Megaprojects as niches of sociotechnical transitions: The case of digitalization in UK construction

Transitions are processes of systemic change where niches peripheral to a sociotechnical regime accumulate momentum, scale up and eventually transform its core. In contrast to this dominant narrative in transitions research, infrastructure systems exhibit the reverse process as change propagates from the regime core to its periphery. We explore this under-researched process in the case of digitalization in UK construction. We analyse six UK megaprojects that span more than 30 years as a single longitudinal embedded case, and show how the adoption of digital technologies driven by regime incumbents, seeds the processes of technology adaptation, aggregation, and system transformation. The adoption of digital technologies by incumbents is necessary to cope with megaproject scale and scope. Their adaptation to technology instigates organizational level change that starts at the regime core, accumulates with each project and makes these changes ripple across the industry and transform it.


Introduction
Sociotechnical transitions are evolutionary processes of systemic change at the organizational field level that alternate between stability and change (Geels et al., 2017Smith et al., 2010. Their study has developed into a substantial research program (Köhler et al., 2019), which describes a diverse set of radical innovations emerging in protected niches, then developing along different directions with some system wide impact (Geels, 2020, Geels and Raven, 2006, Raven and Geels, 2010, Schot and Geels, 2008, 2007, Smith and Raven, 2012. In their development trajectory, sociotechnical niches enter progressively larger market segments and may fundamentally transform sociotechnical systems, i.e. legacy technologies, incumbent actors and the core rules of the regime that underlie system stability and reproduction (Fuenfschilling and Truffer, 2014;Geels, 2011). DiMaggio and Powell (1983, p. 148) refer to the organizational field as a unit of analysis, "a recognized area of institutional life: key suppliers, consumers of resources and products, regulatory agencies and other organizations that produce similar products and services". System change proceeds mostly from small, peripheral niches, or other regimes, to the core of the organizational field (Fuenfschilling and Truffer, 2014). The implication is that small scale options have received far more attention than large scale technological options in transitions literature (Geels, 2020;Kanger et al., 2020;Sovacool and Geels, 2021;Turnheim and Geels, 2019). Nevertheless, infrastructure systems delivered through megaprojects can also be important for transitions (Sovacool and Geels, 2021). Megaprojects, generally cost more than US$ 1 billion (Flyvbjerg, 2017(Flyvbjerg, , 2014Flyvbjerg et al., 2003) and become increasingly relevant Transitions are non-linear processes at the organizational field that result from the interactions of (1) niches, where innovations develop, (2) sociotechnical regimes of established practices and associated rules and (3) long term landscape trends (Geels, 2002;Rip and Kemp, 1998). Transitions have been conceptualised into four stylised pathways: 'substitution', 'transformation', 'reconfiguration', 'de-alignment and re-alignment' (Geels et al., 2016). It is possible that a transition will adhere to one, or shift from one pathway to another. Transitions are initiated when the sociotechnical regime is destabilised through niche innovations, internal regime tensions, landscape trends that put pressure on the focal regime (e.g. climate, economic, cultural, demographic and other), and external influences from other systems, regimes or niches Papachristos et al., 2013;Raven, 2007;Rosenbloom, 2020). A transition proceeds through a process of accumulation of local, peripheral niche activities and subsequently enter increasingly larger (market) niches at the centre of the organizational field (Geels, 2002;Geels and Schot, 2007).

Niche development and empowerment
Niches can act as peripheral, 'protective spaces' where selection criteria are more favourable to new technologies than in the mainstream markets that operate under the dominant regime (Levinthal, 1998;Schot and Geels, 2007). In these shielded markets, the interaction of users with producers, gives rise to mutual learning and expectation articulation processes (Hoogma et al., 2002;Kemp et al., 1998). Niche nurturing involves articulation of actor expectations and visions, development of social networks, and learning processes (Schot and Geels, 2008). Learning processes contribute to niche development and involve first and second order processes. In the early stages of niche development, technology performance criteria are unclear, so technological competition is based on future expectations rather than current performance (Budde and Konrad, 2019;Klepper, 1997;Rosenberg, 1976).
Expectations and visions motivate, engage and align actors to commit their resources towards promising technological fields (Konrad, 2006), and guide niche learning and development (Geels and Raven, 2006). Niche empowerment is integral to transitions because it can influence the regime. Two empowerment processes modulate the interface between niche(s) and the regime (Raven, 2007;Smith et al., 2010;Smith and Raven, 2012): fit and conform and stretch and transform. Niches are empowered through stretch and transformation, by presenting a realistic alternative to problems, instabilities and tensions the regime experiences, such that tech implementation and institutionalization of niche practices becomes accepted by regime actors. In fit and conform, niches are empowered when they improve along established performance dimensions.
The process of technology implementation involves the mutual adaptation of technology and organizations (Leonard-Barton, 1988;Orlikowski, 1992;Tyre and Orlikowski, 1994). Market niches materialize as the product of organizational action (Astley, 1985;McKelvey, 1982). Thus, markets, user preferences and competences may need to be co-constructed with new technologies (Leonard-Barton, 1988;Oudshoorn and Pinch, 2003), to meet various attributes and selection criteria, such as price/performance, and the minimum functionality threshold for a technology in an application domain (Adner and Levinthal, 2001). Technology may have a more aggregate effect at the organizational field level, if technological functionality increases, or its cost is reduced and subsequently reaches larger, mainstream niches and becomes a general-purpose technology (Rosenberg, 1976).

