The next two chapters present the descriptions, data collections and analyses of two case studies on major infrastructure projects: Heathrow Terminal 5 and the Channel Tunnel Rail Link High Speed 1 (also referred to as T5 and HS1). Broadly similar major infrastructure projects in recent years in the UK include Wembley Stadium, High Speed 1 Channel Tunnel Rail Link, Heathrow Terminal 5, London 2012 Olympic Games and Crossrail Project. The best practices and learning from two major infrastructure projects (viz. Heathrow Terminal 5 and Channel Tunnel High speed 1) were presented at two closely connected MPA seminars during the first quarter of 2008. As a consequence some of the ‘elite interviewees’ who were members of MPA suggested a comparative study of these two high-profile projects. After reporting on these two cases, a comparison of the best practices of each is made, and used to test the three research propositions and also to explain the findings from the questionnaire-based survey and PLS modeling (see Chapter 7). Chapter 9 ends with the development of the Assessing Project Excellence (APEX) model.
According to Yin (2003) and Eisenhardt (1989), case studies are the preferred strategies of research on social science topics, especially for ‘how’ and ‘why’ questions, as well as enabling the researcher to focus on a contemporary phenomenon within a real-life environment. This is also supported by Robson (2002) who defines a case study as ‘a strategy for doing research which involves an empirical investigation of a particular contemporary phenomenon within its real life context’. The chosen case studies of Heathrow Terminal 5 (T5) and the Channel Tunnel Rail Link High Speed 1 (HS1) projects are both related to the contemporary phenomenon of efficient transport in a real-life environment. As indicated earlier, the five exogenous constructs as an output from the PLS modelling provided an input to case studies around these constructs. Therefore the case studies are presented and analysed around the five exogenous constructs of the conceptual model (see Figure 6.2). These are:
Quality management systems and procedures
Operational excellence concepts and applications
Self-assessments and knowledge management
Case Study 1: Heathrow Terminal 5 Project (T5)
Heathrow Terminal 5 opened on 27 March 2008 with the very public outcry of the failure of its baggage system. In analysing the root causes of the operational failure during the handover of the project, there are likely to be technical and human errors for both British Airways (BA) and the British Airports Authority (BAA); nevertheless, some of the exemplary best practices and initiatives of the T5 project should not be underestimated or overlooked.
BAA’s Terminal 5 (T5) programme at London Heathrow Airport is currently one of Europe’s largest construction projects. It is expected to cater for approximately 30 million passengers a year and provide additional terminal and aircraft parking capacity. T5 is also expected to feature a world-class transport interchange connecting road, rail and air. The facility represents a £4.3 billion investment for BAA.
To set new standards in delighting the traveller at T5;
To develop and deliver T5 to new industry standards of health safety and security;
To earn the proactive support and trust of key stakeholders;
To leave behind a legacy of quality.
The need for supplier partnerships in line with the T5 Agreement and the complexity of rail, road, construction and systems requirements of the project were additional project objectives. To achieve these challenging targets in money and programme, BAA had to consider a novel contracting and procurement strategy supported by a performance management system and an audit process. The best practices and learning points of the T5 agreement and balanced scorecard-based performance management system have been well documented (Basu et al 2009), but best practices and learning points can also be derived from other constructs of project quality and project excellence.
Quality Management Systems and Procedures
Quality strategy and procedures contribute to the Quality Management System (QMS) framework to set and deliver quality requirements by systematically integrating quality planning, quality assurance, quality control and quality improvement processes in a project. Figure 8.1 shows the structure of QMS adopted in the T5 project.
Quality planning: sets quality objectives and necessary procedures to deliver and assure the objectives and requirements are met.
Quality assurance: puts in place processes and procedures which provide assurance that quality requirements will be met. Quality assurance at the beginning of a project is more cost-effective than control measures used at later stages.
Quality control: focused on verifying quality requirements and standards.
Quality improvement: focuses on increasing the ability to meet quality objectives and requirements, and leads project quality to project excellence.
The primary focus of quality in the T5 project was the quality assurance of suppliers governed by the quality requirements specification and team execution plan. A formal document was drawn up for the quality requirements specification defining quality principles, policies and requirements for both the T5 project teams and first-tier suppliers. The execution planning processes were outlined in another formal document to address key management questions including, ‘Is work complete at handover?’.
The quality control of the T5 project was focused on periodic audit, a well-designed performance monitoring system supported by key indicators and measures and an inspection and test plan when work was completed. Figure 8.2 shows an overview of the quality assurance and control processes of the T5 project.
