Risk Strategy Development

The main focus on Crossrail has been to deliver the end-to-end railway. In order to support delivery during the latter stages of the programme there were a number of changes made to the risk management resources and processes. This paper focuses on the key changes that were made from 2015 to take the programme to completion. Specifically, the key elements are:

• introduction of a ‘programme risk’ approach to focus on the key stages of Crossrail
• centralisation of the risk team,
• centralisation of cost risk from ‘projects’ up to ‘programme’ level,
• changes to how the Anticipated Final Crossrail Direct Cost (AFCDC) was calculated; and
• preparation for demobilisation of risk

Context

When the project was initially set up, a detailed risk management approach was required to ensure that risks for each individual contract were being managed effectively and that the risk forecast was accurately modelled. This has been a key success for Crossrail with all contractors at all levels within the organisation fully engaged in the risk management procedure.
However, as the project focus moved away from individual contracts to a more integrated approach, with greater focus on the interfaces, risk management also needed to change. Hence, Programme Risk Management (PRM) was introduced to integrate and manage risks across the project as a whole, not just within silos. Alongside the changes to programme risk management, risk contingency was centralised on the programme to deal with costs that spanned across multiple contracts.

Introduction of Programme Risk to Focus on the Key Stages of Crossrail

To successfully implement Programme Risk Management there needed to be a greater focus on:

• the Crossrail delivery stages
• aligning risk with the schedule
• changes in corporate governance and specifically to the Risk Sub-Committee

Crossrail Delivery Stages

Crossrail is a large complex programme and programme risk management needs to break down the programme into manageable sections. Crossrail‘s delivery strategy has, since 2010, been built around commissioning the new railway over five distinct stages, planned to take place over 2 years. It therefore requires Programme Risk Management to focus on those schedule activities and the associated risks inherent in delivering each of the five stages. In the earlier years of the project, there was little emphasis placed on the Crossrail stages in terms of a risk management approach. This is simply a reflection of the programme management on Crossrail as a whole, which, until 2015, had focused mainly on the construction of the infrastructure itself.
Each commissioning stage inherently requires multiple entities to come together to deliver it. Therefore the focus of risk management had to move away from individual contract risks, instead establishing overall risks that were applicable to each stage individually. This meant that risks would be identified against multiple contracts and key external stakeholders and would be managed as one entity. Crossrail’s focus therefore shifted to integrating each contract to deliver the key stages, rather than the delivery of individual contracts.

Aligning Risk with the Master Operational Handover Schedule (MOHS)

In the past, risk management focused heavily on cost at the expense of analysing the schedule. Although cost has always remained a priority, programme risk management brought much greater focus onto the schedule, and the key milestones in commissioning each stage opening. The natural strategy for programme risk management was to align efforts on Stage 3, the most complex of all the stages. In order to do so, this required close collaboration with the planning team and the overall MOHS, which was key to defining the risk profile and completing the Quantitative Schedule Risk Analysis (QSRA) that has been undertaken at every Semi Annual Construction Report period since 2010.

Changes to the Risk Sub-Committee (RSC)

Prior to implementing risk management on crossrail there were two Executive level risk registers; the programme risk register and the strategic risk register. These were both reviewed by the RSC. The programme risk register was a combination of business risks and delivery risks and often caused problems in the RSC in terms of focus. To aid a better discussion the programme risk register was separated into two; programme delivery risks and crossrail business risks. Alternating every Period to focus on either the programme delivery risks or the business risks meant that full focus was given to each area and confusion between the two eliminated. Programme delivery risks included detailed risks from across the organisation that were brought together at a higher level and separated into the commissioning stages. To aid this review of programme delivery risks, the schedule was used to highlight risks against the key milestones for each stage. By doing this the RSC was able to review the mitigations to ensure they remained effective and could still intervene if the mitigation was found to be inadequate. In this situation, frequently, the focus of the meeting became about mitigating the risks to ensure delivery of the milestones.

Centralisation of the Risk Team

To support the introduction of programme risk management and to focus the risk team on the commissioning stages, the centralisation of the risk team has proven to be essential. In the past the risk team had been assigned to Sectors (Stations, Systemwide, Technical, Operations etc.) and reviewed the risks against each contract within their Sector accordingly. By centralising the risk team and focusing on risks to the delivery stages, an overlap was created between across each of the sectors. This resulted in a change to the risk profile. The risks identified were now focused around interfaces, integration, access and delays that were potentially impacting the delivery of the stages. As the risks had become more focused they created a higher impact (i.e. to the delivery of a stage). Increased impact then means that the value of the risk increases.
The key benefit to this approach was that risk management became much simpler as the meetings focused on one stage rather than having many meetings that looked across multiple individual contracts. As such, meetings became much more integrated, with multiple contracts and stakeholders working together at the same time. These meetings also enabled contracts to see what the impacts of their works were to a delivery stage. The other key benefit was that fewer risk resources were required, improving the overall efficiency of the team.

Centralisation of Risk Contingency to Programme

In order to support the implementation of programme risk and centralisation of the risk team, a new approach to how Crossrail undertook its Quantitative Cost Risk Analysis (QCRA) was also required. Crossrail removed risk contingency from each contract at Project, Sector and Programme level and centralised all risk to ‘programme’ level. Currently a QCRA is still undertaken at programme level each quarter but, as for reasons outlined later in this paper this will only be until the end of the 2017/18 financial year.
By removing the risk contingency held at project and sector level the need to undertake detailed analysis on each project was eliminated. It also gave the Executive team more control of the costs. The central risk contingency is made up of two parts; delivery risk and pure programme risk.

