Organizing for service delivery projects
Ralph Levene, PhD, ILTM, FAPM
Director: Project Management Programs
Cranfield School of Management
Cranfield University, England
Palie Smart, MSc, PhD
Deputy Director: Innovation Leadership Centre
Cranfield School of Management
Cranfield University, England
David Tranfield, MA, PhD, FRSA, ILTM, FBAM
Director of Research and Deputy Director, Professor of Management
Cranfield School of Management
Cranfield University, England
Proceedings of the PMI Research Conference
11-14 July 2004 – London, UK
In a previous paper (Levene et al., 2002), the authors reported on preliminary results from a research project that focused on project types defined as Service Delivery Focus (SDF) projects. This paper presents the results of the second—and concluding—part of the three-year study that focused on the teamwork and cultural change issues necessary to ensure SDF project success. This study, funded by the British government via the Engineering and Physical Sciences Research Council (EPSRC, 2000), is now complete and its main deliverable—a set of guidelines entitled Organizing construction for service delivery: A managerial approach—has been field tested and is now awaiting publication.
Service delivery focus projects
SDF projects cover a range of project varieties, from Public-Private Partnerships (PPP) in Australia and in Continental Europe to Private Finance Initiative (PFI) projects in the United Kingdom (UK) (Treasury Taskforce, 1997). Throughout these projects, service delivery in the operation phase forms a key success criterion and an important mechanism for reward, whether or not there is any direct participation by public sector bodies.
The key characteristic for these project forms is that a client procures a service from a provider over a specified period of time. This is not to ignore the asset delivery aspect, but that the delivery of any asset is simply a component part of overall service delivery.
SDF projects are not a new phenomenon. Various forms of such projects have been in existence for some time, mainly in sectors such as petrochemicals and water treatment. The project types that best illustrate the phenomenon are Design - Build - Fund – Operate (DBFO), Build – Own – Operate – Transfer (BOOT), and Build – Operate – Transfer (BOT) (Morris, 1995; Woodward, Chan, Smith, & Merna, 1991). The PFI expanded in the UK after the 1997 election; the development of alternative PPP forms, such as the National Health Service Local Improvement Finance Trust (NHS LIFT, 2001) and Schools for the Future (Schools, 2003) appears to be ongoing. In principle, all these contract types were devised to provide clients with the project they required, but had not previously been able to realize due to lack of finance.
Early examples of BOOT and DBFO projects occurred outside the UK. (The Hong Kong Metro in the 1970s is such an example). Perhaps the most famous example of this genre is the Channel Tunnel, developed during the 1980s. While this project is remarkable for its scale and overrun of cost, the second Dartford crossing (a toll bridge over the Thames) was also remarkable, by comparison, in its success. As a result, its concession period was shortened to bring it back to public ownership sooner than originally planned.
Over recent years, the UK—with full government support—has increased its move towards service contracts that enable clients to require from their suppliers both the delivery of a fixed asset plus a guaranteed service over a fixed period of years (e.g., a PFI-type project). This service could be a hospital, a highway, an airport, or an office environment that includes provisions for information technology (IT) systems.
An often-cited classification of SDF projects is the UK Government's Public Private Partnerships – the Government's Approach (HMSO, 2000), which has a list of three types of PPP:
- Private sector (part) ownership of state businesses.
- Sale of government services (i.e., through partnership arrangements).
- Project finance approaches, such as PFI, with public sector authorities purchasing long-term services.
Therefore, PPP's are defined under the three recommended procurement approaches of prime contracting, design, and build, as well as PFI (HM Treasury (Her Majesty's Treasury), 2003).
The context of service delivery—PPP/PFI
This new emphasis on service rather than asset delivery needs a different approach to traditional methods of contracting. Responsibility for full life cycle management requires fundamental changes to the project concept and the design priorities, especially those project priorities of lesser cost. Together with a compressed schedule, these require reassessing in the light of operations with low maintenance and long term cost considerations. Designing in subsequent operability, therefore, becomes of paramount importance when considering the viability of the project. When responsibility continues after asset delivery, the input and requirements of the operations phase and its social organization are vital to the success of the project. For example, the project team must extend from its traditional constitution to include actors (parties) from each phase of the life cycle. Adjusting to this new way of working, in comparison to the traditional (frequently adversarial) approach, requires a change to the fundamental nature of client/contractor relations. This has major implications to the contractor's method of operation and the wider contracting environment that must deliver the various facets of the project. In general, the contracting organization is a legal entity designated as a “special purpose vehicle” (SPV). This designation is frequently constituted by a number of organizations—and in some cases their financial partners—providing expertise to the project.