Niche aggregation and structuration from local to the global level
The process of niche aggregation involves four phases (Geels and Deuten, 2006). In the first (local) phase, new technologies emerge or are introduced through technology absorption and relatively independent new entrants in local practices who create local knowledge for their own purposes. In the second (inter-local) phase, technological knowledge flows initiate in actor networks. Learning and accumulation of experiences has an inter-organizational but still local niche character through people, professional societies and industry associations that stimulate and facilitate the production and circulation of technical knowledge through interaction between ICT firms, users and regulators. When innovation experience is transferred from one project to another then general lessons are developed, and local knowledge is gradually absorbed into generic knowledge (Fleck, 1994).
In the third (trans-local) phase, knowledge production and circulation increase and acquire a global scope as groups of firms coalesce around their collective interests. In this phase a knowledge infrastructure emerges with dedicated journals and conferences for knowledge circulation, and with intermediary actors that perform dedicated knowledge activities at the global level to create, standardise, and distribute knowledge (Kivimaa et al., 2018;Schmidt and Werle, 1998). In the fourth (global/cosmopolitan) phase, dominant cognitive rules and a global stock of knowledge, technology artefacts and standards become established and shape local-level activities. Technological development thus proceeds simultaneously at local and global levels (Geels and Deuten, 2006;Geels and Raven, 2006;Schot and Geels, 2008). Therefore, the aggregation of peripheral niches produces generalizable knowledge through shared cognitive rules, structures and standards (the most tangible and codified outcomes of aggregation), and a system wide impact that catalyses a transition (Geels and Deuten, 2006;Geels and Raven, 2006;Raven and Geels, 2010;Schot and Geels, 2008).

Megaprojects as sociotechnical niches
A motivation to view megaprojects as sociotechnical niches is the argument made in PM literature that projects must be considered as sociotechnical endeavours embedded in complex institutional settings (Biesenthal et al., 2018). Modern infrastructures in transport, energy, and telecommunications are delivered through megaprojects, with investment exceeding $1 billion. Megaprojects are capital intensive, large-scale, complex enterprises where diverse actors collaborate to deliver an intended outcome (Flyvbjerg, 2014). In this respect, megaprojects offer a rich setting for transitions research to study the interplay of institutions, actors and technology due to their longevity, pervasiveness and embeddedness (Blomquist and Packendorff, 1998;Brookes et al., 2017). Their institutional complexity is seen in their front-end management, the promoter's role (Gil and Pinto, 2016), their embeddedness (Blomquist and Packendorff, 1998) and involvement of numerous external stakeholders, such as market and government policy, which are also core regime elements.
Moreover, megaprojects exhibit a range of processes and attributes that are conducive to their conceptualisation as niches for transitions research. Government support and the award of contracts to project consortiums constitute a form of protection from established industry structures, dominant technologies and infrastructures, guiding principles and socio-cognitive processes, and markets and dominant user practices. These were potentially factors to be overcome in the case of the megaprojects and thus a form of shielding had to be put in place. Government support and project procurement conditional on digital technology use was instrumental in bringing about change at the industry level. Then, these megaprojects can be conceptualised as niches created around the digital technologies to provide the "shielding" necessary (A. Smith and Raven, 2012) for it to follow a distinct evolutionary trajectory that was subject to certain technology selection criteria. Selection criteria in the niches-megaprojects as will be discussed in Section 4 were different and evolved progressively with each megaproject. 1 The duration of megaprojects and their delivery phases span across years or even decades, and may give rise to interorganizational relations and learning processes similar to those of permanent organizations that participate in niches. Projects are conduits for knowledge creation due to their transience and interdisciplinary nature (Gann and Salter, 2000;Grabher, 2004;Hobday, 2000). Essential learning processes arise at the interface between a project and the organizations, networks, and institutions in which it is embedded (Prencipe and Tell, 2001). Inter-project learning depends on organizational structures between projects, interproject assimilation practices, and actor alignment that facilitates the relationship with other projects (Lundin et al., 2015;Sense and Antoni, 2003). Aggregation dynamics may develop through organizational adoption and adaptation to technology, and the ways partners relate to projects (Manning and Sydow, 2011). The variety of collaborative paths and projects that may develop over time (Brady and Davies, 2004), may contribute to a variety of scale up, aggregation trajectories, and industry wide impact.
The implication is that no project unfolds in an organizational field vacuum, so it needs to be conceptualized as a history-dependant and institutionally-embedded unit of analysis to explain project success or failure better (Biesenthal et al., 2018;Engwall, 2003). To this end, the study analyses how digital technology implementation in megaprojects aggregates and transforms UK construction.

Rationale and case selection
We investigate the relation between megaprojects and system level digitalization in UK construction with a case study approach (George and Bennett, 2004;Yin, 1984), as is appropriate for the analysis of context rich phenomena (Eisenhardt, 1989). A single embedded case study design is used here (Yin, 1984) to analyse six intertwined infrastructure projects in UK (Engwall, 2003). The UK is an ideal setting to explore the relation between megaprojects as niches of sociotechnical transitions, given its long history in well-documented megaprojects and the institutionalisation of digital delivery in 2016 that was later followed by other European countries, such as France and Germany in 2020. Similarly, despite the unique character of UK construction, the UK BIM standards (PAS 1102 suite) was later published, as the ISO 19,650 series adopted as national standards by ISO member states. This makes the UK a model case for digitalization and megaprojects that has already inspired and influenced policy in other countries, demonstrating transferability of findings.
The six embedded cases presented were selected through theoretical sampling (Eisenhardt and Graebner, 2007;Pettigrew, 1990) as they are particularly instrumental in industry transformation through digitalization. These 6 cases were selected as maximum variation sampling (or heterogeneous sampling) to capture a wide range of perspectives relating to the phenomenon to gain greater insights by looking at it from all angles and searching for variation in perspectives (Levy, 2008). In chronological order, the six projects are: (1) the Channel Tunnel Rail Link, (2) Heathrow Terminal 5, (3) London Olympics, (4) Crossrail, (5) Thames Tideway and (6) High Speed Two. Projects 1-4 are completed, and projects 5-6 are ongoing. These projects have not been treated yet as part of a single longitudinal study on industry transformation, but have been the subject of significant PM-related research (Brady and Davies, 2014;Davies et al., 2009;Davies and Mackenzie, 2014;Dodgson et al., 2015;Gaunt, 2017;Genus, 1997;Zerjav et al., 2021).