1. Delivery risk is available to address specific issues unique to each contract
2. Programme risk is available to address issues that stretch across multiple contracts and/or changes to scope or scope gaps.

Risks from across the stages and contracts are still escalated and reviewed at programme level but the requirement to undertake QCRA at contract level is redundant. Ad hoc QCRA is still undertaken (at contract level) as and when required, for example if there is slippage on the schedule or a number of high-value trends are raised.

Changes to How the AFCDC was Calculated

Prior to the centralisation of risk to programme, the Crossrail Anticipated Final Cost (AFC) was calculated as;

Crossrail AFC = Contract AFC + Unresolved Trend (URTs) + Risk

This somewhat skewed the overall number, as it assumed that a) all URTs would be resolved and b) all URT will be resolved at the cost at which they were raised. URTs are essentially an early warning of a potential contract AFC increase; they’re not a defined cost and are not certain to occur. Therefore the new AFCDC calculation had to exclude new trends raised outside of a quarterly QCRA. New trends should only affect AFCDC when they have been reviewed, accepted and are resolved. To achieve this, the implementation of the change was timed with the planned liabilitisation of a significant proportion of the unresolved trends, which, at the time, were principally against just a few major contracts. The completion of this yielded the new equation of;

Crossrail AFC = Contract AFC + Risk

….. where Risk is reviewed for adequacy against all remaining Unresolved Trend (URTs)

In the new process the main difference is that the Executive Team now defines what URTs are included in the risk forecast. URTs for the main contracts are reviewed and approved or rejected at the Commercial Change Sub-Committee and brought into the Executive Quarterly QRA meeting. As such, there has been a significant increase in the visibility of the trend process at Executive level, affording a certain amount of improved control over costs across the programme.
As Crossrail moves towards contract close-out, part of the Executive Quarterly QRA meeting is committed to reviewing the cost to contract close-out. As such, a professional assessment is made from the commercial team, led by the Commercial Director to establish an estimate of what that likely outturn cost will be based on the information from URTs, risk and other meetings held with Project Cost Engineers. This is also included in the programme risk contingency.

Preparation for Demobilisation of Risk

Crossrail has always undertaken a detailed QCRA each quarter. This QCRA determines the P50, P80 and P95 AFC at Programme level and the Programme S-curve demonstrating likelihood of exceeding the intervention points (see QCRA paper). Over time the difference between the P50, P80 and P95 forecast has been reducing, as is to be expected on a project of the scale of Crossrail. This is due to the increase in cost certainty as we move closer and closer towards completion along with the reduced uncertainty (risk).
As Crossrail entered 2018, the number of new risks being identified across the programme has reduced as each of the contracts progress to completion. Therefore the need to undertake detailed QCRA at programme level every quarter has also now become redundant.
On this basis the last QCRA on Crossrail will be undertaken at the end of 2017/18 financial year. The risk reviews will merge into the Executive periodic review meeting and although risks will continue to be reviewed they will not be modelled.

Lessons Learned

• The main learning for implementation of programme risk management is timing. Due to the size and scale of Crossrail, implementing a significant change to the risk management approach does take time, and Crossrail could have seen more benefits if this had been implemented earlier. For example integration activities, especially between the Stations and Systemwide teams, could have delivered greater benefits if started sooner.
• Due to the complexity and size of Crossrail and the fact that the changes took place in stages there was a point where we were focusing on stages but still undertaking very detailed QCRA’s at project and sector level. The lesson here would be to implement all of the changes at the same time. For this type of situation it is probably easier to implement four things at once rather than over a period of time.
• The level of Executive involvement in the risk management process has created a positive risk culture across the organisation. The executive team has always had a high level of involvement in the QCRA procedure. However, by centralising risk contingency, this has given the executive team much greater visibility and control of the costs and has resulted in much more ‘time on site’ talking directly to the project teams about their issues and challenges.
• At the outset of the Crossrail project, the schedule was very much a separate entity to the risk management procedure. By moving to programme risk management and utilising the key milestones in the schedule it meant that risk and planning were much more closely aligned. Potentially more benefits would have been gained if this had been integrated throughout the lifecycle of the programme.
• Putting the demobilisation plan in place in terms of risk management has forced the team to take a much more proactive approach to the simplification of processes as the programme approaches completion. The Crossrail Executive and RSC have, for years, been used to receiving risk information at certain times and in specific formats. The lesson learnt is that the demobilisation plan has assisted greatly in preparing the Executive for the reduction in the level of information they need to consume and has removed the ‘shock’ of terminating the risk programme altogether in the last months of Crossrail.
• Crossrail’s strategy has always included planned changes to the overall management structure as the programme moves through it’s phases of construction. However, at the functional/department level, there was a failure here to recognise the need to change risk as the programme moved forward. As stated earlier, these changes started in earnest in 2016. The lesson learnt is that Crossrail was probably a year too late and that on any other major programme, there should be a clear strategy from the start as to the anticipated timings to evolving the risk management approach, recognising the lifecycle of the project. The risk management development should align with the rest of the programme management team.