Macro-BIM adoption: Conceptual structures

Bilal Succar; Mohamad  Kassem

Building Information Modelling (BIM) concepts and workflows continue to proliferate within organisations, through project teams, and across the whole construction industry. However, both BIM implementation and BIM diffusion are yet to be reliably assessed at market scale. Insufficient research has been conducted to date towards identifying the conceptual structures that would explain and encourage large-scale BIM adoption. This paper introduces a number of macro-adoption models, matrices and charts (Fig. 1). These models can be used to systematically assess BIM adoption across markets, and inform the structured development of country-specific BIM adoption policies. This research is published in two complementary papers combining conceptual structures with data collected from experts across a number of countries. The first paper “Macro-BIM adoption: conceptual structures” delimits the terms used, reviews applicable diffusion models, and clarifies the research methodology. It then introduces five new conceptual constructs for assessing macro-BIM adoption and informing the development of market-scale BIM diffusion policies. The second paper “Macro-BIM adoption: comparative market analysis” employs these concepts and tools to evaluate BIM adoption and analyse BIM diffusion policies across a number of countries. Using online questionnaires and structured interviews, it applies the models, refines the conceptual tools and develops additional assessment metrics. The two papers are complementary and primarily intended to assist policy makers and domain researchers to analyse, develop and improve BIM diffusion policies.

Automation in Construction 57 (2015) 64–79 Contents lists available at ScienceDirect Automation in Construction journal homepage: www.elsevier.com/locate/autcon Macro-BIM adoption: Conceptual structures Bilal Succar a,b,⁎, Mohamad Kassem c a b c ChangeAgents Pty. Ltd., Melbourne, Australia Centre for Interdisciplinary Built Environment Research (CIBER), University of Newcastle, Australia Technology Futures Institute, Teesside University, United Kingdom a r t i c l e i n f o Article history: Received 30 December 2014 Received in revised form 12 March 2015 Accepted 25 April 2015 Available online xxxx Keywords: BIM readiness Capability and maturity BIM implementation and diffusion Point of adoption BIM framework conceptual reactor BIM diffusion policy development a b s t r a c t Building Information Modelling (BIM) concepts and workﬂows continue to proliferate within organisations, through project teams, and across the whole construction industry. However, both BIM implementation and BIM diffusion are yet to be reliably assessed at market scale. Insufﬁcient research has been conducted to date towards identifying the conceptual structures that would explain and encourage large-scale BIM adoption. This paper introduces a number of macro-adoption models, matrices and charts (Fig. 1). These models can be used to systematically assess BIM adoption across markets, and inform the structured development of countryspeciﬁc BIM adoption policies. This research is published in two complementary papers combining conceptual structures with data collected from experts across a number of countries. The ﬁrst paper “Macro-BIM adoption: conceptual structures” delimits the terms used, reviews applicable diffusion models, and clariﬁes the research methodology. It then introduces ﬁve new conceptual constructs for assessing macro-BIM adoption and informing the development of marketscale BIM diffusion policies. The second paper “Macro-BIM adoption: comparative market analysis” employs these concepts and tools to evaluate BIM adoption and analyse BIM diffusion policies across a number of countries. Using online questionnaires and structured interviews, it applies the models, reﬁnes the conceptual tools and develops additional assessment metrics. The two papers are complementary and primarily intended to assist policy makers and domain researchers to analyse, develop and improve BIM diffusion policies. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Building information modelling (BIM) is the current expression of construction industry innovation, a set of technologies, processes and policies, affecting industry's deliverables, relationships and roles. BIM concepts and tools encourage concurrent revolutionary and evolutionary changes across organisational scales — from individuals and groups; through organisations and project teams; to industries and whole markets [80]. Investigations into BIM implementation across whole markets have been comparatively rare in spite of an ever-increasing range and depth of national BIM initiatives (NBI)s and noteworthy BIM publications (NBP)s [31]. More generally, there has been – and arguably still is – a dearth in investigations covering the diffusion of innovation within the construction industry [87]. Available studies in market-scale BIM implementation and diffusion are dominated by survey ratings generated by commercially-driven service providers. The most prominent of these include: BIM diffusion in the UK, France and Germany [47]; Autodesk software uptake in Europe [3]; BIM diffusion in the U.S. and ⁎ Corresponding author at: ChangeAgents Pty. Ltd., Melbourne, Australia. Tel.: +61 412 556 671. E-mail addresses: bsuccar@changeagents.com.au (B. Succar), m.kassem@tees.ac.uk (M. Kassem). http://dx.doi.org/10.1016/j.autcon.2015.04.018 0926-5805/© 2015 Elsevier B.V. All rights reserved. Canada [48]; BIM diffusion in the UK [60,61]; The Business Value of BIM in Australia and New Zealand [49] among others. While these reports include useful information, they suffer from a number of shortcomings — they: • have unknown, remedial or biased population sampling and data collection methodologies; • do not differentiate between software acquisitions and actual adoption [17]; • mostly neglect non-software aspects of BIM adoption; • are neither based on an existing conceptual framework, nor propose a new one; • do not identify market gaps or reﬂect market-speciﬁc criteria; and • cannot be used by policy makers to facilitate BIM diffusion. In addition to industry surveys, a number of academic investigations covering market-scale BIM implementation and diffusion have been conducted in recent years. These studies covered multiple countries including: Australia [23], China [11], Finland [36], Iceland [34], India [39], South Africa [19], Sweden [71], Taiwan [56], United Kingdom [33], United States [21,38], and multiple markets [75,63,95,98]. While these studies provide more rigorous information than industry reports, and B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 65 Fig. 1. Visual abstract. contribute valuable insights into BIM diffusion trends and paths, they offer little practical assistance to policy maker's intent on assessing current or developing new market-speciﬁc BIM diffusion policies. Based on the aforementioned industry surveys and academic studies; and building-upon published conceptual structures [79,80,83] and earlier investigations [30–32], this research delivers a number of macro-classiﬁcations, taxonomies and models dedicated to assessing and informing the development of BIM diffusion policies. This paper will ﬁrst clarify relevant implementation and diffusion terminology, identify the research methodology, and then introduce ﬁve new conceptual models covering macro-BIM adoption. 1.1. Terms, concepts and their interaction The terms used to describe the act of implementing an innovative system/process are often confused with the terms used to describe the spread of this system/process within a population of adopters — be it within an organisation or across a market. It is therefore prudent to delimit a number of terms before utilising them to clarify larger concepts or propose macro-adoption models. This delimitation is both artiﬁcial and necessary: it is artiﬁcial as other researchers can recalibrate the connotations of the same terms to ﬁt their own unique purposes. It is necessary due to the availability of a large number of relevant diffusion models [66,69,16] which do not differentiate between the stages of implementation – e.g., between acceptance and routinisation as in Cooper and Zmud [13] – the mechanics of diffusion, and the pressures causing the shift from one stage to another. In introducing and delimiting these terms, we also limit ourselves to BIM as an innovative set of tools, processes and policies within the construction industry. This limitation is also both artiﬁcial and necessary: it is artiﬁcial as implementation/diffusion models introduced later are arguably applicable to other innovations within and outside the construction industry (e.g., to GIS and PLM). It is necessary due to the dearth in investigations covering innovation diffusion within the construction industry [87] thus warranting a focused attention on industry-speciﬁc and, by extension, BIM-speciﬁc terms. To avoid confusion, and as a general distinction, this paper differentiates between the notions of BIM implementation as the successful adoption of BIM tools and workﬂows within a single organisation, and ‘BIM diffusion’ as the rate BIM tools and workﬂows are adopted across markets. Both BIM implementation at sub-organisational scales (e.g., individuals and groups) and BIM diffusion across the global construction industry are intently placed outside the scope of this paper. We also make use of the generic term ‘adoption’ to overlay the connotations of implementation and diffusion unto a single word, and we use the term ‘macro’ to focus the readers' attention on large collections of organisational adopters operating within deﬁned national borders (countries). 1.2. Implementation Implementation refers to the wilful activities of a single identiﬁable player1 as it adopts a novel system/process to improve its current performance. More speciﬁcally, BIM implementation refers to the set of activities undertaken by an organisational unit to prepare for, deploy or improve its BIM deliverables (products) and their related workﬂows (processes). BIM implementation is introduced here as a three-phased approach separating an organisation's readiness to adopt; capability to perform; and its performance maturity: • BIM readiness is the pre-implementation status representing the propensity of an organisation or organisational unit to adopt BIM tools, workﬂows and protocols. Readiness is expressed2 as the level of preparation, the potential to participate, or the capacity to innovate. Readiness can be measured using a variety of approaches – productbased, process-based, and overall maturity [70] – and signiﬁes the planning and preparation activities preceding implementation; • BIM capability is the wilful implementation of BIM tools, workﬂows and protocols. BIM capability is achieved through well-deﬁned revolutionary stages (object-based modelling, model-based collaboration, and network-based integration) separated by numerous evolutionary steps [79]. BIM capability cover many technology, process and policy topics and is expressed as the minimum ability of an organisation or team to deliver a measureable outcome; and • BIM maturity (or post-implementation) is the gradual and continual improvement in quality, repeatability and predictability within available capabilities. BIM maturity is expressed as maturity levels (or 1 Depending on the ‘scoping lens’ applied, BIM players are either individuals, groups, organisational units, or whole organisations. BIM players, deliverables and their requirements have been extensively covered in earlier works [79]. 