The contribution margins in service delivery—across all sectors—are potentially better than in asset delivery. Furthermore, construction margins are historically very variable, leading to fluctuations in the level of employment in the industry. This, in turn, contributes to the loss of acquired knowledge and experience. This is also partly responsible for the shortage and the cost of experienced and qualified staff, especially during periods when the industry enjoys a construction boom. The enthusiasm for the UK government's PFI suggests that up to 40% of the industry's turnover will be attributable to 20- to 30-year service delivery projects where the emphasis is on collaborative working relationships, between actors whose contacts in the past have not been harmonious. In addition, and especially important to an industry dependant on project bidding success, the long term guaranteed income and profit that service delivery brings provides a sorely needed stability to these companies.
For construction contractors, the move to a collaborative working relationship with a client requires a major mindset change necessitating radical changes in project and organization structures and processes. Since the nature of this move towards service delivery will, if successful, extend over decades, it is clear that the rigid contractual structures which have been the norm in the industry will necessarily have to be supplanted by active relationship management and realigned to deliver services which, over time, may need adjustment as circumstances change.
Effective teamworking is vital to the success of SDF. It needs to work not only within individual teams, but also across teams, functions, and organizations, and over time. Construction organizations should get involved in production, management, service, and project activities simultaneously.
This variety of activities causes difficulty because not only is it spread across the extent of the organization's business, it also applies to individual projects and the same set of suppliers and customers. Temporary agglomerations of organizations and teams assembled for the performance of an individual project have to operate in this contradictory and conflicting environment. Teamworking (Tranfield & Smith, 2002) as a solution to these difficulties is therefore a practice embedded in, and intrinsic to, industrial networks, albeit one that operates imperfectly.
The SDF marketplace
This study investigated a number of key sectors in the PPP/PFI marketplace. The expenditure by departments of the UK Government is considerable with the highest value of signed PPP/PFI deals covering transport, health, defense, and education. The Scottish Executive performs similar work in Scotland. The values in the table below (Exhibit 1) indicate how important PPP/PFI is to project industries (including construction and IT projects as the major spend areas).
Exhibit 1: PPP/PFI sectors by government department (Office of Government Commerce, London, 2003)
The potential project expense in this area is very large, but the cost to achieve a contract award is also high for both client and contractor (Special Purpose Vehicle – SPV). This has the effect of restricting competition and potentially affecting the parameters that influence the public sector's requirement of value for money. The long lifetimes of such projects (up to 30 years) also has an effect on the constancy of the client's satisfaction criteria, because circumstances will change. Therefore, success criteria for such projects have to encompass both lasting and sometimes changing, objectives and performance specifications. The criteria must also address how challenges are met through the project's lifetime.
The government has attempted to ensure that client team's performances are improved. The current research highlights the competence requirements for client teams. These requirements state that team members must understand the technical aspects (e.g., risk management and design requirements) and the key roles they play in respect to delivering the bid, design, asset, and service. This move is supported by a recent government report (HM Treasury, 2003) that proposes a provision for a more straightforward arrangement of centralized services among client teams. This arrangement would include a system of accreditation for advisors, the specialized procurement of vehicles, and the enforcement of standardized projects. These lessons stem from the results of earlier tranches of projects.
The research project—Building responsiveness into construction
Consequent on this strategic shift towards SDF projects is the need to research and understand requisite organizational configurations to support the changes to the way of working. This EPSRC (2000) study was set up to explore the lessons for teamworking and cultural change in SDF projects (predominantly PPP/PFI) at a senior management level. This has been subtitled Building Responsiveness into Construction (BRiC) (Exhibit 2).
Exhibit 2: Aims of the BRiC research project
The project commenced in January 2001, with a formal launch in March 2001. The overall emphasis of the project has been to concentrate on output that will be valuable to practitioners. Their involvement has been very beneficial to the research process, especially in regards to validating the models produced. This involvement also allowed the conduct and the progress of the project to be continually monitored. The meetings and workshops were held on average every four months and provided a useful networking forum for the project participants.
Research objectives, aims, and purpose
The overall purpose of this research was to establish requirements for creating effective team-based construction organization within SDF projects. To achieve the purpose, this empirical study investigated the effective coordination of construction in PFI/PPP projects, in which design for service delivery is paramount. The main research objective constituted four sub-objectives:
- Key coordinating mechanisms
- Key dependencies that have to be managed
- Key activities, actors, and tools are involved
- Configuration implications for the SPV environment (SPVn, defined later in this paper).
We decided that the research findings would be aimed at developing a processual map and field-tested guidelines that will inform the management and organizations of future construction PPP/PFI projects. In line with a research philosophy of pursuing a practitioner-based enquiry, our approach placed a higher priority on providing evidence of what works when and where, rather than focusing solely upon detailed modeling for the academic community, as found in the explanatory sciences (van Aken, 2001).
