Earned Value Management (EVM) provides a quantitative measurement of the status of a project. This is particularly useful for complex decommissioning work, which can be subject to a large amount of change and uncertainty. However, it does require a well-defined project plan and a large amount of work up front. By Alastair Laird
Ask any project manager about their biggest problems and they will often say schedule and cost overruns associated with inadequate planning and project controls. With Earned Value Management (EVM), success cannot be guaranteed but in most cases surprises can be avoided and early identification of problems can optimise the time left for effective recovery planning. To improve delivery management, project managers need to understand how to effectively monitor and measure the project objectives, cost and schedule. Where complex nuclear decommissioning programmes or baselines are involved EVM is the proven tool.
An Earned Value Management System (EVMS) helps integrate the project work, that is, the scope, cost, schedule and risk, into a single performance measurement baseline (PMB). EVM tracks three key measurements in each reporting period: planned value of the work to be performed, the earned value of the work performed (its budgeted cost) and the actual cost of the work performed. It provides trend analysis, forecast capability of estimated cost at completion, a sound basis for variance analysis and hence informs corrective action planning. Using EVMS allows management to determine "where have we been," "where we are now," and "where are we going." It provides project memory and performance data. A new and initially foreign language, the EVMS terminology needs to be learned and users need to be tuned to its messages — but once the concepts and language become familiar the power of this tool never leaves you.
The American National Standard ANSI/EIA-748-1998 is internationally recognised. It provides guidelines incorporating best practice for the implementation of an effective Earned Value Management system. Its principles are used throughout the US and UK nuclear decommissioning industry. EVM has also been adopted as the standard project management tool for decommissioning in the European Union.
The EVM system
The basic concepts of an EVMS involve planning, integration, performance assessment, analysis of variances and reporting (see Table 1).
The inputs required for an EVMS include:
- Work definition. It is important to clearly define the work and know the end point.
- Work Breakdown Structure (WBS), a way of organising a project into groups of activities and creating hierarchy
- Organisational Breakdown Structure (OBS), a hierarchal breakdown structure for the organisations responsible for delivering the WBS
- Project schedule and critical path
- Time-phased baseline budget. This will be based on a reliable database of a fully-characterised facility and a known end point condition.
- Cost/resource control plan
- Change control plan. This would be triggered for a number of reasons, such as significant variances, changes to project scope or funding. Project managers would then determine what changes were necessary to WBS, OBS and so on to bring the project back under control.
- Risk register. This would include strategic, programme, project and contract-related risks.
The use of Earned Value (EV) will be compromised if any of these inputs are not available or have not been completed appropriately. This will result in the outputs not reflecting the correct current and future status/progress of the project.
With the EVMS much of the work is involved in the planning stage. Before you execute, you need to know exactly what you want. For instance you could spend anywhere from 2-3% of your overall budget to as much as 20-30% of the budget on WBS and OBS and pre-planning. This pre-planning may include detailed investigation, construction of mock-up, and so on.
The process of developing a schedule and cost baseline is interactive. The schedule affects the costs and the costs affect the schedule. Risk and estimating uncertainty also needs to be factored into both the costs and the schedule to add realism and adjust for project manager enthusiasm (optimism bias). The planning concepts of base or P0 (lowest in terms of cost and quickest time) and realistic or P50 (project has 50% chance of being completed on budget and schedule) are vital to understand. A centrally-held management reserve is the normal outcome of client contingency management and a natural attempt to avoid work expanding to fill the time available for its completion – Parkinson’s law. In simple terms project budgets are rarely under-spent, so client behaviours in setting budgets are very important and to do this a robust schedule and cost baseline is key. In the case of decommissioning projects the project manager will often be given a budget of less than P50. The head of a site would hold the contingency. In addition, a national body such as the Nuclear Decommissioning Authority in the UK would hold a management reserve.
Many nuclear decommissioning projects are subject to ‘known unknowns’: expected uncertainties. Assessment and characterization of these is the precursor to robust assessment and allocation of contingency funds. Known unknowns in the decommissioning industry arise from radioactive inventory and contamination associated with operational activities and decommissioning tasks. Monte Carlo analysis is used to assess these uncertainties, to generate a budget and to consider the probabilities and impacts of any discrete risks logged in the register.
Value for money
How can value for money be measured in a business environment with options, uncertainties in planning and also during delivery? The answer lies in the management process, which should include concepts such as stage gating (a decision-making process in which continuation to the next stage is decided by a manager) and decision maturity (understanding the specification and knowing it is not going to change). Governance in investment decisions, stage gating, options selection and decision maturity through design stages all play a part in the front end. These qualities also support defining the right outcome. Having effective delivery metrics and early warnings during execution are the windscreen that EVMS provides to optimise the subsequent journey and enable delivery of the desired outcome.
Management of a project or programme must back up investment decision governance with delivery governance. It should include regular integrated baseline reviews, effective change control, transparent contingency/management reserve drawdown and regular EVMS performance reporting.