2 Deﬁnitions adopted from the e-commerce context as used by the Asia-Paciﬁc Economic Cooperation (APEC), Center for International Development (CID) at Harvard University [12]. 66 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Fig. 2. Point of Adoption model v1.1 (full size, current version). performance improvement milestones) that organisations, teams and whole markets aspire to. There are ﬁve maturity levels: [a] Ad-hoc or low maturity; [b] Deﬁned or medium–low maturity; [c] Managed or medium maturity; [d] Integrated or medium–high maturity; and [e] Optimised or high maturity [81]. 1.3. Point of Adoption The three implementation phases – readiness, capability, and maturity – are depicted in the Point of Adoption (PoA) model (Fig. 2). As explained below, a PoA is a term identifying the juncture(s) where organisational readiness transform into organisational capability/ maturity: As explored in Fig. 2, transformative BIM adoption starts at the Point of Adoption (PoA) when an organisation, after a period of planning and preparation (readiness), successfully adopts object-based modelling tools and workﬂows. The PoA3 thus marks the initial capability jump 3 The Point of Adoption (PoA) is not to be confused with the critical mass ‘inﬂection point’ on the S-curve [67,68]; or with the ‘tipping pint’, the critical threshold introduced by Gladwell [22]. from no BIM abilities (pre-BIM status) to minimum BIM capability (Stage 1). As the adopter interacts with other adopters, a second capability jump (Stage 2) marks the organisation's ability to successfully engage in model-based collaboration. Also, as the organisation starts to engage with multiple stakeholders across the supply chain, a third capability jump (Stage 3) is necessary to beneﬁt from integrated, network-based tools, processes and protocols. Each of these capability jumps is preceded with considerable investment in human and physical resources, and each stage signals new organisational abilities and deliverables not available before the jump. However, the deliverables of different organisations at the same stage may vary in quality, repeatability and predictability. This variance in performance excellence occurs as organisations climb their respective BIM maturity curve, experience their internal BIM diffusion, and gradually improve their performance over time.4 The multiple maturity curves depicted in Fig. 2 reﬂect the heterogeneous nature of BIM adoption even within the same organisation (e.g., sample Organisation X in Fig. 2 has a compiled rating 4 The X-axis in Fig. 2 represents time relative to each PoA, not as an absolute scale. That is, this version of the chart does not represent a snapshot view of compiled capability/maturity at a speciﬁc point in (absolute) time. B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 of 1c, 2b and 3a). This is due to the phased nature of BIM with each revolutionary stage requiring its own readiness ramp, capability jump, maturity climb, and point of adoption. This is also due to varied abilities across organisational sub-units and project teams: while organisational unit A1 (within organisation A) may have elevated model-based collaboration capabilities, unit A2 may have basic modelling capabilities, and unit A3 may still be preparing to implement BIM software tools. This variance in ability necessitates a compiled rating for organisation A as it simultaneously prepares for an innovative solution, implements a system/process, and continually improves its performance. 1.4. Diffusion In contrast to implementation which represents the successful adoption of a system/process by a single organisation, diffusion represents the spread of the system/process within a population of adopters. That is, the diffusion of a solution occurs after the solution has been adopted [64] or what we termed earlier as the Point of Adoption (PoA). However, the mere acquisition of an innovative solution (e.g., a software) “need not be followed by widespread deployment and use by acquiring organisations” [17, p. 256]. E.M. Rogers [67, p. 5] deﬁnes diffusion as the “process by which an innovation is communicated through certain channels over time among the members of a social system”, a deﬁnition that covers the increase in “number of ﬁrms using or owning a technology (inter-ﬁrm diffusion) [and the] more intensive use of the technology by the ﬁrm (intra ﬁrm diffusion)5” [77, p. 919, 44]. Diffusion is also identiﬁed as the third and ﬁnal phase of the well-noted Schumpeterian Trilogy: “invention (the generation of new ideas), innovation (the development of those ideas through to the ﬁrst marketing or use of a technology) and diffusion (the spread of new technology across its potential market)” [77, p. 918]. According to Stoneman [76], as discussed in Mahdjoubi [42, p. 2], diffusion is the phase where the true impact of new technology occurs and thus “the measurement of impact is very much a measurement of how the economy changes as new technologies are introduced and used.” There are numerous studies dedicated to innovation diffusion across a population of adopters [4,29,46,67]. These studies either explain and expand-upon the S-curve diffusion pattern (Cumulative Normal Distribution [68]) consistently encountered when analysing the spread of innovation; or introduce diffusion models that “depict the successive increases in the number of adopters and predict the continued development of a diffusion process already in progress” [41, p. 2]. According to Geroski [20], there are two main types of diffusion models providing insights into the manner and speed of technology adoption — the epidemic model and the probit model. The ‘epidemic’ diffusion model attributes the diffusion of technology (software in particular) to a given population's knowledge of its existence; its comparative beneﬁts; and the spread of its use through word of mouth. As it focuses on a whole population of adopters, the epidemic model is interested in the gradual, unfolding impact of a new system/process on a market through its aggregate use. This contrasts with the ‘probit’ and ‘salience’ diffusion models which focus on the effect of individual decision-making on the spread of innovation [20, p. 614, 78]. This individual decision-making affecting diffusion follows three identiﬁable patterns — contagion, social threshold and social learning [97, p. 4]: ‘Contagion’ represents how an industry player (e.g., an engineering company) adopts an innovative system/process upon contact with another player who has already adopted it; ‘social threshold’ represents how an industry player adopts an innovative system/process when enough similar players have adopted it; and ‘social learning’ 5 To avoid conceptual overlap, the spread of a solution within an organisational unit will not be referred to as intra-diffusion but as improved implementation (or higher level of maturity) across the whole organisation. 67 represents how an industry player adopts an innovative system/process when enough proof is available of prior adopters ﬁnding it worth adopting. These inter-organisational diffusion patterns are further explained by DiMaggio and Powell [14] as reﬂecting two sets of isomorphic pressures — competitive and institutional. Competitive isomorphic pressures are market forces (e.g., supply and demand dynamics) driving organisations towards similarity; while institutional isomorphic pressures involve “organisational competition for political and institutional legitimacy as well as market position” [55, p. 657]. As discussed by DiMaggio and Powell [14], institutional pressures can be understood through their coercive, mimetic and normative effects. That is, organisations may adopt a speciﬁc system/process if it is coerced by either an organisation on which it depends, or the larger society it operates within [65]. It may also adopt the system/process by mimicking other successful organisations which have already adopted it [43]; or by following the industry's norms, standards and regulations [87] which clearly favour the new system/process. These diffusion models, patterns, and pressures have been shown to collectively describe and help predict the incremental diffusion of technological solutions across a population. However BIM is not solely an innovative technological solution proliferating incrementally across the construction industry [18,58,23] but a an organisational and systemic innovation [88] of complementary technologies, processes and policies. While BIM may be initially classiﬁed as a technical innovation [57], it will need to be urgently reclassiﬁed – upon its transformative adoption by organisations – as an organisational innovation characterised by the “generation, acceptance, and implementation of new ideas, processes, products or services” [62,90, p. 2]. As covered in depth in earlier research [85] and brieﬂy explored in Fig. 2, BIM adoption by an organisation pass through three adoption points pertaining to three capability stages. Even if multiple organisations pass through the ﬁrst Point of Adoption (PoA) separating pre-BIM status from minimum BIM capability (Stage 1), the spread of modelling practices among this population does not necessarily or automatically translate into a diffusion of multidisciplinary collaboration or interdisciplinary integration practices (Stages 2 and 3 respectively). Similarly, BIM is not a mere technological solution but reﬂects a combinatory and mutational diffusion of technologies, workﬂows and Fig. 3. The BIM framework conceptual reactor v1.0 (full size, current version). 