Our research was exploratory and inductive, working to identify and articulate significant themes for further study. Our main research methods comprised a critical review of the current literature together with synthesis of findings from empirical work in ten case studies involving thirty-eight collaborating organizations.
We performed an extensive literature review during the first year of the project. Our aim of this was to delineate the conceptual and theoretical concepts underpinning the general area and to create a dataset of relevant academic material, which might inform data collection, analysis, and interpretation at a point later in the project. We examined four main academic areas during our literature review:
- Project management with a strong service focus.
- Sectoral studies of service delivery.
- Organizing and managing service delivery.
- Teamworking and cultural change.
Our literature review was extensive, and by April 2003, the number of entries logged into our Procite™ database reached 1097. The final number of academic journals that we used for our research totaled 240. In addition to these, we used other sources, such as MBA teaching material, to contribute ideas and concepts, including both grey material and sources, which were often identified through the fieldwork. Ideas, concepts, and empirical evidence were extracted inductively from these materials and were synthesized to provide an understanding of the field.
Inductive approach: Grounded in practice, informed by theory
Because of the relative novelty of PFI/ PPP project types, we believed it was pertinent to take an exploratory approach that could inductively draw out the significant themes for further study. This approach was best served by a multiple case study strategy, accessing mainly qualitative data, through semi-structured interviewing and observations. These were supported by documented, company-specific evidence, such as project reports and meeting notes.
Our findings from both the empirical study (Levene et al., 2002) and from the extant literature review were synthesized to ensure the robustness of the resultant deductions, both in terms of academic rigor (validity of the research output) and the relevance of the research to produce practitioner-based guidelines. Our research was not aimed at providing simplistic prescriptive models, which may become obsolete or impractical given the variety of institutional and task environments (Oliver, 1997). Instead, our research output aimed at distilling aggregate patterning that can be used as ideal types and exemplars to help inform future management decisions in regards to organization design within the context of PFI/PPP projects (Greenwood & Hinings, 1993). In this way, the overall research approach can be described as grounded in practice and informed by theory.
Sampling strategy and unit of analysis
Our research sample was selected using a modified snowballing technique, whereby the key interviewees were identified from earlier exploratory and pilot interviews that were conducted with the seven partnering companies. Informants were chosen across organizational and disciplinary boundaries in order to gather evidence from multiple perspectives. The group mostly included senior managers in their role as the architects and engineers of organizational design. Both the client and contracting sides identified these individuals as the first tier of project actors, the core of which formed the special purpose vehicle (SPV). Combined with its non-core component, the project organization in its totality was defined as the SPVn (described later in this paper) (Rowe et al., 2002).
This primarily qualitative study involved 66 in-depth, semi-structured interviews with senior managers from the 7 partnering and 31 other firms. The number of interviewees reflected the need to obtain multiple perspectives on each of the 10 case study projects. On some occasions, managers were able to comment on their involvement in more than one project. This improved the quality and reliability of the data collated.
Two full-time researchers completed a total of approximately 80 hours of interviewing, with each interview lasting around 1½ to 2 hours in length. Wherever possible, interviews were carried out at the company site to allow researchers the opportunity to assess project knowledge in the local environment and interviewees in their settings. The interview sessions were tape-recorded, producing more than 850 pages of transcript. To this was added an additional 100 pages of post-interview field notes. All of this information was included in the final data set (Exhibit 3).
Exhibit 3: Field data
Interviewees were questioned in detail about their specific experiences, involvement, and role in PFI/PPP projects. The studies varied in terms of sector (i.e., transport, health, defense, education, prisons), stage in project life cycle, and asset/service mix.
Data analysis and synthesis
Several processes exist for analyzing and synthesizing data to support practitioners in developing theory and tools from inductive research, based around a grounded theory methodology (Partington, 2000; Strauss & Corbin, 1990). The basic progression of grounded theory study is to change focus “from comparing incidents with one another to comparing incidents with properties of the category that resulted from initial comparisons of incidents” (Partington, 2000, p. 93). There are many variations on this basic theme. We used a two-stage model developed by Eisenhardt (1989) to analyze and synthesize the empirical and theoretical data from the 10 case studies in order to successfully induct the key emerging themes.
First stage: Open and axial coding
Our first steps in using grounded theory involved open coding, or the splitting of data into codes. This process was then re-formed through axial coding, or identifying the major themes (Strauss & Corbin, 1990). We used NVivo™ to assist us in the qualitative data analysis and synthesis. Our content analysis of the 850 pages of interview transcripts and 100 pages of field notes began with a basic coding procedure that identified 400 first-order open codes. These open codes represented the detailed analytical units and the key themes. Our inductive distillation of raw empirical phenomena was achieved through the weighting of instances. This occurred according to whether the interviewees explicitly referred to an instance as significant and the relative length of discussion for a particular instance (Calori, 2000).