Data collection and analysis
The cases offer opportunities for complementary and synergistic data gathering and analysis. The embedded cases of completed projects offer the opportunity to study technology aggregation retrospectively in the context of the overall transition pattern while the study of ongoing cases provides a close-up view on project evolution and the mutual adaptation of technology and organizations (Leonard-Barton, 1988;Tyre and Orlikowski, 1994). The embedded case studies build upon data from different sources that work synergistically together to present a coherent story (Yin, 1984). Data spans 36 years from 1985 to 2021, and have been collected through academic literature, government and industry reports, as well as interviews with eight industry experts, policy makers and practitioners which contribute to the external validity of the research (Sarantakos, 2005). More information about the cases and the associated literature on them analysed is shown in Appendix A for brevity.
To complete the case study data collection, eight interviews were conducted with experts on different cases. This sample was kept small in order to avoid impression management from the retrospective cases (Eisenhardt and Graebner, 2007), but also have enough information to triangulate the rest of the analysis. Interviewees have diverse backgrounds and roles (see Table 1): some have direct involvement in past and ongoing infrastructure projects (Int-2,3,5,6,7), three interviewees (Int-1,4,8) have significant industry and policy experience, and two hold key management roles in ongoing infrastructure projects (Int-2,7). Collectively, all of them have deep involvement in all six cases. The high-quality of the interviewees is shown in their boundary-spanning roles, bridging industry, policy and research that constituted them unique experts on the phenomenon. The coding process followed a combination of deductive coding, to identify linkages to the six megaprojects, and inductive coding to establish the role of institutions, actors and digital technologies (Saldana, 2009).
The cases are explored through a process lens (Langley, 1999;Langley et al., 2013). We first describe the institutional setting for the embedded cases. Then, the managerial choices made and emergent processes in each project are described following guidelines for qualitative enquiry (Denzin and Lincoln, 2017). The longitudinal view on the cases reveals the inter-organizational processes at work and the bidirectional influence of project organization and industry level developments. We use a combination of narrative and visual mapping as the strengths of each method counter the weaknesses of the other (Jick, 1979;Johnson et al., 2017). Diagrams do not constitute theory development on their own, but are part of theory building blocks (Griffin, 1993;Sutton and Staw, 1995), and theorization (Langley and Ravasi, 2019;Swedberg, 2017Swedberg, , 2016Weick, 1995) .

Case institutional setting
The UK construction industry is a regime that has been historically plagued by chronic fragmentation in the coordination and communication of designs and delivery of buildings (Egan, 1998;Latham, 1994), that had resulted in the well-documented performance gap between the design and actual building performance (De Wilde, 2014;Menezes et al., 2012). Such coordination problems are magnified in megaprojects, as each one has its own internal economy, governance structure, and production system ). The consensus view in the industry developed over the past decades is that these challenges have to be addressed (Armitt, 2013;Egan, 1998;Farmer, 2016;Latham, 1994;Sergeeva and Winch, 2020).
In response to this, the 2011 Government Construction Strategy (GCS) focused on the need for widespread adoption of digital technologies such as Building Information modelling (BIM) (Cabinet Office, 2011). Through this milestone development, the Government mandated that BIM Level 2 should be used on all public sector projects starting after April 2016 (GCCG, 2011). This mandate has significantly affected the project delivery of publicly-procured assets such as transportation (Whyte, 2019). It calls for all project and asset information, documentation and data to be digital and specifies the collaborative use of BIM models and digital objects (Papadonikolaki et al. 2019). To support these efforts, the UK government created the UK BIM Task Group which was publicly-funded until 2017. Thereafter a period of intensive standardization followed, with high levels of information exchange, deliberations, and strong leadership. In particular, the British Standards Institute (BSI) issued a suite of Publicly Available Specifications (PAS) number 1192 with parts 2-6 published between 2013 and 2015 (that later became ISO).
In 2013, the Government issued the 'Construction 2025: Industry Strategy' to reaffirm its strong support for BIM. It emphasised the joint commitment and close collaboration of government and industry to the BIM vision and programme. The strategy outlined a vision that all government procured projects would be delivered through a digitally-enabled delivery and lead eventually to a wider offsite manufacturing strategy. In 2016, the Cabinet Office and the Infrastructure and Projects Authority (IPA) issued the 2016-2020 GSC which built upon the 2011 strategy, emphasising BIM and Digital Construction as "an important part of the strategy and is helping to increase productivity and collaboration through technology" (Office, 2016). Presently, BIM is considered the most representative digital technology and information aggregator in construction globally. The Farmer Review (2016) commissioned by the Construction Leadership Council at the request of the UK Government resonated with earlier reports by Wolstenholme et al. (2009), Egan (1998 and Latham (1994) attributing productivity losses and lack of collaboration in construction to a paucity of innovation and widespread skills shortage. The review called for action in light of the newly-announced Thames Tideway and High Speed 2 projects in London. (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994) At the time of construction in the mid-'80 s of the first megaproject studied here, the Channel Tunnel, the UK construction industry was characterised by inefficiency and low digitisation. The discourse on improvements in UK construction included visions of partnering, supply chain management, and lean manufacturing philosophy, all of which were imported from other sectors, such as manufacturing (Bresnen and Marshall, 2001;Pavitt, 1984;Reichstein et al., 2005). In this context, CT was seen as a key infrastructure project for change in UK construction sector. A consortium comprising Bechtel, Arup, Systra, and Halcrow was established to run the $10.3billion Channel Tunnel Rail Link project linking UK to Europe's high-speed rail network . The technical complexity of CT pushed the boundaries of technology innovation in UK construction.

