BOUNDARY MATTERS: THE DYNAMICS OF BOUNDARY OBJECTS, INFORMATION INFRASTRUCTURES, AND ORGANISATIONAL IDENTITIES
by
URI GAL
A Dissertation Submitted in Partial Fulfilment of the Requirements for the Degree of Doctor of Philosophy
Department of Information Systems CASE WESTERN RESERVE UNIVERSITY
May, 2008
CASE WESTERN RESERVE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
We hereby approve the thesis/dissertation of
Uri Gal______
Candidate for the Ph.D.______degree*.
(Signed) Kalle Lyytinen______(Chair of the committee)
Richard Boland______
Susan Leigh Star______
Youngjin Yoo______
(date) 11/02/07
* We also certify that written approval has been obtained for any proprietary material contained therein
2 Table of Contents
Abstract……………………………………………………………………………...6 1. Introduction...... 7 2. Theoretical Foundations...... 15 2.1. Information infrastructures ...... 15 2.2. Organisational identities ...... 22 2.3. Boundary objects ...... 28 3. Research Context ...... 36 3.1. The AEC industry ...... 36 3.2. 2D and 3D modelling technologies in the AEC industry ...... 41 4. Research Methodology ...... 46 4.1. Empirical case: Changes in modelling technologies in the AEC industry ...... 46 4.2. Research Sites: Hoffman Construction Company and A. Zahner Company..... 48 4.2.1. Hoffman Construction Company...... 48 4.2.2. A. Zahner Company...... 48 4.3. Research design ...... 50 4.4. Sampling ...... 51 4.5. Data collection and analysis...... 55 5. Results ...... 59 5.1. Hoffman ...... 60 5.1.1. The poker dealer: Hoffman in typical 2D construction projects...... 60 5.1.2. The dispersed collaborator: Using 3D modelling technologies with Gehry Partners ...... 67 5.1.3. The Air-traffic controller / knowledge broker: The Seattle library project...... 72 5.2. Zahner ...... 76 5.2.1. The Detached Worker: Zahner in typical 2D construction projects ...... 76 5.2.2. The Fixer – Zahner in the construction of the Hunter Museum of American art...... 82 5.2.3. The seamless collaborator: Using 3D modelling technologies with Randall Stout Architects ...... 90 6. Discussion...... 99 7. The Interrelationships of Boundary Objects, Organisational Identities, and Information Infrastructures ...... 113 7.1. How do 2D and 3D modelling technologies, as boundary objects, shape or are shaped by organisational practices and patterns of communication? ...... 113 7.2. How do 2D and 3D modelling technologies shape or are shaped by the identities of the organisations that use them? ...... 115 7.3. What are the implications of changes in modelling technologies for the identities, practices, and interactions of the organisations that share them? ...... 116 8. Contribution to Existing Literature...... 122 9. Implications ...... 132 10. Limitations and Future Research...... 137 Appendix 1. Company Interview Guide………………………….………….……140 References...... 141
3
List of Tables
1. Number of interviews and interviewees ……………….……………………………56 2. A summary of findings for Hoffman ………………….…………………………….99 3. A summary of findings for Zahner ...………………………………….…………….101
4
List of Figures
1. A process model of relationships in 2D-based construction projects………...……….63 2. A process model of relationships during the EMP project……………………...…….71 3. A process model of relationships during the Seattle Library project………………….76 4. A process model of relationships in 2D-based construction projects……………....…81 5. A process model of relationships during the construction of the Hunter Museum…...88 6. A process model of relationships during the AGA project…………………………....97 7. Interrelationships of Boundary Objects, Organisational Identities, and Information Infrastructures……………………………………………………………………….….113
5 Boundary Matters: The Dynamics of Boundary Objects, Information Infrastructures, and Organisational Identities
Abstract
by
URI GAL
In this work I set out to examine the organisational implications of the introduction of
new information systems that are designed for the sharing of information and
collaboration across organisational boundaries. I highlight the social processes within and
across organisations in which information systems are implicated. To do so, I
conceptualise information systems as boundary objects and examine them in relation to the information infrastructures within which they are embedded and the identities of the organisations that share them during a process of technology-enabled change. I present four case studies that describe the introduction of a new collaborative technology, 3D modelling tools, into the architecture, engineering, and construction (AEC) industry, and the accompanying organisational changes. Based on the case studies I suggest that boundary objects are used not only to facilitate cross-organisational communication and collaboration, but also as a resource to form organisational identities. I further suggest the occurrence of a dynamic process whereby changes in boundary objects enable changes in information infrastructures and identities in one organisation. These changes, in turn,
create the conditions for change in bordering organisations through joint boundary
objects and boundary practices.
6 1. Introduction
Today, more than ever, organisations rely on information technologies (IT) to accomplish
their tasks. A vast array of IT are used in organisations to support production activities,
store, retrieve, and transfer information across organisational units, maintain relationships
with customers, suppliers, and other external stakeholders, and facilitate communication
within and across organisational boundaries. Technologies such as enterprise resource
planning systems, supply chain management systems, product lifecycle management
systems, CAD/CAM systems, and e-commerce applications have become an integral part
of organisational functioning to the point that it is hard to imagine organisational life
without them. Furthermore, new technologies have brought with them novel forms of
representing information that are growing richer and more complex.
Given their growing centrality to organisational and inter-organisational activity, it is
becoming increasingly important to understand the organisational changes that are
associated with new IT. Previous studies on the impact of IT on organisations have
emphasised the capacity of technology to cause desired organisational changes (e.g.,
Mukhopadhyay et. al., 1995; Brynjolfsson et. al., 1998) while others have highlighted the
ability of managers to strategically direct organisational change by purposefully
designing and utilising technology (e.g., Henderson & Venkatraman, 1999; Sambamurthy
& Zmud, 2000). Markus and Robey (1988) suggested adopting an emergent perspective
that sees the organisational effects of IT as materialising from the dynamic interplay
among actors, structures, and technology (Barley, 1986; Orlikowski & Robey, 1991;
Orlikowski, 2000). While rooted in different theoretical perspectives, these approaches
generally treat IT as an exogenous force acting upon a single organisation. The approach
7 I take in this dissertation complements this view by observing that the use of IT in an
inter-organisational context places IT among multiple social worlds and can thus mediate
changes between some or all of them simultaneously.
Some existing work on the use of IT in inter-organisational work settings tends to draw
on information processing and transaction cost theoretical approaches. Such work
characteristically emphasises the effect of inter-organisational information systems (IOS)
on the efficiency of inter-organisational coordination, communication sharing, and collaboration. The typical observation of this literature is that organisations move toward more cooperative relationships that are enabled and mediated by the use of IOS (Clemons
& Knez, 1998; Clemons & Row, 1992; Kumar & Van Dissel, 1996). The mentioned economic incentives for doing so are diverse and include the potential for sharing the costs of large investments across organisations, the pooling and sharing of risk, and the access to complementary resources (Guglan & Dunning, 1993). Increased efficiency in inter-organisational coordination and information processing is also often cited as a reason for the use of IOS (Argyres, 1999; Majchrzak et al, 2000; Zhu et al, 2006).
In addition to focusing on the effect of IOS on inter-organisational coordination,
literature on the topic has discussed a variety of related issues such as the role of trust and
control in technological inter-organisational settings (Gallivan & Depledge, 2003), the
importance of IOS ownership structure and potential for information exploitation as
factors influencing a firm’s decision to engage in IOS-related efforts (Han et al, 2004),
and the factors that contribute to the success of IOS (measured in terms of sustained
organisational competitive advantage) (Siau, 2003). While examining a range of subjects
relevant to IOS, literature on the topic has remained largely instrumental. That is, for the
8 most part, it has centred on the enhanced efficiency of inter-organisational information processing and coordination enabled by IOS and on the economic incentives that lead to and benefits associated with the use of IOS. However, it has not paid similar attention to the interrelationships between IOS and the social processes that take place within and across the organisations that use them.
Another body of work that deals with the use of information systems in inter-
organisational settings as well, but that has been more sensitive to the importance of social elements that are involved in such contexts, can be found in computer supported
cooperative work (CSCW) literature. Underlying much of the work in this field is the
emphasis on the potential of information systems to support distributed collaborative
practices that span multiple organisations (e.g., Bannon, 1997; Bannon & Bødker, 1997;
Hanseth & Lundberg, 2001; Schmidt & Simone, 1996). This issue is often framed in this
literature in terms of the development of a collaborative infrastructure to support such
practices. Distributed collective practices signify activities that mediate through
geographical distances, heterogeneous perspectives, and varied semiotic systems, and that
typically cross organisational boundaries (Turner et. al., 2006). The design of IT to
support such practices can be seen as the design of working infrastructures. Such
infrastructures entail the development of technological standards to enable different
applications to seamlessly interact with each other across multiple platforms, and a
system of deeply seated social categories which allocate individuals and groups into
prescribed social spaces and which result from ongoing classification work by multiple
social actors (Bowker & Star, 1999). For example, the functioning of the internet as a
collaborative infrastructural technology relies on the use of specific and widely shared
9 technical standards, such as TCP/IP, and on the articulation of a set of social categories
and the provision of resources, responsibilities, and privileges to different groups of users
based on their location within this system of categories.
A prominent theme in the work on distributed collective practices involves the capacity
of heterogeneous organisations to maintain the integrity of their distinct cultural, lingual,
and socio-cognitive backgrounds while reaching shared understandings and developing a
common ground with their partners for collaboration. Substantial attention has been given
to understanding the nature of work that heterogeneous organisations engage in that
allows them to overcome the fundamental tension between their idiosyncratic
backgrounds and the need to create a shared inter-organisational context. The resolution
of this tension enables these organisations to work together and successfully cooperate
with each other despite their infrastructural differences. Some authors suggested that to
accomplish such collaboration requires the creation of shared artefacts such as
coordination mechanisms (Schmidt & Simone, 1996) or boundary objects (Star &
Griesemer, 1989; Bannon, 1997).
Boundary objects are conceptual or physical artefacts which reside in the interface
between organisations. On the one hand they are flexible enough to contain multiple
meanings which arise from the different organisations that use them. On the other hand, their structure is solid enough to serve as a common reference point to the members of the organisations that use them. On account of this quality, boundary objects can bridge perceptual and practical differences between diverse organisations when they engage in joint practice, and facilitate common understandings and effective cooperation (Star &
Griesemer, 1989; Henderson, 1991; Karsten et. al, 2001).
10 Since it was first articulated by Star and Griesemer, the concept of boundary objects has received considerable attention from researchers from multiple scholarly fields (e.g.,
Bannon, 1997; Briers & Chua, 2001; Carlile, 2002; Subrahmanian et. al., 2003). The common denominator of most of this work is its authors’ agreement on the crucial part that boundary objects play in enabling collaboration to take place among diverse organisations. Accordingly, past work on boundary objects has focused on their role as translation devices that mediate communication processes between two or more organisations with different infrastructural backgrounds and that enable them to collaborate successfully. (e.g., Carlile, 2002; Henderson, 1991; Ackerman & Alverson,
1999; Yakura, 2002). Importantly, for the most part, existing research on boundary objects has attempted to study the specific qualities that enable them to mediate among heterogeneous organisations and identify different types of boundary objects. Such attempts, in turn, emphasised the material aspects of boundary objects and de-emphasised the internal dynamics of interacting organisations and the ways that these dynamics shape and are shaped by the use of boundary objects1. Furthermore, prior research on the topic
has implicitly assumed that organisations and the boundary objects that they employ do
not change over time. In part, this is because the emergence and use of boundary objects
have mostly been examined in a context of relatively stable settings or for short periods
of time.
In this study, I take a more dynamic approach to understanding boundary objects by examining their relationships with the organisations that employ them during a process of
organisational change. I operationalise organisations in terms of their information
1 Similar claims were previously made by Levina and Vaast (2005) as well as Lutters and Ackerman (2007).
11 infrastructures and identities. I define information infrastructure as a system of
standardised practices and modes of communication that emerge in relation to a particular
set of IT artefacts that are deployed within organisational boundaries. An organisational
identity is defined as a symbolic representation of the organisation that is commonly and
continually negotiated among organisational members. I focus on organisational practices
because organisations, as social structures, only exist in and through the ongoing
activities of human actors as they navigate through their local social ecologies (Giddens,
1984). Therefore, social practice is the constituting mechanism through which
organisational institutions are produced and reproduced and technology is appropriated,
given meaning, and integrated in organisational life (Orlikowski, 2000). I focus on
organisational identity because it is enacted through and an integral part of everyday
practices of organisational members. Furthermore, organisational identity is crucial for
understanding the symbolic meanings that members attach to their actions and
surroundings. Observing boundary objects in relation to the information infrastructures
and identities of the organisations that use them will help situate boundary objects in a
rich, practical, and dynamic context, thereby contributing to a comprehensive understanding of their nature.
In this dissertation I centre on the role of IT artefacts as boundary objects. Because of
their inherent ability to embed complex computational and modelling capabilities and their interpretive flexibility (Bijker, 1995), IT artefacts are particularly interesting boundary objects to study. Organisations regularly embed technology into their working practices and into their professional projects (McLaughlin et. al, 1999) and these incorporations of technology take many forms and involve diverse evaluations and
12 attributions of meaning. Although it has been acknowledged that technological artefacts can be used to facilitate communication among diverse organisational groups (e.g.,
Carlile, 2002), past studies on IT artefacts, as boundary objects, have not examined their role in the context of the dynamic changes of the identities and information infrastructures of interacting organisations.
To address these issues, I conducted 4 inductive case studies that explored the relationships between boundary objects and the identities and information infrastructures of the organisations that use them. To account for change over time in boundary objects, I examined an IT-enabled inter-organisational transformation process within the
Architecture, Engineering and Construction (AEC) industry. Traditionally, organisations in the AEC industry have used two-dimensional (2D) modelling technologies, such as
Computer Aided Design (CAD) and paper models, to communicate with one another, exchange information, and negotiate problems during construction projects. However, in the last decade, the AEC industry has undergone significant changes. Most relevant to this study is the introduction of three dimensional (3D) modelling technologies and their use by a growing number of organisations in the industry. Compared to 2D modelling technologies, 3D technologies are characterised by an increased capability to represent and communicate rich and complex information within and across organisational boundaries. The use of 3D modelling technologies can translate into enhanced productivity and accuracy and was additionally found to be associated with significant changes in the management of construction projects, and with shifts in inter- organisational collaboration patterns during construction project (Boland et. al., 2007).
13 By exploring the use of 3D technologies, as a new type of boundary object, and observing the associated changes in organisational practices and communication patterns, and in the way organisations symbolically represent themselves, this study will inquire into the role of boundary objects in general, and IT-based boundary objects in particular, in mediating between and structuring organisations. More specifically, my investigation will be guided by the following questions: 1) How do 2D and 3D modelling technologies, as boundary objects, shape or are shaped by organisational practices and patterns of communication? 2) How do 2D and 3D modelling technologies shape or are shaped by
the identities of the organisations that use them? 3) What are the implications of changes
in such technologies for the identities, practices, and interactions of the organisations that
share them?
In what follows, I first lay out the theoretical foundations of this study. Next, I describe
the research setting and methodology and present the findings of the study. After discussing the findings, I introduce a model that addresses the research questions posed
above and specify the model’s contribution to existing literature. I conclude by
considering the implications of this study, identifying its limitations, and suggesting a
number of possibilities for future research.
14 2. Theoretical Foundations
In order to understand the significance of boundary objects during a process of inter-
organisational change and their relationship to the information infrastructures and
identities of the organisations that share them, it is first necessary to clearly articulate and
define the constructs that I will use in my investigation. In what follows, I will discuss
information infrastructures, organisational identities, and boundary objects and explain
their application in this study.
2.1 Information infrastructures The term infrastructure generally refers to any substructure or underlying system. It
denotes the “basic physical and organisational structures (e.g. buildings, roads, power supplies) needed for the operation of a society or enterprise” (Oxford dictionary) and without which contemporary organisations and societies cannot function (Edwards,
2003).
A common use of the term infrastructure is made by both researchers and practitioners in
the field of information systems. The concept of IT infrastructure is generally used to
describe large and complex technological systems that support the functioning of entire
organisations and that are shared by a large number of people. Conventionally, the idea of
IT infrastructure emphasises the standardisation of systems, data, and communication
across the organisation (Ciborra, 2000); standardised ways of operating are inscribed in
technology, which links applications and people according to predefined notions of
15 business processes, and requires the homogenisation of practices across organisational
units (Ciborra, 2002).
IT infrastructures are generally conceived of as large conglomerations of tangible
technological components and human skills that are combined together to serve the
corporate needs of an organisation. This type of conceptualisation assumes that IT
infrastructures can be differentiated and distinguished from all that is not an
infrastructure, and that since infrastructures can be neatly identified, they can be
controlled and managed in a fairly straightforward fashion (Broadbent et al, 1999). Along
similar lines, it is often argued in the literature that a well organised and thought-out
infrastructure can be sensibly mapped out and delineated, for example, by assessing its
reach and scope. The assumed unproblematic measurability of IT infrastructures is
evident in Broadbent et al’s work (1999). Here the authors suggest three measures to
assess the IT infrastructure capability of a firm: 1) the extent of a firm’s infrastructure
services, determined by the number of firm-wide IT services offered, 2) the provision of
boundary-crossing infrastructure services which are explicitly and actively integrative
and support information flows and transaction processing beyond one functional unit
within a firm, and 3) the firm’s reach and range, which is determined by the firm’s ability
to simultaneously perform transactions on multiple applications, and its capacity to
update its databases across different business units, irrespective of their location.
The presumption that IT infrastructures can be defined and assessed is also demonstrated in Weill and Broadbent (1997) and Henderson and Venkatraman’s work (1994). In these two examples, the authors characterise IT infrastructure by separating the concept into a technical IT infrastructure and a human IT infrastructure. The technical element of the IT
16 infrastructure is described as a set of shared, tangible IT resources forming a foundation
for business applications. These resources include platform technology (hardware and
operating systems), network and telecommunications technologies, data, and core
software applications (Duncan, 1995). The human element of the IT infrastructure
includes human skills, expertise, knowledge, norms, and values relevant to the
functioning of the infrastructural technology (Broadbent & Weill, 1999). Emphasising the importance of the human element of IT infrastructure, Lee et al (1995) argued that human
IT infrastructure needs four types of knowledge and skills to be effective: 1) technology
management skills, 2) business functional skills, 3) management skills, and 4) technical
skills. Similarly, Weill (1992) claimed that the effectiveness and proficiency of the
human IT infrastructure is crucial to the way IT resources are converted into productive
outputs.
Although placing importance on the role of human IT infrastructure to the functioning of
IT resources, the line of thinking outlined above typically treats human actors as mere
technology users which follow certain rationalistic norms (Kling, 1992), and assesses
their importance to the operation of IT infrastructure based on a predetermined set of
skills which they may or may not possess. Furthermore, it explicitly separates the human
elements of the IT infrastructure from its technical elements, as explained by Byrd and
Turner: “the IT infrastructure concept can be divided into two related – but distinct components – a technical IT infrastructure and a human IT infrastructure” (Byrd &
Turner, 2000). While separating the human elements of an IT infrastructure from its
technical elements may be conducive to creating an easily measureable analytical
construct, it contributes to a narrow conceptualisation of the social processes that are
17 involved in the shaping and functioning of information infrastructures, and to an oversimplification of their dynamics.
A number of researchers have outlined an alternative understanding of IT infrastructures that more broadly acknowledges the interconnectivity of human and technical infrastructural elements and that is more sensitive to the social aspects of IT infrastructures. According to this line of thought, information infrastructures extend beyond mere materiality and predefined human skills to encompass social, organisational, and moral elements and considerations (Kling, 1992; Bowker et. al., 1996; Monteiro &
Hanseth, 1996; Star & Ruhleder, 1996; Dourish & Edwards, 2000; Star, 2000; Lee et. al.,
2006; Turner et. al., 2006). Technically, the construction of an infrastructural system requires the establishment of a scheme of protocols and standards that enable the system to be used and seamlessly connect with other systems. Socially, its construction necessitates the elaboration of a system of classifications that symbolically represent and organise most things in society: people, classes, geographical areas, religions, civil status, and so on. These classifications tacitly integrate into the fabric of everyday social life through the practices of the users of the infrastructural system. (Turner et. al., 2006).
They reflect and, in turn, feed into what is considered conventional wisdom concerning gender and race relations, distribution of wealth, appropriate ways of social organising etc. As Edwards observes; “…although ‘infrastructure’ is often used as if it were synonymous with ‘hardware’… all infrastructures… are in fact socio-technical in nature.
Not only hardware but organisations, socially-communicated background knowledge, general acceptance and reliance, and near-ubiquitous accessibility are required for a system to be an infrastructure…” (Edwards, 2003, p. 3).
18 Conceptualised this way, information infrastructures emerge as highly complex systems.
For instance, Ciborra (2000, 2002) suggested that the issue of designing, implementing,
maintaining, managing, and controlling an infrastructure is far from being clear-cut or
straightforward. Instead, he argued that: “Information infrastructures are puzzles, or
better collages, and so are the design and implementation process that lead to their
construction and operation. They are embedded in larger, contextual puzzles and
collages. Interdependence, intricacy, and interweaving of people, systems, and processes
are the culture bed of infrastructure. Patching, alignment of heterogeneous actors, and
bricolage (make do) are the most frequent approaches…” (2000, p. 2-3).
Information infrastructures are evasive phenomena which manifest themselves in ways that are far less tangible and orderly than conventionally assumed. Rather, infrastructures are heterogeneous and dispersed: they encompass both technical and social elements and their boundaries cannot be easily outlined because of the complexity and dynamism of the components that constitute them. This idea is evident in Star and Ruhleder’s work
(1996) where information infrastructures are described as having the following characteristics: they are “sunk” into other structures, social arrangements, information practices, and technologies; they may extend beyond a single event or one-site practice; and they both shape and are shaped by the conventions of a community of practices.
The inherent complexity of information infrastructures is additionally apparent in the
process of their development, which typically involves multiple narrative voices and
groups struggling to shape the standards and classification systems embedded in the
infrastructure to reflect their values, ethical principles, and interests (Star, 1999). For example, in a study of the evolution of the classification of diseases (conceptualised as an
19 information infrastructure) maintained by the World Health Organisation, Bowker and
Star (1994) illustrate how the coding and classification of different diseases are far from being neutral and in fact are profoundly intertwined with the interests and political considerations of the different groups involved in the process (Bowker & Star, 1994).
Another example of the efforts involved in the shaping of information infrastructures is provided by Star (1999) who describes the changes in the choice of race in the US Census forms as a modification in an information infrastructure. In the year 2000, for the first
time, people were able to check more than one racial category. This mundane
infrastructural change was preceded by a heated debate which lasted several years among
different groups of political activists and social justice groups. Some groups argued that
allowing checking multiple categories better reflects the racial diversity in America,
while others claimed that the effects of discrimination will be lost in the count by those
who claim multiple racial origins. (Star, 1999).
In addition to encapsulating negotiations, debates, and sometimes arguments and conflicts among multiple social groupings, information infrastructures in-use, by virtue of
the standards that are embedded in them, can have a significant impact on the nature,
scope, and level of granularity of the information that their users get exposed to
(Monteiro & Hanseth, 1996; Bowker, 2005). As a consequence, they can influence the
way users communicate with one another, engage in social life, and perceive their
environments. For example, in their study of the work of nurses, Bowker et al (1995)
described the relationship between infrastructural changes and the social context within
which they take place. The authors observed changes in the Nursing Interventions
Classification (NIC) and the impacts of these changes on the nurses’ work practices and
20 professional knowledge. The NIC provides a list of over 300 possible nursing
interventions, each is labelled and defined. It links scientific knowledge, practice, and information systems, prescribes scripts for proper social exchange and adequate professional conduct, and formally defines correct technological configurations and uses.
Changes in the NIC involved modifications to the understanding and vision of what nursing is, the allocation of resources essential to the work of nurses, and the rhythm of nurses’ daily activities and work practices. The impact of information infrastructures on communicative activities and work practices is also evident in Hanseth and Monteiro’s work (1997) which discusses the processes through which certain behavioural and communicative patterns are inscribed in different information infrastructures in the
Norwegian healthcare system. Similar ideas are echoed in Borgman’s research which emphasises that the selection, classification, and preservation of information resources in digital libraries can significantly impact the type and amount of information that their users have access to (Borgman, 1997; Borgman, 1999)
Taking these studies into account brings to the fore a crucial point: working
infrastructures inevitably involve the development of standardisation and classification
systems. However, such standardisation stretches beyond technological artefacts,
platforms and protocols to include people’s routines, communicative behaviours, and
work practices (Monteiro & Hanseth, 1996; Bowker, 2005). Accordingly, I conceptualise
information infrastructure as a system of standardised practices and modes of
communication that emerge in relation to a particular set of IT artefacts within
organisational boundaries. Such practices are acquired when actors are inducted into a
community and undergo a process of socialisation whereby they internalise local
21 knowledge, practices, language, and values (Berger & Luckmann 1967). Over time, such
artefacts and associated organisational arrangements and practices become taken for
granted, at which point they recede into the background and become part of the
infrastructure (Star & Ruhleder, 1996; Star, 1999).