Using higher order codes, we clustered more generic categories to produce 40 second-order codes. We abstracted these second-order codes to draw out theoretical constructs that were transferable across the 10 different project contexts (Calori, 2000). Our axial coding was informed by the body of knowledge relating to coordination theory (see below) (Malone & Crowston, 1994), which emphasizes the key project activities performed, the actors (collective and individual) involved, and the resources/tools used to assist in the delivery of key activities. We then developed a third-level coding to group the 40 second-order codes into the three categories of activities, actors, and resources in relation to the construction projects, which informed stage two of the data analysis.
Second stage: Selective coding and activity-focused mapping
During previous stage, we split the data in order to analyze it and then combined it to produce a provisional framework that we refined through a second stage of coding (activity-focused mapping). This second stage is a form of selective coding, where key activities are categorized, the categories are interrelated, and their relative importance rated (Partington, 2000). This stage enabled us to synthesize the empirical findings with the extant literature in order to develop a robust framework for mapping the generic processes associated with coordinating the PFI/PPP projects. We developed a fourth order of coding to enable us to identify the 83 key activities, the associated challenges (or dependencies), and the related actors and resources.
In following the coordination theory recommendations (Malone & Crowston, 1994), we aggregated the 83 activities identified into an Excel™ spreadsheet to produce four activity clusters, as high-level processes with associated dependencies, actors, and resources (Riley & Brown, 2001). We found that these clusters related to delivery of bid, design, asset, and operation. Each activity cluster was hyper-linked to the raw data, which had been coded in NVivo™, into separate Word™ documents. Consequently, we learned that the activities could be mapped back to the raw data to provide a full audit trail from the guidelines back to the original data (Miles & Huberman, 1994).
By using a multi stage approach, we learned that it was possible to provide an audit trail leading from the early open coding (NVivo™) through to the prototype guidelines themselves, ensuring that the findings were able to be validated by practitioners (Tranfield, 2002).
Ensuring validity and reliability of data set and findings
As Yin (1994) emphasizes, the type of study utilizes several techniques in order to improve validity. For example, we established a consensus between the interview data, the detailed field notes, and the corporate documentation about the existing projects and processes. This process informed us about the significance of the case studies and helped us facilitate an across-case analysis (Eisenhardt, 1989).
Our confirmation of the results was enhanced by our use of audit trails (Miles & Huberman, 1994). We used post-interview debriefing notes to capture our main thoughts and any emergent ideas. These methods were used to check the dependability of our findings, alternating between our in-the-field experiences and our follow-up reflections. This enabled us to trace an audit trail along the chain of evidence, which allowed us to develop a construct for ensuring the validity and reliability of our findings.
To further enhance our findings (Yin, 1994), we performed a periodic validation of our findings in group workshops with the managers from the seven partnering organizations. Our experience as reflective field practitioners played an integral part in our research process. This enabled us to use our insights to find gaps in the data collection and to find further research avenues. The implementation of these various tools and techniques allowed a high degree of consensual validity and reflected the closer relationship between academic researcher and industrialist, as indicated by our practitioner-based research (van Aken, 2001).
During the literature search, we discovered the theory of coordination (Malone & Crowston, 1994) was particularly relevant in understanding how organizations cope with extended project lifetimes. It added value in framing:
a) How organizations could interrelate within a long life networked environment.
b) How organizations could understand that current actions can have far-reaching effects and consequences for future project teams.
Both of these aspects affect the cultural and teamworking changes that are needed to ensure success in SDF projects.
The findings that follow cover the network environment of the SPV (SPVn), the specifics of an SDF life cycle, its major delivery processes, and organizational SPV types.
SPVn: The SPV environment
In addition to the complication of the different ways in which constituent organizations make up the SPV and its contractual arrangements, there are other parties who are involved in the project, such as the client, the end-users, and the financial institutions and other stakeholders, such as regulatory authorities. All the parties in the project create a network of relationships—both contractual and informal—that is designated the SPV environment (SPVn).
The SPVn is a networked organization linking various parties in the project supply (value) chain whose purpose is to deliver SDF projects. Indeed, SDF projects are more complex than traditional ADF project types, involving more and diverse parties, activities and resources over its extended project life. The SPVn usually consists of a multi-functional project core team, the SPV itself, and also an extended network of functional, often temporary, teams (Exhibit 4).