A simplified project management office set-up would comprise a project manager and a project controller. The project controller would carry out monthly reviews, looking at actual costs, earned values and variances. The project manager would look at the reports and decide whether to re-plan activities or to effect change control. In order for EVM to work effectively, the organisation’s systems for collecting accurate cost data need to be in place.
One should only plan ahead as far as one can see, so re-planning any ‘out-years’ is a key feature of good baseline governance. Near-term work management is essential for EVMS to work effectively. In some cases it does not make sense to continue with the current plan. For instance, take a decommissioning project in which you plan to dispose of some waste in a local landfill. Landfill prices have recently doubled under national policy. There is no way that you would be able to get back to the baseline plan. As a client it makes sense to go back and change the plans so that the earned value (budgeted cost of work performed) reflects the actual costs.
There are a number of determining factors that need to be in place to ensure successful delivery through implementation of EVM. First and foremost, EVM systems and behaviours should be driven by senior management; the highest levels of the organisation need to understand and drive the assessment and reporting process. Second, key sub-contractor EVM systems need to be good; possibly clients should insist in certification of more strategic suppliers. Third, people involved in key roles need to be competent in the corporate project controls approach and the EVMS.
EVM uses the planned schedule and budget for a project along with an analysis of the actual work that has been performed to produce three values that indicate the progress and performance of the project. These values are planned value, earned value and actual value (see worked example, below).
|EVMS worked example|
Planned Value = BCWS = Budgeted cost of work schedule/What was planned
Example: The project budget is £100,000. At this point in the project, 60% of the work should be complete so BCWS is £60,000. Only 50% of the work is actually completed so BCWP is £50,000. Actual costs (ACWP) is £65,000.
EVM also calculates two variances:
Example: Using the scenario above, the CV is -£15,000 because it has cost £65,000 to complete £50,000 of planned work. The SV is -£10,000 because only £50,000 of the planned £60,000 worth of work has been completed. This project is behind schedule by £10,000 worth of work.
EVM indexes are often used to indicate the performance of the project:
Example: Using the Scenario above, the CPI is 0.77 and the SPI is 0.83, showing that this project is behind schedule and over budget to an even greater extent.
Figure 1 shows a graphical comparison of the budget value of the work schedules, the actual cost of the work completed and the earned value of the physical work actually completed.
A useful method of tracking progress on a project is to track calculated indices such as SPI and CPI versus time. These indices can also be generated as a function of reporting period (monthly), funding cycle (yearly) or project lifecycle. This helps understand the effectiveness of recovery actions at a future point through to project completion (see Figure 2).
Care needs to be taken when using the SPI to monitor the progress of the project as it does not necessarily take into account activities on the critical path, which determine the minimum time needed for the project. It is therefore possible to fail to achieve the critical path schedule but have an SPI > 1.0 because earned value is being taken for non-critical path activities. For this reason, key milestones are almost always incentivized in the nuclear decommissioning industry, with a bonus paid to the contractor for completion on schedule. A report on key milestone progress is an essential complement to all good project reporting.
A key element of any EVMS is the ability to forecast the completion of the project in terms of final cost. The Estimate At Completion (EAC) of a project is the actual cost to date plus an objective estimate of costs for the authorised work that is remaining. There are numerous ways to calculate an EAC. The most commonly-used formula for EAC is:
EAC = ACWP + Estimate To Complete (ETC).
This is the key area where PM judgement is likely to be exercised in terms of recovery plan effectiveness and impact across the project lifecycle. Good PMs will rarely try to hide or mask issues and simply use the available EVM data to inform their assessment. If they get it right EVMS convergence and SPI and/or CPI recovery will become apparent.
Benefits of EVM
There are many benefits of EVMS. Perhaps the biggest benefit of implementing EVM is that it is a single system for tracking scope, time and money. If a common work breakdown structure is used, then benchmarking costs and comparing like-for-like scope becomes a simple task. This is the case within project, within programme, within baseline or at a regional or national programme level. (In this context ‘programme’ covers the entire scope of the decommissioning baselines to achieve the end point state).
For decommissioning, sharing experience and learning how to deal with uncertainty and known unknowns is critical and becoming an emerging area of international decommissioning discussion. The benefit of benchmarking completes the cycle and re-emphasises the need to get the scope definition right during the previous planning step.
Disadvantages of EVM
Implementing EVM can be considered by some to be costly and time-consuming. It often requires the use of specialist software. The cooperative interaction between all elements of the business, and collection of actual data can also be time-consuming.
Most of the work and time is needed in the project planning stage, as having a clearly defined, complete and documented scope of work and objectives is key for EVM.
If all elements required for an effective EVMS are not present, then the earned value analysis data may be misleading.
If the work breakdown structure, scope, schedule and cost are not fully integrated, then simple milestone management and counting actual costs is the default. If work is being conducted that is not in the defined project scope then failure to exercise change control will result in failure to understand the performance metrics.
Alastair Laird, Business Director, Project Time & Cost International Ltd, Innovation Centre, Westlakes Science and Technology Park, Moor Row, Cumbria, CA24 3TP, United Kingdom