Importantly, information infrastructures are contextual because they are meaningful only in relation to an organised system of artefacts and standards such that what seems taken for granted for members of one community, may acquire different meanings to members of other communities. However, despite their contextuality, information infrastructures are not purely local. Although they function in relation to communal standards, no community or organisation exists in complete social isolation. Multiple organisations
always co-exist and interact and therefore do not have clear definitive boundaries (Star &
Ruhleder, 1996; Ciborra, 2000). Hence, infrastructures always overlap to create
interfacing areas. These areas depend on reciprocal recognition and evolution of
practices, discourses, and artefacts across interacting organisations. Therefore, an
important aspect in the constitution of information infrastructures has to do with
organising and generating practices, meanings and roles that form and articulate
organisational identities at the interfacing areas.
2.2 Organisational identities Organisational identity is typically understood in the literature to be an organisation’s
members’ collective understanding of the features that are presumed to be central,
distinctive, and relatively permanent about the organisation (Albert & Whetten, 1985;
Dutton et. al., 1994). Common to most theoretical and empirical accounts of
22 organisational identity is the view that identity is rooted in a deep cultural level of the
organisation (Gioia et. al., 2000). That is, identity is perceived to reside in interpretive
schemes that organisational members collectively construct to provide meaning to their
shared history, experiences, and activities (Ravasi & Schultz, 2006; Gioia, 1998).
Although organisational scholars have mostly accepted and built on the definition of
organisational identity proposed by Albert and Whetten (1985), various views of the
phenomenon have developed over the years which can be crudely grouped into two
principal lines of though (Whetten & Mackey, 2002). Some researchers have emphasised
the functional qualities of organisational collective self-perceptions in fulfilling the basic requirements of individuals as social actors: continuity, coherence, and distinctiveness
(Whetten & Mackey, 2002; Whetten, 2003). According to this view, organisational
identity is inherent in formal and explicit institutional or organisational claims of what
the organisation is about, what values and norms it embodies and represents, and what
types of practices and relationships it engages in. Such claims can be made by openly and
officially declaring the organisation’s membership in specific identity categories or
groups that lend certain cachet to their members. Examples of identity categories can
include for profit versus non for profit organisational purpose, public versus private
ownership, local versus global domain, and transportation versus utilities industry
(Whetten & Mackey, 2002). These formal claims are subsequently internalised by
organisational members and influence their perceptions of the central, unique, and stable
features of their organisation. This is done by providing organisational members with a
coherent and clear narrative which threads the organisation’s culture and history with its
current goals, activities, and relationships. According to this point of view, different
23 organisational members fulfil specific roles in this narrative, which allows them to construct a collective sense of self that is inseparable from the position they occupy in the organisation (Whetten & Mackey, 2002).
Proponents of this perspective typically conceive of organisational identity as a set of durable self descriptions or assertions. These assertions provide a stable sense of self and meaning, and place the individual in a wider social context, thereby tying the ongoing execution of individual roles with the continuing maintenance of the organisation’s collective image. Advocates of this perspective, therefore, observe how deeply held beliefs, represented in formal claims and declarations, tend to change only rarely and never easily (Whetten & Mackey, 2002). Furthermore, they emphasise that core features of organisational identity are resistant to occasional attempts at alteration because of their ties to the organisation’s history. Events that take place outside the organisation and challenge its formal identity claims are likely to facilitate reactions aimed at countering the threats and strengthening personal and collective representations of what the organisation is about and what it stands for (Albert & Whetten, 1985). Some scholars have additionally argued that the main strategy of an organisation is usually geared toward maintaining its identity, especially under threatening conditions of change. These scholars claim that during periods of intensive social, economic, or institutional change, organisations will drop their heaviest cultural anchors in order to resist the currents of transformation, and preserve their self perception (Cohen, 1985).
Although the majority of scholars continue to play up and focus on the seemingly stable and permanent features of organisational identity, acknowledgement of its potentially changing character can be found in recent research on the topic (e.g., Gioia et al, 2000).
24 Such work relies on a wealth of empirical evidence to observe how organisational
members’ beliefs about central and unique features of their organisation may evolve as a
result of events that take place within or outside the organisation (Corley & Gioia, 2004;
Fiol, 2002). Scholars engaged in this research have shifted their focus away from formal
organisational and institutional identity claims as the foundation of an organisation’s
collective self perception. Instead, they give significant attention to the emergent
processes whereby organisational members develop collective understandings of the
main features of the organisation that are presumed to embody what it is about and what
it stands for. According to this view, organisational members negotiate, through
continuous interactions, a shared symbolic representation of their organisation that gives
a sense of meaning to the organisation’s actions, objectives, and existence, and that
distinguishes the organisation from other social entities in its environment (Gioia, 1998;
Gioia et al, 2000). These shared representations may or may not correspond to the
officially proclaimed organisational narrative.
Work adopting this perspective has observed that significant organisational changes, or changes in the organisation’s environment and relationships with other organisations, is likely to require modifications in the way members interpret what is central and
distinctive about their organisation. That is, organisational changes will require members
to actively reinterpret and develop new representations to symbolically characterise their
organisation (Fiol, 1991). Such work typically presumes that shared beliefs will be
continually subjected to revision and reinterpretation as organisational members deal with
changes in their environment. Furthermore, it generally downplays endurance and
stability as core characteristics of organisational identities and suggests that responses to
25 environmental changes may be strategically directed by organisational leaders to
opportunistically promote and shape new conceptualisations of their organisation (Corley
& Gioia, 2004). This idea is demonstrated in Fiol’s study of Tech-Co, a company with a
highly salient and stable organisational identity. During the 1970’s and 1980’s, Tech-Co had a stable organisational identity as an engineering-driven data storage company.
However, during the 1990’s, the computer storage industry as a whole was undergoing
significant changes from a primary hardware, engineering mindset to mindset of
information management and storage solutions. Tech-Co’s leadership recognised these
industry-level transformations and initiated explicit efforts to redefine what Tech-Co
considered itself to be in order to translate and incorporate the changes into the company
(Fiol, 2002).
Work subscribing to this perspective does not refute the relative permanence of formal
organisational identity claims, but emphasises that the meanings associated with these
claims may evolve as organisational members try to cope with changes in their
environment. For example, Gioia et al (2000) differentiate between an enduring identity
and an identity having continuity. The former notion of identity implies that identity
remains the same over time, whereas the latter suggests that identity shifts in its
interpretation and meaning while retaining its formal labels: “the seeming durability of
identity is actually contained in the stability of the labels used by organisation members
to express who or what they believe the organisation to be, but that the meaning
associated with these labels changes so that identity actually is mutable. Therefore, we
reconceptualise organisational identity as a potentially precarious and unstable notion, frequently up for redefinition and revision by organisation members” (p. 64).
26 The conceptualisation of organisational identity that I adopt in this dissertation is
consistent with the point of view that identities may shift and evolve over time as
organisational members adapt to changes in their organisation’s environment. However,
in addition it highlights the relational nature of identities: organisational identity is
constructed not only against a backdrop of members’ shared histories and experiences but
also in the context of multiple interactions in which the organisation is involved with a variety of outsiders such as costumers, competitors, suppliers, and regulatory institutions
(Ashforth & Mael, 1996; Gioia, 2000). Seen this way, organisational identities, or even
identities at the individual level, are social constructions deriving from repeated
interactions with others. As Weick puts it, “identities are constituted out of the process of
interaction. To shift among interactions is to shift among definitions of the self” (Weick,
1995, p.20). Similarly, Giddens recognised that identities must be created and sustained
through interactions with others (Giddens, 1991). This idea is also reflected in Fiol’s
study of an acute care teaching hospital going through changes in its identity in which she
stated, “you can no longer ask only me or look inside of me to understand my identity.
You can also no longer take a single snapshot of me at one point in time and believe you
have captured my identity” (Fiol, 1998, p. 68).
These views of organisational identity suggest that it emerges in a practical, dynamic
context in which organisations interact with each other, and is therefore reflective, to
some degree, of the circumstances in which it is situated. Thus, the mode and scope of
interactions, as well as the parties with which the organisation interacts, play a salient
role in the construction of its identity (Lamb & Davidson, 2005). Because an
organisation’s identity is tied to the dynamic environment from which it materialises, it
27 can vary with the context for which it is expressed (Fiol et. al., 1998; Gioia, 2000). Since
organisations (or different organisational units) typically interact with multiple outsider organisations simultaneously, organisational identity is not uniform but multifaceted; not singular but multiple across different, intersecting practices and discourses; and is constantly in the process of change and transformation2 (Hall, 1996). Accordingly,
identity is best understood as a verb rather than a noun, and is more accurately
conceptualised as a ‘work-in-progress’ rather than a finished product of social processes
(Gioia, 2000). In other words, organisational identities are an ongoing enactment that
unfolds as organisations interact with each other. This interaction is dynamically
constituted by the engagement of organisations in mutual practices. Therefore the
articulation of organisational identities takes place in the interfaces between
organisations, a region where multiple information infrastructures overlap and that is
populated by boundary objects.
2.3 Boundary objects
Boundary objects are conceptual or physical artefacts that reside in the interface between
organisations. On the one hand they are flexible and ambiguous enough to contain
various meanings which arise from multiple organisations and which may be inconsistent with each other. On the other hand they are robust enough to serve as a common reference point to members of more than one organisation when they engage in mutual practice and communication (Star & Griesemer, 1989; Henderson, 1991; Henderson,
1998; Carlile, 1997; Carlile, 2002, Carlile, 2004). For example, a contract is a boundary
2 Thus, an organisation can concurrently exhibit numerous identities to reflect such multiplicity in interactions.
28 objects that is used by multiple parties as a regulating mechanism that defines roles,
allocates responsibilities, and lays out activity plans. The contract acts as a ‘social lubricant’ that enables the disjointed parties to work together. While it is recognised as a
contract to all the parties involved in the interaction, it is sufficiently ambiguous so that
multiple views and interpretations of it can co-exist.
Boundary objects are part of the information infrastructure - they are used by people as
they engage in practice. However, boundary objects are unique because they spread
across two or more information infrastructures. They are therefore an infrastructural
phenomenon that is shared by members of more than one organisation in the course of
their interaction (Bowker, 2005; Bowker & Star, 1999; Star, 2000). Accordingly, they are
located at the interface of two or more information infrastructures - a region where multiple infrastructures overlap.
Boundary objects are a critical component in the process of communication and joint
action. Where mutual, inter-organisational action is required, lack of boundary objects
would lead to efforts directed at creating them (Clark, 1996). The reason for this is that
they enable a space in which trajectories of more than one organisation can meet and
within which different organisations can develop reciprocal social practices. This
reciprocity implies that organisations are able to cooperate without having a unified set of
meanings they attach to their actions, communications, or to the artefacts that they share.
The notion of boundary objects was first introduced by Star and Griesemer (1989). In
their work, the authors develop the concept of boundary objects in relation to Actor
Network Theory (ANT). ANT is interested in explaining the stability of everyday
phenomena or artefacts in terms of heterogeneous socio-technical networks that are put
29 together to control a wide range of influencing factors. Therefore the stability and form of artefacts should be seen as a function of the interaction of heterogeneous elements as these are shaped and assimilated into a network (Law, 1987). The creation of these networks is conceptualised as an act of “heterogeneous engineering” in which bits and pieces from the social, the technical, the conceptual, and the textual are fitted together and given a specific meaning, or are translated, into a network by a “heterogeneous engineer” or an “entrepreneur”. Empirical applications of ANT have typically traced back and analysed the translation of human, social, and technical elements into a single heterogeneous network. However, in their work, Star and Griesemer (1989) explicitly adopt an ecological analysis which takes notice of the fact that multiple translations can occur simultaneously as various heterogeneous networks are assembled, and of the reciprocity that characterises such situations. According to the authors, there is an indefinite number of ways in which entrepreneurs from each cooperating social world may set up their heterogeneous networks and therefore an indeterminate number of possible coherent sets of translations. Thus, each heterogeneous element can belong to multiple networks, undergo multiple translations, and contain multiple meanings at the same time. The notion of boundary objects was created to describe such elements.
In their paper, Star and Griesemer described the attributes of boundary objects that enable them to serve as translation devices: they inhabit several intersecting social worlds and satisfy the informational requirements of each; they are weakly structured in common use and become strongly structured in individual-site use; and they have different meanings in different social worlds but their structure is common enough to more than one world to make them recognizable and function as a means of translation (1989). The authors
30 concluded that the creation and management of boundary objects is a key process in
developing and maintaining coherence across intersecting social groupings (Bowker &
Star, 1999).
Following Star and Griesemer’s work, the concept of boundary objects has been applied
in various contexts such as design teams (Henderson, 1991; Henderson 1998;
Subrahmanian et. al., 2003), new product development (Carlile, 2002, 2004; Bechky,
2003), knowledge brokering activities (Pawlowski & Robey, 2004), technological change in accounting systems (Briers & Chua, 2001), production and manufacturing systems
(Garrety & Badham, 2000; Karsten et. al., 2001), implementation of information systems
(Yakura, 2002), boundary spanning activities (Levina & Vaast, 2005), and IS
development projects (Sasped & Salter, 2004; Gasson, 2006).
These studies, and others, have expanded the original conceptualisation of boundary
objects in a variety of ways. For example, Garrety and Badham (2000) distinguished
between primary and secondary boundary objects in socio-technical projects, the former being the technology itself – the material artefact around which activity is organized, and
the latter being other physical or abstract entities that enable communication across social
communities (e.g. contracts). Brier and Chua (2001) discussed ideal or visionary
boundary objects. Visionary boundary objects are conceptual in nature and therefore
cannot be argued against (e.g. institutionalised organisational “best practices”).
A number of researchers focused on the role of boundary objects in enabling cross-
boundary organisational learning and knowledge sharing. Carlile (1997) specified the
critical features that make boundary objects effective at enabling cross-boundary
communication and knowledge sharing: first, boundary objects must specify or represent
31 the different types of knowledge that are involved or are at stake; second, boundary
objects must be tangible (easy to manipulate) and concrete (resemble the actual problem
or object that they represent); third, boundary objects must be accessible and contain over
time the current concepts, ideas, and data that are relevant to a certain cross-boundary
interaction, and; fourth, boundary objects must be sufficiently loosely defined to be
playable in practice. Wenger (2000) suggested that boundary objects can be used as a
means to bridge across boundaries between communities of practice in order to enhance
inter-organisational learning activities. Bechky (2003) discussed the role of boundary
objects in reducing misunderstandings among heterogeneous professional groups in a production company. By utilising boundary objects, members of the different professional groups created common ground which transformed their understanding of the product and the production process, and enhanced cross-boundary knowledge sharing.
Pawlowski and Robey (2004) described how boundary objects were implicated in a process of knowledge brokering between IT professionals and business users in a large manufacturing and distribution company. The authors proposed that knowledge brokering activities were conditioned by structural conditions (e.g., the structure of the IT organisation) and technical conditions, which included IT systems that served as boundary objects that were used to transfer business and IT knowledge across units in the organisation.
Other researchers suggested expanding the analysis of boundary objects beyond their
structural properties to include the type of practices in which they are implicated when
examining their role in cross-boundary communication and knowledge sharing. For
example, Levina (2005) and Levina and Vaast (2005) proposed that the usefulness of an
32 object in enabling cross-boundary collaboration is not inherent in the object’s physical
properties. Rather it is associated with the ways in which the object is used during a process of collective-reflection-in-action, and with the extent to which it acquires a common identity across contexts.
Different artefacts in different contexts have been studied as boundary objects. For
example, Yakura (2002) looked at timelines (i.e. a graphical representation of a set of
temporal units in the lifetime of a project) as boundary objects, and demonstrated their
ability to reconcile diverse socially constructed temporal arrangements. Henderson
(1991) emphasised the role of visual representations as boundary objects in the world of
design engineers. Karsten et. al. (2001) studied the role of technical specifications of a
paper machine delivery project as a boundary object and argued that “when knowledge is
made available to others in a boundary object, it provides a basis for perspective taking”
(Karsten et. al., 2001, p. 95). They further maintained that a boundary object is always
located in a larger system of interconnected boundary objects that influence its
interpretation.
Some authors tried to demonstrate the variability of boundary objects and to outline their
evolvement. For example, Carlile (2002, 2004) distinguished among different types of
boundaries – syntactic, semantic, and pragmatic - that require different types of boundary
objects. He argued that as the novelty of the situation increases, organisations will face
more pragmatic boundaries. In such situations organisations will need boundary objects
which allow them not only to transfer or translate knowledge, but also to transform it in
order to effectively share and assess their knowledge at the boundary. Thus, Carlile
anticipates that boundary objects may be dynamic, but does not elaborate on the change
33 process. Subrahmanian et al (2003) discussed changes in design and manufacturing teams
and the consequent affects that they have on boundary objects. They claimed that changes
disrupt common grounds among organisations and therefore open a debate on the role of
existing boundary objects.
Although the use of different types of boundary objects has been studies in a variety of
organisational and social contexts, most research on boundary objects has examined them in a context of relatively stable settings or for short periods of time. A few exceptions
hint at the potential for dynamic change in boundary objects, but do not explore that
dynamic process (Subrahmanian et al 2003; Carlile, 2002, Carlile, 2004). In addition,
research on boundary objects has generally focused on the material aspects of boundary objects and not explored the social dynamics within and across interacting organisations
(Luttes & Ackerman, 2007), or the changing organisational identities and information infrastructures that are associated with dynamic boundary objects. In short, the relationship among information infrastructures, organisational identities and the creation, maintenance, and change of boundary objects has remained largely unexplored.
To address these gaps in the literature, I will inductively examine boundary objects in
close proximity to the organisations that use them, during a change process that
encompasses multiple organisations. In particular, I will inquire into the processes
whereby boundary objects interrelate with the practices and patterns of communication of
the organisations that use them, and the processes whereby boundary objects interrelate
with the identities of the organisations that use them. Furthermore, to account for change
in boundary objects and its possible implications for the organisations that use them, my
research will be conducted in the context of a technology-enabled inter-organisational
34 change process that has taken place in the AEC industry. In the last 15 years, multiple organisations in the industry have adopted a new type of collaborative technology in their practice and used it as a new kind of boundary object to communicate with their counterparts during construction projects. In what follows, I describe the AEC industry, elaborate on the abovementioned changes it has gone through, and explore the manner in which two organisations in the industry have adapted to and coped with these technological changes.
35 3. Research Context
3.1 The AEC industry The setting for this research is the AEC industry. The AEC industry is one of the largest in the American economy and globally. In the USA alone, it encompasses tens of thousands of professional organisations such as architects, engineers, designers, contractors, and constructors. The main undertaking of these organisations is the design, engineering, and construction of civic structures. The design and construction of such structures is a complex undertaking involving the interaction and coordination of the skills, knowledge, and efforts of a large number and variety of professional organisations.
Some of these organisations acting for the customer / owner of the building (i.e. architects and general contractors) are concerned with defining the project scope and ensuring that performance requirements are met in the constructed facility. These organisations focus on quality control, general supervision, and monitoring functions during construction. Some organisations are associated with the constructor (i.e. contractors) and are involved either with the planning, directing, and management of the construction process or with the actual work effort itself. Other organisations (i.e. labour and material suppliers) are concerned with supporting the construction effort through the acquisition and supply of essential resources needed for the projects (Sweet, 1994).
The traditional and established approach to construction in the AEC industry usually follows a linear form of the building process. Most construction is performed in the context of the organisational and contractual relationships among owner, consultant, and construction agents. Normally, a consulting architectural or engineering group, acting on behalf of the owner of the building, performs a project analysis and design and provides
36 overall management and supervision of the contract. The contractor, generally referred to
as the general contractor or construction manager, working under the specific terms of the
contract assumes the detailed control, coordination, and management of the productive
effort required to build the facility. In most cases, the contractor subcontracts out specific
portions of the project to specialised subcontractors.
In this traditional organisation of the building process, the construction process is mostly separated from the preceding planning, design, and engineering processes. The contract
documents that are created in the earlier stages do not refer to specific construction
methods and leave the selection of these and the solution of related field engineering problems to the contractor and subcontractors working on site. In some cases, the design details are formulated without considering the selection of materials and their associated cost. This established contractual and organisational environment tends to enforce a segmentation of the building process and is often the source of conflict between the architect or engineer and the contractor and subcontractors (Halpin & Riggs, 1992).
Increasing the fragmentation of this process and the potential for conflicts and
misunderstandings is the fact that many of the involved organisations in the building
process typically come from different professional backgrounds and may have different
technical expertise, organisational practices, and specialised jargons that they rely on to
accomplish their tasks. In addition, they are likely to have diverse ways of implementing
their knowledge and skills, and unique ways of representing themselves and what they do
(Carrillo & Anumba, 2002). Furthermore, these organisations are prone to regard and
participate in the construction process from different points of view and with different
technical and organisational responsibilities.
37 Given this diversity and multiplicity of perspectives, boundary objects are an important
means for maintaining effective collaboration during construction projects. A prominent
boundary object that has been traditionally used in the AEC industry comes in the form
of 2D representations such as CAD models and paper drawings (e.g., contract
documents). CAD models and paper drawings are an integral part of the construction
process and are used from its inception until its completion by most involved
organisations. Early design concepts of the building and the different systems and
components that are embedded in it are represented in models that are used to assist the design development. Later in the process, models are used to support construction
activities and communicate information across organisational boundaries. The capacity of
CAD models and paper drawings to accommodate different views that originate from the different organisations that use them during the construction process qualifies them as
boundary objects. While the models are shared by multiple organisations and used by them to support collaboration and coordination, different organisations are likely to have their own specified language and practices in light of which they understand and use the models. Thus, the models are not owned exclusively by any one organisation but instead simultaneously shared by multiple organisations and used by them in joint practice.
Approximately 15 years ago a significant change process started taking shape in the AEC
industry; 3D modelling tools were introduced into the construction practice as a vehicle
to both assist the design process and coordinate construction activities during the building phase. Compared to 2D modelling technologies, 3D technologies are characterised by an advanced capacity to represent and communicate rich and complex information within
38 and across organisational boundaries. Such capabilities can translate into enhanced productivity and accuracy.
3D modelling tools were first introduced into the industry by architect Frank Gehry in the beginning of the 1990’s (Boland et. al., 2007). Gehry is a renowned architect who is internationally recognised for his exceptional building designs such as the Guggenheim
Museum in Bilbao Spain and the Disney Concert Hall in Los Angeles. These buildings drastically break from conventional architectural norms and are characterised by unstandardised curvilinear geometrical shapes. These unique buildings are designed with
3D modelling software, CATIA. Originally used in the aeronautics and automotive industries, CATIA was first introduced into the AEC industry by Gehry Partners (Frank
Gehry’s architecture firm) when the company was working on a large fish sculpture for the Barcelona Olympic Games in 1992. Because of the complex structure of the sculpture, Gehry Partners used 3D modelling technologies instead of traditional 2D technologies in the design and construction of the sculpture. The project was completed on schedule and was regarded a success. Having this first positive experience with the new technology, Gehry Partners has since increased its use of 3D modelling tools and incorporated them in most of its construction projects. Importantly, the use of 3D modelling technologies by Gehry Partners has gone beyond the sheer design and planning of the structure to be built. It has also been used as a central coordinating mechanism among engineers, subcontractors and construction managers during the actual construction process.
Initially, the majority of organisations that participated in Gehry Partners’ 3D-based projects had never experienced working with 3D modelling tools. 2D modelling
39 technologies, specifically 2D CAD, have been in use in the AEC industry since the
1960’s and have since been established as the absolute consensus and norm. However, since the beginning of the 1990’s a growing number of organisations in the industry started using various forms of 3D modelling tools and incorporating them into their practice. Today, while working with 2D models is still considered the norm in the AEC industry, the use of 3D tools is continually spreading and is no longer as uncommon as it was over a decade ago when they were initially introduced.