The quality improvement process was driven by audit reports and performance management RAG (red, amber and green) reports.
The quality procedures were continuously reviewed, with a formal structure of forums which allowed the information and decision process to be cascaded up and down the project quality organisation. The forums also enabled behavioural changes through the participation of quality personnel and leaders at all levels at relevant stages of the review process. Table 8.1 summarises the purpose and membership of the different quality forums in the T5 project.
Source: Terminal 5 Project.
Quality Audit and Compliance
Both audits and health checks have two primary objectives (Wateridge, 2002). The first is to ensure that the project is running satisfactorily according to the procedures and criteria. The second is to learn from practices of a completed project in order to improve the management of future schemes. Regardless of whether the project audit is conducted mid-term or at its conclusion, the procedure is similar and there is no shortage of audit processes or checklists. However, the audit and self-assessment practices of the T5 project have created a fresh approach involving all key stakeholders.
In the T5 project audits were carried out broadly in two areas — product audit and process audit. Product audits were aimed to review the status and readiness of product specifications by project teams (e.g. rail, systems, baggage, energy centre, airfield, site and logistics etc.) and product areas (e.g. structure and fabric, rail systems, passenger equipment, utilities, civil and earthworks and so forth.).
Process audits aimed to review the process conformance by project teams to provide assurance for gateway decisions. In addition, as shown in Figure 8.3, the audit activities were extended to supplier assessment and failure investigation.
Level 1 audits: designed for auditing the majority of smaller projects (c.300).
Level 2 audits: designed for auditing the top 30 projects as agreed by the T5 quality management executive.
Level 3 audits: conducted on critical projects by the full-time quality management team.
The levels of audit engagement were determined by the Quality Management Executive (QME) which comprised the engineering director, construction director and quality director. As shown in Figure 8.4, the QME assessed the level of engagement according to the importance and priority of sub-projects by taking into account the type of commercial contract, the business risk, supplier performance data and engineering criticality results.
The audits were conducted each month, and the selection of assessments at each level depended primarily on the decisions made during ‘milestone review meetings’ (‘gateway decisions’) for process compliance and ‘readiness for key construction events’ for product compliance. The robustness of the T5 audit processes is underpinned by the decisions made by the QME to determine appropriate levels, risk focus, safety, gateway decisions and readiness for key construction events. The audit results are also linked to performance management and improvement priorities.
Each audit was led by an appointed auditor who was responsible for the detailed planning of the appraisal, involving the audit manager as appropriate. Guidance and reference documents were used in planning the different types of assessment. The requirement for the pre-audit meeting was agreed between the auditor and the auditee. Pre-audit meetings proved to be helpful to both parties as they provided the auditee with a copy of the checklist in advance.
In addition to publishing ongoing audit results, there was also an annual overview of these results which were published along with KPI findings. A sample annual audit results overview (23/03/05 to 29/03/06) included:
71 product audits completed
66 process audits completed
9 readinesses for key stage events completed.
The report also showed red, amber and green findings. For example, process audits included 58 red findings and 133 amber findings. The audit findings were categorised, and reported with green status and typical assurance requirements, as shown in Table 8.2, as an example of product audits for 2005/6.
Source: Terminal 5 Project.
Suppliers need to ensure that they audit their own activities in order to provide an adequate level of assurance for both on- and off-site activities.
Project teams need to enhance supervision to ensure that production and conformance controls are implemented as planned.
The objective of auditing was not just to generate findings, but to bring about improvement in compliance with project requirements. Most of the findings relating to process non-compliance fell into the amber category. Red risks were very serious and this category was not used lightly; the failing was deemed sufficiently serious that project management considered stopping production until corrective action was taken. These findings were notified to appropriate project leaders as soon as possible in advance of the formal audit reports. Data received from auditees in response to audit findings were reviewed by the auditor or the audit manager. After satisfactory mitigation, the result was communicated to the action owner and then the audit finding was considered closed.
The KPIs are selected as high-level quality indicators to steer the major project objectives and requirements. The KPIs, supported by linked key measures, provide overall snapshots to direct the project through enablers, monitoring progress or assuring results. The performance data are the metrics which are measured for each part of the project by team members, including suppliers, to monitor performance as a target or planned versus actual. Finally, the key measures are the chosen ten measures to be reported and published regularly.
Table 8.3 shows an example of metrics for the manufacturing and assembly stage of the project. Each KPI is linked to relevant key measures and each key measure is supported by a number of performance data.