68 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 1 Macro-BIM Adoption models, matrices and charts. Adoption model title A Diffusion Areas model (Fig. 4) B Macro-Maturity Components model (Fig. 6) C Macro-Diffusion Dynamics model (Fig. 7) Accompanying matrix or chart Intended use + applicable organisational scales (OScales) Diffusion Areas matrix (Table 2) + Diffusion Areas sample chart (Fig. 5) Macro-Maturity matrix (Table 11) Establish the diffusion areas to be assessed [Applicable at OScales 1–10] Assess the BIM maturity of countries holistically using a comparative matrix or granularly using component-speciﬁc metrics [Applicable at OScales 1–7] Assess and compare the directional pressures and mechanisms affecting how diffusion unfolds within a population [Applicable at OScales 1–7; another version at OScales 9–12] Identify, assess and compare the actions policy makers take (or can take) to facilitate market-wide adoption [Applicable across all OScales] Assess and compare the roles played by different stakeholder groups in facilitating diffusion within and across markets [Applicable at OScales 1–7; another version at OScales 9–12] Macro-Diffusion Dynamics matrix (Table 12) D Policy Actions model (Fig. 8) Policy Actions matrix (Table 13) + Policy Action Patterns sample chart (Fig. 9) E Macro-Diffusion Responsibilities matrix (Table 14) Macro-Diffusion Responsibilities model (Fig. 10) protocols [52,96]. This multi-stage, multi-component nature of BIM – resembling a complex adaptive system [28] – prevents the effortless application of technology-centric diffusion modelling and invites the development of more representative BIM adoption models. 1.5. Diffusion modelling and adoption models This paper differentiates between ‘diffusion modelling’ and ‘adoption models’. Diffusion modelling uses mathematical means to understand the “patterns innovations follow as they spread across a population of potential adopters over time” [17, p. 256]. It serves in understanding the social forces underlying technology diffusion [10]; predicting the diffusion of products across a market [40]; describing the time/speed of cumulative adoption of a speciﬁc innovation [24]; deciphering why some innovations are ‘imitated faster’ in some markets [45]; or establishing the impact regulation has on innovation diffusion [87]. Adoption models are conceptual structures describing how adoption – a term overlaying the deﬁnitions of implementation and diffusion – occurs across a population of organisations. Adoption models do not employ mathematical formulae to explain past or predict future diffusion patterns but use inductive inference to generate graphical representations that reduce topic complexity and promote understanding [53]. Each adoption model is formulated through a process of identiﬁcation, classiﬁcation and clustering, which simplify a large system by decomposing it into smaller sub-systems [54]. From a utilitarian perspective, adoption models provide a set of tools to assess and develop policies which encourage implementation and facilitate diffusion. Before introducing ﬁve macro-BIM adoption models, the next section clariﬁes the research methodology underlying their development. [50,51] and includes three repetitive stages – description, explanation and testing. First, the description stage develops a description of reality; identiﬁes phenomena; explores events; and documents ﬁndings and behaviours. According to Dubin [15, p. 85], “the more adequate the description, the greater is the likelihood that the units derived from the description will be useful in subsequent theory building.” Second, the explanation stage builds upon descriptions to infer a concept, a conceptual relationship or a construct; and then, develops a framework or a theory to explain and/or predict behaviours or events. In essence, the explaining stage develops a testable theoretical proposition which clariﬁes what has previously been described. Third, the testing stage inspects explanations and propositions for validity; tests concepts or their relationships for accuracy; and tests predictions against new observables. Each macro-BIM adoption model, presented in this paper, follows a similar cyclical path to that described by Meredith [50] — from describing; to explaining; to testing; and then back to describing. First, a description of each macro-BIM adoption model is generated through a process of inductive inference [53], conceptual clustering [54] and reﬂective learning [92,94]. Second, conceptual models are developed to 2. Research methodology This article is built-upon and further extends the BIM Framework [79] by employing existing conceptual constructs – terms, classiﬁcations, taxonomies, models and frameworks – to identify, explain and test new constructs. This cumulative theory-building exercise is summarised in the BIM Framework Conceptual Reactor (Fig. 3) incorporating the Normal Research Cycle by Meredith [50]. The conceptual reactor (Fig. 3) represents how the BIM framework can be continuously extended according to evolved research aims and objectives (input 1). By integrating existing conceptual structures (input 2) with new knowledge gained through literature reviews, and data collection (input 3), the reactor can then generate new conceptual structures (output) after passing through an iterative, three-stage theory-building process. This process has been identiﬁed by J. Meredith Fig. 4. Diffusion Areas model v1.0 (full size, current version). 69 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 2 Diffusion areas matrix (with sample granular metrics within each diffusion area). Technology Cumulative capability increase → Integration 3TE: Integration Technologies Rate of adoption of network-based interchange solutions (e.g., model servers); rate of proliferation of real-time network-based integration across disparate systems Collaboration 2TE: Collaboration technologies Rate of inter-organisational adoption of model-sharing software and middleware tools (e.g., Navisworks, Vico and Ecodomus) Modelling 1TE: Modelling technologies Rate of intra-organisational adoption of BIM software tools (e.g., Revit and Tekla) and their underlying hardware and network requirements Process Policy 3PR: Integration Processes Rate of adoption of integrated supply-chain processes across the whole supply chain; rate of proliferation of interdisciplinary workﬂows across all project life cycle phases 3PO: Integration Policies Rate of adoption of integrated supply-chain standards, protocols and contractual agreements; rate of proliferation of interdisciplinary educational programmes 2PR: Collaboration processes Rate of inter-organisational adoption of project BIM roles (e.g., Information Manager); rate of proliferation of multidisciplinary model-based workﬂows; rate of proliferation of new collaboration-centric business models 1PR: Modelling processes Rate of intra-organisational BIM roles (e.g., model manager, and BIM trainer) and model-based workﬂows 2PO: Collaboration policies Rate of inter-organisational adoption of modelling standards and collaboration protocols; rate of proliferation of collaboration-centric contractual agreements and educational programmes 1PO: Modelling policies Rate of intra-organisational adoption of modelling standards (e.g., naming standards, shared parameters, level of details, and property sets) and ﬁle exchange protocols visually explain the knowledge structures. Third, each model is tested through either a focus group, peer-review or questionnaire. The conceptual reactor with its core three-stage approach reﬂects the researchers' underlying retroductive research strategy which follows a similar three-step approach. First, “the research starts in the domain of actual, by observing connections between phenomena […]. To do so, as a second step, researchers build a hypothetical model, involving structures and causal powers located in the domain of real, which, if it were to exist and act in the postulated way, would provide a causal explanation of the phenomena in question. The third step is to subject the postulated explanation to empirical scrutiny” [35, p. 635]. This retroductive research strategy represents a “logic of enquiry associated with the philosophical approach of Scientiﬁc Realism” Blaikie [6, p. 108]. Similar to deductive research, retroduction “starts with an Fig. 5. Diffusion Areas Comparison sample chart v1.0 (full size, current version). 70 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 includes market subdivisions, sectors, industries and specialities (OScales 1–7); the Meso-cluster includes project-centric organisational teams (OScale 8); and the Micro-cluster includes organisational subdivisions, groups, and individuals (OScales 9–12). Although the models proposed are applicable at a number of organisational scales, the focus of this paper is exclusively on BIM adoption at the macro-cluster, and speciﬁcally at OScale 2 (deﬁned markets or countries). 3. New model list After clarifying the terminology used in this research, and identifying the methodology adopted in generating new conceptual constructs, this section introduces ﬁve macro-BIM adoption models (Table 1). 3.1. Model A: diffusion areas Fig. 6. The Macro-Maturity Components model v1.2 (full size, current version). Earlier versions of this model were presented in 2011–2013 at a number of industry events. This version was ﬁrst published as Item 26 on the BIM Framework blog — July 20, 2014. Component numbers (I–VIII) has since been added to improve visual clarity; they are not intended to imply priority or precedence. observed regularity but seeks a different type of explanation”. Through retroduction, events are explained by postulating and identifying structures and causal powers capable of generating them [72]; and by locating the “real underlying structure or mechanism that is responsible for producing the observed regularity” [6, p. 25]. Retroduction uses “creative imagination and analogy to work back from data to an explanation” and involves the “building of hypothetical models as a way of uncovering the real structures and mechanisms which are assumed to produce empirical phenomena” [6, p. 25]. In constructing these hypothetical models, ideas are “borrowed from known structures and mechanisms in other ﬁelds” [2, p. 2]. Models are clarity-improvement tools. By generating adoption models, this paper thus introduces an artiﬁcial reconstruction of reality [51, p. 307], a hypothesis to be used in assessing and comparing BIM implementation/diffusion across countries. 2.1. Built-in research limitations BIM implementation and diffusion can be analysed across varied organisational scales. In previous papers [81,80], we have identiﬁed twelve organisational scales (OScales) spread across three organisational clusters. These scales and clusters are intended to balance the dual notions of ﬂexibility, to cater for the uniqueness of each OScale; and uniformity, to cater for the similarity between them. The Macro-cluster This macro-adoption model clariﬁes how BIM ﬁeld types (technology, process and policy) interact with BIM capability stages (modelling, collaboration and integration) to generate nine areas for targeted BIM diffusion analysis and BIM diffusion planning (Fig. 4). The nine diffusion areas, explored in Table 2, can be assessed independently or collectively. For example, the diffusion of BIM software tools within a population (modelling technologies [1TE]) can be assessed separately, and using different assessment methods, than establishing the proliferation of integrated project delivery contracts (integration policies [3PO]). Also, the diffusion of multidisciplinary BIM educational curricula (collaboration policies [2PO]) can be assessed separately, or in combination with, the proliferation of collaborative BIM roles and responsibilities (collaboration processes [2PR]). The nine diffusion areas, their structured subdivisions and combinations, provide an opportunity for granular assessments of BIM diffusion within a population of adopters. Rather than being treated uniformly as a single set of data, or separated into disparate topics without an underlying conceptual structure, the Diffusion Areas' model (Fig. 4) allows the generation of targeted ratings for comparative market analysis — as exempliﬁed in Fig. 5. 