It is important to recognise that the move from using 2D modelling tools to 3D tools has much greater significance to organisations in the AEC industry than a mere technological one. Modelling technologies are a central element in the coordination efforts during construction projects and play a critical role in structuring construction activities (Beck,
2001). A significant proportion of the communication among different organisations in the course of the construction process takes place through the sharing and exchange of
CAD or paper models rather than verbally. For example, it is highly uncommon for meetings to take place during the construction process without at least some of the parties using models to communicate their viewpoint, concerns, or requests to their peers. Many times, the models are the main medium through which opinions are expressed, views are exchanged, and information is shared in an inter-organisational setting. Therefore, the models constitute a significant arena through which organisational interactions take place. As Henderson notes, “sketching and drawing constitute the basic component of communication…Coordination and conflict take place over, on, and through the drawings. These visual representations shape the structure of the work, who may participate in the work, and the final products of design engineering. They are…the locus
40 for practice-situated and practice-generated knowledge” (Henderson, 1991, p.449). She
further points out that “…drawings are the building blocks of technological design and
production…because they are developed and used through interaction, these visual
representations act as the means for organizing the design to production process, hence
serving as a social glue between individuals and between groups” (Henderson, 1991,
p.449).
Given the importance of 2D CAD and paper models, as boundary objects, to the
organisation and coordination of construction projects, a move to using 3D models can
have a vast significance to the manner in which construction projects are organised, to the
way in which professional organisations interact with one another through the new
boundary objects, and to the manner in which organisations construct their identities vis-
à-vis their counterparts. Therefore, an exploration into how this process (i.e., of moving from using 2D models to 3D models) unfolds can shed new light on how boundary objects, information infrastructures and organisational identities interrelate. In the next section I elaborate in further detail on the nature of 2D and 3D technologies and the differences between them.
3.2 2D and 3D modelling technologies in the AEC industry
As described, modelling technologies are foundational to the accomplishment of work in
the AEC industry. Most information pertinent to the construction of buildings is represented in models which are designed, exchanged and analysed by multiple organisations throughout the construction process. Models have traditionally been used as a visual aid by designers and engineers. For centuries, paper drawings and concrete
41 small-scale physical models were used to represent buildings for construction (for
instance, some of the great cathedrals which are among the most complex buildings in the
world were constructed using such modelling methods).
Since the 1960’s, 2D CAD tools have become an integral component of the work of
engineers and designers. Design engineers began to use 2D CAD tools primarily as an
electronic draft board. The use of CAD significantly enhanced the accuracy and
efficiency of paper drawings, particularly when engineers were able to develop drawings
based on existing ones (Baba & Nobeoka, 1998). 2D CAD also enabled engineers and designers to transfer data from one site of application to other sites. In the early 1970’s, a
new breed of CAD tools came into being - 3D mechanical CAD was first developed for use in the aircraft and automobile industries. The application of this technology in these industries, however, only extended to defining the outlines or curved surface of a product and did not include representing the interior of the product in terms of digital data. In the late 1980’s more advanced versions of 3D CAD were commercialised that enabled to digitally represent all aspects of a product, including its interior (Aoshima et. al, 1999).
Over the last decade the use of such digital 3D tools (e.g., CATIA and Rhino software3) has become more commonplace in AEC due to these technologies’ lowered cost and improved functionality. Whereas 2D CAD tools can essentially be seen as a linear extension of the capabilities embedded in 2D paper drawings, 3D modelling tools entail capabilities that go far beyond that. 3D tools allow more daringly shaped structures to be built cost effectively. 3D tools make possible a full visualisation of designs in actual scale, and support simulation as well as integration and coordination of detailed design
3 For more information on Catia and Rhino, see http://www.3Ds.com/en/brands/catia_ipf.asp, http://www- 306.ibm.com/software/applications/plm/catiav5/, and http://www.rhino3D.com.
42 information for digital manufacturing (Boland et al, 2007). With 3D representations,
designers and engineers can view a model of a product from various perspectives and
angles in a way that captures its entire form. A 2D design only enables a view of the product from a few fixed angles and does not allow for full visualization. Further, with complex products, 2D models can become extremely complicated and difficult to
understand. Because physical artefacts are always 3D, 3D models can better capture their contour even when they are complex. In addition, 3D models not only provide full form geometric representations of objects, but also allow designers to fly through and around
3D digitally-computerised objects. This allows designers to see how various components of the product fit together, what are the distances between different shapes and objects, and view the relative positioning of different planes and edges with few visual barriers.
This way, designers can increase the accuracy of the product design which can translate into reduced errors in manufacturing (Yap et. al, 2003).
Image 1. Examples of 2D (left) and 3D (right) computerised models.
When fully developed, a digital 3D representations is a complete digital prototype of a
building that acts like the actual building (Saggio, 1997). A digital 3D prototype can
43 reflect all elements of the design and construction process and permits different tasks and
elements of the building to be dynamically interconnected so that multiple actors can
coordinate and share their knowledge more openly (Lacourse, 2001). Digital 3D also
enhances designers’ cognition by allowing multiple aspects of a building to be displayed
in a single representation and explored interactively (Baba & Nobeoka, 1998). It also
enables designers to visually move back and forth between views of detail and views of the whole in a hermeneutic process of learning (Boland et al, 2007). In addition, digital
3D affects how design information is used by project stakeholders, making previously tacit design knowledge explicit (Yap et al., 2003) and enabling richer, more detailed
knowledge transfer and interactions.
Using object-oriented technology, where data can be captured in data-objects, 3D
modelling allows objects to assume and simulate solid properties such as mass, volume
and density (Yap et. al, 2003). In 3D modelling software, graphic representations are
accompanied by parametric-mathematical representations that accurately describe the
represented object. 3D modelling can synchronise both representations in the sense that
adjusting one type of representation will cause the other type of representation to change
accordingly, and in real time. Further, as new graphic representations of product
components are created, they are automatically structured as data-objects with their own
unique set of numerical parameters and attributes. The dual representation of 3D objects
allows capturing rich and accurate design information such as measurements, shape,
colour, volume etc. These are important to communicate the nature of the designed object
among designers and across organisations.
The dual-representational information can also be stored, modified, and later reused
44 through a repository (Toupin, 2001). Designers can easily manipulate complex surface geometries, more freely experiment with non-traditional shapes, and quickly explore their costs and structural implications (Lindsey, 2001). Changes to sections of an integrated 3D model can be automatically propagated throughout the whole design (Argyres, 1999) and show consequent changes to adjacent sectors of the building. This helps manage complex designs and understand interactions between design components (Greco, 2001). For all these reasons, 3D representations have the potential to dramatically reduce the cost, effort, and error rates of design and construction projects (Koutamanis, 2000).
45 4. Research Methodology
4. 1 Empirical case: changes in modelling technologies in the AEC industry
My research focused on the replacement of 2D modelling technologies by 3D
technologies by various organisations in the AEC industry and on the way that these technologies were used as a new form of boundary object to represent knowledge and facilitate collaboration during construction projects. I additionally observed the associated changes in the infrastructural practices and identities that these organisations have experienced.
I utilised data from multiple case studies that centre on two organisations - Hoffman
Construction Company (hereafter Hoffman), a general contractor, and A. Zahner
Company (hereafter Zahner), a metal fabricator – and their use of 2D and 3D
technologies during their involvement in multiple construction projects. For each of the
two organisations, I initially established a baseline case, or rather, an ideal-type case that describes their typical work practices, interactions, and identities during construction projects in which conventional 2D technologies are used. Such interactions and patterns of engagement do not characterise any one specific project in which either Hoffman or
Zahner have been involved. Rather, they are reflective of these organisations’ work practices and collaboration patterns that have emerged and matured as the use of 2D technologies in the AEC industry has become institutionalised and accepted as the norm.
I then used these two ideal-type cases as a benchmark against which I evaluated Hoffman
and Zahner’s work practices, cross-organisational interactions, and identities as they
started using 3D modelling technologies in their construction projects. This was done by
46 focusing on each of the organisations’ interactions with other actors during two different projects in which each of them was involved (two projects per organisation and a total of four projects) and in which 3D technologies were used to a varying degree. These construction projects demonstrate the transitions that each of the organisations experienced when it started to incorporate 3D technologies to replace 2D technologies in its construction practice. They were therefore used a key domain to understanding how changes in IT-based boundary objects are associated with changes in organisational identities and information infrastructures.
It should be noted that my use of the terms information infrastructure and organisational identity does not confer to all the employees of either Hoffman or Zahner. My research focuses on the association of changes in modelling technologies with information infrastructures and organisational identities. Therefore, in the presentation of the case studies and in their analysis, the two concepts will only apply to those people within the two organisations that actively interact with the technology in question, namely, 2D and
3D modelling tools. Therefore, my claims for established practices and identities, or for changes in practices and identities in the two organisations, only pertain to those people who actually participate in construction projects on behalf of their organisation and actively engage with modelling technologies in the process of doing so. It is quite likely, for example, that whereas the practices and identity of the people that use the technology change when 3D tools replace 2D tools, the practices and identity of other groups of
47 employees working at Zahner or Hoffman (such as accountants, human resource managers, or receptionists) remain stable4.
4.2 Research sites: Hoffman Construction Company and A. Zahner Company
4.2.1 Hoffman Construction Company
Hoffman Construction Company is a general contractor that has its main offices in
Portland, Oregon and employs approximately 100 people. The company is widely
regarded as one of the leading and most experienced general contractors in the AEC
industry. As a general contractor, Hoffman is responsible for ensuring effective
collaboration among the various stakeholders during construction projects and for
managing and mediating communications between subcontractors and the architectural
team.
Hoffman was established in 1922 and played a key role in Portland’s building boom
around that time. During the late 1940’s and 1950’s the company started getting involved
in large commercial projects and provided strong support for the growth of the Northwest
wood products industry. In the early 1970’s Hoffman first introduced computerised technologies such as 2D CAD into its practice, which served the company as it became involved in a number of projects of massive size and complexity such as the construction of nuclear and coal-fired power plants, and a submarine base for the US navy. During the
1980’s and the 1990’s the company continued to gain a reputation as a general contractor
4 This is consistent with my previous theorising of organisational identities which claims that organisations may exhibit multiple concurrent identities to reflect the multiple relationships in which its different units are engaged.
48 that specialises in large and complex construction projects. It lived up to this reputation
when it started using advanced 3D modelling technologies in the late 1990’s during its
involvement in the construction of the EMP building in Seattle, Washington, USA which was designed by Frank Gehry. Since then the company’s use of 3D technologies has increasingly grown although it still uses conventional 2D technologies in its more traditional construction projects.
4.2.2 A. Zahner Company
A. Zahner Company is a metal fabrication firm with its main offices in Kansas City,
Missouri. The company is well known in the AEC industry for its advanced technological and design capabilities and for its innovative approach to metal fabrication. Many of the company’s works have been designed by leading architects in the field such as Frank
Gehry and Randall Stout whose designs are famous for their unique geometrical shapes and use of metal. The company employs 150 people and is expected to grow to approximately 220 employees by the end of summer 2007.
Zahner was established in 1897, and has been a family owned business for four
generations. In its early years, the company leveraged the increasing demand for housing
in the Mid-West of the USA which was brought about by the population spread to the
west, and the growing use of steel in the construction industry, to significantly expand its
business. The company continued to grow as it started manufacturing and installing metal
siding panels for civic buildings and industrial plants in the 1950’s and 1960’s.
Zahner’s use of computerised technologies began in the 1980’s with the
commercialisation of the PC. During this decade the company introduced 2D CAD into
its design and manufacturing processes. In 1989, the company first started collaborating
49 with Frank Gehry in the construction of the Frederick R. Wiseman Museum in
Minneapolis, Minnesota, USA. During work on this project and on the Museum of
Science and Industry in Tampa, Florida, USA, which was also designed by Gehry and built around the same time, the company was introduced into the world of 3D design (it is important to note, however, that it still did not utilise 3D modelling technologies in its
practice. Also, Gehry himself was still using physical models in his design process and
did not start incorporating computerised 3D technologies into his practice before the beginning of the 1990’s).
Being exposed to Gehry’s work and exceptional designs had made Zahner consider
adopting 3D modelling technologies into its own practice, which it did in the end of the
1990’s during its involvement in the construction of the EMP building in Seattle,
Washington, USA which was designed by Frank Gehry. In the following years the
company gradually incorporated 3D tools into its practice and they are now an integral
part of its design and fabrication processes. Today Zahner is recognised as a leader in the
AEC industry in designing and fabricating metal using 3D technologies such as Rhino,
CATIA, and Pro-Engineer.
4.3 Research design
Four inductive case studies (Yin, 2003) were conducted to study boundary objects in an
environment in which they play a salient role. Each case study comprises a narrative of a
construction project in which either Hoffman or Zahner were involved5. A case study is
most suitable when “how” or “why” questions are posed and when the investigator has
5 As noted above, in addition to narratives of four actual construction projects, an ideal-type case description of a conventional 2D-based construction project is presented for each organisation.
50 little or no control over complex events which take place in some real-life context (Yin,
2003). Given the nature of my research questions and since I had no influence over the proceedings of any of the construction projects in which any of the companies had been involved, I chose to use this method. Furthermore, because there was not a significant research base pertaining to my research interests, I used an inductive logic to yield new understandings of the relationships among boundary objects, information infrastructures, and organisational identities instead of trying to validate existing models or theories.
4.4 Sampling This study was carried out in the context of a broader research project that was conducted by a group of researchers of which I was a member at Case Western Reserve University.
The aim of the project was to examine the pattern of innovations in the AEC industry following the implementation and application of 3D modelling technologies by Gehry
Partners (see Boland et al, 2007). Our interest in the AEC industry developed when the business school at which we worked moved to a new building that was designed by Frank
Gehry in 2001 (See image 2, below).
51
Image 2. The Weatherhead School of Management at Case Western Reserve University in Cleveland, Ohio.
One of the researchers in our group was a member of the faculty steering committee that was responsible for liaisoning with Gehry Partners and for representing the school’s interests and requirements vis-à-vis the architectural team throughout the design and construction processes. While fulfilling his role, this researcher developed personal relationships with a number of people from Gehry Partners who later granted us access into the company to collect data. Through the people we interviewed at Gehry Partners, we received additional contacts of people and companies that had worked or were working with Gehry on various projects such as the EMP building in Seattle, Washington and the STATA centre at MIT in Boston, Massachusetts. The common theme to all these buildings was the significant use that was made of 3D modelling technologies during their design and construction. These newer contacts later referred us to additional companies that they knew had used 3D modelling tools in their practice. As our project moved forward, we collected a sizeable pool of interviews with dozens of people from a wide variety of organisations in the AEC industry that, in one way or another, had been involved with the use of 3D technologies (see table 1 below).
52 The selection of the organisations that I focused on in this dissertation was based on my
own theoretical interests and practical considerations. During the writing of my
dissertation proposal I went through an iterative process that included developing and
refining my theoretical interests and vocabulary as well as getting increasingly
familiarised with the nature of the AEC industry and of some of its main actors. As this
process progressed, I identified qualities of a research site that would be helpful in
answering my research questions. First, given my interest in the relationship between
changes in the use of modelling technologies (conceptualised as boundary objects) and
the information infrastructures and identities of technology-using organisations, I wanted
to examine firms that had had significant experience in using conventional 2D modelling
technologies in their practice in addition to having experienced using 3D modelling tools
in some of their construction projects. Second, to increase external validity, I wanted to
study firms in two different professional positions within the industry’s network of
relationships to observe the interaction of boundary objects, information infrastructures,
and organisational identities under different technological, organisational, and social
conditions. Third, to account for my theoretical interest in observing the implications of changes in technological boundary objects for the organisations that use them, within each firm I sampled two case studies (i.e., two construction projects in which the firm had been involved) that reflected a significant variance in the use of modelling technologies6.
I thus followed a theoretical replication (Yin, 2003; Eisenhardt & Graebner, 2007) in that in each case a different type, or combination of types, of IT-based boundary objects was primarily used. For each of the two firms I relied on three cases (one ideal-type case and
6 As explained above, I conducted these two case studies in addition to establishing an ideal-type case for each of the companies that describes their involvement in 2D-based construction projects.
53 two actual cases) where the use of either 2D modelling technologies, 3D technologies, or
a combination of 2D and 3D technologies was prevalent.
Another consideration in the selection of cases for each company was maintaining
temporal continuity between them. That is, the selected cases constitute a continuous
chronological path in the lives of each of the two companies; the involvement of each
company in the second actual construction project, immediately followed its involvement
in the first one, which, in turn, represents the company’s first (in Hoffman’s case) or one
of the first (in Zahner’s case) significant breaks from its previously established 2D-based
practice, as represented in its ideal-type case. The rationale for doing so was to try and
maintain a certain level of continuity in laying out each of the companies’ narrative
concerning its use of 2D and 3D modelling tools and the observed associated changes.
Hoffman’s first actual case is a description of the first project in which the company was
involved and in which significant use of 3D modelling technologies was made. It thus represents a major break from the company’s previously established information infrastructure and identity that characterised its involvement in 2D-based construction
projects. Hoffman’s second actual case describes a project that immediately followed the
first one and where both 3D and 2D technologies were used. This case therefore allowed
me to follow Hoffman’s information infrastructure and identity as they continued to
unfold. Zahner’s first actual case study is a description of one of the first complex, large-
scale projects in which the company took part and during which it heavily used 3D
modelling technologies, although most of the other participants used 2D technologies7.
7 As noted above, although Zahner’s initial exposure to 3D design methods took place in the late 1980’s, its first actual use of 3D modelling technologies took place in the end of the 1990’s.
54 The case illustrates the changes in the company’s information infrastructure and identity compared to the idea-type case which describes Zahner’s typical involvement in 2D-
based projects. The second actual case is a depiction of a construction project that
followed the previous one and where Zahner, as well as all other major participants, used
3D modelling technologies. This case enabled me to continue to examine Zahner’s
changing information infrastructure and identity.
4.5 Data collection and analysis
93 semi-structured interviews were conducted with 52 interviewees from 16
organisations in the context of our research project (see table 1 below). For the purpose of
this dissertation, the majority of these interviews were used to provide background information about the AEC industry, its typical infrastructural practices, the prominent actors operating in it, and the main technologies that they use. Out of the 93 interviews,
18 were conducted with five Hoffman employees: an executive, two project managers, a detailer, and a surveyor. 16 additional interviews were conducted with six Zahner employees: two project managers, as assistant project manager, a lead designer, a project engineer, and the company’s CEO. Interviewees were selected based on their level of familiarity with the AEC industry and its institutionalised practices and on their involvement in their company’s deployment of 3D modelling technologies. Thus, all interviewees are experienced veterans in their fields who had significant knowledge of their company’s ways of operating and of the technologies that it uses in its construction projects. Furthermore, interviewees from Hoffman and Zahner were selected if they had been involved in their company’s 2D as well as 3D construction projects.
55
Table 1. Number of interviews and interviewees Interviews Interviewees Hoffman 18 5 Zahner 16 6 Remaining 14 firms8 59 41 Total 93 52
Interviews with Hoffman and Zahner employees formed a critical component of my data
and were the focus of my analysis. The first meeting with Hoffman took place in
December 2003 during a visit to their offices in Seattle, Washington. Since then we have
conducted two additional visits to their office, most recently in February 2007. The first
meeting with Zahner took place in the summer of 2002 during a visit to their offices in
Kansas City, Missouri. Since then I have conducted two additional visits to their office,
most recently in March 2007. During those visits, methods for data collection included
observations of meetings and work on construction sites, review of construction
documents and models, and office and on-site interviews.
Interviews lasted between forty five minutes to two hours and were all recorded and
transcribed. My main interest in collecting the data was to understand Hoffman and
Zahner’s construction practice and explore how it played out during their interactions
with other organisations during construction projects. I was particularly keen to
understand how the conditions for the enactment of Hoffman and Zahner’s practices,
interactions, and identities were reshaped with the introduction of 3D technologies into
their practice. Interviews were analysed by relying on established methods for handling
8 Members from the following firms were interviewed: Gehry Partners (Los Angeles), Hunt Construction Company (Indianapolis), Mariani Steel Fabricators (Toronto), Cleveland Fire Prevention Bureau (Cleveland), Desimone Engineering (New York), Donnelly Concrete (Cleveland), Spark Steel (Toronto), Columbia Wire and Iron (Portland), GQ Contractors (Cleveland), Skanska Construction USA (Boston), Coop Himmelblau Architects (Vienna), Etalliers Jean Nouvel (Paris), Westbrooks Architects (Cleveland), and Magnusson Klemencic Associates (Seattle).
56 qualitative data (Eisenhardt, 1989; Yin, 2003) and followed these procedures: The first stage involved preparing detailed write-ups for each interview. Based on these transcripts
I constructed profiles of six construction projects in which Hoffman and Zahner were involved. Each profile included a story line, or narrative, of a construction project that consisted of a description of the main actors in the project, their main actions and interactions, and the technologies that they used to communicate with each other. Within each construction project I next sought to instantiate the three theoretical constructs which had initially shaped my inquiry, namely, boundary objects, information infrastructures, and organisational identities. To establish construct validity, this was done by matching interviewees’ descriptions and data from observations and documents with the theoretical characterisations of each construct. For example, if a technological artefact was referred to by an interviewee as an enabler of communication between diverse organisations, this artefact would be labelled as a boundary object (e.g., “…we frequently exchanged 2D CAD models with other subcontractors. These models were our main means for communicating with them.”). Similarly, an interviewee’s reference to an organisational practice that had remained stable over an extended time period and become part of the company’s established way of doing things would warrant labelling this practice as part of that company’s information infrastructure (e.g., “...typically we would interact only with a limited number of subcontractors. There just isn’t a need for more than that when the building is all in right angles and simple and everything is in 2D.
That was normally the case but things started to change when 3D tools were introduced.”).
57 Subsequently, within each project I looked for concrete examples of the relationships among the three theoretical constructs, specifically, for changes in organisational practices and identities that were associated with the use of the new 3D boundary object.
Then I mapped these interactions onto a more consistent framework for understanding the relationships between technologies (conceptualised as boundary objects) and the organisations that use them and formulated the process models that show the main configurations of interactions between actors and boundary objects in each case. This was followed by conducting a cross-case comparison of the relationships among the three constructs and outlining the observed patterns of change in practices and identities in a theoretical model9.
9 As an evidence of the model’s reliability, some of the interviews used here were reviewed and analysed by three additional members of the research project of which I was a member, who have reached conclusions that are consistent with the pattern of organisational changes presented here (see, Boland et. al., forthcoming).
58 5. Results
The cases below illustrate the changing dynamics of interactions among actors in six
construction projects in which Hoffman and Zahner were involved (two ideal-type
project descriptions and four actual projects). The first three cases depict in temporal
order the changing nature of Hoffman’s infrastructure and identity as the company starts
using 3D modelling technologies. The first case describes Hoffman’s typical
infrastructural practices and identity prior to using 3D technologies. The second case (i.e.,
the first actual case) demonstrates how Hoffman’s infrastructure and identity start
changing when it takes part in a project where 3D modelling technologies are used. The
third case (i.e., the second actual case) shows how Hoffman’s identity and infrastructure
continue to evolve when it partakes in a project where a combination of 2D and 3D
technologies is used10. The last three cases describe in chronological order Zahner’s
changing information infrastructure and identity as the company starts using 3D
modelling technologies in its practice. The first case describes Zahner’s typical
infrastructural practices and identity prior to using 3D technologies. The second case (i.e.,
the first actual case) illustrates how Zahner’s practices and identity change when the
company participates in a project where it used 3D technologies but the rest of the
construction team used 2D technologies. The third case (i.e., the second actual case)
demonstrates how Zahner’s practices and identity continue to evolve as the company
takes part in a construction project where most of the involved parties use 3D modelling
technologies11.
10 To ensure construct and face validity, a report containing the findings from these cases was reviewed by one key informant from Hoffman who confirmed the findings’ faithfulness. 11 As explained above, claims for (changes in) organisational identities and infrastructures only pertain to those people within Hoffman and Zahner who actively interact with 2D and 3D modelling tools.
59
5.1 Hoffman
5.1.1 The poker dealer: Hoffman in typical 2D construction projects
The use of 2D models in the AEC industry is a standardised and taken for granted aspect
of the relationships among architects, general contractors, and subcontractors. The
infrastructural practice in a typical construction project starts with an architectural team
working to create a set of 2D drawings, which are called contract documents. These documents convey the architect’s design intent and indicate the performance and quality requirements of the building. Many of the systems that must be included - mechanical, electrical and structural - are shown schematically, but not in elaborate detail. The architects then deliver the contract documents to Hoffman, which distributes them to the subcontractors that are involved in the project. The subcontractors submit back to
Hoffman specific work plans and shop drawings. The latter are detailed depictions of the manufactured items that go in the building and the systems that tie into them. These drawings, and not the contract documents, eventually become the blue prints used for construction. Hoffman has to ascertain that the shop drawings are in compliance with the contract requirements and industry standards and then forwards them to the architects for approval. When the drawings are approved, they are sent back to the subcontractors that can begin their work.
60
Image 3. One of Hoffman’s 2D projects: An office building in Portland, Oregon.