Exhibit 4: The SPVn stakeholders and major relationships
When several teams are operating within the SPVn, two important entities govern the effectiveness of the project:
a) The team/organization responsible for the realization of the project—the capital asset company (CapCo).
b) The team/organization responsible for the operation of the asset—the operations company (OpCo).
The project life cycle
The main focus in most of the literature on the progression of a project through its life cycle is the mechanics of managing projects. The main differentiator between project types is the variation in the phases (by number and name) and the balance of time from phase-to-phase. Several models have been used to describe construction projects, including the PPP/PFI projects. For example, the process protocol lists 10 phases, the Treasury Taskforce guidelines (Treasury Taskforce, 1999) state 14, British Standards Institute standard BS6079 (British Standards Institute, 2000) records 11, and the Office of Government Commerce Competency Framework (Office of Government Commerce, London, 2000) puts forward 3. For the most part, these different models are subjective; each serves the purposes for which the models were produced. For the most part, each neglects the operations phase.
After reviewing these models, we decided to simplify the model to the four phases: feasibility, inception, realization, and operations (FIRO) (Exhibit 5). The simplification of these phases delineates the major milestones that form the gates between them.
Exhibit 5: The FIRO model
Project gates & milestones
The key milestones listed below (Exhibit 6) are highlighted as gates in the process because of their impact upon activities and actors, particularly in regard to impacting a change in the role of specific actors. The table below illustrates these milestones. This research shows the most important activities and the actors who execute them.
Exhibit 6: Key project milestones
Key milestones as hard gates
The gates noted above are understood as hard gates, as opposed to soft gates, which imply that there is a scope for tolerance on completion of activities. Hard gates must be addressed before the next phase can begin.
In reviewing some of the case studies, we discovered that some projects showed some flexibility, such as pre-close works or phased signing off of works packages. This does not compromise the hard nature of these gates. For example, the phased signing off of works means that the phase gate is merely staggered: for the individual actors (e.g., the end users or work teams on a particular section of building or site), there is a clear transition point, and from an project overview perspective, there are final dates.
Transcending delivery processes - The Laminar model
The high-level clusters derived from the interview data consisted of a number of lower-level activities and tasks, which we equated to the activity (or job) described by the British Standards Institute (2000) as “an operation or process consuming time and possibly other resources” (p. 2).
The project phases—FIRO—are well bounded by milestones that are specific, have contractual significance, and are seemingly unambiguous. The research has identified four clusters of activities that transcend the phase boundaries. These comprise linked activities that are not exclusively limited to particular phases, though these activities mostly occur at different periods in the project life and are associated with particular phases. These clusters could be considered as groups of process-oriented activities, with each cluster having a common purpose. The clusters are identified as:
- Delivery of bid
- Delivery of design
- Delivery of asset
- Delivery of service.
Exhibit 7 shows, conceptually, the time span of each activity cluster. This evokes a layered or laminar view of the four major delivery processes of an SDF project.
Exhibit 7: The Laminar model
Delivery of bid
Project teams must complete this cluster by the end of inception. This cluster comprises the preparation of the tender document, its advertisement in the Official Journal of the European Communities (OJEC)), and the responding bid by the tenderers. The major part of this cluster occurs during feasibility but carries on through inception, as negotiation between client and potential SPV contractors continues.
The bidding process involves a competitive process of potential buyers and suppliers. Performance in this lamina was highlighted in terms of the costs associated with a bid being delivered that was suitable for bidder and client. One project studied estimated bidding costs at £6 million.
Key activities of this phase include:
- Managing bid costs
- Setting a realistic and affordable solution
- Managing changes in the bid
- Building bid teams
- Building confidence in the project between the parties.
Delivery of bid
This cluster starts in the feasibility phase when the client considers the output specification for the design activities. The specification is reviewed during the inception phase. The potential contractors carry out asset design activities in preparation for the bid and consider their design in relation to the effect on operations and maintenance.
Key activities include:
- The forming and aligning of specifications
- Managing changes in design specifications
- Configuration of design for whole life.
Delivery of asset
This activity cluster focuses on the construction process. This phase requires considerable interaction with the design process for both constructability and whole-life implications. It may extend into the service phase for staged completions of the asset and for changes to the asset, as required by the client, for conversion or refurbishment of existing assets.
Key activities include:
- Preparatory work (e.g., timetabling of works packages)
- The monitoring of asset delivery
- The managing of changes/variation in this delivery.
Completion of a single asset is not invariably an objective. The objective is the phased delivery of sections of an asset, delivery of a number of assets over time, or the conversion or refurbishment of existing assets.