3.2. Model B: macro-maturity components The macro-maturity components model identiﬁes eight complementary components for measuring and establishing the BIM maturity of countries and other macro-organisational scales: Objectives, stages and milestones; Champions and drivers; Regulatory framework; Noteworthy publications; Learning and education; Measurements and benchmarks; Standardised parts and deliverables; and Technology infrastructure (Fig. 6). Macro-maturity components are assessed using the BIM Maturity Index (BIMMI) which includes ﬁve maturity levels: [a] Ad-hoc or low maturity; [b] Deﬁned or medium–low maturity; [c] Managed or medium maturity; [d] Integrated or medium–high maturity; and [e] Optimised or high maturity [81]. When applying the BIMMI, assessments can be made holistically (low detail ‘discovery’ assessments) or granularly (higher detail ‘evaluation’ assessment). ‘Discovery’ assessments are beneﬁcial for comparing the relative maturity of each macro-component against the Table 3 Availability of capability stages to guide market adoption metric. a (low) b (medium–low) There are no capability stages Capability stages are deﬁned yet lack separating lack of ability internal consistency or well-deﬁned from heightened boundaries (overlap with each other) proﬁciency c (medium) d (medium–high) Capability stages are Capability stages are well-deﬁned and consistent yet are not integrated integrated with objectives and with objectives and milestones milestones e (high) Capability stages are dynamically optimised in response to changes in other macro-maturity components Other granular metrics include: the availability of long-term objectives to guide market adoption; and the availability of maturity milestones to guide market adoption. 71 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 4 Availability of a policy driver metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) There is no designated policy driver; market may include volunteer champions There is a designated policy driver; driver may not be inﬂuential or is not supported by a clear mandate The designated driver is inﬂuential with a clear wide-reaching mandate Designated driver's activities are integrated with other macro-components Driver's role no longer required due to system/process infusion across the market Other granular metrics include: driver inﬂuence; driver mandate clarity; driver competency; and leadership style. Table 5 Procurement policy metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) Procurement policies do not include any requirements for digital workﬂows or model-based deliverables Procurement policies include basic requirements for digital workﬂows and model-based deliverables Procurement policies include detailed requirements for digital workﬂows and model-based deliverables Model-based deliverables and digital workﬂows are integrated into all procurement policies Procurement policies are continuously optimised to reﬂect industry best practices for model-based deliverables and digital workﬂows Other granular metrics include: contractual coverage of digital workﬂows and model-based deliverables; extent of handover protocols for information-rich models; and proliferation of integrated project delivery. Table 6 Noteworthy publications relevance metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) The noteworthy publication includes out-dated information which is no longer usable or useful The noteworthy publication is relevant, current and contains actionable information The noteworthy publication is highly-relevant, well-cited and well-used in comparison to other similar-topic NBPs The noteworthy publication is authoritative and impactful and considered a reference (among other references) The noteworthy publication is the most authoritative document covering a speciﬁc topic Other granular metrics include: distribution of noteworthy publications according to knowledge clusters and labels. Table 7 BIM infusion into tertiary curricula metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) BIM is not included in the curricula BIM is taught in separate learning unit(s) or introduced into existing units without altering their formal (pre-set) delivery structures or pre-BIM learning objectives Unit structure(s) and learning objectives are formally altered to accommodate BIM tools, workﬂows and deliverables Unit structure(s) and learning objectives are integrated with and complementary to all other BIM-infused units BIM tools and workﬂows are inseparable from the unit's structure and learning objectives Other granular metrics include: multi-disciplinary integration of curricula; use of simulated design, construction and operation environments; and expertise of learning providers. other seven components — as represented by the Macro-Maturity Matrix (Table 11); while ‘evaluation’ assessments allow the detailed analysis of each component using specialised metrics only applicable to that component. Below is explanation of the eight macro-maturity components including sample granular component-speciﬁc metrics (Tables 3–10). 3.2.1. Objectives, stages and milestones This component represents the availability of clear BIM-speciﬁc policy objectives, intermediate capability stages, and measureable maturity milestones separating current status from a quantiﬁable future target. BIM policy objectives, stages and milestones may exist separately or found embedded within a country's wider construction strategy. For the purposes of macro-maturity assessment, more-granular metrics can be used to evaluate objectives within their respective contexts, analyse the clarity of pre-determined stages, and compare the duration/effort separating different milestones (Table 3). system/process to a population of potential adopters. As early adopters [67], champions can be individuals promoting a new software solution; a community of practice promoting a new process; or an industry association promoting a new standard. While champions are ‘volunteer experimentalists’, drivers are ‘designated executors’ of a top-down strategy (refer to Fig. 7) with a mandate to stimulate the adoption of a designated technology, process or policy. Drivers may be individuals, groups, institutions or an authority intent on communicating, encouraging and monitoring the adoption of a system/process (refer to Fig. 8). The positive impacts of champions/drivers on innovation have been explored in numerous studies [7,27,59,67] especially if they exhibit clustering and reach characteristics [73]. For the purposes of macro-maturity assessment, the availability of champions/drivers within a market signals higher maturity when compared to markets lacking champions/ drivers, or where champions/drivers do not exhibit clustering and reach characteristics. Additional granular metrics can be used to evaluate the competency of individual drivers [86] or the championship/ leadership style across markets (Table 4). 3.2.2. Champions and drivers This component represents the individuals, groups and organisations undertaking the task of demonstrating the efﬁcacy of an innovative 3.2.3. Regulatory framework This component describes the contractual environment, intellectual property rights, and professional indemnity insurance underlying 72 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 8 Project performance benchmark metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) There are no common or mandated project performance benchmarks Project performance benchmarks are deﬁned/agreed by industry associations or mandated by regulatory bodies Project performance benchmarks are centrally collated and accessed by stakeholders Project performance benchmarks are integrated with other organisational and team benchmarks Project performance benchmarks are continuously optimised to reﬂect emergent technologies, workﬂows and protocols Other granular metrics include: organisational capability benchmarks and individual competency benchmarks. Table 9 Availability of an elemental classiﬁcation system metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) There are no market-speciﬁc elemental classiﬁcation system There is a number of market-speciﬁc elemental classiﬁcation system A uniﬁed elemental classiﬁcation system is standardised and centrally managed by a dedicated authority The standardised elemental classiﬁcation system is integrated with software tools and speciﬁcation/costing regimes The standardised elemental classiﬁcation system is continuously reviewed and optimised to reﬂect international best practices Other granular metrics include: availability of national object libraries and availability of standardised model uses. Table 10 Central model repository metric. a (low) b (medium–low) c (medium) d (medium–high) e (high) There is no central repository for data-rich 3D models There is an optional or feature-poor central repository for data-rich 3D models There is a central and mature system for submitting and querying data-rich 3D models The central model repository is integrated with multiple data sources, infrastructure models, procurement systems, ﬁrst responders and the internet of things (IoT) The central model repository is continuously optimised to improve stakeholder accessibility and allow innovative uses Other granular metrics include: data openness requirements; availability of e-submission systems; and software availability and affordability. Fig. 7. Macro-Diffusion Dynamics model v1.1 (full size, current version). An earlier version of this model was ﬁrst published as Episode 19 on BIMThinkSpace.com — July 12, 2014. B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 73 For the purposes of macro-maturity assessment, this component clariﬁes whether digital workﬂows and model-based deliverables are included as learning topics within education/training programmes. Additional metrics can be used to evaluate how BIM concepts, tools and workﬂows are infused into curricula [25, p.8]; if varied learning requirements of professionals, paraprofessionals and tradespeople are met [1]; and whether these learning/education resources are affordable and accessible (Table 7). 