Since the contract documents do not specify in detail the construction process, it remains unclear which components of the building need to get built first. In principle, each subcontractor wants to be the first to lay its equipment and product so that they serve as a standard for other subcontractors. However, this is done strategically as different subcontractors try to estimate their potential risks and benefits. Project manager 1 explained:
Who’s going to step up and take the lead? Normally the one who steps up first is the one who thinks they have the most to gain by establishing the location of their parts first. But they won’t necessarily want to do the smaller things because they usually have to take second priority to other bigger things by other subcontractors, so they don’t want to go to the trouble of laying all those out first until someone else has put all their stuff in and then they’ll layout their stuff around that. So everybody wants to wait until somebody else has done something.
Establishing coordination for construction is a highly spontaneous and emergent process.
However, Hoffman tries to regulate this procedure and introduce order into it. Project manager 2 provided the following example:
61 If the electrical contractor has a 4 inch conduit that he wants to route down a corridor and the duct work contractor has a 24 inch duct that he wants to route down a ceiling, and both of them are bringing their contract documents and showing how the engineers have told them to do that, we have to direct traffic and come up with a hierarchy of who goes first and who goes where.
The recurrent use of 2D contract documents and shop drawings has standardised the form and channels for inter-organisational communication and reduced the need for organisations to interact face to face (Boland et. al., 2003). Because the contract documents produced by the architect are subsequently translated into shop drawings that are individually created by different professional organisations, each organisation ends up using separate models to support their respective construction activities. Consequently, each organisation operates in a partially isolated silo with little interaction with or visibility into the processes of other organisations. This generates a loosely-coupled system of communication flows among heterogeneous organisations that, nonetheless, have to maintain a high level of coordination. Therefore, Hoffman’s role in 2D-based projects consists primarily of making sure that collaboration takes place among organisations. A Hoffman detailer commented:
We have a fragmented collection of data points that originate from different subcontractors' shop drawings that have to be assimilated into a task. It is our responsibility to make sure that this is done accurately and efficiently.
Accordingly, Hoffman’s infrastructural practice relies, to a large extent, on the use of 2D
models as boundary objects to coordinate construction efforts and reconcile various and,
at times, conflicting perspectives that participating organisations have. Project manager 2 explained
We do a lot of interdisciplinary coordination. We have these things called shop drawings. Those are used primarily by the mechanical, electrical, sometimes structural engineers
62 and the general contractors. What happens is that everybody wants to take a straight shot from point A to point B of what they need to put in, but we need to work together to come up with a scheme that makes the most sense for everyone. We end up doing this in a more organised way. Everyone involved brings their drawings and we roll up our sleeves and work through the nuances… and then people leave with that collective information and go revise their own drawings.
Figure 1 below illustrates the established pattern of relationships among different
organisations in 2D-based projects. It reflects Hoffman’s central position in this network
of relationships and the extensive use that is made of 2D models (i.e., contract documents
and shop drawings) to coordinate the construction process. The arrows in the figure
denote the transfer of information across organisations through 2D models. These models
act as immutable mobiles (Latour, 1987) in the sense that the information inscribed in
them remains unchanged as they are transferred across organisational boundaries. This way they are used by different organisations in an attempt to influence the sequence of, and their own role, within the construction process, particularly during the coordinated shop drawings process.
Figure 1. A process model of relationships in 2D-based construction projects
63 As I premised above, Hoffman’s organisational identity is predicated upon the
interactions in which it is involved. In 2D-based projects, the nature of these interactions
noticeably resembles those that characterise a poker game. Each actor’s actions are
calculated based on the anticipated actions of other actors in an effort to increase their
own benefits and minimise their losses. Furthermore, the actors are wary of the
information that they share with others. From a subcontractors’ perspective, each
subcontractor is reluctant to share its shop drawings with other subcontractors as they
may contain proprietary knowledge and because it is not incentivised to do so. Similarly,
architects are reluctant to release fully detailed information in their contract documents
due to liability concerns and to avoid contradicting their own modelling and
measurements. Project manager 2 provided the following anecdote:
Architects are taught, the term is called 'do not double dimension', and they are taught never to give the same information more than once in a set of drawings to avoid potential conflicts: reduce the information content to reduce the probability of contradictions.
In a system in which actors only cautiously and strategically share information with each
other, Hoffman is the only party that has access to everybody else’s information bases: it
is the only one to receive all the architect’s contract documents and all the
subcontractors’ shop drawings. Moreover, Hoffman too, is strategic about sharing
information with others. It has to make sure that it does not reveal any of the other actors’ proprietary knowledge, and that it does not unnecessarily burden different subcontractors with overwhelming amounts of information. Thus, Hoffman is situated at a central node
in a system composed of itself, the architects, and the subcontractors; it is the only party
to have access to everybody else’s information and is in control of the information that
everybody else gets exposed to. Much like a poker dealer, Hoffman sits at a table while
64 each party lays out their cards for construction. Each player keeps their proprietary
information to themselves and tries to negotiate the sequence of construction and the
location of their components for their own benefit. Hoffman’s identity is derived from
this prominent position within this system of relationships. When prompted to describe
Hoffman’s identity, Project manager 2 provided this allegory:
Basically we're more like dealers. Everybody is sitting around a table and we're playing poker. We are the dealer. We are making the rules, we're controlling the flow of the cards.
5.1.2 The dispersed collaborator: Using 3D modelling technologies with Gehry Partners In the summer of 1997 Hoffman became involved in the construction of the Experience
Music Project (EMP) in Seattle, Washington, which was designed by Frank Gehry. This was the first complex, large-scale, 3D-based construction project in which Hoffman took part.
Image 4. The experience music project in Seattle, Washington.
65 As noted, Gehry started using CATIA in his practice in the beginning of the 1990’s. The
EMP building was one of the first projects in which Gehry used CATIA models as contract documents, and insisted that all the main subcontractors adopt CATIA.
During the construction process, infrastructural practices that are usually taken for granted within the AEC industry were disrupted. The loosely-coupled inter-organisational system associated with the established 2D document-driven construction process, was reshaped into a more tightly-coupled and interdependent system. In this system, new levels of knowledge sharing and communications were reached among the participating organisations. One reason for these changes had to do with the overarching use of the
CATIA technology. Since Gehry Partners imposed 3D CATIA models on Hoffman and on the prime subcontractors, they all had to learn how to use the software in one way or another. They were all influenced by it and had to integrate it into their practice:
CATIA was imposed on the team. In some ways it was an inefficient process because we had to inherit an entire learning curve at once. Structural engineer, mechanical engineer, prime subcontractors, Hoffman, owner, client. A whole list of people had to learn how to use this tool in one way or another. (Project manager 2)
3D CATIA models are extremely rich and accurate digital representations of the building that embed multiple layers of information that are meaningful to different professional organisations. Therefore, the same computerised model can be used by multiple subcontractors to support their construction activities, thereby eliminating the need to create separate shop drawings. Also, 3D models are transparent such that changes that are made in one portion of the model are immediately visible to other users of the model.
Additionally, any changes to the model are checked automatically and on the fly to verify that they do not clash with other elements in the model, thus enhancing inter-
66 organisational coordination and allowing new levels of interactivity. A Hoffman detailer
elaborated on this issue:
In 2D, drawings aren’t put together in one framework whereas in 3D it’s all on one visual platform so you can see how things combine together. In a 3D environment you can see all the conflicts in the model upfront. In 2D you have to do all the layouts first and check each model separately and manually, and then, at the end, come to this conclusion of, well these things don't match up, they don’t line up. Putting together a 3D model is a good way of having visibility into discrepancy issues, constructability issues, and better understanding of how different models come together.
A second reason for the emergence of a more tightly-coupled system during EMP
concerns the nature of the task. The high complexity of the building combined with the
fact that most of the participants had not been previously involved in such a project or
used CATIA facilitated much tighter collaboration throughout the construction process.
The pre-established 2D contract language and associated infrastructural practices of loose coupling did not provide actors with sufficient information to understand and negotiate their new roles and tasks (Boland et. al., 2003). Therefore, they relied extensively on each
other to inform one another about their designs and work during the construction process.
Project manager 1 remarked:
During the construction of EMP we met much more frequently than we would on a typical [2D-based] project. Often there would be a problem on how to support a particular curved wing or something made up of warped metal sheets or something, and there was no definition of how to do that. And so we would meet and come up with ideas on how to do it. And those ideas would bounce from the structural engineer, to the architect, to us, and we would all have ideas and we would all contribute using the CATIA model. Using the CATIA model to describe your ideas made it easier to reach a solution.
To make the project’s joint construction and design effort as productive as possible,
Hoffman organised the main subcontractors into eight design/construction coordination
teams, which met regularly. Those subcontractors included Zahner, which was
67 responsible for fabricating and installing the exterior metal cladding system; Columbia
Wire & Iron Works, which produced curved structural steel members; Benson Industries,
which fabricated and installed the exterior glazing; and Permasteelisa Cladding
Technologies, which installed the glass-and-steel exterior skin. Project manager 1
elaborated on the nature of the inter-organisational meetings that involved these
organisations:
The team meetings were a collaborative process in its pure form. They were typically held once a week. We had one Team 1 meeting each week and one Team 2 meeting each week. Team 1 was the exterior cladding, which was done by Zahner, team 2 was the structural steel, team 3 was the interior exhibit areas, all the way to team 8 which was the technology group. Hoffman ran these meetings. For example, the team 2 meetings. Team 2 was comprised of Bob Park who is the owner of the steel fabrication company (Columbia Wire & Iron Works), his detailing team, his installation team, and then myself, the structural engineer (Magnusson Klemencic Associates), the architect, and the owner. The structural steel design was the critical path of the schedule for almost six months. So we were literally getting a piece of the design done on a certain day with Bob and the next day we were fabricating the steel. As soon as we got the design done we were fabricating. It was essential that that process be as efficient as possible.
In this newly formed interdependent system, Hoffman’s infrastructural practices and
typical role as a general contractor markedly changed. Instead of managing a linear 2D-
based communication process, the company became more involved in the construction
process and stepped out of its traditional organisational bounds. One example of this has
to do with the use of CATIA by the subcontractors. During the construction of EMP, the
3D CATIA-generated contract documents that were created by the architectural team
served as the central component around which construction efforts revolved (instead of
the shop drawings which are produced by the subcontractors in a 2D-based process).
However, since most of the subcontractors were not familiar with CATIA, Hoffman hired
CATIA workstations and operators and placed in them in a few of the major
68 subcontractors. This was unusual as the general contractor now had its own employees
located in the subcontractors whereas usually the parties remain clearly separated. Project
manager 1 provided the following example:
We hired a glass contractor, they did all the glass work, some of which was complex and curvy. To figure out how that works with the metal skin, we needed to have the glass work done in CATIA. That wasn’t something the contractor anticipated when they bid the job to us. They thought, ok you give us the information and we will somehow get it into 2D CAD and manipulate it. But we found that there were enough conflicts that they would find that would require 3D modelling. We found that [2D] sketches were not working and needed to be adjusted, and the only way to do that was to build all the glass in CATIA. So we ended up hiring a CATIA guy to work at their office to do that. If the geometry of the building was simple enough that it didn’t require that, this was probably an expense you wouldn’t go through. But the geometry of the building was so complex that was probably the only way to do it.
A second way in which Hoffman became more involved in the construction process
concerns the design of the 3D models. The initial models that were produced by Gehry
Partners often were not sufficiently detailed to support construction. Therefore, Hoffman
had to step in to further refine and develop these models and transform them from design
schematics to actual construction documents, sometimes in collaboration with the
subcontractors. Project manager 1 provided an example of such a situation:
Gehry's people didn’t have the surfaces of the building trimmed to fit the sidewalk in the CATIA model. Some of our first metal panels we had were hitting the ground. And we said, what’s going on here? Didn’t you have those trimmed? And they said, no, we had extended the surfaces too long and we figured you’d trim them off. We took the CATIA model and, with the metal fabricator, set the lines and gave them elevations and essentially built the surfaces, and made cut lines around the building where the sidewalks are.
Working on the models required extensive collaboration efforts from Hoffman and the
involved parties that extended well beyond the requirements of a typical 2D construction
project:
69 We would sometimes have three or four meetings a day with six to 12 people from different subcontractors. To compare, a typical 2D project would have one meeting a week. In a 2D process, shop drawings would get approved prior to construction. Here things were being designed on the fly. One reason is that the models were so complicated and difficult to understand and we wanted to make sure we were on the same page with everyone else. The second reason was that the design wasn’t done, sometimes it just didn’t exist. For example, in some cases the metal skin of the building was shooting 20 feet away from the concrete and this information did not exist. And we met to figure out how to correct it. (Project manager 2)
Changing the architects’ contract documents is something general contractors normally
refrain from doing. Usually, due to liability issues, the general contractor prefers not to
meddle with the contract documents and instead receives them from the architects and
sends them to the subcontractors. Thus, the Hoffman’s practice during the construction of
EMP was very unusual:
The idea that we added information to the [3D] model is different than what a true general contractor does, because a true contractor doesn't touch any of the product. They just want to shuffle the cards and deal them out and dictate the sequence. But we rolled up our sleeves and got involved because there was a lot of information that we wanted to make sure was done right. We are more involved in this process than we might be other wise. (Project manager 1)
Figure 2 below illustrates the emerging pattern of relationships among the different
organisations during the EMP project. The grey areas in the subcontractors’ rectangles
denote the CATIA operators that were hired and financed by Hoffman. The bold arrow
signifies that the 3D model was initially created by Gehry Partners. The double headed
arrows signify the transparency and interactivity that were allowed by CATIA. They
underscore the flexibility of the 3D models, which stands in contrast to the stringent
nature of the 2D paper models that are normally used. The figure also reflects the
centrality of the 3D model as a key coordinating mechanism.
70
Figure 2. A process model of relationships during the EMP project
As Hoffman’s infrastructural practices and interactions changed, so did its identity.
Hoffman’s poker dealer identity drew on the distinct and central position that the company had in 2D-based project settings, and on the relationships with other organisations that were associated with this position. On EMP, Hoffman’s interactions significantly changed as the company became more involved in the construction process and as boundaries between Hoffman and the subcontractors blurred. Accordingly,
Hoffman was now characterised as a dispersed collaborator rather than a poker dealer:
You could say that on EMP we became more of a dispersed collaborator. Although in essence our responsibilities did not change, our people were now dispersed across several organisations. Also, whereas before we were calling the shots, on EMP it was more of a collaborative effort which included Hoffman and the subs. (Project manager2)
71 5.1.3 The Air-traffic controller: The Seattle library project
Almost immediately after finishing work on the EMP project, Hoffman got involved in the construction of the Seattle public library in Seattle, Washington, which was completed in the winter of 2003.
Image 5. The Seattle public library in Seattle, Washington.
Here, again, Hoffman faced a new situation. The building was structurally very complex and designed by another internationally renowned architect, Rem Koolhaas. Koolhaas has been a prominent figure in the AEC industry since the early 1980’s and has designed a number of highly acclaimed buildings such as the Guggenheim museum in Las Vegas and the Educatorium at Utrecht University in the Netherlands. Like Frank Gehry,
Koolhaas used 3D modelling tools to design the Seattle library. However, different from
Gehry, due to liability concerns, Koolhaas refused to share the 3D models with Hoffman or with the subcontractors. He was only willing to release a limited number of models, and even then it was done on a ‘use at your own risk’ basis. Given the enhanced collaboration and coordination that they experienced while previously working in a 3D
72 environment, some of the people at Hoffman did not perceive this decision to be
altogether conducive to the efficient management of the construction process.
The architects hated giving us their 3D models because they knew they would keep changing stuff. If architects could get educated to feel more comfortable with their models and download them to the people that they work with, everybody’s life would be easier. Gehry is very much an exception that way. (Surveyor)
Because of Koolhaas’ decision, all official communication between the architects and the
other participants in the project was in 2D and paper-based. However, large portions of
the building were made of glass that was designed in highly complex shapes. This
required very high tolerances and accurate modelling and measurement.
The library was a very complicated project. We had a lot of sloping slabs and weird angles, and no forgiveness - very high tolerances because of the glass and the complex geometries. There were warped surfaces, like sushi mats, where in some areas all the lines are straight but in others they are curved – a surveyor’s nightmare. And there were four surfaces like this. There were thousands of surveying points for each such surface. It took us weeks to do that. (Surveyor)
In the absence of initiative from the architect and building upon its recently acquired
experience from the EMP building, Hoffman worked actively to create its own 3D
environment to accommodate this complexity, allow for accurate enough designs, and enhance collaboration among subcontractors. To accomplish that, Hoffman leveraged a
3D model of the building that had been originally created by the curtain wall contractor, a
German company by the name of Seele. This model was the starting point for a more comprehensive three-dimensional modelling of the building that was led by Hoffman.
Importantly, this three-dimensional modelling effort was only contributed to by Hoffman and the subcontractors. For the most part, the architects were using their own 3D models
73 and relying on 2D models to communicate with the rest of the construction crew. Project
manager 2 commented:
We didn’t submit 3D electronic models to the architectural team for review. Everything got turned back into paper as far as the administrative hand off from one entity to another went. The paper was the bible that we used but we used these [3D] models to generate the paper.
In this situation, Hoffman’s practices and role as a general contractor changed again. The company took some lessons it had learned from the EMP building and implemented them with its subcontractors. It was more proactive in its efforts to create a tightly coupled system to enhance collaboration. For instance, unlike EMP, Hoffman made a strategic choice to recruit subcontractors that had had prior experience with 3D technologies.
Thus, several of the main subcontractors that had collaborated with Hoffman on the EMP project, were also recruited to work on the Seattle library, such as the detailer (Angle detailing), the mechanical contractor (McKinstry), and the structural engineer
(Magnusson Klemencic Associates). In addition, Hoffman set up a 3D AutoCAD platform (a 3D modelling technology that is similar to the CATIA platform that was used during the construction of EMP) that served as a central translation device that mediated all communication between itself and the subcontractors and among the subcontractors:
On EMP, all the models came from Gehry, he was guiding everything, the structural engineers, everyone was onboard there. On the library project the creation of the 3D models did not involve the architect. It was initiated by us and pushed by some of the subcontractors that collaborated with us. (Project manager 2)
The use of 3D AutoCAD-generated models enabled Hoffman to create a collaborative
project management context for subcontractors. Furthermore, to enhance communication
with and among the subcontractors, Hoffman centralised surveying activities during the
74 project. Usually, the general contractor does some surveying to provide the
subcontractors with basic information that they can use to produce their final measures.
Each subcontractor applies this information to generate their own distinctive x, y, z grids
for the parts they place in the building. But on a complex project like the Seattle public
library, there is a lot more detail to the layout, a large number of potential interactions between components, and therefore basic surveying practice becomes insufficient.
Consequently, Hoffman did extensive surveying work and distributed the information to the subcontractors, helping them to set the layout to their installed components.
Importantly, Hoffman used a centralised surveying system wherein all subcontractors shared the same x, y, z measurement space. This positioned Hoffman as a major provider of information that all the subcontractors had to rely on to carry out their work:
The amount of surveying work we did on the library project was extraordinary. Typically the subcontractors rely on us to some extent for this information but also have their own people who do surveying. Here we were the only ones feeding this information to them because without it they would virtually be lost. (Surveyor)
Figure 3 below illustrates the pattern of relationships among the participating
organisations during the construction of the Seattle public library. The bold arrow
denotes that the 3D environment was generated by Hoffman and the dashed arrows
signify the surveying information that was provided by Hoffman to the subcontractors.
The figure again reflects the centrality of the 3D AutoCad environment to the
coordination of construction activities between Hoffman and the subcontractors.
75
Figure 3. A process model of relationships during the Seattle Library project
Hoffman’s identity was reshaped again as the conditions for its ongoing enactment changed. By playing a proactive role in the construction process, Hoffman brought more
knowledge to bear and was able to act like an air-traffic controller. Discussing the 3D
AutoCAD environment that Hoffman set up and the surveying information they provided
to the subcontractors, Project manager 2 explained:
We become more like air-traffic controllers or knowledge brokers. We are still dealing with traffic, but we are using this radar tied to a computer and a radio to be more effective. We are putting radios in all the cars and being able to call the guy coming down the street and say hey you need to change lanes before you get to the centre section.
5.2 Zahner
5.2.1 The Detached Worker: Zahner in typical 2D construction projects
Zahner’s involvement in 2D construction projects typically starts when the company
receives a set of contract documents which are created by the architects, from the general contractor. The drawings convey the architects’ deign of the building in general terms
76 without specifying the materials to be used, the construction techniques, or the sequence for construction. Based on these drawings, the company makes a bid for the project by
submitting to the general contractor a set of 2D shop drawings (CAD models and paper
drawings) that detail the parts the company intends to fabricate, the materials from which
the parts are made, and the way in which the parts interface with other systems in the
building. The shop drawings are subsequently reviewed by the general contractor to
ascertain that they meet the industry standards and maintain the integrity of the
architects’ general design intent. When the drawings are approved, they are sent back to
Zahner, which can then start fabricating its parts for construction.
Image 6. One of Zahner’s 2D projects: A residential building in NYC.
With buildings that are structurally simple and that are designed and modelled by the
architectural team in 2D, Zahner’s modelling and fabrication of its own parts is typically
also done in 2D, usually using CAD. Furthermore, during such construction projects, most inter-organisational communication, coordination, and sharing of information are carried out through the exchange of 2D models. In most cases, these communication flows adhere to very specific patterns that reflect the functional structure of the building.
77 That is, a subcontractor would typically only interact and exchange information with other subcontractors whose manufactured systems directly interface with its own, and even then, this would be done through the mediation of the general contractor. Ordinarily,
Zahner’s manufactured metal parts interface with the structural steel system, the glazing, and sometimes the concrete, drywall and air conditioning systems. Accordingly, in designing, fabricating, and laying out its parts, Zahner has to maintain some level of cooperation and mutual communication with the subcontractors who are responsible for manufacturing those systems. A lead designer at Zahner commented on the matter:
The 2D drawings are mainly done to communicate with other subcontractors. We typically interact with the drywall systems, glazing, and interior steel. We need to make sure the building metal envelope stays watertight so we need to make sure there is continuity between our parts and these trades’ parts.
Thus, the pattern of relationships between Zahner and other subcontractors follows a
rather orderly inter-organisational structure that is derived from the design of the
building. This pattern of relationships is conducive to limiting the scope of inter-
organisational communication and the quantity of the content exchanged as it narrows
down the channels for inter-organisational relationships. Another reason for the restricted
interactions among the subcontractors has to do with the contractual arrangements among
the parties involved in the construction project. Contractually, the formal responsibility
for maintaining effective coordination among the different subcontractors lies on the
shoulders of the general contractor. Similarly, each subcontractor is contractually liable for manufacturing and installing only their own systems and parts. This creates a situation in which each organisation, be it the architects, the general contractor, or any of the subcontractors, has a negative incentive to engage in any activities that even slightly
78 exceed their formal and documented responsibilities. In such circumstances, Zahner, as
well as any of the other subcontractors, invests most of its time and effort in its own
modelling and fabrication processes, while limiting its interactions and exchange of information with other parties to the required minimum. As noted, these interactions only take place with those subcontractors whose parts and systems directly interface with those of Zahner, and through the mediation of the general contractor. The company’s
CEO made this observation:
We are finding that everyone wants a structured process that follows an organisational chart. This is a system that is based on an ongoing hand-off process where each subcontractor passes their models on to the subcontractors that they interface with and no one wants to breach their boundaries. They all want a hierarchical lay out: I am here, the general contractor is here, here are all the different subcontractors, they all have their areas of responsibilities and do not blur them.
Many times, the nature of these inter-organisational relationships is characterised by a
culture of risk aversion on the part of multiple organisations, and explicit efforts to pass
on their responsibilities to their counterparts. Zahner’s CEO provided the following
example involving the role general contractors play in construction projects:
The general contractors typically do not take responsibility. It always blows me away. So many of them do not take responsibility for knowing the coordinate points, knowing the geometry of the building. It does not have to do with them not having the technological capabilities to do it. They do not have the mindset to do it that way because it takes on risk. They want to put the risk somewhere else, with the architects, with the subcontractors. They do not take on any risk.
An additional reason for the limited direct communication across organisational
boundaries during 2D construction projects has to do with that, for the most part, the
structural simplicity of the building allows for it to be reliably captured in 2D models.
79 Therefore, the exchange of models across organisational boundaries often is sufficient for keeping the construction process going without having to supplement the model-based information exchange with face to face coordination meeting. A project manager at
Zahner gave this example:
We are involved in a lot of 2D projects. For instance, a high-end luxury residential building in New York City. It is a simple job. The structure itself is flat and simple. Measurements are in perpendicular lines. Everything we fabricated was actually done from the 2D shop drawings. The way our stuff relates to what other subcontractors do, it is all 2D. All the drawings and information we received from the architect and the general contractor were in a 2D format. With very simple geometries, everything can be easily described in 2D and there is no real need for any intensive collaboration beyond that.