The overall performance of asset delivery in the case projects was mixed. Some of the early projects were poor, suffering because of delays in asset provision due to sub-contractor failure and poor scheduling of works; other projects involved the on-time delivery of assets.
Delivery of service
This cluster starts at the beginning of the project, not just at the start of this operations phase. For SDF projects to be successful, the involved project parties must direct their attention to the levels of service and to their performance criteria. Success starts with a good client specification and continues with a design that maximizes up-time during operations. Success is intrinsic to construction quality.The activities associated with the actual delivery of a service were many and varied, occurring throughout the project life.
Key activities include:
- Selection of service providers
- The monitoring of delivery-of-service performance
- The managing of variations and service-level expectations.
Metamorphosis of the Laminar model
Each cluster interrelates with its neighboring cluster, and aggregately, these clusters metamorphose from one to the other. Within these dimensions there exists an extensive interchange of activities, roles, and resources between the actors.
SPV Organizational Issues
Our research suggests that the pivotal position of the SPV as manager of the project deserves greater consideration in terms of its composition and key relationships.
The significance of the SPV consortium has been highlighted in other research, and the AMA Market Research (2001) typology suggests that there are two types of SPV:
- Where the financier is member of the SPV (e.g., an equity holder)
- Where the financier is financing but not a part of the SPV (non-equity holder) (AMA Market Research, 2001).
However, these SPV configurations are but two of many; the development of further variations (for example, where public sector SPV partners are involved) means that these typologies are evolving.
Types of SPV: Light-touch and heavy-touch
Our research focused on strategic issues rather than specific detailed configurations. What emerged was an understanding that the impact of configuration of SPV team varies between two idealized types: light-touch and heavy-touch. Exhibit 8 outlines the essential characteristics of these types. The main difference between these two types is that the heavy-touch tends to operate in executive mode.
Working in executive mode can be defined as:
- Putting into effect and taking responsibility for activities such as allocating work to sub-contractors; this also involves monitoring activities that have been/are being carried out according to requirements
- Accepting risk
- Taking responsibility for activity completion.
By contrast, the non-executive mode involves:
- Overseeing and rubberstamping outcomes; that is, being persuaded that works have been allocated, rather than personally carrying them out
- Risk-passing, where the risk is managed by another who is acting in executive mode
- Awareness of activity completion, whether by standardized reporting or other informal methods; awareness does not mean taking or accepting risk.
Exhibit 8: Generic types of special purpose vehicle (SPV)
The key point is that any SDF project requires a shared understanding of who does what when, and who cannot do what when. This is made clear in relation to the role of the SPV itself. Problems arise because project teams lack a shared understanding (between SPV, other project teams, and their respective stakeholder/parent organizations) about the various operating modes of the SPV.
Examples of the light-touch and heavy-touch approaches
One funder (a member of more than a dozen SPV boards) learned from his experience with SPVs that these projects vary because of:
1) The size of the project
2) The complexity of the project
For example, a simple project, such as a single school, might only have 2-3 people in the SPV: a full-time on-site project manager and part-time specialists, such as a finance specialist. In one project studied, the CapCo manager noted how the SPV tended to pass risk down to the sub-contractors.
In contrast, more complex projects require that more people are involved. For instance, a hospital PPP/PFI SPV included 35 specialists in quality and design, among other disciplines. In another hospital PPP/PFI, the team size was 15 people. In the above-mentioned funder's experience, in more technically complex projects the SPV becomes, as he notes, “a mirror of the project team.” In more complex projects, as in the first hospital PPP/PFI mentioned above, the funder says the SPV “very actively managed the project,” unlike simpler projects, where the management of the project is vested in the main sub-contractor.
Constancy of the SPV
What an SPV can be, in theory, is the one constant project team of an SDF project. This has meant that the SPV is perceived as taking a lead actor role. However, as has been found in practice, the SPV team may not be able to fulfill a proactive role because of risk concerns and lack of resources. Indeed, it is not necessarily important for an SPV team to be especially proactive, as risks can be passed on to relevant contractor teams. Keeping the SPV team together would be an important objective through delivery of asset; continuing with the same team members through the operational phase is likely to be much more difficult (see legacy of the project team below).
The ideal situation is where the SPV team and the other teams/stakeholder organizations are aligned in their objectives and are fully tuned to the delivery processes of the Laminar model.
The issues that follow focus on aspects of SDF team implementation, in particular to the concept of the team realizing the requirements of service delivery through design.
Project Management Issues
Although the Laminar model takes a strategic view of the team's execution of an SDF project, there are general project management issues that need to be understood in order to grasp the team's coordination of these projects. A key metric for PPP/PFI type projects is the availability of the service. The team's failure to reach the defined levels has potentially catastrophic effects upon end users and the SPV, which would lose revenue. This focuses attention on the designs that configure the asset, the organization structures, and the processes that ensure delivery of reliable assets and services.