3.2.6. Measurements and benchmarks This component represents market-wide metrics for benchmarking project outcomes and assessing the capabilities of individuals, organisations and teams. The availability of market-speciﬁc – or the formal adoption of international – benchmarks and metrics signiﬁes a market's ability to assess and potentially improve its performance. Additional granular metrics are proposed in Table 8. Fig. 8. Policy Actions model v1.4 (full size, current version). collaborative BIM projects. Information-rich, model-based deliverables require more detailed contractual, project and process management protocols than their pre-BIM counterparts. Responsibilities pertaining to shared models (e.g., elemental authorship and model ownership), collaborative processes (e.g., overlapping project phases and early involvement of subcontractors), and prescriptive protocols (e.g., data exchange structures and information delivery standards) add layers of complexity to team interactions. This complexity and varied risk environment can be mitigated by the availability of a regulatory framework clarifying the rights, responsibilities and liabilities of varied project stakeholders across overlapping – and even concurrent – project lifecycle phases. For the purposes of macro-maturity assessment, the availability of a regulatory framework – addressing procurement, workﬂows, deliverables, and stakeholder rights – signals higher maturity. More-granular metrics can be used to evaluate the proliferation of these sub-components across markets (Table 5). 3.2.4. Noteworthy publications This component represents publically-available documents of relevance, developed by inﬂuential industry stakeholders, and intended for a market-wide audience. As covered in detail in Kassem et al. [31, 82], noteworthy BIM publications (NBP)s pertain to three knowledge content clusters (guides, protocols and mandates) and eighteen knowledge content labels (e.g., report, manual, and contract). For the purposes of macro-maturity assessment, this component clariﬁes the availability of noteworthy BIM publications within a speciﬁc market as a sign of maturity. Additional metrics can be used to evaluate the distribution of NBPs according to knowledge clusters/labels or the relevance of each NBP when compared to similar publications from other markets (Table 6). 3.2.7. Standardised parts and deliverables This component represents the standardised, data-rich model parts6 (e.g., walls, beams, HVAC units, doors and furniture) which populate object-based models. It also represents model uses,7 the standardisable deliverables from generating, collaborating-on and linking objectbased models to external databases. For the purposes of macro-maturity assessment, the availability of standardised parts and deliverables signals a mature market. Additional granular metrics are proposed in Table 9. 3.2.8. Technology infrastructure This component refers to the availability, accessibility and affordability of hardware, software and network systems [12]. It also refers to the availability, usability, connectivity and openness of information systems hosting data-rich three-dimensional models. Additional granular metrics are proposed in Table 10. 3.2.8.1. Macro-maturity matrix. The macro-maturity matrix (Table 11) provides a summary of the eight macro-maturity components (Fig. 6) mapped against the ﬁve level of the BIM maturity index. The macro-maturity matrix (Table 11) can be used in identifying the comparative BIM maturity across markets. The matrix aggregates a number of sub-topics within each component and is thus suitable for low-detail ‘discovery’ assessment (Granularity Level 1), where the contents of each cell represents – partially or fully – the current maturity status. More detailed ‘evaluation’ assessments (Granularity Level 2)8 require the integration of a large number of metrics unique to each component (refer back to Tables 3–10). The macro-maturity components identify the areas to be addressed by stakeholder groups (refer to Model E). While each component can be measured using the ﬁve-level index, it can also be transformed into a set of development activities that can be targeted for completion by policy makers.9 3.3. Model C: macro-diffusion dynamics According to Geroski [20, p. 621], “the real problem may not be understanding how the process of diffusion unfolds, but understanding how it starts”. To allow a clearer understanding of from-where and how a diffusion starts to unfold within a population, this macro6 3.2.5. Learning and education This component represents market-wide educational activities covering BIM concepts, tools and workﬂows. These educational activities are either delivered through tertiary education, vocational training or professional development; either as competency-based or coursebased learning models [93,84]. Also typically referred to as elements, components, objects or families. Model uses can be speciﬁc to the design phase (e.g., immersive environments), construction phase (e.g., construction logistics and ﬂow), operation phase (e.g., asset tracking), or across all project lifecycle phases (e.g., cost-planning and lean modelling). 8 The varied applications of the four granularity levels and their applicability across organisational scales have been discussed in detail in Succar [81], Table 2. 9 For an example of a similar approach, please refer to http://www.bimtaskgroup.org/ work-streams-wps/. 7 74 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 11 Macro-maturity matrix at granularity level 1. a b c d Low maturity Medium–low maturity Medium maturity Medium–high maturity High-maturity BIM objectives and stages are continuously reﬁned to reﬂect advancements in technology, facilitate process innovation, and beneﬁt from international best practices Driver(s) role is diminished, replaced by optimised systems, standards and protocols The regulatory framework is continuously reﬁned to reﬂect technological advancements and optimised collaborative workﬂows NBPs are continuously optimised to reﬂect international best practices I Objectives, stages and milestones There are no market-scale BIM objectives or well-deﬁned BIM implementation stages or milestones There are well-deﬁned macro-BIM objectives, implementation milestones and capability stages BIM objectives, stages and milestones are centrally managed and formally monitored BIM objectives and stages are integrated into policies, processes and technologies and manifest themselves within all other macro-maturity components II Champions and drivers III Regulatory framework There are no identiﬁable market-wide champions or BIM implementation drivers There is no formal BIM-era regulatory framework There are one or more volunteer champions and/or informal BIM drivers operating across the market There is a formal regulatory framework addressing basic BIM-era rights and responsibilities of a number of stakeholders There is a uniﬁed task group or committee driving BIM implementation/diffusion across the market The formal regulatory framework covers all BIM-era rights and responsibilities of all stakeholders Driver(s) coordinate all macro-adoption activities, minimise activity overlaps, and address diffusion gaps The regulatory framework is integrated into all requirements, roles, processes and deliverables IV Noteworthy publications There are no – or a small number of – noteworthy BIM publications (NBPs) across the market V Learning and education There are many NBPs with overlapping knowledge content; some NBPs are redundant or collectively include knowledge gaps BIM learning topics are identiﬁed and introduced into education/training programmes; BIM learning providers are available across a number of disciplines and specialties NBPs are developed and/or coordinated by a single entity thus minimising overlaps and knowledge gaps BIM learning topics are mapped to current and emergent roles; BIM learning providers deliver accredited programmes across disciplines and specialties NBPs are authoritative, interconnected and integrated across project life cycle phases and the whole construction supply chain BIM learning topics are integrated across educational tiers (tertiary, and vocational) and address the learning requirements of all industry stakeholders Formal metrics are used to benchmark project outcomes and assess the abilities of individuals, organisations and teams across the market Standardised metrics are used to centrally benchmark project outcomes; certify the abilities of individuals, organisations and teams; and accredit learning programmes, software systems and project delivery mechanisms Standardised object libraries are available and used; service delivery model uses and operational data requirements are formally deﬁned and used across all project lifecycle phases Standardised metrics and benchmarks are integrated into project requirements, workﬂows and deliverables; consistently used in deﬁning and procuring services; and used to prequalify the abilities of individuals, organisations and teams Standardised object libraries, service delivery model uses, and operational data requirements are integrated into, procurement mechanisms, project workﬂows and lifecycle facility operations The technology infrastructure is of high quality and affordability enabling the efﬁcient exchange, storage and management of complex, federated models among dispersed project teams The technology infrastructure is uniformly accessible and interoperable allowing real-time network-based integration across disparate systems and data networks VI VII BIM learning topics are neither identiﬁed nor included within legacy education/training programmes; learning providers lack the ability to deliver BIM-infused education Measurements There are no market-wide and metrics applied in benchmarks measuring BIM diffusion, organisational capability or project performance Standardised parts and deliverables VIII Technology infrastructure There no market-speciﬁc object libraries (e.g., doors and windows); service delivery model uses (e.g., clash detection) and operational data requirements (e.g., COBie) Object libraries are available yet follow varied modelling and classiﬁcation norms; service delivery model uses and operational data requirements are informally deﬁned and partially used Non-existent, inadequate or unaffordable technology infrastructure (software, hardware and networks) as to prohibit widespread BIM adoption The technology infrastructure is of adequate quality and affordability to enable BIM implementation within organisations and diffusion across varied market sectors e BIM learning topics are infused (not separately identiﬁable) into education, training and professional development programmes Standardised metrics are continuously revised to reﬂect evolving accreditation