Figure 4 below illustrates the established pattern of relationships among different organisations in 2D-based projects. It reflects Zahner’s position in this network of relationships, the limited scope of interactions and communication between Zahner and other organisations, and the extensive use that is made of 2D models to communicate and exchange information across organisational boundaries. The arrows in the figure denote the transfer of information across organisations through 2D models. The shade of the arrows and rectangles represents the level of intensity of interactions between Zahner and its counterparts. Darker shades indicate tighter collaboration and exchange of information between Zahner and the involved organisation whereas lighter shades of arrows and rectangles signify that Zahner has less interaction with and access to the processes and information of the involved organisation. For example, the dark shade of the structural steel subcontractor’s rectangle indicates that the level of this organisation’s interactions with Zahner is higher than that between Zahner and the electrical subcontractor whose rectangle’s shade is lighter. Thus, the figure demonstrates that Zahner mainly
80 collaborates and directly exchanges information with the general contractor and to a lesser degree with the structural steel, drywall, glazing and concrete subcontractors. The light rectangles representing the architect, and the electrical, plumbing, and carpentry subcontractors, as well as the light shade of arrows connecting to them, signify that
Zahner has almost no direct interactions with these organisations during the construction project.
Figure 4. A process model of relationships in 2D-based construction projects
In line with my previous theorising, the interactions in which Zahner is involved provide the context for the enactment of its organisational identity. In 2D-based construction projects these interactions are characterised by limited connectivity and exchange of information among the involved parties. For the reasons detailed above, each organisation makes an explicit effort to operate within its legal bounds as specified in the contractual arrangements, thereby contributing to an inter-organisational environment in
81 which most subcontractors have very little knowledge of and visibility into the perspectives and work practices of other organisations that participate in the construction
process. Within this system of relationships, Zahner’s capability to holistically grasp the
construction process as well as the design of the building being constructed is minimal.
Accordingly, Zahner’s organisational identity resembles that of a detached worker. A
project manager at the company provided this allegory:
This type of relationships between us, other subcontractors, the general contractor, and the architect, creates an environment where I am working in my little detached world and all I want from you is the four or five single points that I can relate to. Basically all you do is handing concepts and ideas on paper to someone else and saying here it is. You do not really care what they do with it later just as they do not really care what we do with the models they give us. Imagine reading a chapter in a book without knowing the whole narrative – we have no knowledge of what came before it or what comes after it.
5.2.2 The Fixer – Zahner in the construction of the Hunter Museum of American art
In 2003 Zahner became involved in the construction of the Hunter museum of American
Art in Chattanooga, Tennessee. This was one of the first complex, large-scale,
construction projects in which Zahner took part, and during which the company actively
used 3D digital modelling technologies. The Hunter Museum of American Art was
designed by the architect Randall Stout. Before opening his own architecture office, Stout
had worked for Gehry Partners and is considered among many to be one of Gehry
protégés. Similar to Gehry, Stout’s designs often challenge conventional architectural
aesthetics and his buildings are known for their dynamic forms and unique geometrical
shapes. Like Gehry, Stout also makes use of 3D modelling tools in his design process, as
was the case in the construction of the Hunter museum.
82
Image 7. The Hunter museum of American Art in Chattanooga, Tennessee.
In many ways, the inter-organisational dynamics during the construction of the Hunter
museum were uncharacteristic. The previously established system of relationships associated with the use of 2D models and characterised by limited inter-organisational communication and interaction was thrown off balance. The major coordinating task usually assumed by the general contractor was significantly changed during the construction process, as were Zahner’s infrastructural role and practices.
The Hunter museum of American art is a highly complex building with very complicated
geometries and warped interior and exterior surfaces (see image 7 above). Stout designed
the building using a digital 3D modelling technology named Rhino, although the official
contract documents that he distributed to the general contractor and subcontractors were
in 2D. This created a significant challenge for the construction crew because the building
design was so complex to the point that it became impossible to reliably and
comprehensively model it in a 2D space. In fact, many of the 2D models that were
provided to the construction crew by the architect were not accurate enough to be used
for construction, as was acknowledged by a project manager at Zahner:
83 We didn’t think 2D was going to work. We anticipated many problems. You have to get the steel very accurate. We talked to the general contractor and said look, you need to set points, establish some points for the steel, and that will be a work point. That never happened, and in some cases the steel was six inches out.
A project engineer at Zahner commented on the same matter:
Whenever we extrapolated from these 2D drawings we found that there were some areas that were not accurate. Because in 2D you don’t really see what happens as you’re passing through space. We found some steel on the interior actually interfering with some of the interior walls.
In addition to the official 2D models, Stout also shared a 3D computerised model of the
building with the construction crew. However, this model was not fully formed and did not contain detailed information about the surfaces, connections, and materials embedded in the building. Furthermore, the 3D model was not made part of the contract documents package and was not legally binding. Therefore it did not receive the general contractor’s full attention and was not regularly used by it in the construction process. A project manager at Zahner commented:
Early in the design phase it became apparent that we were not going to get from the architects the master 3D model which defines all the surfaces, intersections, and the fine details. We got the framework, the beginning of that, but it was not in detail. The 3D model never really got updated. The information we received would be displayed in a 2D format or a written format.
Thus, a major difficulty emerged early on in the construction process. On the one hand,
the building design involved highly complex geometries that could not be accurately
captured in a 2D space or reliably conveyed in 2D models. On the other hand, the 3D
Rhino model that was provided by the architect was not fully formed and did not contain sufficient detail to support construction. Furthermore, the majority of subcontractors and the general contractor did not have sufficient 3D capabilities to understand, incorporate,
84 or interact with the 3D model that was provided to them by the architect, and did not
possess the required expertise to effectively work in a 3D environment. For example, the
general contractor, a large firm by the name of EMJ Corporation, had very limited
previous experience with using 3D technologies and, according to Zahner’s CEO, had
never been involved in a construction project that was as complex as the Hunter museum.
Given the central coordinating role that the general contractor is supposed to play during
the construction process, this was a sign of what was to come.
Among all the participants in the construction of the Hunter museum, Zahner had the
most advanced 3D capabilities. Initially being exposed to 3D technologies during the late
1990’s, by 2003 Zahner had accumulated some experience in using 3D modelling tools to
design and fabricate metal parts for structurally complex buildings. Therefore, when it
was given the 3D model from Stout, the company incorporated the measurements from it
into its own 3D Rhino model that it had created to design its parts for the building. A
project manager at Zahner explained:
Because the 2D models were not accurate enough, what we ended up doing was developing a 3D model. We had to take the architect’s model, rationalise the geometry, because maybe some of it isn’t constructible because maybe two pieces clash with each other. We would smooth that out, talk to the architect about it, make sure the design concept is there, and build.
The 3D model Zahner had created was frequently updated and entailed relatively detailed information about the various aspects of the building. Thus, an unusual situation was created in which Zahner, rather than the general contractor or the architect, had the most accurate and detailed model of the building that could actually be used for construction.
The rest of the subcontractors only had the 2D models to rely on whereas these models only provided a fragmented and inaccurate vision of the building design. As a result,
85 different subcontractors frequently came to Zahner asking for help with their construction
and measurements. A lead designer at Zahner explained:
What happened on Hunter was that we were the only subcontractor with 3D capabilities and we had a 3D model and therefore everyone came to us whenever they needed dimensions; we were a central source of information for them, above and beyond of what our scope was on that job.
This posed a challenge for Zahner as well as for the different subcontractors because
most of them had very limited knowledge of 3D technologies and little experience, if any,
in using them. A project manager at the company provided the following example:
Now you have to take a dry wall subcontractor and he is going to have to build this structure and he doesn’t work in 3D. He has to try to combine his structures to the exterior steel so it joins together. They don’t have 3D capabilities and his framing can be a few inches off and he has to figure that out. What happens at that point is that they come to us and to help them figure it out.
Importantly, not only the subcontractors lacked sufficient 3D capabilities to
independently develop, use, and manipulate their own 3D models, so did the general
contractor. As a result, the general contractor often came to Zahner asking for assistance with their coordination tasks. A lead designer at Zahner gave this example:
The general contractor didn’t have a 3D capability and therefore had no way of reading our models. They had to trust us that our models were going to work. For instance, when the steel fabricator couldn’t use their 2D information, what ended up happening was that we were forced to make the steel fabricator go out and actually survey the steel to determine if their system was going to work with the concrete wall. The surveying was actually done by another company that was hired by the general contractor. But we, not the general contractor, provided them with what we wanted surveyed, particular points located in a 3D space that came from our model.
86 The fact that they had to step in and manage a process that was contractually assigned to
the general contractor left many people at Zahner frustrated. A project manager from the
company elaborated on this point:
On Hunter, the general contractor only existed in 2D. And yet the geometry was very complicated. You can see how everybody went to the only player who had 3D capabilities. To me, that’s to the fault of the general contractor. How on earth could you get involved in such a project and not realise early on, hey we have to get up to speed. And they were asked to. They were told, or at least suggested, what software and hardware to buy, who you can talk to to get it operated. It didn’t happen and as a result we got into a position which we were not contractually obligated to be in.
Zahner’s CEO commented on the same issue:
With Hunter it all goes back to the general contractor. If the general contractor who is supposed to be in control of constructing the building doesn’t have a base knowledge of 3D technologies, that’s when people have no choice but to migrate to who ever has the information, which was us.
Zahner’s 3D capabilities and the general contractor and subcontractors’ lack of such
capabilities had drastically changed the nature of the previously established inter-
organisational dynamics during the construction process, and Zahner’s position within it.
Instead of mainly handling the design, fabrication, and laying out of its own metal parts
and systems, the company had to attend to the tasks of other organisations and serve as a
major provider of information for the rest of the construction crew. A project manager
illustrated this point:
We normally wouldn’t figure out where the drywall goes or where the glass interfaces with something. It got to a point where we told people what the height of the doors was. We had all the information because we were the only ones working with the 3D model. There were cases where they had walls with very unique angles and the general contractor told the drywall guy if this isn’t up by Friday I’m kicking you off the job. So what does he do? He only knows it starts here and ends there and that it’s curved but he’ll go to Zahner to get the detailed information. So my foreman has to be out there to
87 take measurements off the interface and say this isn’t right and so on. Everything, the most minute piece of information, is filtered through us and our 3D model.
Figure 5 below illustrates the pattern of relationships among the different parties involved
in the construction of the Hunter museum. It reflects Zahner’s central position in this
network of relationships, the extensive use that was made of the 3D model that it
produced, and the model’s importance in enabling the different subcontractors’
construction efforts. The figure also highlights the general contractor’s unusually
marginal position relative to the rest of the construction crew. The single-headed arrows
coming out of the model and pointing to the subcontractors and general contractor signify
that these organisations were, for the most part, passive users of the model and did not contribute to its generation or updating.
Figure 5. A process model of relationships during the construction of the Hunter Museum
As Zahner’s practices and interactions changed, so did its identity. Zahner’s detached
worker identity drew on the company’s position within an infrastructural inter-
88 organisational system of relationships that was associated with the use of 2D modelling
technologies. Within that system Zahner was mainly responsible for designing,
fabricating, and laying out its metal parts while maintaining a minimal level of
collaboration with its counterparts. During the construction of the Hunter museum,
Zahner’s inter-organisational relations significantly changed as the company served as a
central coordinating entity and a major source of information for the rest of the
construction crew. Due to the general contractor and subcontractors’ lack of 3D
capability, Zahner had to step up and assume responsibilities that went above and beyond
those which the company was contractually obligated to. A project engineer at the
company offered this example:
We were the default contact for the general contractor whenever there was a problem that needed to be solved. In a way we were doing their job. We had to report that to the general contractor and it got to a point that I would be modelling specific things in 3D just so I could show them the discrepancies, which is not the role we wanted to have or were contractually obligated to play. But when we see mistakes we’ll call attention to them.
As a result of these new relationships, the company was now characterised as a fixer
rather than a detached worker, as was illustrated by one of its project managers:
In a normal situation, the general contractor oversees all the bits and pieces. You have a structural steel subcontractor, a glass guy, a dry wall guy, there’s us who do the exterior cladding. The general contractor is in the middle coordinating everything. The subcontractors are there and somewhat interact and do their work. On Hunter, with the general contractor not having 3D capabilities, and other subcontractors not having 3D capabilities, you could see how imbalanced the system was. Everything was tilting towards getting the information from Zahner, instead of being centralised around the general contractor. We had to provide the drywall guy all the information from our 3D model, and the same goes for the glazing people, the concrete and the structural steel. The burden was on us to make sure everything was running smoothly. We were there as an all-around fixer, taking care of everybody else’s problems.
89 5.2.3 The seamless collaborator: Using 3D modelling technologies with Randall Stout Architects
As noted above, Zahner had started using 3D modelling technologies in the end of the
1990’s when it collaborated with Frank Gehry during the construction of the EMP
building. Since then, the company had gradually incorporated 3D tools into its shop and
its design process as was evident in the construction of the Hunter museum. In 2006, after
wrapping up its work on the Hunter Museum, Zahner became involved in the
construction of the Alberta Art Gallery (hereafter, AGA) in Edmonton, Canada.
Image 8. The Alberta Art Gallery in Edmonton, Canada.
Similar to the Hunter museum, this building was also designed by Randall Stout who was
the lead architect on this project. However, different from the Hunter museum, on the
AGA project 3D technologies were comprehensively used by all the major participants
from the initial pre-fabrication design phase, through the bidding phase, and all the way
to the construction phase.
90 During the construction process of AGA, Zahner’s interactions with the architects, the
general contractor, and the subcontractor were significantly different from the ones that
characterised the 2D-based projects in which the company had previously been involved,
as well as from the ones in which some use of 3D technologies was made, such as the
Hunter museum. Compared with the company’s limited communication and exchange of information with its counterparts that were associated with the use of 2D modelling technologies, Zahner’s scope of interactions with other organisations during the construction of the AGA was considerably broader. These interactions also differed from those that characterised Zahner’s involvement in the Hunter museum job; instead of serving as a focal point for inter-organisational coordination and information sharing, attending to other organisations’ tasks and problems and thereby exceeding its contractual responsibilities, Zahner collaborated with other organisations in a more balanced manner and focused on its own tasks as a metal fabricator.
In many respects, the construction of the Hunter museum was a learning experience for
Zahner that offered two important and interrelated lessons. The first is that serious
difficulties emerged out of the discrepancies between the complex design of the museum
and the limited capacity of most subcontractors to effectively work in a 3D environment.
The second is that the fact that most subcontractors and general contractor worked in a
2D environment whereas Zahner worked in a 3D environment had caused serious
coordination and communication problems among the involved organisations. These
understandings had caused Zahner to be more assertive in having a say in the selection of
the subcontractors for the AGA project. A lead designer at Zahner explained:
91 Based on the Hunter experience we tried to emphasise to the architects and general contractor in the AGA project that it is very important for all the main participants to have 3D capabilities. This includes the general contractor, glazing, structural steel. When you are dealing with complex structures that are indescribable in 2D, this is absolutely essential.
A project manager at Zahner added:
Part of the requirements for bidding on the AGA job that we were pushing for involved subcontractors having the ability to interpret, understand, and participate in the 3D digital definition part of this process. They had to have the ability to look at the model, talk about it, modify it, and query it. We wanted it to be part of the culture of this project.
Accordingly, all the main players that participated in the project had had significant experience in operating in 3D environments. For example, the structural engineer, a company named De-Simone, is a “3D veteran” that had participated, among other 3D- based projects, in the construction of the PBL building at Case Western Reserve
University, a highly complex building that was designed by Frank Gehry; and the associate architects, HIP Architects, is a 3D-savvy firm that specialises in the use of
Revit, an innovative 3D design tool.
Compared to the Hunter project, Stout’s contractual approach on the AGA project was
more explicitly 3D aware. Similar to the Hunter museum, the AGA’s complex design
was modelled using the 3D platform, Rhino. However, unlike his approach on Hunter,
and deviating from the established practice in the AEC industry, Stout incorporated the
3D master model of the building into the contract documents package thereby affording it
formal status. A project manager at Zahner commented:
3D models were part of the contract documents on this project. Stout pushed for that because these structures are so complicated you simply cannot say I’ll give you 2D details for every possible condition out there. And some subcontractor is not going to be
92 able to see the whole picture. You have got to have the ability to manipulate the model and see things three-dimensionally.
This time, unlike the Hunter job, the 3D Rhino model went beyond crudely specifying the surfaces of the building and was in fact a fully formed and detailed representation of the building. In addition, Stout insisted that all the major participants in the construction process have significant 3D capabilities and use the 3D models to support their respective construction activities and communicate with each other. A project manager at Zahner made this observation:
Around the time of the Hunter museum job the understanding wasn’t necessarily there that you would have to have as many players as possible that are 3D digitally manipulating, reviewing, and investigating information as it is now. On the AGA job it was a heavy requirement for Stout in deciding who was going to be involved in the project.
A project engineer at the company added:
Stout has gone from designing very complex structures and having very limited knowledge of 3D and the practical implications of using 3D for the construction process, to still doing very complex structures but with fuller understanding of 3D technologies, and the importance of having the main players in the construction project with that capability.
Although Stout used Rhino to design his 3D model of the building, other participants in
the project used different types of 3D modelling platforms to design their own parts and
systems for the building. For example, the structural engineer, De-Simone, used CATIA while Zahner used a 3D modelling tool called Pro-Engineer to design its metal panels.
Due to the multiplicity of technological systems, their inter-operability became an
important issue during the construction process. To address this issue, another 3D
platform, Revit, was used to convert models between the different tools involved in the
93 project. Since Revit exports file in a universal format, any file that needed to be viewed in
a third party program (e.g., CATIA or Rhino) was first imported into Revit which then
exported it in a format that could be read by any other program. This way, all the
stakeholders could read information from and feed information into the Rhino model that
was created by Stout.
The nature of 3D models, along with the fact the all major subcontractors as well as the general contractor and the architect used and incorporated the models in their practice, had enabled the creation of a collaborative inter-organisational environment. Within this environment, the participants in the construction process were able to achieve higher levels of information sharing and coordination than was previously possible when 2D technologies were used or when only Zahner had 3D capabilities. Zahner’s CEO noted:
When 3D tools were used by all the major subcontractors on the AGA job, the hierarchical organisational system that emerged with the use of 2D technologies and that everybody was so used to completely changed. It was not a hand off process anymore where everyone was looking to avoid taking risks. Instead, we were able to collaborate and exchange information much more efficiently.
The 3D models that were used by Stout during the AGA project are extremely rich representations of the building design that hold multiple layers of information concerning
the different aspects of the building. Thus, the same model can be used by multiple
organisations thereby placing them within the same 3D computerised environment and
reducing barriers among them. Another feature of the models which helped facilitate
inter-organisational collaboration is that they are transparent and interactive such that
changes that are made in one segment of the model are immediately visible to other users
of the model, and their impacts on other aspects of the model are checked automatically.
94 An additional characteristic of the 3D models is that their users share and operate within the same x,y,z coordinate system. In construction projects in which 2D technologies are used, different organisations typically employ their own surveyors and develop separate measurements for the installation of their parts. During the construction of the Hunter museum, different subcontractors came to Zahner asking for help with their measurements when their surveying information was not sufficient to support their construction activities. However, they still maintained their separate x,y,z, coordinate systems (or rather, x,y coordinate systems). During the construction of the AGA, the use of 3D models by all the major subcontractors and the general contractor placed all of them within the same 3D x,y,z space, and all their measurements were calculated relative to a shared absolute 0,0,0 point which they shared.
Sharing the same 3D technological platform and x,y,z coordinate system with other subcontractors has considerably improved Zahner’s ability to coordinate its construction process with other subcontractors as was illustrated by a project engineer at Zahner:
For example, the structural steel people also used our models. We connect directly to structural steel and there is a very important interface between us. So we would create boundaries for our panels and then coordinate that information with the architect and the structural steel contractor. And the steel contractor would sometimes produce elements that tie to our panels by relying on the shared dimensions, the shared x,y,z data. This really helped a lot in coordinating the process with them.
The use of 3D models by all the major subcontractors and the general contractor also enabled the different parties to clearly visualise and surface common issues and problems during the construction process, thereby facilitating their joint resolution. A project engineer at Zahner provided this example:
95 AGA is probably 4 times as complicated as Hunter but the process was so much smoother and more balanced. For example, we had a meeting to discuss the sprinklers in the building for the fire alarm system. They brought up a 3D model, the master model, and they had the sprinklers in the model in a very complex geometric shape. They couldn’t do that if they didn’t have this 3D capability and they couldn’t put a simple concept as a sprinkler system into this complex geometry. The conversation went on for about an hour and a ton was resolved. Having everyone sharing the 3D model is a huge step.
Compared to the company’s situation on the Hunter job, Zahner was in a more balanced
position within the inter-organisational system composed of the subcontractors, the
general contractor and the architect. In this system, Zahner could focus on its role as a
metal fabricator instead of having to provide information to support the construction of
other subcontractors. In large part this was because the general contractor on the AGA
project had substantial 3D capabilities and was able to communicate with the rest of the
construction crew and coordinate their activities through the 3D platform. A project
manager at Zahner explained:
On AGA the general contractor was at the core of the job feeding information out and coordinating, pulling information in and getting different trades to interrelate with one another through the 3D model. It all started with the general contractor. We did what we had to do to complete our work but it all tied back to the general contractor and to the architect who had full 3D capabilities.
Figure 6 below illustrates the pattern of inter-organisational relations among the
participants in the AGA project. The figure illustrates the importance of 3D models in
organising and coordinating construction activities, and in facilitating inter-organisational collaboration, problem solving, and sharing of information. It also highlights the general contractor’s central position relative to the rest of the construction crew, and its
importance in the management of the construction process. Further, it demonstrates that
96 Zahner was back playing its role as a subcontractor interacting with other stakeholder
through the use of 3D models.
Figure 6. A process model of relationships during the AGA project.
As the inter-organisational practices in which Zahner was involved changed, so did its identity. The 3D computerised environment and the x,y,z coordinate system that the subcontractors and general contractor shared increased Zahner’s exposure to the processes and activities of other subcontractors that it collaborated with, and made its interactions with them, as well as its own construction activities more efficient. A project
manager at the company provided this explanation:
3D models tell a story. They tell you how different parts connect. It makes it pretty clear to the glass, and steel and concrete guys, what am I doing, where am I going with this. What happens when we all come together. You get exposed to the whole story instead of just reading one chapter without knowing what comes before it and what comes after it. 3D technology has made it easier and quicker to get to that point.
97 As a result of these new relationships, the company was now characterised as a seamless collaborator. Zahner’s increased ability to communicate, collaborate, and share information with other subcontractors stimulated the following allegory from a project manager at the company:
On AGA, cross-organisational boundaries, as they existed in 2D projects, almost melted away. Because we were all on the same 3D platform we had a much better picture of what other subcontractors were doing and could better collaborate with them. You could say that we were a seamless collaborator and that was possible to a large part because of the connectivity and visibility that you get from the 3D technologies.
98 6. Discussion
The case studies presented above display an interesting pattern of evolvement in the
identities of each of the two organisations as the modelling technologies that they use and
the associated organisational practices that they rely on to accomplish their construction
and coordination tasks change. As can be seen from table 2 below, the established
information infrastructures and identities of both organisations, which were associated
with the use of 2D modelling technologies, undergo transformations when 3D
technologies are used in their construction projects. In each of the six cases, the use of a
different type of IT-based boundary object together with the distinct organisational
practices and pattern of inter-organisational communications through which the use of the
boundary objects unfolds, are associated with unique organisational symbolic
representations:
Table 2. A summary of findings for Hoffman
2D Based Project EMP Seattle Library
Main Boundary 2D AutoCAD models and paper 3D CATIA models 3D CATIA models and 2D Object drawings drawings
Example We do a lot of interdisciplinary In 2D, drawings aren’t put together We didn’t submit electronic coordination. We have these things in one framework whereas in 3D submittals to the design team called shop drawings. Those are it’s all on one visual platform so for review. Everything got used primarily by the mechanical, you can see how things combine turned back into paper as far electrical, sometimes structural together. In a 3D environment you as the administrative hand off engineers and the general can see all the conflicts in the from one entity to another contractors. What happens is that model upfront. In 2D you have to went. The paper was the bible everybody wants to take a straight do all the layouts first and check that we used but we used shot from point A to point B of what each model separately and these [3D] models to generate they need to put in, but we need to manually, and then, at the end, the paper. work together to come up with a come to this conclusion of, well scheme that makes the most sense these things don't match up, they for everyone. We end up doing this don’t line up. Putting together a 3D in a more organised way. Everyone model is a good way of having involved brings their drawings and visibility into discrepancy issues, we roll up our sleeves and work constructability issues, and better through the nuances… and then understanding of how different people leave with that collective models come together. information and go revise their own
99 drawings.