Channel tunnel (CT)
Back then, the method to produce and share design information was through 2D plans (Harty and Whyte, 2010). Digital innovations introduced from the US were a radical departure from prior established ways of working (Genus, 1997). They enabled the exchange of building information through Computer-Aided Design (CAD) applications, the creation of a virtual building through software before commencement of on-site work (Harty, 2005), and limited error-prone human interventions (Eastman, 1999). As Int-6 explained, 3D modelling tools (Pöttler, 1992) and other computer-based tools to deal with interoperability issues amongst different systems were tested in CT for project planning, cost management, procurement systems, and for facility management after construction.
As Int-6 explained, the implementation of 3D CAD in drafting departments required broader organizational adaptations and support, as its repercussions reverberated throughout the inter-organizational network of project actors. Instead of 2D plans to be used and adjusted on-site, 3D CAD required full design specification upfront, which left no room for design alterations during construction. Innovations used in CT paved the way for two subsequent CT phases in 2003 and 2007 (Arup, 2004;Pollalis and Georgoulias, 2008). The CT project informed the Latham (1994) and Egan (1998) reports that called for increased supply chain integration and collaboration and digital technology use to improve industry performance (Papadonikolaki and Wamelink, 2017). (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008) T5 was a $8.5 billion project for a new terminal at Heathrow airport to increase its annual capacity from 67 million to 95 million passengers. Due to its size, the T5 project became a program of industry-wide change in the UK . British Airports Authority (BAA) committed to 3D CAD implementation for coordination of design and construction, and information management across the whole project (Harty, 2005). BAA in consultation with partners pursued an integrated team approach to technology introduction and use in T5, that substituted some traditional drafting practices. This was shaped by T5 project managers who drew on the previous Heathrow T4 project (Harty, 2005). Personnel transfer from CT to T5 facilitated the further "evolution of digital technologies there" (Int-6). This approach was enabled by inter-compatible design and drafting software packages, mediated by a document management database.

Heathrow terminal 5 (T5)
The use of BIM in T5 aimed to create a single-model environment to improve coordination in the construction process, systems integration and information flows across actors, so that problems could be identified prior to work on-site where errors are costly . In T5, BIM can be described as a systemic innovation, as its consequences spread across multiple partner tiers (Harty, 2005). BIM is not a singular, stand-alone technology like CAD, but a more platform-like and thus disruptive innovation at the organizational level (Morgan, 2019). It is the result of evolving efforts by industry consortia, such as BuildingSMART to standardize building information (East and Smith, 2016).
The benefits of BIM were enough to convince project engineers and drafters they had to change their practices but were not enough to drive technology implementation at the organizational level. Considerable work was necessary to organise a coherent and inclusive system of technologies and practices in alignment to the diverse expectations and visions of the project (Harty, 2005). Negotiations between project actors indicated the software packages as necessary to align the BAA vision and the visions of services engineers and drafters. However, existing software packages could not facilitate the BAA vision and meet the selection criteria of project actors. This required that software had to transform from being primarily a design tool for 3D design to a manufacturing and control tool. Such digital innovations were highly influential in allowing the design to be engineered: for instance, model buckling analysis and sophisticated wind analysis were made possible on the complex roof structure of the terminal (Arup, 2006). T5 project activities shaped software package functionality and triggered a cycle of mutual technology and adaptation (Morgan, 2019). Technologies were incorporated into practices that already utilized a diverse network of material and digital artifacts. Moving towards a digitally orientated process was not just a simple case of technology substitution but fundamentally changing the practices they constituted (Harty, 2005;Harty and Whyte, 2010). However, configurations of practices already in place in T5 proved robust so the more initiatives to change challenged established practices and ways of work, the more they were resisted (Harty and Whyte, 2010). Eventually, the digital link between design and fabrication had to be dropped due to such difficulties.