It is important to recognise that all metrics must be tried and tested with worked out examples. In addition they should be validated by collecting trial data under different conditions before they are communicated to the project team. It is helpful to provide a guidance note for each metric which can then be explained to team members in workshops to gain their understanding and acceptance. A similar process was followed for T5 performance metrics and a Quality KPI Workbook was prepared. The workbook contained a description and definition of each indicator and measure supported by guidance notes and individual or team responsibilities. For example, each of the KPIs and performance measures (also called data table heading) were supported by guidance notes for data collection and reporting as shown in Table 8.4 for the KPI Verifications planned and work supervised.
Source: Terminal 5 Project.
In order to clearly assign responsibility and accountability for each KPI, a simple RACI (responsible, accountable, consult and inform) format was used. Each team member or leader, either as an individual or as a team, was aware of their role as a sponsor (responsible), owner (accountable), contributor (consult) or participant (inform). The roll-out and implementation of the balanced scorecard-based performance management for the T5 project was enabled and enhanced by two major initiatives of the project. These were the ‘T5 Agreement’ and ‘Four-tiered approach of quality culture’. As part of the ‘T5 Agreement’ contractors were responsible for the relevant key performance measures.
Source: Terminal 5 Project.
Each project team (such as airfield, baggage, rail and so on) recorded, measured and monitored each performance measure and, on a monthly basis, the 10 key performance measures were reported as a balanced scorecard (Kaplan and Norton. 2004). Table 8.5 shows an example of a balanced scorecard.
Source: Terminal 5 Project.
The overall T5 results for key performance measures are also presented graphically as the quality management profile, as shown in Figure 8.6.
The key performance measures provide a snapshot of the operation of each project team which were also highlighted by RAG colour codes according to their status with regard to targets. However, improvement projects were acted upon more by individual performance measures at the specific project level. The most significant contributors to improvement projects were Non-conformance reports (NCRs). There were nine performance measures related to NCRs as part of one KPI, viz. compliance assured. These measures enabled the quantification of a part of ‘cost of poor quality’ (COPQ) given by estimated cost of NCRs. Root cause analyses by type of non-conformance and supplier led to continuous improvement in design, processes and savings. Figure 8.7 shows an example of NCR report analysis.
Overall c.6000 non-conformance reports were raised on T5, but the accumulative cost of non-conformance was only 0.6 per cent of the budget. Analysis of the data showed that 70 per cent of the total cost of non-conformance resulted from just 150 reports. A no-blame culture resulted in the speedy and effective resolution of all issues.
As discussed above, KPIs and key measures of the T5 project were customised to meet the requirements of the T5 Agreement and the complexity of the project spanning across rail, road and air infrastructures. However, the key balancing principles of the four aspects (financial, customer, internal processes and learning and growth) of the Kaplan and Norton balanced scorecard (Kaplan and Norton, 2004) have been incorporated in the T5 KPIs as shown in Figure 8.8.
In the Kaplan and Norton balanced scorecard, the enabling or leading indicators are ‘Financial’ and ‘Learning and growth’. In the T5 balanced scorecard, the enabling indicators are ‘Benchmarks agreed’ (which also include some financial benchmarks) and ‘Verifications planned and work supervised’ (containing supervisors training). As regards the lagging or results indicators, ‘Handover agreed and work complete’ in T5 relates to the ‘Customer’ aspect of Kaplan and Norton, while the T5 KPIs, ‘Inspected and protected’ and ‘Compliance ASSURED’ relate to the ‘Internal process’ aspect of Kaplan and Norton.
Source: Based on Kaplan and Norton (2004).
On a closer analysis, all the key measures as a group in each of the T5 KPIs do not conform to a specific aspect of the Kaplan and Norton balanced scorecard as shown in Figure 8.8. For example ‘Total estimated cost of NCRs’, which is a key measure of the KPI ‘Compliance assured’ also relates to the financial aspect. The matching of the T5 metrics is more appropriate at the level of key measures, as shown in Table 8.6.
Source: Terminal 5 Project.
Arguably there are some gaps in the T5 KPIs and key measures related to the ‘Financial’ and ‘Growth’ (innovation) aspects, but the manufacturing and assembly stage KPIs would not be expected to address this.
The metrics of the T5 balanced scorecard have been designed to reflect specific requirements of the project as enablers, as well as showing results leading to continuous improvement. The experience of the project team indicates that NCR-related data has been most effective in the identification of the cost of poor quality, to improve design and processes by analysing root causes by task or supplier, and also to attract the attention of the project board.
It is evident from preceding analysis that the fundamental principles of the balanced scorecard have been gainfully adopted and customised to the performance management systems of T5, in order to meet the specific requirements of this complex major project. The best practices of project performance management arising from this case study include:
Encouraging supplier partnership and proactive involvement of contractors in monitoring and improving project quality and conformance to standards.