requirements and international best practices Standardised object libraries, service delivery model uses and operational data requirements are continuously optimised and realigned to improve usage, accessibility, interoperability and connectivity The technology infrastructure is intuitive and ubiquitously accessible allowing seamless interchange between all users, virtual systems and physical objects across the whole lifecycle Table 12 Macro-diffusion dynamics matrix. Diffusion dynamic Macro-actor, transmitter Pressure mechanism Pressure recipient, potential adopter Isomorphic pressure type Top-down Government or regulatory body Downwards Coercive; normative Middle-out Large organisation or industry association All stakeholders falling within the circle of inﬂuence of the authority exerting pressure Governments and authorities in other markets Smaller organisations further down the supply chain; members of industry associations Governments and regulatory bodies within the market Other large organisations and industry bodies within or outside the market Larger organisations and industry bodies Other small organisations Horizontal Downwards Upwards Horizontal Bottom-up Small organisation Upwards Horizontal mimetic Coercive; normative; mimetic Normative Mimetic; normative Normative Mimetic; normative 75 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 Table 13 Policy actions matrix. Approaches [1] Passive Activities [A] Make aware: the policy player informs stakeholders Communicate about the existence of, or the importance, challenges and business beneﬁts of, a system/process through formal and informal communications [B] Engage Encourage: the policy player conducts workshops and networking events to encourage stakeholders to adopt the system/process [C] Monitor Observe: the policy player observes as (or if) stakeholders have adopted the system/process [2] Active [3] Assertive Educate: the policy player generates informative guides to educate stakeholders of the speciﬁc deliverables, requirements and workﬂows of the system/process Incentivise: the policy player provides rewards, ﬁnancial incentives and preferential treatment to stakeholders adopting the system/process Prescribe: the policy player details the exact system/process to be adopted by stakeholders Track: the policy player surveys, tracks and scrutinises how/if the system/process is adopted by stakeholders adoption model identiﬁes three diffusion dynamics — top-down, bottom-up and middle-out (Fig. 7). The three diffusion dynamics introduced in Fig. 7 embody horizontal and vertical mechanics, and a combination of isomorphic pressures – coercive, mimetic and normative – allowing innovation to contagiously pass from ‘transmitters’ to adopters [78,14,11]. Horizontal mechanisms represent the mimetic effects organisations have on their peers; while vertical mechanisms represent the upward and downward pressures (normative and coercive) organisations have on non-peer organisations across the supply chain. These dynamics, mechanics and pressures are combined in Table 12. Enforce: the policy player includes (favours) or excludes (penalises) stakeholders based on their respective adoption of the system/process Control: the policy player establishes ﬁnancial triggers, compliance gates and mandatory standards for the prescribed system/process The three dynamics discussed in Table 12 identify the how the adoption decision taken by one player inﬂuences the adoption decisions of other players. For example, the early adoption of a policy player (an authority) of an innovative policy in one market encourages later adopters to make “the same choices as early adopters without having gone through the same investment in learning by experience” [20, pp. 618– 619,74], a process often referred to as the ‘information cascade’ or ‘bandwagon effect’ [20,43]. As explored by Simmons and Elkins [74, p. 174], policy players of a speciﬁc market “pay deliberate attention to foreign models and their outcomes […as…] foreign models can encourage or expedite adoption by inserting a policy innovation on a Fig. 9. Policy action patterns sample chart v1.1 (full size, current version). 76 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 The three approaches within each activity signify an increase in the intensity of policy maker's involvement in facilitating BIM adoption, from a passive stance to more assertive actions. Also, the three activities signify a progression from clarifying the availability, beneﬁt or necessity of a new system/process, to assessing adoption behaviours, challenges and outcomes. Each of the nine resulting policy actions can be further divided into smaller policy tasks. For example, the incentivise action [B2] can be subdivided into incentivise tasks – make tax regime favourable for BIM adoption, develop a BIM procurement policy, and introduce BIMfocused funding [8,9] – that can be undertaken by policy makers. The nine actions can also be applied to the eight Macro-Maturity Components (refer back to Model B). That is, a policy maker my Educate [A2] stakeholders of the need for – or Prescribe [A3] the necessity of – Measurements and Benchmarks (Component VI). It may also Track [C3] the development of Educational Curricula (Component V), or Enforce [B3] the use of standardised parts and deliverables (Component VII). These activities, actions and tasks can be used as a template to structure a policy intervention, or as an assessment tool to compare policy actions across different countries (Fig. 9). The Policy Action Patterns sample chart (Fig. 9) allows a quick comparison of diffusion actions undertaken by policy makers in different markets. Fig. 10. Macro-Diffusion Responsibilities model v1.1 (full size, current version). legislature's agenda. A foreign model may also offer a ready-made answer to ill-deﬁned domestic pressure for “change” and “innovation.” Or it may legitimate conclusions or predispositions already held or add a decisive data point in the evaluation of alternatives [5].” That is, the adoption of a BIM diffusion policy by one authority within a speciﬁc market may result – through mimetic and normative pressures – in the adoption of similar BIM diffusion policies by other authorities in different markets. These top-down, bottom-up and middle-out dynamics are not independent: the diffusion of innovation at the lower-end of the supply chain (e.g., within smaller organisations) will lead to the development of a diffusion phenomenon at the macro-scale. Similarly, the diffusion of innovation at the higher-end of the supply chain will inﬂuence the behaviour of smaller organisations and individuals operating at the micro-scales [68, p. 13,28]. 3.4. Model D: policy actions Information provision by policy makers to a target population of potential adopters – highlighting the advantages of an innovative system/ process – will not necessarily encourage implementation or speed-up diffusion [77]. However, policy makers may affect the adoption of an innovative solution through “a judicious mix of information provision and subsidies” [20, p. 621]. This macro-adoption model focuses on the actions a policy maker takes to inﬂuence the market-wide adoption of an innovative system/ process. The Policy Actions model (Fig. 8) identiﬁes three implementation activities (communicate, engage, monitor) mapped against three implementation approaches (passive, active and assertive) to generate nine policy actions. The Policy Actions model (Fig. 8) identiﬁes nine actions (squares) and represents the relation between them (directional arrows and dotted connecting lines).10 The policy actions are brieﬂy explained in Table 13. 10 The Policy Implementation Actions model is a visually-enhanced ‘concept map’ with concepts represented as squares, relations represented as dotted lines, and textual labels clarifying the ontological relation between concepts [89,26]. That is, action A1 (Make Aware) is followed by either A2 (Educate), B1 (Encourage) or B2 (Incentivise). To disallow a counter-intuitive bottom-up use of this model, top-down and horizontal–diagonal arrows are added. For more information covering how concept maps are used to graphically represent BIM Framework parts, please refer to Succar [79, p. 368]. 3.5. Model E: macro-diffusion responsibilities This macro-adoption model (Fig. 10) analyses BIM diffusion through the roles played by industry stakeholders as a network of actors [37,91]. It ﬁrst identiﬁes nine BIM player groups (stakeholders) distributed across three BIM ﬁelds (technology, process and policy) as deﬁned within the BIM framework [79]. The nine player groups are: policy makers, educational institutions, construction organisations, individual practitioners, technology developers, technology service providers, industry associations, communities of practice, and technology advocates. The nine player groups11 belong to either BIM ﬁeld or their overlaps. Table 14 provides a succinct description of each player group followed by how this subdivision can be used in evaluating BIM diffusion within and across different markets. Each of the nine player groups identiﬁed in Fig. 10 includes a number of player types. For example, player group 3 (construction organisations) is composed of varied player types including: asset owners, architects, engineers and project managers. Also, player group 4 (individual practitioners) is composed of professionals, associated professionals and tradespeople. These distinctions between player groups, player types and unique players (e.g., a speciﬁc person, group, association, company or university) allow the targeted assessment and comparison of stakeholders' involvement. For example, this macro-adoption model can be used to: • compare the BIM diffusion activities of one player group to other groups within the same market. An example assessment question would be: “Which player group played a more leading BIM diffusion role in ‘Country A’: Education Institutions or Industry Associations?” • compare the BIM diffusion activities of two or more player types within the same player group. For example: “How does the role played by asset owners in BIM diffusion differ from the role played by large contractors?” • compare the BIM diffusion activities of players pertaining to the same player type across different markets. For example: “Is the BIM diffusion role played by large contractors in ‘Country A’ similar to the role played by large contractors in ‘Country B’?” • isolate BIM players by their group/type and analyse their BIM diffusion 11 Pending further research, the tenth player group at the intersection of the three ﬁelds is intentionally excluded from this model. B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 77 Table 14 Macro-diffusion responsibilities matrix (player groups with sample players — market scale). Policy ﬁeld Process ﬁeld Technology ﬁeld 1 Policy makers Authorities involved in