London olympics 2012 (2005-2012)
London Olympics transformed Stratford in east London, into a 2.5 km 2 Olympic Park with an athletics stadium, aquatics centre and velodrome. The London Olympics project (LO) drew on the lessons of T5 with digital technologies and software tools used for coordination at each phase of system integration project design and documentation management, e.g. in ProjectWise Davies and Mackenzie, 2014). At the same time, systems integration related to collaborative teamwork and managing uncertainty by mutual adjustment and collection of new information. Legacy issues were widely considered in LO, described as a 'field-configuring event' (Thiel and Grabher, 2015). Several senior managers from London Olympics transferred their experience and innovative ideas on digital innovation to Crossrail, including Andrew Wolstenholme chief executive at Crossrail and former programme director at T5 (Wolstenholme et al., 2009).
Simultaneously, the UK government took steps to promote digitalization of UK construction through institutional projects (Holm, 1995). This included Avanti, a research project whose aim was to enable effective team collaboration through technology (Morgan, 2017). It was an inter-institutional collaboration amongst the Department of Trade and Industry supported by large UK construction firms, universities and R&D organizations, and BuildingSMART (formerly International Alliance for Interoperability). Funded by the UK government, Avanti drew several major industry partners together to create a culture, processes and digital tools to enable team collaboration (Morgan, 2017) through the use of two-dimensional (2D) digital design.
Avanti became the basis of the BIM British Standard BS1192. Such was the industry enthusiasm for Avanti, that it continued to be supported after government funding had run out. The Avanti principles were subsequently used in the Victoria Station Upgrade project to manage information flows and partner collaboration and "get people to trust that data" (Int-3). The significance of Avanti lay not only in its influence in establishing data standards but also in addressing cultural change: "in an industry where adversarial practices ruled…we paid a lot of attention to establishing a collaborative culture in Avanti, with hindsight this was as great an achievement of the project as some of the more technical advances". (Int-8).

Crossrail (2008-2022)
Crossrail is the biggest civil engineering project in Europe and the basis for several digital innovations in construction (Arup, 2012). Crossrail will provide high-frequency suburban passenger service crossing London from west to east. Crossrail started with 2D deliverables before the 2011 UK BIM mandate, but was completed using BIM and 3D digital deliverables. It was amongst the first UK projects to become PAS1192-compliant and use BIM as a digital platform for other innovations, such as laser scanning and augmented reality. However, the BIM Level 2 criteria were seen as obscure and too "open to interpretation, to be legally enforceable, but we were all aiming towards that 2016 deadline" resulting in standards not "as clear and concise as we would want" (Int-3), which created challenges in the project and technology uptake. The innovation strategy followed at Crossrail received considerable scholarly and practitioner attention (DeBarro et al., 2015). The Innovate 18 programme, was set up in association with Imperial College as a formal element of Crossrail delivery aiming to create a collaborative innovation culture across the project, partners and suppliers (DeBarro et al., 2015;Pelton et al., 2017).
The leadership shown by the government was catalytic to reconfigure work with digital innovations and "really changed where we were." (Int-3). From the perspective of the private sector, the government both local and national did some brilliant work to modernise planning policy and drive "digital transformation in large-scale infrastructure through the BIM policy" (Int-5). To further digital innovation in the industry, a BIM Academy was established in partnership with Bentley software to support the Government Construction Strategy, increase the use of BIM and create a lasting legacy of best practices in the industry (Munsi, 2012). The Academy training helps industry constructors to transfer their knowledge to other major projects such as HS2 and TT. Both LO and Crossrail had a strong emphasis on passing on lessons learnt to later infrastructure projects. This is particularly evident in HS2 and Thames Tideway (Pelton et al., 2017). Many senior managers from the London Olympics and Crossrail went on to work at TT or HS2, thereby transferring knowledge on digital technologies between these megaprojects.

Thames tideway (TT) (2012-2023)
The TT project is a tunnel running mostly under the tidal section of the River Thames through central London to capture, store and convey almost all the raw sewage and rainwater that currently overflows into the river. The TT is the first megaproject to adhere to BIM Level 2 standards. It is a landmark development in digital modelling, requiring that all project actors use digital model-based delivery (Gaunt, 2017). Significant efforts were made during initial procurement to assess the risks implicit in BIM delivery. However, while contractors were required to use BIM delivery, the TT team chose to adopt a 'technology agnostic' approach during tendering. The reason was prior experience from projects including Crossrail "where they had imposed a particular bespoke software or the client hosted everything' meaning that 'the risk was with the client." (Int-7). The TT team defined project deliverables from a digital blueprint and the BIM Execution Plan but encouraged contractors to use tools they already had expertise in so that they would "be able to go back to their own home organisations and utilise the latest, greatest software, where possible." (Int-7).
The TT team is also heavily involved in the Infrastructure Client Group to promote digitalization in construction and align "to the wider UK rollout, especially infrastructure" (Int-7). The aim is to draw on the lessons of the Crossrail and Olympics projects, create a digital learning legacy for construction, pass on lessons to HS2 and share best practices. 2016 marked a turning point with less government involvement but industry sentiment was that government support was still necessary "to try and drive that leadership from the top" (Int-3). Initially, there were great expectations that the industry would follow through with best practices in BIM Level 2 and the insurance sector would play a key part, but it became apparent that time was necessary "to see that properly emerge. I believe that's starting to happen, however." (Int-1).

High speed two (HS2) (Phase 1: authorised in 2017 2029-2033)
The first phase of HS2 is a major high-speed rail line that links London and Birmingham. Building upon developments from previous projects, but also benefitting by movement of experienced staff from LO and Crossrail, HS2 will collect and utilise knowledge to construct the digital twin of the asset, that mirrors the physical infrastructure and enables scenario simulations. This is one of several industry initiatives in Digital Built Britain, that "can certainly go a long way, and that would probably be HS2's digital legacy." (Int-2). In this regard the HS2 team works closely with government departments, with major infrastructure programmes, and professional institutions to try and "be proactive in developing standards and making sure that our requirements, our expectations from our supply chain, are reasonable." (Int-2).
The HS2 project has redefined digital innovation in UK construction by "looking at developing competencies which are not just limited to BIM in a sense that it's been defined as just a modelling or a CAD tool." (Int-2). There is a growing awareness of the digital legacy the project creates, and it becomes clear that it is necessary to "look at this as a wider data management piece" (Int-2). In this respect, standards like ISO19650 are a good starting point because they look at "at information management in its wider sense and its wider context" (Int-2). They are also pioneering solutions that benefit the entire industry and aim to create lasting change such as the HS2 BIM upskilling platform (HS2, 2019).
Although regulation activities stopped with the 2016 mandate, organizations have since then increasingly committed to digital practices. The perception in the industry changed from "Digital is something interesting. It's something on the side," to "It's core to our business … It's not a case of digital is a nice thing you do in your spare time." . This indicates the increased legitimacy of digital competences and practices. Despite the marked change, the five years from 2011 to 2016 "was an incredibly short amount of time to change an industry which, […] I think that hasn't entirely happened yet. I think we still need a lot more leadership. We need a lot more guidance." (Int-3). Indicative of the progress made is the refinement of ISO19650 which still is "not as enforceable in the client world as I think it should be." (Int-3).