Information Directing traffic, managing and Enhanced involvement, stepping Supplying information, Infrastructure / coordinating construction out of established organisational enhancing collaboration Changes in activities bounds. Information Infrastructure12 Example For example, if the electrical Gehry's people didn’t have the The amount of surveying contractor has a 4 inch conduit that surfaces of the building trimmed to work we did on the library he wants to route down a corridor or fit the sidewalk in the CATIA project was extraordinary. above a ceiling and a duct work model. Some of our first metal Typically the subcontractors contractor has a 24 inch duct that he panels we had were hitting the rely on us to some extent for also wants to route down a ceiling ground. And we said, what’s going this information, but also have and both of them are bringing their on here? Didn’t you have those their own people who do contract documents and showing trimmed? And they said, no, we had surveying. Here we were the how the engineers have told them to extended the surfaces too long and only ones feeding this do that, we have to direct traffic and we figured you’d trim them off. We information to them because come up with a hierarchy of who took the CATIA model and, with without it they would virtually goes first and who goes where. the metal fabricator, set the lines… be lost. gave them elevations and made cut lines around the building where the On EMP, all the models came sidewalks are. from Gehry, he was guiding everything, the structural The idea that we added information engineers, everyone was to the [3D] model is different than onboard there. On the library what a true general contractor does, project, the creation of the 3D because a true contractor doesn't models did not involve the touch any of the product. They just architect. It was initiated by us want to shuffle the cards and deal and pushed by some of the them out and dictate the sequence. subcontractors that But we rolled up our sleeves and collaborated with us. got involved because there was a lot of information that we wanted to make sure was done right. We are more involved in this process than we might be other wise. Organisational Poker Dealer Dispersed collaborator Air-traffic controller Identity
Example Basically we're more like dealers. You could say that on EMP we We become more like air- Everybody is sitting around a table became more of a dispersed traffic controllers or and we're playing poker. We are the collaborator. Although in essence knowledge brokers. We are dealer. We are making the rules, our responsibilities did not change, still dealing with traffic, but we're controlling the flow of the our people were now dispersed we're using this radar tied to a cards. across several organisations. Also, computer and a radio to be whereas before we were calling the more effective. We are putting shots, on EMP it was more of a radios in all the cars and being collaborative effort which included able to call the guy coming Hoffman and the subs. down the street and say hey you need to change lanes before you get to the centre
12 The practices that characterise Hoffman’s involvement in 2D-based projects (e.g., directing traffic) represent the company established information infrastructure. The company’s practices during its involvement in the EMP and Library projects represent changes in its information infrastructure.
100 section.
Table 3. A summary of findings for Zahner
2D Based Project Hunter Museum AGA
Main Boundary 2D CAD models and paper 2D CAD models, 3D CAD 3D CATIA models and 2D Object drawings models drawings
Example The 2D drawings are mainly done Because the 2D models were not For example, we had a meeting to communicate with other accurate enough, what we ended to discuss the sprinklers in the subcontractors. We typically up doing was developing a 3D building for the fire alarm interact with the drywall systems, model. We had to take the system. They brought up a 3D glazing, and interior steel. We need architect’s model, rationalise the model, the master model, and to make sure the building metal geometry, because maybe some of they had the sprinklers in the envelope stays watertight so we it isn’t constructible because model in a very complex need to make sure there is maybe two pieces clash with each geometric shape. They couldn’t continuity between our parts and other. We would smooth that out, do that if they didn’t have this these trades’ parts. talk to the architect about it, make 3D capability and they couldn’t sure the design concept is there, put a simple concept as a and build. sprinkler system into this complex geometry. The conversation went on for about an hour and a ton was resolved. Having everyone sharing the 3D model is a huge step. Information Designing, manufacturing, and Exceeding contractual Effectively collaborating with Infrastructure / installing parts. Limited obligations: assisting and other organisation in a more Changes in cooperation with other sharing information with other balanced way Information subcontractors subcontractors and the general Infrastructure13 contractor Example We are finding that everyone wants We normally wouldn’t figure out For example, the structural steel a structured process that follows an where the drywall goes or where people also used our models. We organisational chart. This is a the glass interfaces with connect directly to structural system that is based on an ongoing something. It got to a point where steel and there is a very hand-off process where each we told people what the height of important interface between us. subcontractor passes their models the doors was. We had all the So we would create boundaries on to the subcontractors that they information because we were the for our panels and then interface with and no one wants to only ones working with the 3D coordinate that information with breach their boundaries. They all model. There were cases where the architect and the structural want a hierarchical lay out: I am they had walls with very unique steel contractor. And the steel here, the general contractor is here, angles and the general contractor contractor would sometimes here are all the different told the drywall guy if this isn’t up produce elements that tie to our subcontractors, they all have their by Friday I’m kicking you off the panels by relying on the shared areas of responsibilities and do not job. So what does he do? He only dimensions, the shared x,y,z data. blur them. knows it starts here and ends there This really helped a lot in and that it’s curved but he’ll go to coordinating the process with Zahner to get the detailed them. information. So my foreman has to
13 The practices that characterise Zahner’s involvement in 2D-based projects (e.g., limited cooperation with other subcontractors) represent the company established information infrastructure. The company’s practices during its involvement in the construction of the Hunter museum and AGA represent changes in its information infrastructure.
101 be out there to take measurements off the interface and say this isn’t right and so on. Everything, the most minute piece of information, is filtered through us and our 3D model. Organisational Detached worker The fixer The seamless collaborator Identity
Example This type of relationships between On Hunter, with the general On AGA, cross-organisational us, other subcontractors, the contractor not having 3D boundaries, as they existed in 2D general contractor, and the capabilities, and other projects, almost melted away. architect, creates an environment subcontractors not having 3D Because we were all on the same where I am working in my little capabilities, you could see how 3D platform we had a much detached world and all I want from imbalanced the system was. better picture of what other you is the four or five single points Everything was tilting towards subcontractors were doing and that I can relate to. Basically all getting the information from could better collaborate with you do is handing concepts and Zahner, instead of being them. You could say that we ideas on paper to someone else and centralised around the general were a seamless collaborator and saying here it is. You do not really contractor. We had to provide the that was possible to a large part care what they do with it later just drywall guy all the information because of the connectivity and as they do not really care what we from our 3D model, and the same visibility that you get from the do with the models they give us. goes for the glazing people, the 3D technologies. Imagine reading a chapter in a concrete and the structural steel. book without knowing the whole The burden was on us to make narrative – we have no knowledge sure everything was running of what came before it or what smoothly. We were there as an all- comes after it. around fixer, taking care of everybody else’s problems.
Prior to working with Gehry Partners, Hoffman’s organisational identity reflected a
traditional general contracting company. This identity was essentially defined by the
company’s use of 2D CAD and paper models as boundary objects, and its engagement in
associated infrastructural industry practices. The use of 2D boundary objects in
conjunction with carrying out infrastructural boundary practices (i.e., practices that are
carried in relation to, or in common with, practices that take place in bordering
organisations) such as sequentially managing communication channels between
subcontractors and architects and coordinating the subcontractors’ work, manifested in an
inter-organisational system whose nature resembled that of a poker game. This system of
relationships constituted the interface between Hoffman and its adjacent organisations
102 and provided the context for the ongoing enactment of the company’s identity of a poker dealer.
However, Hoffman’s poker dealer identity began to change when the company started interacting with Gehry Partners during work on EMP. The use of 3D tools as boundary objects that was imposed by Gehry Partners and the change in Hoffman’s infrastructural boundary practices associated with the use of this technology, reconfigured the interface between Hoffman and its adjacent organisations, and changed the conditions for the enactment of the company’s organisational identity. Hoffman was no longer able to draw on 2D boundary objects to form its relationships (i.e., its boundary practices) with architects and subcontractors in a way that would reaffirm its poker dealer identity. The previously established linear-sequential construction process was reshaped into a tighter collaborative system. In this system, traditional organisational distinctions shifted, organisational boundaries blurred and, consequently, existing organisational identities were renegotiated. Rather than having a distinct, and somewhat removed position in the construction process, as was the case in a 2D-based process, Hoffman now played a more involved role. This was manifested in the company having its own CATIA operators embedded in the subcontractor’s offices, in it taking a hands-on approach with the 3D models that were produced by Gehry Partners, and in it intensively interacting with its subcontractors. These newly formed practices provided the context for the enactment of
Hoffman’s dispersed collaborator identity.
Similar to EMP, 3D tools were also used on the Seattle Library project, however, in a different way. Unlike Gehry Partners, Rem Koolhaas was not willing to share his 3D models with Hoffman and the subcontractors. As a result, boundary practices common to
103 the architects, Hoffman, and the subcontractors changed again, as did the conditions for
the enactment of Hoffman’s organisational identity. Under these new conditions, 3D
tools were no longer used as boundary objects between the architect and the rest of the
construction team - paper documents were used instead. However, due to the structural
complexity of the building, Hoffman initiated the creation of its own 3D models. Thus,
3D tools were used as a boundary object between Hoffman and the subcontractors, and
among the subcontractors. But whereas on EMP the 3D models were generated by the architects, here the models were developed by Hoffman in collaboration with the subcontractors in an effort to create a shared platform to enhance their collaboration. The company now played a more proactive role which was expressed both in them providing surveying information to the subcontractors, and in creating and managing the shared
AutoCAD platform. Within this newly formed pattern of relationships, Hoffman’s
organisational identity was reshaped yet again. Rather then being a dispersed collaborator
across organisations, Hoffman conceived of itself as a central coordinator and supplier of
knowledge.
Similar to Hoffman, before starting to use 3D modelling technologies in its practice,
Zahner’s identity reflected a traditional subcontractor in the AEC industry. This identity was based on the infrastructural boundary practices in which Zahner was involved during its construction projects and on the use of conventional 2D modelling technologies as boundary objects. The use of such boundary objects combined with the associated limited sharing of information and scant collaboration among multiple subcontractors during construction projects, gave rise to an inter-organisational environment in which individual subcontractors were only partially aware of the concerns, perspectives,
104 problems, and activities of other subcontractors. Instead, each subcontractor focused
primarily on manufacturing and installing its own parts while interacting with its
counterparts only to the extent necessitated by its contractual obligations and the practical
requirements of the job. This system of restricted inter-organisational coordination constituted the interface between Zahner and its adjacent organisations during 2D-based construction projects and provided the context for the enactment of its identity of a detached worker.
However, as was the case with Hoffman, Zahner’s identity started to change when the
company incorporated 3D technologies into its practice and interactions as a new type of
boundary object. Zahner’s utilisation of 3D modelling tools on the Hunter museum
project and the profound changes in the company’s pre-established boundary practices
associated with the use of this technology, reconfigured its interface with its neighbouring organisations and changed the conditions for the enactment of its organisational identity. During its involvement in the construction of the Hunter museum,
Zahner no longer operated in relative isolation from other subcontractors or centred on carrying out its own responsibilities as a metal fabricator as was the case in 2D-based construction processes. Instead, the company was sucked into a position in which its responsibilities exceeded far beyond its own expectations, past experiences, and contractual obligations. Zahner’s expertise in the use of 3D tools and possession of the only comprehensive 3D model of the building, combined with the lack of 3D technological proficiency in any of the other involved organisations and the structural complexity of the building, had caused other subcontractors and the general contractor to turn to Zahner for help with their tasks. As a result, Zahner had to act as a central
105 coordinator and supplier of information for the rest of the construction crew and assist in
solving other organisations’ difficulties and problems. These newly formed boundary
practices provided the context for the enactment of Zahner’s fixer identity.
Similar to the Hunter museum, 3D tools were also used during the AGA project.
However, the associated inter-organisational pattern of relationships was markedly
different. Most importantly, the architect, Randall Stout, had made an explicit decision
that all the main subcontractors on the job should have significant knowledge of 3D
modelling technologies and experience in working in 3D construction environments.
Furthermore, Stout incorporated his detailed 3D model of the building into the contract
documents package thereby warranting it a formal and legally-binding status. These two
actions combined with the overarching use that was made of 3D tools as boundary
objects during the construction process had enabled the creation of an inter-organisational collaborative environment that was characterised by enhanced coordination, transparency, and sharing of information. In this environment, Zahner’s boundary practices changed again from those that characterised traditional 2D-based projects as
well as from those that characterised the Hunter museum job. Compared to the limited
interaction and visibility that exemplified 2D-based jobs, Zahner’s scope of collaboration
with its counterparts was much broader. This was possible due to the fact that all the
prime subcontractors, in addition to the general contractor and the architects, were
operating within a shared 3D space. Compared to its involvement on the Hunter museum
project, Zahner’s interactions during the AGA project were more balanced; because the
company was not the only professional entity which was 3D savvy, it did not have to
allocate significant resources to assist with the construction tasks of other organisations
106 and was able to focus on its own responsibilities as a metal fabricator. The enhanced
collaboration and balanced interactions constituted Zahner’s boundary practices during
the AGA project and provided the context for the enactment of its identity of a seamless
collaborator.
Building on the case studies of both organisations, and in accordance with previous
research (Boland et al, 2007), a claim could be made that the use of 3D modelling technologies as an integral part of the construction process in the AEC industry is associated with significant changes in the way construction projects are organised and the manner in which different organisations relate to each other and carry out their tasks during construction projects. In the cases of both Hoffman and Zahner, the ongoing and taken for granted use of 2D modelling technologies as a means for supporting cross organisational information flows (i.e., as boundary objects) during the construction process has been associated with a set of infrastructural processes of executing professional tasks and interacting with other organisations (i.e., boundary practices).
Hoffman’s use of 2D CAD models and paper drawings was linked with the practice of directing traffic and coordinating the construction activities of the rest of the construction crew. Zahner’s use of 2D tools was linked with the practice of limited cross- organisational connectivity and cooperation. In both cases, the combination of a specific type of boundary object and certain practices through which the use of the boundary object unfolded facilitated the emergence of distinct organisational identities: Hoffman’s poker dealer and Zahner’s detached worker.
Despite the different positions that they occupy within the AEC’s network of
relationships, their varied technological backgrounds, and the diverse social and political
107 perspectives from which they engage in professional interactions, Hoffman and Zahner
present a similar trend of organisational transformations. Although the local conditions
and contingencies differ to some extent, in examples from both organisations the
introduction of 3D modelling technologies is associated with a break from the traditional
industry practices of loosely-coupled coordination, limited cooperation, and restricted
cross-organisational visibility. The growing use of 3D tools is linked, in the cases of both organisations, with the emergence of an inter-organisational construction environment that is characterised by increased information sharing, transparency, and collaboration.
The identities that the two organisations exhibited in projects where 3D technologies were used reflect this trend. Hoffman’s identity of a poker dealer that characterised its engagement in 2D-based projects gave way to the dispersed collaborator identity during its involvement in the 3D-based EMP project. Similarly, Zahner’s detached worker identity was later reshaped into the seamless collaborator identity when the company participated in the 3D-based AGA project.
However, as the cases above demonstrate, the mere introduction of 3D modelling tools as
boundary objects in construction projects does not necessarily lead to enhanced cross-
organisational collaboration and knowledge sharing. In two of the four cases of actual construction projects (Hoffman’s Seattle library and Zahner’s Hunter museum), 3D tools were used to support some aspects of the design and construction processes. Despite that,
the patterns of inter-organisational relationships in these two examples were not as
collaborative and harmonious as they were in the cases of the EMP building and the
AGA. During the construction of the Seattle library, the architect was mostly
marginalised from the construction process due to his unwillingness to share his 3D
108 information with the rest of the construction crew, prompting Hoffman to take action and
develop its own 3D modelling environment. The construction of Hunter museum, in turn,
was underscored by the incapacity of most subcontractors and the general contractor to
effectively utilise 3D technologies in a design environment that was too complex to be
reliably modelled in 2D technologies. This required Zahner to step in and assume
responsibilities that it was not contractually obligated to take on.
A number of explanations can be offered to make sense of the differences between EMP
and AGA on the one hand, and the Seattle library and the Hunter museum on the other.
First, the pervasiveness of the use of a boundary object by interacting organisations has
real and significant implications for the impact it may have on the nature of their
interactions. Both on EMP and AGA, 3D technologies were comprehensively used
throughout the construction process as this was officially mandated. In both construction projects, 3D computerised models were made part of the contract documents package and
therefore awarded a legally binding status. In an industry such as the AEC where
organisations traditionally tend to avoid unnecessary risks and minimise deviations from
established contractual conventions, the importance of this can hardly be overstated. On
EMP for instance, the formal status of the 3D CATIA models sent ripples through most
of the construction crew and those subcontractors that had not already had the expertise
and know-how to utilise the technology, soon had to catch up. Furthermore, on AGA and,
to some degree, on EMP, a decision was made early on by the architects to only recruit
contractors and subcontractors that are experienced in the use of 3D technologies. For
example, on AGA, substantial knowledge of 3D modelling was a prerequisite during the
bidding process. Therefore, only 3D savvy organisations were hired for the job. On EMP
109 this was only partly the case. Whereas Hoffman was recruited for the job based on its
openness to the idea of comprehensively using 3D (its bid for the job was done in 3D),
most other subcontractors were hired whether or not they had extensive experience in
using 3D tools (the repercussions of this decision were later felt when Hoffman had to hire external CATIA operators to help various subcontractors with their 3D tasks).
Nevertheless, the fact remains that both on AGA and EMP, 3D tools were the main medium (i.e., boundary object) through which multiple organisations communicated and shared knowledge with each other. In comparison, both on the Hunter museum and the
Seattle library, 3D-based boundary objects were not comprehensively used by all the involved actors. In these two projects, 3D tools were not made part of the contract documents package, not awarded a legally binding status, and their use was not officially mandated.
A second and related explanation for the difference between EMP and AGA on the one
hand, and the Seattle library and the Hunter museum on the other, has to do with the use
of a shared technological platform by the involved actors. As described above, the
recurring use of 2D modelling tools as boundary objects in the AEC industry has been
associated with the emergence of an institutionalised pattern of inter-organisational
relationships during construction projects, as well as with the development of distinct
construction methods used by various actors in the industry. The institutionalisation of
these artefacts and practices in the AEC industry is profound and manifested in formal
arrangements (such as contractual forms that reflect and reinforce the abovementioned
inter-organisational loose coupling) as well as in informal organisational attitudes,
identities, and mindsets (such as actors’ reluctance to share information with each other
110 during the construction process). The implications of the introduction of 3D modelling tools as a new type of boundary object are therefore not solely technological. The affordances of 3D modelling tools combined with the hyper complex structures whose design and construction they supported, challenged existing routines and traditions in the
AEC industry. The technology’s capacity to converge multiple subcontractors into a shared 3D space in which cross-organisational transparency and inter-dependence are increased, does not readily coincide with the previously established practice of limited knowledge sharing and loose coupling. Therefore, in the two projects where both 2D and
3D technologies were used (i.e., the Seattle library and the Hunter museum) some difficulties occurred during the construction process. Whereas some actors operated in a
3D environment and were ready to embrace their potential practical implications (i.e., enhanced collaboration, knowledge sharing, etc.), others still existed in a 2D world and subscribed to the guiding principles that dominated it (i.e., limited interactions, restricted sharing of knowledge etc.). For example, Koolhaas’ decision not to share his 3D information with the rest of the construction crew during the construction of the Seattle library combined with the complexity of the building created a modelling gap that had to be filled. This forced Hoffman to step in, assume a major coordinating responsibility, and effectively marginalised Koolhaas and diminished his impact on the construction process.
On the Hunter museum the situation was even more acute. There, the architect’s decision not to fully share his 3D information with the rest of the construction crew despite the complexity of the building, was aggravated by the general contractor and most of the subcontractors’ lack of substantial 3D expertise. Whereas the architect and the general contractor tried to maintain an inter-organisational system of loose coupling, delegation
111 of responsibilities, and limited knowledge sharing, Zahner was the only entity that was
3D savvy and that was willing to engage in more collaborative relationships with its
counterparts. The construction process therefore was rife with conflicts,
misunderstandings, and inefficiencies as mismatches between the 2D and 3D
representational systems and their associated practices repeatedly surfaced.
On EMP and AGA on the other hand, most organisations used 3D tools and adjusted to
the associated practices of enhanced knowledge exchange and communication. On EMP,
this was reflected in the unusually high number of inter-organisational coordination
meetings during which information was shared and operative collective decisions were
made. On AGA, the participating actors made sure that their different 3D platforms were
able to exchange information with each other by adopting another 3D platform, Revit, to convert models between the different tools involved in the project. The readiness of the multiple parties in these two projects to use 3D tools and engage in practices that correspond to these tools’ affordances, contributed to more efficient inter-organisational coordination, exchange of knowledge, and collaboration.
112 7. The Interrelationships of Boundary Objects, Organisational Identities, and
Information Infrastructures
Based on the six case studies and on the discussion that followed them it is now possible to propose a model that delineates the relationships among boundary objects, organisational identities and information infrastructures (see figure 7 below). In outlining the relationships among the three constructs, I address the research questions that I posed at the outset of the dissertation.
Figure 7: Interrelationships of Boundary Objects, Organisational Identities, and Information Infrastructures.
7.1 How do 2D and 3D modelling technologies, as boundary objects, shape or are shaped by organisational practices and patterns of communication?
With regard to boundary objects and organisational practices, the cases demonstrate the tight connection between the objects that are commonly used by multiple organisations
113 and the practices through which these organisations relate to each other. That is, the cases
show that boundary objects are implicated in boundary practices which, in turn, rely on
boundary objects. In other words, organisations interact by engaging in boundary
practices which enable and are enabled by the use of boundary objects. On the one hand
boundary objects are enabled and rendered meaningful when organisations engage in
boundary practices. For example, during Hoffman’s work on EMP, 3D models gained
their significance when they were used in common practice by multiple and diverse
organisations such as Hoffman and Gehry Partners. On the other hand, the flexibility and
ambiguity which characterise boundary objects enable diverse organisations to engage in
shared boundary practices. For example, the richness of the 3D models that were used on
the EMP job and the AGA project and their ability to embed complex and multi-layered
The mutual constitution of boundary objects and practices makes up the interface, which is the overlapping area of at least two information infrastructures14. In each of the case
studies one would be hard pressed to describe the interface shared by the multiple
organisations without taking into account both the boundary practices and the objects that
constitute it. For example, understanding the poker game inter-organisational
environment which characterises traditional construction projects requires more than analysing the features of 2D CAD and paper models, and examining the linearity of inter- organisational practices in separation from each other. It is the interlacing of 2D models
and linear coordination, which makes up the interface that is shared by organisations. On
the one hand, the immutable nature of 2D models and their restricted capacity to hold and
transfer rich and complex information limits the ability of organisations to directly
14 In figure 7 above, the oval containing boundary objects and boundary practices should have been the same area signifying the interface. We kept them graphically separate to prevent from cluttering the model
114 interact with or be aware of the activities of other organisations. This in turn supports
sequential interactions that are mediated by a ‘governing hand’, or rather, a ‘dealer’. On
the other hand, such sequential interactions that are characterised by interdependency,
mutual suspicion, and strategic sharing of information, encourage the use of technologies that limit the degree of inter-organisational visibility and that enable organisations to keep private their proprietary knowledge and processes.
7.2 How do 2D and 3D modelling technologies shape or are shaped by the identities
of the organisations that use them?
With regard to boundary objects and organisational identities, the six cases demonstrate
the importance of the technologies that organisations use to communicate with each other
and the mutual practices through which such communications unfold in the development
of organisational symbolic representations. The cases indicate that the ongoing
constitution of boundary practices and boundary objects provides the context for the
enactment of organisational identities. As explained above, in this dissertation
organisational identities are framed as relational constructs. That is, organisational
identity is based not only on the experiences and social and cultural similarities that
organisational members share, but also on the type of interactions that the organisation is
involved in and on the nature of the stakeholders with which it interacts. However, since I
conceptualise organisational interactions as the overlapping of multiple information
infrastructures that is made up of the mutual constitution of boundary object and
boundary practices, the latter two constructs are hypothesised to play a major role in the
articulation of organisational identities. In each of the case studies, Hoffman and
115 Zahner’s organisational identities are intimately intertwined with the nature of the
companies’ relationships with other organisations, which in turn are associated with their use of particular IT-based boundary objects. For example, Hoffman’s identity of a poker dealer was based, to a large degree, on its boundary practices which involved mediating and coordinating sequential communication and construction processes. However, these practices supported and were supported by Hoffman and other organisations’ use of 2D
CAD and paper models as boundary objects. Thus, it was the mutual constitution of practices and objects that provided the context for the enactment of Hoffman’s organisational identity. Similarly, Zahner’s identity of a detached worker was intertwined with the company’s involvement in infrastructural boundary practices which included partial collaboration and sharing of information with other organisations during the construction process. However, these practices supported and were supported by Zahner and these organisations’ use of 2D CAD and paper models as boundary objects.