mandating, regulating or facilitating the adoption of innovative systems/ processes across an industry or whole markets e.g., the BIM Task Group in the UK and BCA in Singapore 2 Educational institutions Universities and other learning institutions developing and/or delivering educational programmes and related material 3 Industry organisations Organisational players involved in deploying innovative systems/ processes for commercial advantage e.g., AECOM and Multiplex 4 Individual practitioners Practitioners (including students/trainees) involved in learning or applying innovative systems/ processes 5 Technology developers Software, hardware and network solution providers with offerings targeted at whole industries or speciﬁc sectors, disciplines and specialties e.g., Autodesk, Leica and Acconex 6 Technology service providers Commercial companies bridging the sales/services gap between technology providers and end users Policy-process overlap Process-technology overlap Policy-technology overlap 7 Industry associations Associations representing the interests of their individual/ organisational members within a speciﬁc industry, sector, discipline or speciality e.g., AIA, ACIF and APCC 8 Communities of practice An informal grouping of individual practitioners with a common interest in a speciﬁc software, hardware or network solution e.g., Revit user groups and SmartGeometry 9 Technology advocates A formal grouping of individuals and organisations focused on the development/ promotion of technology-centric standards and policies e.g., buildingSmart and Australian Computer Society activities. For example: “What is the role played by Industry Association X in facilitating BIM diffusion within its membership base?” and diffusion strategies; policy makers can use these concepts and knowledge tools to either assess their ongoing BIM adoption efforts or to structure the development of new ones. 4. Conclusion This paper introduced numerous new concepts, models and decision support tools for macro-BIM adoption assessment and planning. It ﬁrst presented a number of delineations between readiness, capability and maturity; between implementation and diffusion; and between diffusion modelling and adoption models. Second, it introduced the Point of Adoption (PoA) concept and linked it to previous BIM capability/ maturity research. Third, it clariﬁed the research methodology, introduced the BIM Framework conceptual reactor, and discussed the research's underlying retroductive strategy. Fourth, it extended the BIM Framework by introducing ﬁve new adoption models, matrices and charts applicable across multiple organisational scales (Table 1): Model A identiﬁed nine areas for targeted BIM diffusion assessment and planning; Model B introduced eight components and a number of granular metrics for assessing and comparing the BIM maturity of countries; Model C identiﬁed three directional dynamics that clarify how diffusion unfolds within a market; Model D deﬁned three activities, three approaches and nine actions for assessing, comparing and planning adoption policies across markets; and Model E deﬁned nine groups to be used in analysing the diffusion activities/roles played by industry stakeholders. Based on the above deliverables, this research – presented in two complementary papers – contributes to domain knowledge by: • setting the scene for macro-BIM adoption assessment based on an established framework with a large set of interconnected terms, classiﬁcations, taxonomies and models; • refocusing the discussion away from software acquisition/implementation as a singular criterion for BIM diffusion surveys and studies; • overlaying the concepts of BIM implementation and BIM diffusion into a single term thus generating a uniﬁed view (Fig. 2) for establishing and comparing the readiness, capability and maturity of organisations; • introducing ﬁve macro-adoption models, their companion matrices and charts to be used in assessing and comparing BIM adoption across countries; • identifying multiple avenues for domain researchers to adapt, improve or correlate adoption models; each model represents a separate opportunity for data collection and additional conceptual investigation; and • informing the development of country-speciﬁc BIM implementation Research is currently being conducted to apply these concepts and tools across a number of countries. The results of these applications, and the conceptual calibrations that ensue, will be published in an upcoming paper. The deliverable of this research will instigate discussions among policy makers, encourage additional BIM implementation/ diffusion research, and hopefully contribute to the improvement of BIM adoption policies across a number of markets. References [1] AIA-CA, BIM in Practice — BIM Education, a Position Paper by the Australian Institute of Architects and Consult Australia, http://www.bim.architecture.com.au/2012. [2] D. Atkinson, Approaches and Strategies of Social Research, Essay for Reasearch Methods Class, RMIT, http://minyos.its.rmit.edu.au/~dwa/Methods.html2011. [3] Autodesk, Autodesk Survey Shows Building Information Modelling Continues to Gain Momentum Across Europe. News Room, http://bit.ly/AutodeskSurvey20112011. [4] F.M. Bass, Comments on “a new product growth for model consumer durables the bass model”, Manag. Sci. 50 (12_Suppl.) (2004) 1833–1840. [5] C.J. Bennett, How states utilize foreign evidence, J. Public Policy 11 (01) (1991) 31–54, http://dx.doi.org/10.1017/S0143814X0000492X. [6] N.W.H. Blaikie, Designing Social Research: The Logic of Anticipation, Polity Press, 2000. [7] B.A.G. Bossink, Managing drivers of innovation in construction networks, J. Constr. Eng. Manag. 130 (3) (2004) 337–345, http://dx.doi.org/10.1061/(ASCE)07339364(2004)130:3(337). [8] J. Boya, Q. Zhenqiang, J. Zhanyong, Based on game model to design of building information modeling application policy, Paper Presented at the Fifth International Conference on Intelligent Systems Design and Engineering Applications (ISDEA), Zhangjiajie, China, 2014. [9] M. Braddeley, C.-Y. Chang, Collaborative Building Information Modelling (BIM), Insights from Behavioural Economics and Incentive Theory, Royal Institute of Chartered Surveyors (RICS), 2015. [10] J.C. Brancheau, J.C. Wetherbe, The adoption of spreadsheet software: testing innovation diffusion theory in the context of end-user computing, Inf. Syst. Res. 1 (2) (1990) 115–143. [11] D. Cao, H. Li, G. Wang, Impacts of isomorphic pressures on BIM adoption in construction projects, J. Constr. Eng. Manag. (2014) 04014056, http://dx.doi.org/10.1061/ (ASCE)CO.1943-7862.0000903. [12] CID, Readiness for the Networked World: A Guide for Developing Countries, Information Technologies Group, Center for International Development, Information Technologies Group, Harvard University, Cambridge, MA, 2014. [13] R.B. Cooper, R.W. Zmud, Information technology implementation research: a technological diffusion approach, Manag. Sci. 36 (2) (1990) 123–139, http://dx. doi.org/10.1287/mnsc.36.2.123. [14] P.J. DiMaggio, W.W. Powell, The iron cage revisited: institutional isomorphism and collective rationality in organizational ﬁelds, Am. Sociol. Rev. (1983) 147–160. [15] R. Dubin, Theory Building, Free Press, New York, 1978. [16] K.J. Fadel, User adaptation and infusion of information systems, J. Comput. Inf. Syst. 52 (3) (2012) 1. [17] R.G. Fichman, C.F. Kemerer, The illusory diffusion of innovation: an examination of assimilation gaps, Inf. Syst. Res. 10 (3) (1999) 255–275, http://dx.doi.org/10.1287/ isre.10.3.255. 78 B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 [18] S. Fox, J. Hietanen, Interorganizational use of building information models: potential for automational, informational and transformational effects, Constr. Manag. Econ. 25 (3) (2007) 289–296. [19] T. Froise, W. Shakantu, Diffusion of innovations: an assessment of building information modelling uptake trends in South Africa, http://bit.ly/1v62nSH2014. [20] P.A. Geroski, Models of technology diffusion, Res. Policy 29 (4) (2000) 603–625. [21] B. Gilligan, J. Kunz, VDC use in 2007: signiﬁcant value, dramatic growth, and apparent business opportunity, TR1712007. 36. [22] M. Gladwell, The Tipping Point: How Little Things Can Make a Big Difference, Abacus, London, 2001. [23] N. Gu, K. London, Understanding and facilitating BIM adoption in the AEC industry, Autom. Constr. 19 (8) (2010) 988–999, http://dx.doi.org/10.1016/j.autcon.2010.09. 002. [24] V. Gurbaxani, Diffusion in computing networks: the case of BITNET, Commun. ACM 33 (12) (1990) 65–75, http://dx.doi.org/10.1145/96267.96283. [25] HEA, Embedding Building Information Modelling (BIM) Within the Taught Curriculum, Higher Education Academy, BIM Education Forum, York, United Kingdom, 2013. [26] R.R. Hoffman, G. Lintern, Eliciting and Representing the Knowledge Experts, The Cambridge Handbook of Expertise and Expert Performance, Cambridge University Press, New York, 2006. [27] J.M. Howell, C.A. Higgins, Champions of technological innovation, Adm. Sci. Q. 35 (2) (1990) 317–341. [28] S. Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software, Simon and Schuster, 2002. [29] S. Kale, D. Arditi, Innovation diffusion modeling in the construction industry, J. Constr. Eng. Manag. 136 (3) (2010) 329–340, http://dx.doi.org/10.1061/(ASCE)CO. 1943-7862.0000134. [30] M. Kassem, S. Leusin, (Producer). (2014, October 16, 2014). Strategy for the diffusion of building information modelling in Brazil, experiences exchange in BIMbuilding information modelling (Apoio aos Diálogos Setoriais UE-Brasil, Fase II). Retrieved from www.ﬁesp.com.br/arquivo-download/?id=171359. [31] M. Kassem, B. Succar, N. Dawood, A proposed approach to comparing the BIM maturity of countries, Paper presented at the 30th International Conference on Applications of IT in the AEC Industry — CIB W078, Beijing, China, 2013. [32] M. Kassem, B. Succar, N. Dawood, Building information modeling: analyzing noteworthy publications of eight countries using a knowledge content taxonomy, in: R. Issa, S. Olbina (Eds.), Building Information Modeling: Applications and Practices in the AEC Industry, University of Florida: ASCE Press, 2014 (in production). [33] F. Khosrowshahi, Y. Arayici, Roadmap for implementation of BIM in the UK construction industry, Eng. Constr. Archit. Manag. 19 (6) (2012) 610–635, http://dx. doi.org/10.1108/09699981211277531. [34] I.B. Kjartansdóttir, BIM adoption in Iceland and its relation to lean construction, Master's of Science in Construction Management, Reykjavík University, Reykjavík, 2011. [35] B. Leca, P. Naccache, A critical realist approach to institutional entrepreneurship, Organization 13 (5) (2006) 627–651, http://dx.doi.org/10.1177/1350508406067007. [36] T. Lehtinen, Advantages and Disadvantages of Vertical Integration in The Implementation of Systemic Process Innovations: Case Studies on Implementing Building Information Modelling (BIM) in the Finnish Construction Industry(Master's Thesis) Aalto University, 2010. (Retrieved from http://lib.tkk.