Fig. 1. Timeline of digital technologies in UK construction influencing and being influenced by institutions, actors and megaprojects as sociotechnical systems (dashed lines indicate authorisation dates).
Despite the UK government leadership, the "local government doesn't seem to actually drive the mandate, that it was pushed through for 2016. So, we can already see setbacks. We want it to work, but we have really setbacks that we need to overcome" (Int-4). Industry digitalization requires collective action "So, those early adopters in terms of procurement and use of BIM used -to the majority, again, to stage open book tender. So, technology could actually assist but, again, it needs the combination of that collaborative work that we need to bring together." (Int-4). For example, in some of the pilot projects in the Ministry of Justice "they actually brought in the supply chain, and even the operators, and they were involved in the dialogue very early on in the design stage to be able to really." (Int-4). Further evidence of industry change can be found in 'Project 13 ′ , a recent industry-led collaborative endeavour that seeks to develop a new business model, based on an enterprise, not on traditional transactional arrangements, to improve UK construction. amongst the ambitions of the Infrastructure Client Group (ICG) is to "establish standards that all ICG members will sign up to and enforce across all of their projects".

Aggregation of digital technologies
To address the first question, which focuses on system-wide transformational impacts in UK construction, the analysis indicates that the evolution of digital technologies and related standards in UK construction has gone through several stages, punctuated by regulation, and marked by the move of scholars and practitioners towards a broader scope for BIM Morgan, 2017). Drawing on the preceding sections, Fig. 1 summarises the process of aggregation of digital transition in UK construction and juxtaposes (1) rules and institutions, (2) actors and social groups and (3) digital technologies with the temporal sequence of megaprojects, and policy reports. By viewing infrastructure projects as niches, a key contribution of this study is apparent. Our analysis suggests that the transformation process ripples outwards from the regime core to its periphery. This is interesting but in addition one could argue that this transition does not adhere closely to one of the proposed stylised pathways . Rather, we distinguish three phases ( Fig. 1) in this process for which a tentative case can be made as to its correspondence to a pathway.
In the first phase, digital technologies are adopted to enhance design and communication processes in construction supply chains so that the production and delivery onsite of complete blueprints is done with minimum errors. The incremental adoption of technologies in existing design and construction incumbent practices resembles the reconfiguration pathway where innovations are adopted in a symbiotic way in a regime to solve local problems, in our case the challenges that actors in individual megaprojects face. This process triggers subsequently further adjustments in the basic architecture of the regime. The second phase has elements of transformation pathway, as innovations are still being developed and refined and at the same time regime actors modify the development direction of their capabilities to align them with new standards they should adhere to. This was in part because technologies where not endemic to construction. Section 4 documents how industry incumbents engaged with software vendors in the development of digital technologies, and with government initiatives in the development of relevant industry standards. The third phase has elements of substitution pathway in that BIM solutions are available to industry actors and are developed sufficiently and specifically to their needs. To the extent that actors commit the necessary resources, BIM can potentially substitute completely previous practices and processes industry that actors have in place in their construction supply chains. Table 2 shows the sociotechnical transitions in digital technologies through UK megaprojects.

Megaprojects in sociotechnical transitions processes
To address the second question on how technologies in megaprojects drive sociotechnical transitions and vice versa, we start from framing infrastructure megaprojects as niches where incumbent actors engage in joint endeavours, standardization, and cooperation with the government. The scale of megaprojects magnifies coordination and performance problems due to industry fragmentation and intensifies partner tensions. Technology adoption triggers the digitalization process and the six megaprojects drive the aggregation Table 2 Transition pathways of digital technologies through UK megaprojects along three categories (institutions, actors and technologies), based on framework by Geels et al. (2016). process. They provide the impetus for digital technology implementation, a test bed for the refinement of its selection criteria, and the development of new regulations and standards. Sociotechnical changes at the firm and regime levels influence mutually constitutive relations between institutions, organizations, projects and actors (Morgan, 2019;Shibeika and Harty, 2015). The flow of key personnel across projects and institutions and their boundary-spanning capabilities drives and is driven by digital innovation and standardization (Koskinen, 2008;Levina and Vaast, 2005). Standardization is punctuated by two milestones, the Egan Report (1998) and the 2011 mandate for digital project delivery. Particularly after 2011, the Crossrail, TT and HS2 consortia formed strategic alliances with the UK government to drive digital innovation and standardization which were essential for digital technology evolution.