Providing indicators and measures within three main themes as enablers, monitoring progress and showing results along the project life cycle, right up to the handover and completion of work.
The metrics and processes are validated and then embedded by extensive discussions with stakeholders followed by documentation, communication campaign and training workshops.
The ongoing testimony of NCRs, supported by the estimation of the cost of non-conformance and improvement projects based on root cause analysis, is a strong point of the process and opens up opportunities for Six Sigma and innovation.
Suppliers should be empowered to own the monitoring and improvement process using their performance data. Metrics should be customised within the framework of the Kaplan and Norton balanced scorecard.
This case study is an important first step in providing support towards measuring and improving quality standards in major projects. An initial research (Basu, 2008) indicates that in spite of formal quality plans supported by PRINCE2 and ISO 9000, many projects managed to ‘tick many boxes’ but failed to deliver expected quality criteria. The performance management system of the T5 project, having learnt from other major ventures, has established a ‘best practice’ of the application of a balanced scorecard approach in major projects by involving key stakeholders and contractors.
The efficacy of the T5 performance management system complements the audits and also underlines the gap in the five exogenous constructs of the conceptual model (see Figure 6.2). In addition it bolsters the need for performance management as an additional construct. This is in agreement with the conclusions on additional latent variables by Temme, Paulssen and Dannewald (2008).
The T5 project was given the go ahead on 20 November 2001 after 46 months of a public enquiry and consultation, the longest in British history. The project leadership was taken up by BAA (the owner/operator of Heathrow Airport) and Terminal 5 was built for British Airways as the project client and customer. During the long enquiry period, BAA went through training initiatives on other airport projects (such as Hong Kong airport) and previous large-scale infrastructure projects (such as the Jubilee Line Extension and the Channel Tunnel).
Figure 8.9 shows the high-level organisation structure of the T5 Project and the structure of quality management within this organogram.
Although BAA already possessed core project teams and technical leadership formed by their own employees, the project deployed around 100 key contractors and consultancy firms including three major contractors. Fundamental to the success of the supplier partnership was the T5 Agreement. The idea was to free consultants and contractors from the financial risk and in return empower them to manage and mitigate this risk. To this end, BAA took out a single insurance policy to cover the whole of the multibillion pound project. For the workers too, the project set the benchmark for construction pay. The ideas for this agreement emerged while plans for T5 were being scrutinised under the long-running public enquiry. As a result of the T5 agreement, each contractor must be totally open in their business dealings and firms like Balfour Beatty and Crown House shared information with each other, even though they were working on different parts of the project.
The head of quality, as part of the technical leadership department of BAA, was responsible for enabling and assuring the process quality of the scheme. Project group leaders of four main project groups were responsible for delivering the product quality of the completed work.
Each project group (such as rail and tunnel) had its own quality coordinators who worked alongside the quality coordination manager, who in turn reported to the head of quality. As part of the T5 agreement, suppliers within supply chain organisations provided QA/QC (quality assurance/quality control) resources who liaised with quality coordinators of the respective project teams. The audit schedule manager reported to the head of quality and managed both multidiscipline audit teams and independent surveillance teams. In order to clearly assign responsibility and accountability for each function, a simple RACI (responsible, accountable, consult and inform) format was used. Each team member or leader, whether as an individual or as a team, was aware of their role as either a sponsor (responsible), owner (accountable), contributor (consult) or participant (inform).
The role of the integration director appeared to be complex. This position was reported to be responsible for assuring ‘cost, time, quality and safety’ at Terminal 5 and also to manage the integration of 100 IT systems to airport operations, flight information boards, human resources information and administration systems (T5 Supplement, December 2006). To supervise the integration, a separate test centre was constructed and a series of trial runs were planned to ensure that all passenger services worked on the day T5 opened. There were also five large-scale passenger trials to ‘flood’ T5 with 2000 people acting as travelling passengers. The integration director was also the managing director of Heathrow airport, and acted as the link between the T5 project management (BAA) and client (BA).
In spite of these preparation plans, Terminal 5 opened on 27 March 2008 with the very public failure of its much-lauded baggage system. The terminal’s sole resident, British Airways (BA), had failed to properly prepare its employees for the handover of 70 per cent of its operations to the new building. With staff unable to get into position to carry out their tasks, it was not long before the baggage system choked and had to be shut down. According to one newspaper report,
BA and BAA ran 75 trials with a total of 15,000 would-be passengers, yet due to the need to keep the rest of organisation running concurrently, only 50 (5%) or so of the baggage handlers were fully trained on T5 equipment. Training for the majority of the 6,500 BA employees involved a day tour of the terminal and a couple of hours of watching reels about the impact of the change. (Financial Times, 5 April 2008)
Operational Excellence Concepts and Applications
The literature review clearly indicated the benefits of applying operational excellence concepts in major projects. These concepts include TQM, Six Sigma, Lean Thinking and supply chain management.