Therefore, it was the interlacing of these practices and objects that provided the context for the enactment of Zahner’s organisational identity.
7.3 What are the implications of changes in modelling technologies for the identities,
practices, and interactions of the organisations that share them?
Concerning the consequences of change in modelling technologies, as IT-based boundary
objects, for the identities and practices of the organisations that use them, the cases
illustrate that a change in the technologies that are shared by multiple organisations is likely to be associated with changes in these organisations’ boundary practices, and therefore with modified conditions for enacting their identities. This was evident when
116 Hoffman’s poker dealer identity, which was based on the use of 2D models and
engagement in sequential construction processes, was going through transformations
when the company started working on EMP. The use of a new technological boundary
object in conjunction with the involvement in different boundary practices altered the
interface between Hoffman and its neighbouring organisations and enabled the
company’s dispersed collaborator identity to emerge. This was also evident when
Zahner’s detached worker identity, whose enactment was interrelated with the company’s
use of 2D modelling technologies and engagement in associated boundary practices such
as limited inter-organisational coordination and collaboration, was going through changes
when the company became involved in the construction of the Hunter museum. Zahner’s
utilisation of a different modelling technology as a boundary object to communicate and
exchange information with its counterparts along with its involvement in different
boundary practices reconfigured the interface that it shared with its bordering
organisations and enabled its identity of a fixer to emerge.
Importantly, the six case studies highlight that organisational changes associated with
dynamic boundary objects are not always limited to one organisation and can in fact cross
organisational boundaries. For example, the introduction of 3D technologies by Gehry
Partners and their use as boundary objects instigated a cascade of changes in other organisations that interacted with Gehry Partners (Boland et. al., 2007). The use of 3D tools by Gehry Partners changed not only their internal practices, but also their boundary practices – the way they communicated and cooperated with general contractors, engineers and subcontractors. It should be noted that Gehry Partners’ ability to impose their own boundary objects and dictate their use on the rest of the construction team
117 during the EMP project should not be taken for granted but examined in the context of
the relationships among these organisations. Gehry’s position and reputation as a leading
architect in the AEC industry invested him with significant symbolic power (Bourdieu,
1991), which he exercised over the other participants in the EMP project. Had Gehry enjoyed a less favourable status, or had it been a different, less reputable architect who
was trying to impose the use of a new technology in a way that could significantly change long established practices, this attempt might not have been as successful. Nonetheless,
Gehry’s ability to impose the use of 3D tools on the construction team and change mutual boundary practices, consequently changed the conditions for the ongoing enactment of other organisations’ identities, as was evident in Hoffman’s case whose identity changed when it started interacting with Gehry partners during EMP. Accordingly, the model proposed above suggests that boundary objects are part of a dynamic system whose elements are bound up in reciprocal relationships. When change occurs in one of the elements, it will carry over to the others, generating cascading transformations. Notably, these changes are not always accommodated within a single organisation and can carry over to neighbouring organisations through the use of shared boundary objects and engagement in mutual boundary practices.
A point should be made here with regard to the capacity of IT-based boundary objects to
influence organisational practices and drive organisational change. The claim I wish to
make here should not be misconstrued so as to mean that technological artefacts can
unilaterally determine their use patterns, the extent and nature of inter-organisational
collaborations, and the formation of organisational identities. The trajectory of causality I
wish to draw is more subtle than that. 3D technologies indeed play a major role in the
118 case studies presented above and I do maintain that the observed changes in Hoffman and
Zahner’s organisational practices and identities could not have taken place without the technological transformations that afforded new types of information sharing and
collaborative practices. However, while 3D technologies play a key role in the cases,
technology should be viewed in relation to the practices through which it is enacted. In
other words, technological boundary objects and boundary practices should be recognised
as mutually constitutive elements that can only be jointly understood. While 3D
technologies enabled and promoted the observed organisational changes (rather than
caused them), the nature and scope of these changes cannot be understood by solely
analysing the technological difference between 2D and 3D tools. These changes were
realised through the practices of organisational actors which allowed new identities and
collaboration patterns to emerge. At the same time, it should also be recognised that actors do not have unlimited freedom and choice with regard to their identity formation and use of technology. They operate within the constraints and scripts offered by the technological boundary objects with which they interact. Thus, I hold that technological
boundary objects were a necessary but not sufficient condition for new organisational
practices and identities to emerge.
Finally, a word concerning the generalisability of the proposed model is in order.
Although the theoretical principles that are incorporated in the model were developed in
the context of the technological and organisational changes that took place in the AEC
industry, their applicability extends beyond this immediate circumstance. The
relationships described in the model also apply to other inter-organisational contexts
which exhibit characteristics that are similar to those of the AEC industry. In such
119 contexts, knowledge, resources, and expertise are distributed across multiple
organisations that come from diverse social and technological backgrounds and that
nonetheless have to collaborate to achieve a mutual task, and do so by communicating by
means of some shared artefacts. For example, system development projects typically
include multiple organisational entities that have diverse backgrounds and professional
and technological skills. Nevertheless, these entities have to work together in order to
accomplish a common task and do so by using a variety of artefacts such as entity
relationships diagrams, data flow diagrams, or requirements specification documents.
Applying the model described above to such situations may produce interesting insights
concerning the processes whereby multiple organisations enact their practices and
identities through the use of mutual artefacts and engagement in shared practices during
the course of a system development project.
That said, the main strength of qualitative case studies (and this study is no exception) is
that their findings can be used to apply to and develop different theoretical domains
(Silverman, 2001; Yin, 2003). Therefore, as was pointed out at the outset of the
dissertation, the case studies that served as the basis of this work were chosen based on their capacity to contribute to developing meaningful theoretical inferences about the interrelationships of technological artefacts (conceptualised as boundary objects), and the social processes that take place within and across the organisations (conceptualised in terms of their identities and infrastructural practices) that use them during a process of organisational change. Accordingly, the model I put forward should be generalised beyond the particular characteristics of the organisations and technologies that were
studied here and examined in relation to other theoretical frameworks that discuss the
120 concepts of organisational identity, organisational change, technological change, and boundary objects. I do this in the next two sections.
121 8. Contribution to Existing Literature
None of the elements contained in the theoretical model outlined in figure 7 is new.
Organisational identities, information infrastructures, and boundary objects have each been extensively researched as is documented in the literature review at the beginning of this dissertation. Furthermore, some of the relationships depicted in the model have received some attention, although not in a way that connects all three constructs together.
For example, the relationship between boundary objects, institutionalised organisational practices, and patterns of communication has been studied by Star and Bowker for some time now. In their work, the authors examine the tensions associated with integrating multiple information infrastructures (and the practices, communicative patterns, and classification systems embedded in them) and the significance of boundary objects in such situations (Star & Griesemer, 1989; Star, 1999; Bowker & Star, 1999; Bowker,
2005). However, their work does not explicitly examine how social, group, or organisational identities may be implicated in these processes.
The relationship between boundary objects (in particular, IT-based boundary objects) and
organisational identities has been studied by a number of researchers. For example,
Levina (2005) argued that boundary objects can lead to the development of a collective identity across organisations, and consequently to improved collaboration. However, in
her study organisational identities were not conceptualised as relational. Furthermore,
standardised interactions among organisational groups were not examined, and the issue
of change in boundary object was not considered. Levina and Vaast (2005) focused on
boundary objects and practices and argued that nominated boundary objects only become
boundary-objects-in-use within joint fields of practice that are produced by “boundary-
122 spanners-in-practice”. They did not, however, consider how organisational identities may
be implicated in the processes they described and did not take into account the possibility
of change in boundary objects.
Several researchers have examined the effect of the introduction of new IT artefacts on organisational identities although without explicitly conceptualising such artefacts as boundary objects. For example, Walsham (1998), Walsham and Barrett (1999), and
Lamb and Davidson (2005) described in detail the transformations in the identities of groups associated with the introduction of new IT. These studies, however, did not focus on the boundary practices that are associated with the use of the new IT. In addition, their analysis of identities was not based on a relational interpretation. Furthermore, Lamb and
Davidson (2005) emphasise the capacity of new IT to shape identities by providing new venues for identity representation, such as project websites. Alternatively, in model proposed here I highlight changes in practices that are associated with new boundary objects, which enable new forms of identities to emerge.
In sum, earlier studies have not systematically examined the relationships among
boundary objects, information infrastructures, and organisational identities. Furthermore,
they have not looked at changes in boundary objects and their implications for the
organisations that use them (specifically in terms of their identities and information
infrastructures). Therefore the model presented here supplements these studies by
suggesting that information infrastructures, boundary objects, and organisational
identities constitute a dynamic system whose constituent elements can not be examined in
isolation from each other. Furthermore, it provides a framework for examining the
dynamic relationships among these elements and delineates the consequences of changes
123 in boundary objects for the identities and infrastructures of the organisations that share them.
The juxtaposition of boundary objects, information infrastructures, and organisational identities and their conceptualisation as interrelated elements that constitute a holistic system is of particular importance and relevance to the study of organisational identities.
As mentioned before, existing literature on organisational identity treats the concept as a symbolic representation that is continually negotiated among organisational members to give a sense of meaning to the organisation’s actions, objectives, and existence (Gioia,
1998; Gioia et al, 2000). This representation is recurrently produced as organisational members deal with the local conditions, contingencies, and uncertainties of their everyday lives and environments and as they work to make sense of and narrativise their experiences (Moscovici, 2000). Emphasising the fluidity and flexibility of this representation, literature on organisational identity has expanded the locus of investigation from the internal core features and qualities that characterise an organisation to encompass its environment and external stakeholders and its relationships with them in order to capture the varied processes through which a collective sense of self is formed (Emirbayer, 1997; Hall, 1996; Fiol, 2002; Corley & Gioia, 2004). The provided reasoning for doing so is that a shared organisational identity is fashioned not only based on the experiences and practices that take place within an organisation and that are shared by its members, but also in the context of a larger social ecology of which the organisation is part. Therefore, the nature of this ecology and its composition, as well as the nature of the organisation’s relationships with it are integral to the process of shaping its identity.
124 In that regard, the contribution of my work is twofold. First, the model I propose provides
a clear vocabulary to understand, discuss, and analyse the relationships that take place
between an organisation and its external stakeholders. Building on the case studies, my
work suggests that inter-organisational relationships can be effectively analysed in terms of the practices and objects that constitute them. Using the terms ‘boundary objects’ and
‘boundary practices’, the model put forward above explicitly takes into account both the
social relationships through which inter-organisational associations are forged,
maintained, change, and eventually cease, and the objects that are intertwined with and
mediate these associations.
Existing literature on organisational identity tends to focus on inter-organisational
relationships, or boundary practices, and the way these relationships are entangled with
the process of identity construction in which organisational members engage (Emirbayer,
1997; Hall, 1996). However, less attention is given in the literature to the objects that are
shared by interacting organisations and that mediate their communication. The findings
presented earlier demonstrate that in order to appropriately understand inter-
organisational interactions, boundary objects, boundary practices, and their interlacing
need to be taken into account. Therefore, when setting out to analyse organisational
identity, researchers need to be aware not only of the content and form of the
communication that takes place between an organisation and its external stakeholders.
They should also be mindful of the range of objects that are situated at the boundary
between an organisation and its environment and that mediate this communication. The
reason is that such objects are not merely neutral passage points or conduits through
which information is transferred across organisational boundaries. The structure and
125 characteristics of these objects can play a significant part in shaping the content that is
exchanged across organisational boundaries, the nature of inter-organisational
relationships, and ultimately, the formation of organisational identities.
A Second contribution my work makes is by calling specific attention to the role that technology plays in the structuring of organisational identities. Conceptualising
technological artefacts as boundary objects, the model I propose indicates that technology
that is used to facilitate communication or interactions between diverse organisations is
likely to play a significant role in the construction of these organisations’ identities. The
idea that the technologies that organisational members use may be influential in shaping
their organisation’s identity is not altogether novel. For example, in their study of
research scientists in oceanography and marine biology, Lamb and Davidson (2005)
suggest that scientists’ professional identity is influenced by their use of technology in
three ways: first, development of new core technologies can create new areas of expertise and thus enhance their scientific identity; second, coordination technologies can provides new forums, such as project websites, for identity enactment; and third, dissemination technologies such as email or project websites enable new ways to represent scientific concepts and enhance scientists’ professional identity. Thus, in their work, Lamb and
Davidson focus primarily on the capacity of technologies to provide new venues to symbolically represent professional identities.
Alternatively, my work conceptualises organisational identities as relational constructs
and emphasises the role that technological artefacts play in shaping the interfaces between organisations, namely, the manner in which organisations interact with one another. By providing new channels and possibilities for communication, collaboration,
126 and exchange of information across organisational boundaries, technological artefact can
change the conditions for inter-organisational relationships. This, in turn may alter the
context for the enactment of the identities of the interacting organisations. This point is
particularly relevant in today’s business and organisational environments given that a
growing number of organisations all across the world are becoming increasingly inter-
connected with their counterparts through a wide range of information and
communication technologies. In this context, understanding the nature of the
relationships between such technologies and the identities of the organisations that use
them is as important as it has ever been.
The findings presented above also contribute to a richer understanding of the capacity of
organisations to effectively deal with changes that emanate from radical technological
innovations in their environments. Distinguished from incremental innovations, radical
innovations have been defined as possessing the following three characteristics: (1) they represent a significant departure from existing conventions and are dissimilar from current structures and processes (Zaltman et al. 1973); (2) their deployment necessitates assimilating new cognitive frames (Bijker 1992); and (3) they influence future innovations and are therefore transformative for surrounding structures or processes
(Dahlin & Behrens 2005, Dosi 1982). 3D modelling technologies correspond to these three characteristics because (1) they are significantly different from existing 2D technologies both technologically and in terms of their broader affordances; (2) their deployment calls for the enactment of a new inter-organisational collaborative framework that is associated with novel organisational attitudes, identities, and mindsets; and (3) their utilisation has been shown to influence the trajectory of multiple subsequent
127 innovations within a network of distributed organisations in the AEC industry (Boland et.
al., 2007).
Because of their transformative nature, radical innovations pose a significant challenge
for adopting organisations. As described above, the deployment of 3D tools in AEC construction projects contested deeply seated social conventions, patterns of interactions,
systems of allocating responsibilities among multiple organisations, and work practices.
In the context of these changes, Hoffman and Zahner’s ability to continually and
effectively perform their jobs stands out. It could be argued that both organisations have
been successful in adapting to the new ways of operating in 3D-based construction
environments. Despite the considerable technological, organisational, and social
adjustments that they had to make, both organisations were able to maintain a high level
of performance that helped the projects in which they were involved to be completed.
Moreover, in two of the four cases (the Seattle library and the Hunter museum), Hoffman
and Zahner effectively stepped in to assume broader roles and responsibilities than they
were contractually obligated to in order to assist with the collective construction effort.
An interesting aspect of Hoffman and Zahner’s ability to cope with radical changes in
their environments has to do with the manner in which these changes were represented
and made sense of within these two organisations. Of specific interest is the unfolding
narrative that members of each of the two organisations developed to portray both the
change process in which they were involved and the role that they played in this process.
Sensemaking involves the retrospective development of a plausible narrative to explain
what people have done, and the reasons for them having done so (Weick, 1979). Explicit
efforts at sensemaking typically occur when the current state of the world is perceived to
128 be different from the expected state of the world, or when there is no readily available
way to engage the world because of its perceived newness (Weick et. al., 2005). When
the situation feels different, this circumstance is experienced as a situation discrepancy
(Orlikowski & Gash, 1994) or interruption; an expectation of continuity is breached,
ongoing collective action is disrupted, and efforts are made to construct a plausible sense
of what is happening (Weick et. al., 2005). This sense of plausibility normalises the
breach, familiarises the seemingly unfamiliar, and enables collective action to resume
(Weick et. al., 2005).
Hoffman and Zahner’s way of dealing with the changes in the technologies that they had
used and with their associated practices followed a similar pattern. The efforts of
members of both organisations to narratively reconstruct the events of their involvement
in the abovementioned four projects can be seen as an attempt to emphasise the saliency of the familiar, enduring, and consistent elements in their practices and roles across these projects, and underplay the saliency of the unfamiliar, unstable, and novel aspects in their practices and roles. For example, a recurrent experience during my interviews with multiple members from Hoffman and Zahner was their initial reluctance to recognise the scope of the changes that their organisations has to adjust to during work on projects in which 3D modelling tools were used. For example, in interviews with two Hoffman project managers I asked about the differences between their company’s involvement in
conventional 2D-based projects and the Seattle library and the EMP building. Both project managers initially stated that there were no significant differences between these projects in terms of their roles, practices, or responsibilities. As project manager 1 put it:
“We are a general contractor. Whether we use 2D technologies or 3D technologies we are
129 still a general contractor and our job is still to manage and coordinate a complex
construction process. We still have to make sure that the different subcontractors talk to
each other and get their job done. I do not see any real difference in our role or
responsibilities.” The tactic of trying to explicitly and directly gauge the differences
between traditional 2D-based projects and projects in which 3D tools were used yielded
similar results with various Zahner employees who also emphasised that their
professional role and responsibilities have remained the same, namely, to fabricate metal parts for buildings. The magnitude of the changes that both organisations have gone
through only started to emerge when each of the respondents was asked to describe in
detail his or her company’s involvement in specific construction projects. Only when
recalling concrete instances of engaging with other companies during the construction
process and of using 2D and 3D tools to support their practice, and when the differences
among those instances were mirrored back to them by the interviewer, were the
interviewees willing to accept the possibility that significant changes in their roles and
practice had in fact occurred.
The retrospective production of a narrative that emphasises the familiar and the stable is
interwoven with the continuous process of identity construction (Mills, 2003). Just as it is
important for people to develop a plausible story of the events that are happening around
them, it is important for them to construct a coherent image that characterises their own
involvement in these events. The processes that develop and maintain this self image are
posited to operate in the service of the need for self-consistency, which is the desire to
sense and experience consistency and continuity (Weick, 1995). The efforts of employees
from Hoffman and Zahner’s to emphasise the enduring aspects of their practices and
130 roles, also included efforts directed at highlighting the durable portions of their organisations’ identities. Thus, as the quote from the Hoffman project manager above illustrates, not only the company’s practices were perceived to have remained stable, so was its identity of a general contractor: “Whether we use 2D technologies or 3D technologies we are still a general contractor”. As before, only when going through concrete instances of engagement in practices, using the technology, and interacting with other organisations, did this overarching identity start to get unpacked and the differences in the two organisations’ representations of themselves across different construction projects emerge.
In addition to demonstrating sensemaking in practice, this example points to an interesting possibility that organisations may exhibit multiple nested identities. Nested identities are lower and higher order identities such that the latter encompasses and provides the context for the former (Medrano & Gutierrez, 2001). On one level, organisations may have a canopy identity that is derived from a formal recognition of their line of work, their area of expertise, and their institutionally recognised roles. On another level, organisations may develop additional identities that are nested within the broader identity and that are reflective of the local contingencies of the environments within which they exist; the dynamic circumstances and interactions within which they have to navigate. Whereas the former level of identity is relatively stringent, the latter is more flexible to accommodate the varying demands and reflect the changing circumstances in which the organisation operates.
131 9. Implications
A number of implications can be drawn out of my study. The first stems from the finding
that boundary objects have greater significance than previously indicated. As noted
above, past research has primarily focused on boundary objects as translation devices that
allow for cross-organisational knowledge sharing and communication. My study, by
highlighting the social dynamics of the organisations that use boundary objects, illustrates
that whereas boundary objects indeed play a significant role in the interactions between
organisations, they are also an essential resource used for the symbolic structuring of
organisations. This is due to the important role boundary objects play in the formation of organisational identities. This point suggests that the adoption of new inter-organisational information systems can not only improve the efficiency of information sharing and coordination across organisational boundaries (Clemons & Row, 1992; Argyres, 1999;
Majchrzak et al, 2000), but potentially also bring about fundamental changes in the identity and practices of the adopting organisations. Therefore, organisations that implement inter-organisational technologies need to be mindful and strategic about the possible consequences of such technologies, not only for their relationships with other actors, but also for their own identity and practices.
Second, and following from the previous implication, my study suggests that the choice
of boundary objects to communicate and mediate between different organisations or
social contexts can be used to purposefully influence the shaping of practices and
identities of adjacent organisations and social groups. That is, organisations may position
specific types of objects at their boundaries with different actors in an effort to influence their pattern of relationships with these actors (i.e., their boundary practices) and
132 consequently, these actors’ institutionalised practices and identities. For example, the
decision to implement a new supply chain management system by a manufacturing firm
may be intended not only to improve the efficiency of its interactions with its customers
and suppliers, but also to purposefully alter its customers and suppliers’ institutionalised
practices and organisational identities. In my research I have come across such a situation
when a large international manufacturing firm that I studies implemented a new web-
based supply chain management system to replace its previous traditional phone-and-fax-
based system of managing its relationships with its suppliers. The purpose of the new
system was not only to improve the firm’s supply chain efficiency, but also to
deliberately instil significant changes in their suppliers’ institutionalised practices, and in
the manner in which the suppliers perceived their work, their relationships, and
themselves. Managers at the manufacturing firm often referred to this effort by stating
that they wanted to turn their suppliers that relied on such traditional communication
methods from ‘mom and pop’ businesses to more modern organisations.
Third, although the idea that IT artefacts form an important class of boundary objects has
been previously discussed (e.g., Karsten et. al., 2001), their relation to the organisations
that draw upon them, in particular to their identities and practices, has remained largely
unexplored. My study suggests that the use of IT artefacts goes beyond sheer instrumental considerations of efficiency or effectiveness and cannot be fully comprehended when such logic is applied. IT artefacts should be understood not only as physical objects that can enter an organisation and change it in any number of ways, where some may prove to be more effective than others. IT artefacts also need to be construed as elements around which organisations develop institutional practices and
133 patterns of inter-organisational relationships, which serve as the basis for the enactment
of their identity. Therefore they have a significant symbolic importance for organisational
members. When dealing with resistance to IT-based change, managers need to be aware
that organisational actors’ reluctance to use the new technology may have to do not only
with the necessity of simply having to change their work routines or acquire new
expertise, but possibly also with the difficulties of having to change their established
identities.
Fourth, my study implies that organisational change associated with mobilising new IT
artefacts can be fruitfully understood in the context of reciprocal and dynamic inter-
organisational relationships that are constituted through the use of shared boundary
objects and engagement in mutual boundary practices. Some organisational change
literature has suggested ways to improve the change process and its organisational
consequences (Labianca et. al., 2000; Ward & Elvin, 1999; Hsiao & Ormerod, 1998).
Process-oriented studies have examined the actual process of organisational change (e.g.,
Van De Ven & Poole, 1995). Information systems researchers have examined the ensuing contradictory impacts of IT artefacts (e.g., Robey & Boudreau, 1999). What is striking in the extant literature is the scant attention paid to the fact that organisations are situated within dynamic and reciprocal social systems that are mediated and supported by boundary objects. Understanding organisational change in such a dynamic context is particularly important given that contemporary organisations are increasingly connected through a myriad of IT artefacts that serve as boundary objects. My theoretical perspective complements existing theories of organisational change by explicitly incorporating such dynamic and reciprocal inter-organisational social contexts.
134 Importantly, the theoretical model and cases presented above suggest that attempts to
initiate IT-enabled organisational changes are likely to instigate further changes in
neighbouring organisations through mutually shared boundary objects, which could have
unexpected consequences. Therefore, managers of organisational change initiatives should take into account the broader socio-technical ecology that may be affected as a result of the initial change. Furthermore, anticipating and mitigating resistance to change within the implementing organisation may be insufficient when a change in technology is likely to instigate a cascade of changes in neighbouring organisations, and when those organisations are not ready to embrace such changes. In such cases, an organisation’s efforts to minimise internal resistance to change are likely to be futile as the source of resistance may lie beyond that organisation’s boundaries and immediate reach. Therefore, management may need to enlist the assistance and cooperation of neighbouring organisations and develop much broader and systemic approaches to technology-enabled organisational change.
Finally, my study highlights the importance of boundary research; research that focuses
on the processes, artefacts, and resources that are situated between and shared by multiple
social or organisational contexts. Either explicitly but mostly implicitly, the vast majority
of organisational academic research and practitioner interest has centred on the practices,
technologies, and social and psychological processes that take place within relatively
clearly demarcated social or organisational units. Since most individuals typically
identify themselves as members of certain social groupings such as work organisations,
clubs, or ethnic and national associations, the salience of such social frameworks in academic and practitioner discourses has been accepted as natural. Accordingly, such
135 concepts as organisational culture, identity, structure, design, and strategy and the manner
in which they emerge, change, and develop within organisational boundaries have
received the bulk of researchers’ attention.