Rules and institutions Actors and social groups Digital technologies and STS
To understand how the digitalization process begun at the regime core and rippled out to the periphery of the construction industry, one has to understand that large incumbent firms in UK construction were in a position to impose their project related requirements to their 1st and 2nd tier suppliers. These requirements emanated from government strategic procurement policies, institutional developments and standards, for example, the Avanti project (2001)(2002)(2003)(2004)(2005) set the foundation for PAS1192 standards issued during 2013-2015. The regime/organizational field core to periphery dynamics then unfold over time because BIM adoption and implementation at the project level require collaboration between partner firms (Harty, 2005) and trigger organizational changes (Leonard-Barton, 1988;Morgan, 2019;Peansupap and Walker, 2007;Tyre and Orlikowski, 1994). The establishment of local standards of practice and organizational changes that initially span design to actual construction and post commission work, eventually aggregate to institutional changes that ripple across the industry (Boland et al., 2007). They have system wide impact as they change established organizational practices and technology selection criteria, and stimulate the development and use of sector specific software packages that are considered the norm today.

Theoretical implications
The case offers a clear example of how incumbents in capital-intensive industries survive the challenge an innovation may present (Ansari and Krop, 2012;Bergek et al., 2013;Berggren et al., 2015;Cohen and Tripsas, 2018;Hill and Rothaermel, 2003). This would suggest that incumbents' chances of survival during a transition are better than is implied in most transitions research. Indeed, industrial organization research indicates industry incumbents have better survival chances than entrants of an industry in its mature phase (Agarwal et al., 2002;Agarwal and Gort, 1996;Christensen et al., 1998;Klepper, 1997Klepper, , 1996Malerba and Orsenigo, 1996;Suarez and Utterback, 1995). The importance of this is significant in terms of transitions research as large incumbent organizations often have a large societal impact (Matten and Crane, 2005), that can stall or accelerate a transition.
The case of megaprojects deviates in a number of ways from a narrative where new entrants and outsiders challenge dominant incumbent actors and regimes. Megaprojects are not peripheral outsiders but instead protected innovative niches at the centre of the organizational field and their effect ripples outward towards its periphery. This provides fertile ground for transitions research that may uncover a greater variety in aggregation patterns and transition pathways (A. Smith and Raven, 2012). First, the positive role of incumbents that catalyse rather than raise barriers to change indicates that scholars should analyse symmetrically niche-to-regime activities and regime-to-niche activities. This may include the strategic reorientation of incumbent actors in the focal regime or capital-intensive industries.
Second, the technology implementation and use in megaprojects is in stark contrast to the system innovation pattern that begins with diffuse niche visions, open-ended experimentation, expectations and diverse search and technology development processes and develop to present a credible alternative to the dominant regime J. Schot and Geels, 2008;A. Smith and Raven, 2012;Turnheim and Geels, 2019). This may be a feature pertinent to infrastructure projects that need prior consensus and agreement to start, where failure is simply not an option, in part because of their high profile. Moreover, the digitalization process seems to become broad and divergent rather than become more specific and convergent in scope, as suggested in the literature (J. Schot and Geels, 2008). Digital technologies become linked to multiple possibilities e.g. advanced applications of BIM, additive manufacturing, artificial intelligence and robotics, the internet of things, big data analytics and blockchain technology.
Finally, bi-directional exchange between PM research and transition research offers some promising avenues through structuration theory (Bresnen et al., 2005;Floricel et al., 2014;Manning, 2008;Windeler and Sydow, 2001), and multi-level theorization that includes the firm level and the institutional -organizational field level (Gann and Salter, 2000;Lundin et al., 2015;Sydow and Braun, 2018;Winch, 1998), to examine temporalities and temporary/permanent organizing interaction at macro-meso-micro levels, the relation of projects to institutions (Biesenthal et al., 2018;Qiu et al., 2019;Söderlund and Sydow, 2019;Winch and Maytorena-Sanchez, 2020), and project governance (Ahola et al., 2014;Müller et al., 2016;Too and Weaver, 2014). With macro scale initiatives for macro scale changes as a starting point, researchers in transitions research and PM could look for other similar theoretical tensions and use them to stimulate better theorizing (Swedberg, 2017;Weick, 1995), and the development of more encompassing theories (Poole and van de Ven, 1989).

Implications for policy and practice
Our cases show the role of strong guidance, consensus, early direction-setting, and sustained resource commitments, e.g. national R&D funding, national strategic procurement for transition processes. The industry reports exemplify the strategic consensus and the vision for the industry (Egan, 1998;Farmer, 2016;Latham, 1994;Wolstenholme et al., 2009). They provide a strong impetus for change supported by government level standardisation initiatives such as the Avanti project, that is catalysed by the six megaprojects. These projects serve as milestones because of the significance of the technology verification and exemplification they enable, but also because their design implications were systematically replicated in later projects, supporting stepwise changes in the overall system trajectory. These megaprojects together with policy on digital delivery of projects (announced in 2011 and enforced in 2016) catalysed the speed of technology diffusion and organizational change and its directionality. To this end policy acted as the protective shielding of the megaproject-niches.
Across the 30-year longitudinal process of industry digital evolution studied here, we see the rich interplay of government and other policy bodies adopting role of 'client' or funder for the megaprojects and that of policy producer (for example through the Latham and Egan reports and BIM Mandate). Our data suggests that it is this interplay of roles that guides the evolution of industry practice, and is guided by it (Morgan, 2019). The recognition of these roles and the mutually constitutive relationship between them is significant in developing effective policy interventions that encourage evolutionary processes.
The purposeful character and focus of industry change efforts indicates that each megaproject can be considered as a niche for analysis purposes . Megaprojects with the extensive range of organisations and institutions involved, can function as institutional change levers or even political projects by invoking the collective action of political actors (Holm, 1995). For example, megaproject executives were aware of and keen to leave a 'digital legacy' (Munsi, 2012). It would seem that the proposition that single projects are not so important for niche development does not apply in our case (Hoogma et al., 2002).
While the sheer scale of megaprojects may seemingly provide little scope to consider agency, managerial cognition in fact plays a key role in identifying capabilities suitable for the environment where an organization operates (Lavie, 2006). The case of Wolstenholme is illustrative in this respect. Managerial cognition enables organizations to reconfigure their capabilities through substitution, evolution or transformation (Cohen and Tripsas, 2018;Lavie, 2006). A leader's interpretations of the external landscape will affect how they see their own organization's abilities to respond to it (Milliken, 1987). The cases highlight that the match between organizational capabilities and a market opportunity is not sufficient for change if leadership beliefs e.g. of Wolstenholme, are not aligned with the opportunity Kaplan, 2013, 2009;Tripsas and Gavetti, 2000).