The T5 project did not apply the Six Sigma approach. It is arguable that if Bechtel had been chosen as a major contractor, then the likelihood of applying Six Sigma in T5 would have been high, and this could have enhanced the interface between BAA and BA during the launch of T5. Limited green belt training workshops were conducted for Heathrow operations but these were not extended to the T5 project teams. The presence of a TQM culture was evident in the four-tier approach to training but it was different from the rigour of Six Sigma deployment. A feasibility study of the application of EFQM was conducted at the early stage of the project and then aborted. This study report was not available for this research, and the reason for the failure to apply EFQM was not explained.
The much-applauded T5 Agreement made with suppliers underpins some key ingredients of both Lean Thinking and supply chain management in major projects. A supply chain management function was incorporated into the T5 Project organisation to make the best of the supplier partnership in the procurement of supply and services. In addition, this T5 Agreement:
had an emphasis on teamwork and joint problem solving;
had a joint risk management, problem solving and planning infrastructure;
stated that the suppliers’ profits were ring-fenced on top of openly audited costs, eliminating claims for extra payments;
was deliberately written in a non-adversarial style.
The use of prototyping activities. The roof of the T5 building, with a span of 150 metres, is one of the largest in Europe. BAA decided to trial the erection off-site, documenting the lessons learnt, leading to a three-month schedule saving when the roof was erected in situ. This use of prototyping is one of the core techniques of agile thinking.
The use of Lean principles. Work on the site was arranged according to a ‘demand-led’ or ‘pull’ system, where materials are delivered on a just-in-time basis. This constitutes one of the central tenets of Lean Thinking.
The use of iterative techniques. By using an expert joint review team, BAA is able to give approval for the construction of a feature to start before its design is complete. This is similar to agile software-development techniques where expansion starts before all the requirements are known.
Self-assessments and Knowledge Management
There were no specially designed self-assessment checklists in the T5 project to identify gaps, quality processes and required skill levels. A copy of the audit checklist was sent in advance to the auditee by the auditors, and the pre-audits following the checklist were the only considered form of self-assessment. This process was limited to the compliance of audit requirements, and therefore did not offer a comprehensive self-assessment method. The absence of an EFQM-based evaluation process also showed the weakness and vulnerability in quality procedures in areas which were not covered by quality KPIs or audits.
As described earlier, the training and communication aspects of the T5 project were enabled and enhanced by its two major initiatives. These were the T5 Agreement and the four-tiered approach of quality culture. As a result of the T5 Agreement, BAA shouldered the financial risk and expected the consultants and suppliers to work together. People from all stakeholders were encouraged to raise issues at the earliest opportunity. This in turn enhanced the reporting and discussions on performance and non-conformance issues. ‘When you align people’s objectives, stuff happens. The agreement has allowed us to work with our consultants and suppliers in a refreshing new way’, observed Andrew Wolstenholme, the T5 project director.
This four-tier approach is an ongoing process and is primarily driven by focussed discussion groups and workshops. The stakeholders engagement and commitment process is supported by the project executive commitment to engage with project leadership and suppliers (principals), and to introduce a RFT quality concept to ensure their buy-in and commitment. The culture and behaviour change process has been iterative, comprising regular workshops, briefing, awareness and feedback on quality KPIs and an RFT behavioural change programme. Quality certificates were awarded for site supervisors, and eventually site workers, who attended RFT workshops. This was further supported by the third-tier communication campaign, which included quality logo branding, quality commitment workshops, quality booklets, a quality walkabout and quality awards and posters. The fourth tier on quality best practice started with research and interviews with experts to establish best practices and align them with quality KPIs. This was followed by supervisor training and workshops to ensure understanding and ownership from supervisors.
There was a common safety education programme that aimed to make every individual accountable for safety. Everyone on site, from designers and managers, to supervisors and operatives, was given training on how to work safely. The project was supported by a journal called T5 which published selected articles to communicate general progress and best practices. The project team also shared best practices with industry (namely the BMW Mini plant in Oxford), professional bodies (the Major Projects Association) and major infrastructure projects (the Channel Tunnel Rail Link High Speed 1 Project).