Design for serviceability
By reverse engineering the needs of the project, the project's main focus becomes the concept of serviceability, which in turn embraces:
- Fitness of purpose of the facility (asset)
- Reliability of the processes of operation
These issues can be distilled into some of the characteristics of good project management practice:
- Project definition: Bid delivery as expected, with implications for delivery of design and the delivery of the asset and service.
- Review procedures: Continued tracking and checking of progress and objectives, especially those that related to the delivery of the asset and service.
- Leadership and team maintenance: In regards to the legacy of the project team.
This latter concept has been interpolated from the research data. It is unlikely that anyone involved in the early stages of an SDF project will still be involved at the close of the project, which can extend to more than thirty years after project commencement. Project team membership is frequently transitory and it seems likely that a mechanism is needed to transfer team knowledge and values from generation-to- generation to its successor. This is a new area for teamwork and needs further research,
One of the key drivers that we identified for PPP/PFI is whole-life costing. The Commission for Architecture and the Built Environment (CABE) (2002) PFI design recommendations speak of the need for design for manufacture rather than manufacture for design (CABE, 2002; Office of Government Commerce, London, 2002). This is exemplified by the activity clusters and their interrelationship across project phases, but most importantly by how the delivery of design is affected by other activity clusters. This implies that design for serviceability focuses on the needs of the operations phase. The concept of serviceability therefore not only influences the formulation of the bid, but has greater influence on the service-related decisions of asset design. The need is to involve parties from later operations phases.
From the above, it can be seen that the challenges which require managing also involve ensuring that information has been shared and that the various teams have aligned objectives (i.e., common knowledge and shared characteristics, aspects of the coordination theory) (Malone & Crowston, 1994). This also ensures that the key actors agree to the configuration of asset and service, particularly as many of the key actors involved in these decisions do not have the technical background to inform their decision-making. We found that a recurring theme in our research was the difficulties in estimating maintainability and operability
Tools to facilitate design for serviceability
Tools and mechanisms of value to SDF projects and that draw on coordination theory are:
- Shared ideology: Ensuring that all team members understand what/who/when/how with what
- Standardization of processes and outputs: Regulating what is to be done when and by whom
- Mutual adjustment: Improvising in response to challenges as each emerges
The following are examples of coordination tools that were used in some of the case studies:
- Collocated teams, particularly collocated design teams, using simulation tools and standardized design specifications.
- Key managers: On a government project, the head of the design team was retained after the realization phase in order to provide continuity for design decisions.
- Information systems: Systems were designed to make data readily accessible.
In addition to these tools, one of the projects that we studied used the standardization of processes/outputs to good effect. This standardization was seen as being difficult for the diversity of actors involved. The experience of PPP/PFI, however, led the organizations, particularly those contractors involved in the delivery of asset (as CapCo or part of a construction joint venture), to develop their own standardized systems to calculate whole life costs.
We observed that a hindrance to integration has been the difficulty of sharing information between the various disciplines involved in the design and implementation phases of capital projects, due to different understandings and priorities, and because of the absence of suitable data exchange software. An interesting initiative is the International Association for Interoperability (IAI, 2004). This initiative is designed to enable the standardization and exchange of data between the various systems used in the design and execution phases of projects. This may be the beginning of the solution to life-cycle costing/life-cycle management of assets.
Mutual adjustment and shared ideology provides a particular challenge for teams working on SDF projects in that design considerations are not immediately apparent to all actors in the SPVn. The CapCo's primary concerns are to deliver assets to time and meet criteria for certification. This traditional mindset, with its embodied design and quality implications, can negate the optimization of operability and maintainability. Breaking down these traditional beliefs is a major cultural challenge. Our hope is that the guidelines produced from this research will help to engineer the change.
The main contribution of this research has been a set of guidelines to help managers involved in SDF projects to formulate relationships between teams within the SPVn. These guidelines have evolved through a number of editions, each of which was scrutinized by its project partners. These guidelines were not designed as a methodology, but as a framework for executing SDF projects. These guidelines were written in a style suitable for practitioners and real world application, not for researchers and research work. Each section is preceded by a summary of actions to be done and actions to watch against.
The guidelines – content
The document, totaling 216 pages, is divided into four main sections, with appendices:
Part 1 - Strategic operational shifts
Part 2 - Strategic organizational considerations
Part 3 - Operationalizing effective coordination in the SPVn
Part 2 concentrates on the design influences that shape the SPVn. It focuses on three main issues:
- The project life cycle
- Team structural issues
- Team support issues
We are currently preparing a summary document, in brochure format, aimed at the industry involved in SDF projects.
