogy. Technical Communication Quarterly, 10,

129e148.

Exploring whole system design 179