Future research avenues
The sociotechnical perspective and analysis of projects as part of broader system level changes may be of interest to PM scholars that view projects primarily as distinct units of analysis, and vice versa transition scholars may stand to benefit from the vast PM literature. The potential for bidirectional cross overs is considerable. For example, the subject of interorganizational relations and actor alignment is important from a PM perspective but underdeveloped (Sydow and Braun, 2018). In this regard, PM research can benefit from a socio-technical perspective (Biesenthal et al., 2018), in order to highlight the role of organizational inertia and path dependency, and broader social, economic and historical processes that lie beyond the project management scope.
Research can extend beyond the study of individual projects to a project lineage, knowledge transfer from one project to another, to permanent organizations and the institutional context in which they are embedded (Engwall, 2003;Windeler and Sydow, 2001). The analysis of inter-project knowledge transfer could benefit from transitions research on how knowledge about project coordination practices accumulates, aggregates and changes the organizational field, and, in turn, it shapes these practices (Geels and Deuten, 2006;J. Schot and Geels, 2008;A. Smith and Raven, 2012). Similarly, transitions research can draw on the vast PM literature on factors that inhibit knowledge development and transfer between niche projects (DeFillippi and Arthur, 1998;Gann and Salter, 2000;Turner, 2002, 2001;Koskinen et al., 2003;Williams, 2008).
Moreover, another bidirectional avenue between transitions research and PM research lies at the institutional level of analysis (Fuenfschilling and Truffer, 2014). PM scholars advocate for more fine-grained analyses of projects in their institutional context (Dille and Söderlund, 2011;Engwall, 2003;Grabher, 2004;Morris and Geraldi, 2011). The successful completion of projects and megaprojects in particular, requires considerable institutional entrepreneurship and strong project ownership or sponsorship (Biesenthal et al., 2018;Brady and Davies, 2014). This raises also the issue of megaproject timing in relation to a favourable or unfavourable institutional context (Dille and Söderlund, 2011), and is resonant with recent calls to explore the role of large-scale technological options for low carbon transitions (Sovacool and Geels, 2021;Turnheim and Geels, 2019), and the importance of windows of opportunity, timing and sequence for transitions and projects.
In summary, bidirectional exchange between PM research and transition research offers some promising avenues through structuration theory (Bresnen et al., 2005;Floricel et al., 2014;Windeler and Sydow, 2001), and multi-level theorization that includes the firm level and the institutional -organizational field level (Lundin et al., 2015;Sydow and Braun, 2018;Winch, 1998), to examine temporalities and temporary/permanent organizing interaction at macro-meso-micro levels, the relation of projects to institutions (Söderlund and Sydow, 2019), and project governance (Ahola et al., 2014). This opens a wider vista of research on the interface between PM and transitions research on megaprojects and a contextual understanding of how they are situated in a large project-based context.

Conclusions
The framing of digital technology adoption in UK construction as a sociotechnical transition process with megaprojects framed as niches, offers an interesting alternative to the dominant narrative in transition research where sociotechnical processes at the periphery of an organizational field, accumulate momentum, scale up and eventually transform it. In contrast to this dominant narrative, the case analysed in this paper indicates that infrastructure systems may exhibit the reverse process: from core to periphery. The analysis of UK megaprojects shows how digital technologies introduced in construction, shape and are shaped by the interplay of institutional and organizational change processes of incumbents at the regime core. They are shown to enable and drive, rather than inhibit the aggregation trajectory of project based digital innovation. The trajectory is underlined by actors whose mobility across megaprojects and institutions is instrumental in the development of digital technology standards and regulation.
The sociotechnical perspective and analysis of projects as part of broader system level changes may be of interest to project management scholars that view projects primarily as distinct units of analysis. The implication of this study is to highlight potential conceptual cross overs between PM and transition research. Understanding the inter-relationships amongst key infrastructure projects, institutions, actors and how they influence digital innovation is a timely subject for both project management and transition research. It will help prepare for and identify patterns and opportunities to manage the unprecedented pace of emergent digital technologies that influence the industry.
These findings are also valuable for other sectors. The built environment allows us to study this relatively slow transformation over three decades and identify mechanisms and inter-relations that are hardly noticeable in other sectors, where the pace of innovation is more accelerated. In this respect this single embedded case may be shed light on contemporaneous projectification processes in other systems. Digitalization has been seen as a catalyst and an important means to address Grand Challenges such as Climate Change and Clean Growth. This study offers transferable findings for energy and pharmaceuticals disciplines that are also highly-regulated with a strong regime element that is challenged through digitalization and policy acting as protective shields of niches.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability
The data that has been used is confidential.  D.E. Papadonikolaki et al.