An alternative way of thinking about these phenomena is to recognise the importance of
organisational boundaries to their articulation. Boundaries are laden with symbolic meaning and significance that are of fundamental importance to the way in which
organisational members go about performing their jobs, interacting with one another, and
engaging in communicative activities with members of other organisations. The symbolic
meaning inherent in organisational boundaries is neither coincidental nor inconsequential.
Organisational members invest a lot of effort and resources in boundary work trying to
shape the area that lies between the organisation and its periphery. By designing and
manipulating a range of boundary resources that are situated between the organisation
and its environment (e.g., websites, advertisements, technologies that are used to
communicate with other organisations, corporate buildings, etc.) members act to present
their organisation in a certain manner and influence others’ perceptions and
interpretations of the organisation.
Boundaries therefore are more than a mere means to delineate the scope of an
organisation’s jurisdiction, influence, or responsibility. Through continuous social
practice, organisational members can transform boundaries into integral resources by
means of which they communicate with, present themselves to, and understand members
of other organisations (Brown & Duguid, 1994). Consequently, examining these
boundary resources and practices and their organisational implications should be of
particular importance and relevance for researchers.
136 10. Limitations and Future Research
There are several limitations related to this study. First, I examined a relatively small fraction of an ongoing, large scale change process in the AEC industry. A more methodical pursuit of my theoretical interests would necessitate broadening of the scope of research both in terms of the period of time covered and the number of actor examined. Doing so would contribute to formulating a more valid theoretical model.
Second, in this study I relied heavily on semi-structured interviews as a data collection method. Although interviews provide detailed, contextual information concerning organisational processes and subjective personal impressions of organisational members, they do not produce the same narrative-like rich descriptions that often result from ethnographic studies and studies that rely on different forms of lengthy participant observations. While such descriptions are not always necessary, they would have been helpful to this study because of its subject matter. Concepts such as information infrastructures and organisational identities are inherently loosely-structured, elusive, and manifest themselves in multi-faceted ways. Therefore they are hard to gauge merely by means of post-hoc verbal accounts of organisational members. The use of other data- collection methods such as those described above could have improved the suggested model’s construct validity and reliability and offered a means to more richly illustrate the formation of organisational practices and identities. Finally, and related to the previous point, my study relied extensively on interviewees’ accounts of past events. In some cases, interviewees described events that took place up to ten years prior to their interview. As a result, some important details could have been left out. A good way to
137 remedy this would be to collect more data from different sources whose reliability is not
affected by the passing of time such as, documents or archival data.
The findings of this study suggest several paths for future research. First, modelling and
representational technologies are just one IT tool that can be employed to mediate
communications in inter-organisational environments. There is a need to examine the role
of other mediating devices, and in other contexts, to assess the soundness of the model
proposed here. Second, boundary objects are part of a system whose elements are
connected in reciprocal relationships. When change occurs in one of the elements it will
carry over to the others, spanning multiple organisations. In an increasingly inter-
connected world, studies that follow the trajectory of such inter-organisational change
processes and examine their patterns can be particularly relevant. Finally, efforts by
organisational groups to retain their identities can play a significant role in shaping their adoption and use of new technologies. Future studies could illustrate how this process unfolds, and identify conditions that generate different patterns of this process.
138 Appendix 1. Company interview guide
Bio.
Personal background. Past work experience with Hoffman – position, responsibilities.
Company history
What did it do? Which industries did it operate in? What were its core competencies? What technologies did it use, and how? (CATIA?) What were routine practices / interactions?
Information on specific projects
Before using 3D (and/or working with Gehry) Working with 3D (and/or working with Gehry) After working with 3D (and/or working with Gehry), or, subsequent projects with 3D (and/or Gehry)
What was the background of the project? How did the company get involved in it? What was the Scope of the project? Who were the major players in the project? Who were the major players from the company? What role did the company play? What were channels of communication / mechanisms of coordination / negotiation among the players…?
Has the company changed as a result?
Have you changed your practices (use of CATIA)? The way you approach a project / communicate with other players? The way you perceive your job / responsibilities? The way the company perceives itself?
139 References
Ackerman. M. S and Halverson. C (1999) Organizational memory: processes, boundary objects and trajectories. Proceedings of the 32nd Hawaii International Conference on System Sciences.
Albert, S. & Whetten, D. A. (1985) Organisational Identity. In: Research in Organisational Behaviour, Cummings, L. L. & Staw, B. M. (Eds.). Greenwich, CT: JAI Press.
Aoshima, Y. & Nobeoka, K. & Takeda, Y. (1999) Benefiting from Task-Redefining Process Technology: The influence of an organizational process on the introduction of 3D-CAD.
Argyres, N. S. (1999) The impact of information technology on coordination: Evidence from the B-2 “Stealth” bomber. Organization Science, 10, 2, 162-180.
Ashforth, B. & Mael, F. (1996) Organisational Identity and Strategy as a Context for the Individual. In: Advances in Strategic Management, Baum, J. A. C & Dutton, J. E. (Eds.), 13, 19-64. Greenwich, CT: JAI Press.
Baba, Y. & Nobeoka, K. (1998) Towards knowledge-based product development: the 3D CAD model of knowledge creation. Research Policy, 26, 6, 643-659.
Bannon, L. & Bødke, S. (1997) Constructing common information spaces. Proceedings of the 5th European conference on computer supported cooperative work, 81-96.
Barley, S. R. (1986) Technology as an occasion for structuring: Evidence from observations of CT scanners and the social order of radiology departments. Administrative Science Quarterly, 31, 1, 78-109.
Bechky, B. A. (2003) Sharing Meaning Across Occupational Communities: The Transformation of Understanding on a Production Floor. Organization Science, 14, 3, 312-330.
Beck, P. (2001) The AEC dilemma: Exploring the barriers to change. Leadership and Management in Engineering, Apr, 31-36.
Berger, P. L. & Luckman, T. (1967) The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Anchor Books.
Bijker, W.E. (1992) The social construction of fluorescent lighting, or how an aircraft was invented in its diffusion stage. In: Shaping technology / building society: Studies in sociotechnical change, W.E. Bijker and J. Law (eds.), The MIT Press, Cambridge Ma.
140 Bijker. W. E. (1995). Of bicycles, bakelites and bulbs: Toward a theory of sociotechnical change. Cambridge MA: MIT Press.
Boland, R.J., Lyytinen, K. & Yoo, Y. (2003) Path Creation with Digital 3D Representation: Networks of Innovation in Architectural Design and Construction. Sprouts: Working Papers on Information Environments, Systems and Organisations, 3, 1, 1-28.
Boland, R.J., Lyytinen, K. & Yoo, Y. (2007) Wakes of Innovation in Project Networks: The case of digital 3-D representations in architecture, engineering and construction. Organization Science. Borgman, C. (1997) From acting locally to thinking globally: A brief history of library automation. Library Quarterly, 67, 3, 215-249. Borgman, C. (1999) What are digital libraries? Competing visions. Information processing and management, 35, 227-243.
Bourdieu, P. (1991) Language and symbolic power. Harvard University Press.
Bowker, G. C., Timmermans, S. & Star, S. L (1995) Infrastructure and organisational transformation: classifying nurses’ work In: Information Technology and Changes in Organisational Work, Orlikowski, W., Walsham, G., Jones, M.R. & Degross, J (Eds.), pp. 344-369. Chapman & Hall.
Bowker, G. C. (2005) Memory practices in the sciences. The MIT Press.
Bowker, G. C. & Star, S. L. (1994) Knowledge and infrastructure in international information management: Problems of classification and coding. In: Information acumen: The understanding and use of information in modern business, Bud, L. (Ed.). London: Routledge.
Bowker, G. C. & Star, S. L. (1999) Sorting Things Out: Classification and its Consequences. The MIT Press Cambridge
Briers, M. & Chua, W. F. (2001) The role of actor-networks and boundary objects in management accounting change: a field study of an implementation of activity-based costing. Accounting, organisations & Society, 26, 237-269.
Broadbent, M. & Weill, P. & St. Clair, D. (1999) The Implications of Information Technology Infrastructure for Business Process Redesign. MIS Quarterly, 23, 2, 159-182.
Brown, J. S. & Duguid, P. (1994) Borderline Issues: Social and Material Aspects of Design. Human-Computer Interaction, 9, 1, 3-36.
Brynjolfsson, E. & Hitt, L. M. (1998) Beyond the productivity paradox. Communications of the ACM, 41, 8, 49-55.
141
Byrd, T. A. & Turner, D. (2000) Measuring the flexibility of information technology infrastructure: Exploratory analysis of a construct. Journal of Management Information Systems, 17, 1, 167-208.
Callon, M. & Law, J. (1989) On the construction of sociotechnical networks: content and context revisited. Knowledge Society, 9, 57-83
Carlile, P. R. (1997) Understanding knowledge transformation in product development: Making knowledge manifest through boundary objects. The University of Michigan.
Carlile, P. R. (2002) A pragmatic view of knowledge and boundaries: Boundary objects in new product development. Organisation Science, 13, 4, 442-455.
Carlile, P. R. (2004) Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organisation Science, 15, 5, 555- 568.
Carrillo, P. & Anumba, C. (2002) Knowledge management in the AEC sector: an exploration of the mergers and acquisitions context. Knowledge and Process Management, 9, 3, 149-161.
Ciborra, U. C. (2000) From Control to Drift: The Dynamics of Corporate Information Infrastructures. Oxford University Press.
Ciborra, U. C. (2002) The Labyrinths of Information: Ccenging the Wisdom of Systems. Oxford University Press.
Clark, H. H (1996) Using language. Cambridge (England); New York: Cambridge University Press.
Clemons. E. K & Knez. M (1988) Competition and cooperation in information systems innovation. Information and Management, 14, 1, 25-35.
Clemons. E. K & Row, M. C (1992) Information technology and industrial cooperation: the changing economics of coordination and ownership. Journal of Management Information Systems, 9, 2, 9-28.
Cohen.P.A (1995) The symbolic construction of community, London: Routledge.
Corley, K. G & Gioia, D. A. (2004) Identity ambiguity and change in the wake of a corporate spin-off. Administrative Science Quarterly, 49, 173-208.
Dahlin, K.B., and Behrens, D.M. "When is an invention really radical?: Defining and measuring technological radicalness," Research Policy (34:5), June 2005, pp 717-737.
142 Dosi, G. (1982) Technological paradigms and technological trajectories: A suggested interpretation of the determinants and directions of technical change. Research Policy, 11, 3, 147-162.
Dourish, P. & Edwards, K. (2000) A tale of two toolkits: Relating infrastructure and use in flexible CSCW toolkits. Computer supported cooperative work, 9, 33-51.
Duncan, N. (1995) Capturing flexibility of information technology infrastructure: a study of resource characteristics and their measure. Journal of Management Information Systems, 12, 2, 37-57.
Dutton, J. E & Dukerich, J. M & Harquail, C. V (1994) Organizational Images and Member Identification. Administrative Science Quarterly, 39, 239-263.
Edwards, P. M. (2003) Infrastructure and modernity: Force, time, and social organization in the history of sociotechnical systems. In Modernity and Technology, T. J. Misa & P. Brey & A. Feenberg (Eds.) Cambridge, MA: M IT Press.
Eisenhardt, K. M. (1989) Building theories from case studies. Academy of Management Review, 14, 4, 532-550.
Eisenhardt, K. M. & Graebner, M. E. (2007) Theory Building from Cases: Opportunities and Challenges. Academy of Management Journal, 50, 1, 25-32.
Emirbayer. M (1997) Manifesto for a relational sociology. Manuscript, New School for Social Research, Department of Sociology, University of Chicago.
Fiol, C. M. & Hatch, M. J. & Golden-Biddle, K. (1998) Organisational Culture and Identity: What’s the Difference Anyway?. In: Identity in Organisations: Developing Theory through Conversations, Whetten, D. & Godfrey, P. (Eds.). thousand Oaks, CA: Sage.
Fiol, C. M. (1991) Managing Culture as a Competitive Resource: An Identity-Based View of Sustainable Competitive. Journal of Management, 17, 1, 191-211.
Fiol, C. M. (2002) Capitalizing on Paradox: The Role of Language in Transforming Organizational Identities. Organization Science, 13, 653-666.
Gallivan, M. J. & Depledge, G. (2003) Trust, control and the role of inter-organizational systems in electronic partnerships. Information Systems Journal, 13, 159-190.
Garrety, K. & Badham, R. (2000) The politics of socio-technical intervention: an interactionist view. Technology Analysis & Strategic Management, 12, 1, 103-118.
Gasson. S (2006) A genealogical study of boundary-spanning IS design. European Journal of Information Systems, 15, 26-41.
143
Giddens, A. (1984) The Constitution of Society: Outline of Theory of Structuration. University of California Press.
Giddens, A. (1991) Modernity and self identity: Self and society in the late modern age. Cambridge: Polity Press.
Gioia, D. A. (1998) From individual to organisational identity. In: Identity in Organisations: Developing Theory through Conversations, Whetten, D. & Godfrey, P. (Eds.). thousand Oaks, CA: Sage.
Gioia, D. A. & Schultz, M. & Corley, K. G. (2000) Organizational Identity, Image, and Adaptive Instability. Academy of Management Journal, 25, 1, 63-81.
Greco, J. (2001) Catia Version 5 Release 7. Cadence, 16, 12, 26-31.
Guglan. P & Dunning, J. H (1993) Technology based cross-border alliances. In: Multinational strategic alliances, Culpan, R. (Ed.) Howarth Press Inc., Binghamton, NY.
Halpin, D. W. & Riggs, L. S. (1992) Planning and Analysis of Construction Operations. New York : Wiley
Hall, S. (1996) Who needs social identity? In: Questions of Cultural Identity, Hall, S. & Gay, P.D. (Eds.) Sage Publications.
Han, K. & Kauffman, R. J. & Nault, B. R (2004) Information exploitation and inter- organizational systems ownership. Journal of Management Information Systems, 21, 2, 109-135.
Hanseth, O. & Monteiro, E. (1997) Inscribing behaviour in information infrastructures standards. Accounting, Management, and Information Technology, 7, 4, 183-211.
Hanseth, O. & Lundberg, N. (2001) Designing work oriented infrastructures. Computer supported cooperative work, 10, 347-372.
Henderson, J. C. & Venkatraman, N. (1999) Strategic alignment: Leveraging information technology for transforming organizations. IBM Systems Journal, 1999, 38, 2/3, 472-485.
Henderson, K. (1991) Flexible sketches and inflexible data-bases: Visual communication, conscription devices and boundary objects in design engineering. Science, Technology& Human Values, 16, 4, 448-473.
Henderson, K. (1998) The role of material objects in the design process: a comparison of two design cultures and how they contend with automation. Science, Technology & Human Values, 23, 139-174.
144 Hsiao, R. L.& Ormerod, R. J. (1998) A new perspective on the dynamics of information technology-enabled strategic change. Information Systems Journal, 8, 1, 21-52.
Karsten, H., Lyytinen, K., Hurskainen, M. & Koskelainen, T. (2001) Crossing boundaries and conscripting participation: Representing and integrating knowledge in a paper machinery project. European Journal of Information Systems, 10, 89-98.
Kling, R. (1992) Behind the Terminal: The critical role of computing infrastructure in effective information systems’ development and use. In Challenges and Strategies for Research in Systems Development. W. Cotterman & J. Senn (Eds.) New York, John Wiley.
Koutamanis, A. (2000) Digital Architectural Construction. Automation in Construction, 9, 347-360.
Kumar. K & Van Dissel. H. G (1996) Sustainable collaboration: Managing conflict and cooperation in inter-organizational Systems. MIS Quarterly, 20, 3, 279-300.
Labianca, G., Gray, B. & Brass, D. (2000) A grounded model of organisational schema change during empowerment. Organisation Science, 11, 2, 235-257.
Lacourse, D. (2001) Catia V5 r6 Serves up robust tools for mechanical engineers. CADalyst, 18, 12, 32-36.
Lamb, R. & Davidson, E. (2005) Information and Communication Technology Challenges to Scientific Professional Identity. The Information Society, 21, 1-24.
Latour, B. (1987) Science in Action: How to Follow Scientists and Engineers Through Society. Cambridge : Harvard University Press
Law, J. (1987) Technology and heterogeneous engineering: The case of portuguese expansion. In: Social construction of technological systems: New directions in the sociology and history of technology, Bijker, W. E & Hughes, T. P. & Pinch, T. J. (Eds.). Cambridge, Mass.: MIT Press
Law, J. & Callon, M. (1992) The Life and Death of an Aircraft: A Network Analysis of Technical Change. In: Shaping Technology / Building Society: studies in sociotechnical change, Bijker, W. E. & Law, J (Eds.). Cambridge, Mass.: MIT Press.
Lee, D. M. S & Trauth, E. M & Farwell, D. (1995) Critical Skills and Knowledge Requirements of IS Professionals: A Joint academic / industry investigation. MIS Quarterly, 19, 3, 313-340.
Lee, C. P. & Dourish, P. & Mark, G. (2006) The human infrastructure of cyberinfrastructure. Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work, Banff, Alberta, Canada.
145
Levina, N. (2005) Collaborating on multiparty information systems development projects: A collective reflection-in-action view. Information Systems Research, 16, 2, 109-130.
Levina, N. & Vaast, E. (2005) The emergence of boundary spanning competence in practice: Implications for implementation and use of information systems. MIS Qaurterly, 29, 2, 335, 363.
Lindsey, B. (2001) Digital Gehry. Basel; Boston: Birkhäuser.
Majchrzak. A & Rice. R. E & Malhotra. A & King. N (2000) Technology adaptation: The case of a computer-supported inter-organizational virtual team. MIS Quarterly, 24, 4, 569-600.
Markus, M. L. & Robey, R (1988) Information technology and organizational change: Casual structure and theory and research. Management Science, 34, 5, 583-598.
McLaughlin, J. & Rosen, P. & Skinner, D. & Webster, A. (1999) Valuing technology: organisations, culture and change. Routledge.
Medrano, J. D. & Guiterrez, P (2001) Nested identities: National and European identity in Spain. Ethnic and racial studies, 24, 5, 753-778.
Mills, J. H. (2003) Making sense of organizational change. Routledge: London & NY.
Monteiro, E. & Hanseth, O. (1996) Social shaping of information infrastructure: on being specific about the technology. In: Information Technology and Changes in Organisational Work, Orlikowski, W., Walsham, G., Jones, M.R. & Degross, J (Eds.), pp.325-343. Chapman & Hall.
Moscovici, S. (2000) Social representations: explorations in social psychology. Polity Press.
Mukhopadhyay, T., Kekre, S. & Kalathur, S. (1995) Business value of information technology: A study of electronic data interchange. MIS Quarterly, 19, 2, 137-157.
Orlikowski, W. & Robey, D. (1991) Information technology and the structuring of organizations. Information Systems Research, 2, 2, 143-169.
Orlikowski, W., and Gash, D. (1994) Technology frames: Making sense of information technology in organizations. ACM Transactions on Information Systems, 12, 174-207.
Orlikowski, W. (2000) Using technology and constituting structures: A practice lens for studying technology in organizations. Organization Science, 11, 4, 404-429.
146 Oxford Dictionary: www.askoxford.com
Pawlowski, S. D. & Robey, D. (2004) Bridging user organisations: Knowledge brokering and the work of information technology professionals. MIS Quarterly, 28, 4, 645-672.
Ravasi, D. & Schultz, M. (2006) Responding to organisational identity threats: Exploring the role of organisational culture. Academy of Management Journal, 49, 3, 433-458.
Robey, D. & Boudreau, M. C. (1999) Accounting for the contradictory consequences of information technology: Theoretical directions and methodological implications. Information Systems Research, 10, 2, 167-187.
Saggio, A. (1997) Frank O. Gehry: Architecture Residuali Torino.
Sambamurthy, V. & Zmud, R. W. (2000) Research commentary: The organizing logic for an enterprise's IT activities in the digital Era - a prognosis of practice and a call for research. Information Systems Research, 2000, 11, 2, 105-115.
Sasped, J. & Salter, A. (2004) Postcards from the Edge: Local Communities, Global Programs and Boundary Objects. Organization Studies, 25, 9, 1515-1534.
Schmidt, K. & Simone, C. (1996) Coordination mechanisms: Towards a conceptual foundation of CSCW systems design. Computer supported cooperative work: The journal of collaborative computing, 5, 155-200.
Siau, K. (2003) Inter-organizational systems and competitive advantages: Lessons from history. Journal of Computer Information Systems, fall, 33-39.
Silverman, D. (2001) Interpreting qualitative data: Methods for analysing talk, text and interaction. Sage Publications.
Star, S. L. & Griesemer. J. R. (1989) Institutional ecology, ‘translations’ and boundary objects: Amateurs and professionals in Berkeley’s museum of vertebrate zoology, 1907- 1939. Social Studies of Science, 19, 387-420.
Star, S. L. & Ruhleder, K. (1996) Steps toward an ecology of infrastructure: Design and access for large information spaces. Information Systems Research, 7, 1, 111-134.
Star, S. L. (1999) The ethnography of infrastructure. American Behavioural Scientist, 43, 3, 377-391
Star, S. L. (2000) Infrastructure and ethnographic practice. Scandinavian Journal of Information Systems, 14, 2, 107-122.
Subrahmanian, E., Monrach, I., Konda, S., Granger, H., Milliken, R. & Westerberg, A. (2003) Boundary objects and prototypes at the interfaces of engineering design. Computer Supported Cooperative Work, 12, 185-203.
147
Sweet, J. (1994) Legal aspects of architecture, engineering and the construction process. Minneapolis/St. Paul : West Pub. Co.
Toupin, L.A. (2001) Knowledgeware makes reusing existing designs a snap. Design News, 56, 20, 42-53.
Turner, W. & Bowker, G. & Gasser, L. & Zackland, M (2006) Information infrastructures for distributed collective practices. Computer supported cooperative work, 15, 93-110.
Van De Ven, A. H. & Poole, M. S. (1995) Explaining development and change in organisations. Academy of Management Review, 20, 3, 510-540.
Walsham, G. (1998) IT and Changing Professional Identity: Micro-Studies and Macro- Theory. Journal of the American Society for Information Science, 49, 12, 1081-1089.
Walsham, G. & Barrett, M. (1999) Electronic Trading and Work Transformations in the London Insurance Market. Information Systems Research, 10, 1, 1-22.
Ward, J. & Elvin, R. (1999) A new framework for managing IT-enabled business change. Information Systems Journal, 9, 3, 197-221.
Weick. K. E. (1979) The social psychology of organizing. McGraw-Hill Inc.
Weick, K. E. (1995) Sensemaking in organizations. Thousand Oaks, CA: Sage.
Weick, K. E. & Sutcliffe, K. M. & Obstfeld, D. (2005) Organizing and the process of sensemaking. Organization Science, 16, 4, 409-421.
Weill, P. (1992) The relationship between investment in information technology and firm performance: A study of the valve manufacturing sector. Information Systems Research, 3, 4, 307-333.
Weill, P. & Broadbent, M. (1997) Leveraging the new infrastructure: how market leaders capitalize on information technology. Harvard Business School Press.
Weill, P. & Subramani, M. & Broadbent, M. (2002) Building IT infrastructure for strategic agility. MIT Sloan Management Review, 44, 1, 57-65.
Wenger, E. (2000) Communities of practice and social learning systems. Organisation, 7, 2, 225-246.
Whetten, D. A. (2003) A social actor conception of organizational identity. Thousand Oaks, CA: Sage.
148 Whetten, D. A. & Mackey, A. (2002) A social actor conception of organizational identity and its implications for the study of organizational reputation. Business & Society, 41, 4, 393-414.
Yakura, E. K. (2002) Charting time: Timelines as temporal boundary objects. Academy of Management Journal, 45, 5, 956-970
Yap, A. Y. & Ngwenyama, O. & Osei-Bryson, K. (2003) Leveraging knowledge representation, usage, and interpretation to help reengineer the product development life cycle: visual computing and the tacit dimensions of product development. Computers in Industry, 51, 1, 89-110.
Yin, K.R. (2003) Case Study Research: Design and Methods. Sage Publications.
Zaltman, G., Duncan, R., and Holbek, J. Innovations and organizations John Wiley and Sons, New York, 1973.
Zhu, K. & Kraemer, K. L & Gurbaxani, V. & Xu, S, X. (2006) Migration to open- standard inter-organizational systems: Network effects, switching costs, and path- dependency. MIS Quarterly, 30, 515-539.
149