ﬁ/Dipl/2010/urn100299.pdf). [37] H.C.J. Linderoth, Understanding adoption and use of BIM as the creation of actor networks, Autom. Constr. 19 (1) (2010) 66–72, http://dx.doi.org/10.1016/j.autcon. 2009.09.003. [38] R. Liu, R. Issa, S. Olbina, Factors inﬂuencing the adoption of building information modeling in the AEC Industry, Paper presented at the Proceedings of the International Conference on Computing in Civil and Building Engineering, 2010. [39] A. Luthra, Implementation of Building Information Modeling in Architectural Firms in India(Master's of Science) Purdue University, 2010. [40] V. Mahajan, E. Muller, F.M. Bass, New product diffusion models in marketing: a review and directions for research, J. Mark. 54 (1) (1990) 1–26. [41] V. Mahajan, E. Muller, F.M. Bass, New product diffusion models in marketing: a review and new directions for research, J. Mark. 54 (1) (1990) 1. [42] D. Mahdjoubi, Schumpeterian Economics and the Trilogy of ‘Invention–Innovation– Diffusion’, www.gslis.utexas.edu/~darius/schump/schumo.htm1997. [43] E. Mansﬁeld, Technical change and the rate of imitation, Econ. J. Econ. Soc. (1961) 741–766. [44] E. Mansﬁeld, Intraﬁrm rates of diffusion of an innovation, Rev. Econ. Stat. (1963) 348–359. [45] E. Mansﬁeld, The diffusion of ﬂexible manufacturing systems in Japan, Europe and the United States, Manag. Sci. 39 (2) (1993) 149–159. [46] E. Mansﬁeld, J. Rapoport, A. Romeo, S. Wagner, G. Beardsley, Social and private rates of return from industrial innovations*, Q. J. Econ. (1977) 221–240. [47] McGraw-Hill-Construction, The business value of BIM Europe, SmartMarket Report2010. [48] McGraw-Hill-Construction, The business value of BIM in North America: multi-year trend analysis and user ratings (2007–2012), SmartMarket Report2012. [49] McGraw-Hill-Construction, The business value of BIM in Australia and New Zealand, SmartMarket Report2014. [50] J. Meredith, Theory building through conceptual methods, Int. J. Oper. Prod. Manag. 13 (5) (1993) 3. [51] J.R. Meredith, A. Raturi, K. Amoako-Gyampah, B. Kaplan, Alternative research paradigms in operations, J. Oper. Manag. 8 (4) (1989) 297–326, http://dx.doi.org/10. 1016/0272-6963(89)90033-8. [52] C. Merschbrock, B.E. Munkvold, Succeeding with building information modeling: a case study of BIM diffusion in a healthcare construction project, Paper presented [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] at the System Sciences (HICSS), 2014 47th Hawaii International Conference on, Jan. 6–9 2014. R.S. Michalski, Concept learning, in: S.S. Shapiro (Ed.)Encyclopedia of Artiﬁcial Intelligence, vol. 1, Wiley, New York 1987, pp. 185–194. R.S. Michalski, R.E. Stepp, Clustering, in: S.S. Shapiro (Ed.)Encyclopedia of Artiﬁcial Intelligence, vol. 1, Wiley, New York 1987, pp. 103–111. M.S. Mizruchi, L.C. Fein, The social construction of organizational knowledge: a study of the uses of coercive, mimetic, and normative isomorphism, Adm. Sci. Q. 44 (4) (1999) 653–683. M. Mom, M.H. Tsai, S.-H. Hsieh, On decision-making and technology-implementing factors for BIM adoption, Paper presented at the International Conference on Construction Applications of Virtual Reality (CONVR2011), Weimar, Germany, 2011. M.E. Murphy, S. Wardleworth, Implementing innovation: a stakeholder competency-based approach for BIM, Constr. Innov. 14 (4) (2014)http://dx.doi. org/10.1108/CI-01-2014-0011. A. Mutai, Factors Inﬂuencing the Use of Building Information Modeling (BIM) Within Leading Construction Firms in the United States of America(Doctor of Philosophy) Indiana State University, Terre Haute, 2009. C.H. Nam, C.B. Tatum, Leaders and champions for construction innovation, Constr. Manag. Econ. 15 (3) (1997) 259–270, http://dx.doi.org/10.1080/014461997372999. NBS, International BIM Report 2013, NBS, Newcastle upon Tyne, UK, 2013. NBS, International BIM Report 2014, NBS, Newcastle upon Tyne, UK, 2014. OECD, Oslo Manual: Guidelines for Collecting and Interpreting Innovation Data, Organisation for Economic Co-operation Development, OECD publishing, 2005. K. Panuwatwanich, V. Peansupap, Factors affecting the current diffusion of BIM: a qualitative study of online professional network, Paper presented at the Creative Construction Conference, Budapest, Hungary, 2013. V. Peansupap, D.H.T. Walker, Factors enabling information and communication technology diffusion and actual implementation in construction organisations, Electron. J. Inf. Technol. Constr. 10 (2005) 193–218. J. Pfeffer, G.R. Salancik, The External Control of Organizations: A Resource Dependence Perspective, Stanford University Press, 2003. J.L. Pierce, A.L. Delbecq, Organization structure, individual attitudes and innovation, Acad. Manag. Rev. 2 (1) (1977) 27–37, http://dx.doi.org/10.5465/AMR.1977. 4409154. E.M. Rogers, Diffusion of Innovations, Free Press, New York, 1995. E.M. Rogers, U.E. Medina, M.A. Rivera, C.J. Wiley, Complex adaptive systems and the diffusion of innovations, Innov. J. Public Sect. Innov. J. 10 (3) (2005) 1–26. V.L. Saga, R.W. Zmud, The nature and determinants of IT acceptance, routinization, and infusion, Paper presented at the Proceedings of the IFIP TC8 Working Conference on Diffusion, Transfer and Implementation of Information Technology, 1993. Y. Saleh, M. Alshawi, An alternative model for measuring the success of IS projects: the GPIS model, J. Enterp. Inf. Manag. 18 (1) (2005) 47–63, http://dx.doi.org/10. 1108/17410390510571484. O. Samuelson, B.-C. Björk, Adoption processes for EDM, EDI and BIM technologies in the construction industry, J. Civ. Eng. Manag. 19 (Suppl. 1) (2013) S172–S187, http://dx.doi.org/10.3846/13923730.2013.801888. A. Sayer, Method in Social Science: A Realist ApproachRetrieved from http://bit.ly/ gHz6Md1992. M.A. Schilling, C.C. Phelps, Interﬁrm collaboration networks: the impact of largescale network structure on ﬁrm innovation, Manag. Sci. 53 (7) (2007) 1113–1126. B.A. Simmons, Z. Elkins, The globalization of liberalization: policy diffusion in the international political economy, Am. Polit. Sci. Rev. 98 (1) (2004) 171–189. P. Smith, BIM Implementation — Global Strategies, Paper presented at the Creative Construction Conference 2014, Prague, Czech Republic, 2014 (http://www. sciencedirect.com/science/article/pii/S1877705814019419). P. Stoneman, Handbook of the Economics of Innovation and Technological Change, Blackwell, Cambridge, MA, 1995. P. Stoneman, P. Diederen, Technology diffusion and public policy, Econ. J. (1994) 918–930. D. Strang, Adding social structure to diffusion models: an event history framework, Sociol. Methods Res. 19 (3) (1991) 324–353. B. Succar, Building information modelling framework: a research and delivery foundation for industry stakeholders, Autom. Constr. 18 (3) (2009) 357–375. B. Succar, Building information modelling maturity matrix, in: J. Underwood, U. Isikdag (Eds.), Handbook of Research on Building Information Modelling and Construction Informatics: Concepts and Technologies, Information Science ReferenceIGI Publishing 2010, pp. 65–103. B. Succar, The ﬁve components of BIM performance measurement, Paper presented at the CIB World Congress, Salford, United Kingdom, 2010. B. Succar, BIM Knowledge Content Taxonomy v1.3(PhD Thesis) Building Information Modelling: Conceptual Framework and Performance Improvement Models, School of Architecture and Built Environment, University of Newcastle, NSW Australia, 2013. (submitted for publication). B. Succar, Building Information Modelling: Conceptual Constructs and Performance Improvement Tools(PhD Doctor of Philosophy) University of Newcastle, Newcastle, NSW, 2013. (Retrieved from http://bit.ly/BSuccar-Thesis). B. Succar, W. Sher, A competency knowledge-base for BIM learning, Paper presented at the Australasian Universities Building Education (AUBEA2013), Auckland, New Zealand, 2013. B. Succar, W. Sher, A. Williams, Measuring BIM performance: ﬁve metrics, Archit. Eng. Des. Manag. 8 (2) (2012) 120–142, http://dx.doi.org/10.1080/17452007.2012. 659506. B. Succar, W. Sher, A. Williams, An integrated approach to BIM competency acquisition, assessment and application, Autom. Constr. 35 (2013) 174–189, http://dx.doi. org/10.1016/j.autcon.2013.05.016. B. Succar, M. Kassem / Automation in Construction 57 (2015) 64–79 [87] J. Taylor, R.E. Levitt, Inter-organizational knowledge ﬂow and innovation diffusion in project-based industries, Paper presented at the 38th International Conference on System Sciences, Hawaii, USA, 2005. [88] J.E. Taylor, R.E. Levitt, A new model for systemic innovation diffusion in projectbased industries, Paper presented at the Project Management Institute International Research Conference, 2004. [89] S.O. Tergan, knowledge with computer-based mapping tools, Paper presented at the ED-Media 2003 World Conference on Educational Multimedia, Hypermedia & Telecommunication, University of Honolulu, Honolulu, HI, 2003. [90] V.A. Thompson, Bureaucracy and innovation, Adm. Sci. Q. 10 (1) (1965) 1–20, http://dx.doi.org/10.2307/2391646. [91] A. Travaglini, M. Radujković, M. Mancini, Building information modelling (BIM) and project management: a stakeholders perspective, Organ. Technol. Manag. Constr. Int. J. 6 (2) (2014) 1001–1008. [92] K. Van der Heijden, C. Eden, The theory and praxis of reﬂective learning in strategy making, in: C. Eden, J.-C. Spender (Eds.), Managerial and Organizational Cognition: Theory, Methods and Research, Sage, London 1998, pp. 58–75. 79 [93] R.A. Voorhees, Competency-based learning models: a necessary future, New Dir. Inst. Res. 2001 (110) (2001)http://dx.doi.org/10.1002/ir.7. [94] D.H.T. Walker, L.M. Bourne, A. Shelley, Inﬂuence, stakeholder mapping and visualization, Constr. Manag. Econ. 26 (6) (2008) 645–658. [95] A.K.D. Wong, F.K.W. Wong, A. Nadeem, Attributes of Building Information Modelling Implementations in Various Countries, Architectural Engineering and Design Management, Special Issue: Integrated Design and Delivery Solutions, 62010. 288–302. [96] Y. Yoo, J. Richard, J. Boland, K. Lyytinen, A. Majchrzak, Organizing for innovation in the digitized world, Organ. Sci. 23 (5) (2012) 1398–1408, http://dx.doi.org/10. 1287/orsc.1120.0771. [97] H.P. Young, Innovation Diffusion in Heterogeneous Populations: Contagion, Social Inﬂuence, and Social Learning, Center on Social and Economic Dynamics at Brookings, Washington, D.C., 2006. (in collaboration with the University of Oxford and the Santa Fe Institute). [98] Z. Zahrizan, N.M. Ali, A.T. Haron, A. Marshall-Ponting, Z. Abd, Exploring the adoption of building information modelling (BIM) in the Malaysian construction industry: a qualitative approach, Int. J. Res. Eng. Technol. 2 (8) (2013) 384–395.

RELATED PAPERS

RELATED TOPICS

Log In

Macro-BIM adoption: Conceptual structures

Macro-BIM adoption: Conceptual structures

Related Papers

RELATED PAPERS

RELATED TOPICS