AMA Market Research. (2001). The 2000 PFI report. Cheltenham, England: AMA Market Research.
British Standards Institute. (2000). BS 6079-2:2000 Project Management–Part 2 Vocabulary. London: British Standards Institute.
CABE, (2002, Sept. 2002) Client guide: Achieving well designed schools through PFI. Retrieved 25 Mar 2003 from www.cabe.org.uk
Calori, R. (2000). Ordinary theorists in mixed industries. Organizational Studies, 21(6), 1031-1057.
Eisenhardt, K. M. (1989, October). Building theories from case study research. Academy of Management Review, 14(4), 19.
EPSRC. (2000). Teamworking and cultural change in construction: Developing a service delivery approach. Engineering and Physical Sciences Research Council Grant Award GR/N33263
Greenwood, R., & Hinings, C. R. (1993). Understanding strategic change: The contribution of archetypes. Academy of Management Journal, 36(5), 1052-1075.
HM Treasury (2003). PFI: Meeting the investment challenge. London: Her Majesty's Treasury.
HMSO. (2000). Public Private Partnerships – the Government's Approach. London: The Stationery Office
IAI. (2004). Introduction to the International Alliance for Interoperability. Retrieved 27 Jan 2004 from http://cig.bre.co.uk/iai_uk/iai/
Levene, R. J., Smart, P., Rowe, A., Tranfield, D., Corley, J., & Deasley, P. (2002). Service delivery – Preparing for a new future in projects. Proceedings of the Project Management Institute Research Conference 2002, 331-341. Newtown Square, PA: Project Management Institute
Malone, T. W., & Crowston, K. (1994, March). The interdisciplinary study of co-ordination. ACM - Association for Computing Machinery Computing Surveys. 36(1), 87-119
Miles, M., & Huberman, A. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage Publications.
Morris, P. G. W. (1995). Build-own-operate(-transfer), BOO(T), In J. R. Turner (Ed.), The commercial project manager (pp130-137) Maidenhead, England: McGraw-Hill.
NHS LIFT (2001). National Health Service Local Improvement Finance Trust Retrieved 4 Dec 2003 from www.doh.gov.uk/nhslift/Office of Government Commerce, London. (2000). A competence framework for creating effective PFI project teams. Unpublished draft.
Office of Government Commerce, London. (2002). Standardisation of PFI contracts—general. London: Office of Government Commerce.
Office of Government Commerce, London. (2003). PFI Signed Projects List Retrieved 23 Oct 2003 from http://pfi.ogc.gov.uk/stats.asp
Oliver, C. (1997). The influence of institutional and task environment relationships on organizational performance. Journal of Management Studies, 34(1), 99-125.
Partington, D. (2000). Building grounded theories of management action. British Journal of Management, 11(2), 91-102.
Rowe, A., Smart, P., Tranfield, D., Levene, R. J., Corley, J., & Deasley, P. (2002), Exploring design for serviceability in construction: A case for relationship management in a web of networks. Proceedings of the 9th European Operations Management Conference: Operations Management and the New Economy. Copenhagen, Denmark. (2-4 June), Vol. 2, 1215-1228.
Riley, M. J. and Brown, D. C. (2001), Case Study of the Application of BPR in an SME Contractor. Knowledge and Process Management, 8(1), 17 Schools (2003). Schools for the Future – Building Bulletin Retrieved 4 Dec 2003 from www.teachernet.gov.uk/management/resourcesfinanceandbuilding/funding/capitalinvestment/
Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory processes and techniques. The Conditional Matrix (pp. 158-175). London: Sage Publications.
Tranfield, D. (2002). Formulating the nature of management research. European Management Journal, 20(4), 378-382.
Tranfield, D., & Smith, S. (2002). Organisation designs for teamworking. International Journal of Operations and Production Management, 21(9), 1-28.
Treasury Taskforce. (1997). Partnerships for prosperity: The private finance initiative. London: Her Majesty's Treasury.
Treasury Taskforce. (1999). A step by step guide to the PFI procurement process. London: Her Majesty's Treasury.
van Aken, J. E. (2001, October). Management research based on the paradigm of the design sciences: The quest for tested and grounded technological rules. Eindhoven Centre for Innovation Studies: Working Papers Series. Eindhoven, The Netherlands: Eindhoven Centre for Innovation Studies. (Paper 01.11. 16).
Woodward, D. G., Chan, S. L. F., Smith, N. J., & Merna, A. (1991, November). The model concession agreement and financing issues. Project, 23-26.
Yin, R. K. (1994). Case study research, design and method (2nd ed.). London: Sage Publications.