Integrated Cost-Schedule Risk Analysis

David Hulett

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781317115298
• Edition: 1
• Topic: Business
• Subtopic: Bookkeeping & Budgets
• Index: Business

Project managers tend to believe their cost estimates - whether they have exceeded budgets in the past or not. It is dangerous to accept the engineering cost estimates, which are often optimistic or unrealistic. Though cost estimates incorporate contingency reserves below-the-line, these estimates of reserves often do not benefit from a rigorous assessment of risk to project costs. Risks to cost come from multiple sources including uncertain project duration, which is often ignored in cost risk analyses. In short, experience shows that cost estimating on projects is rarely successful - cost overruns routinely occur. There are effective ways to estimate the impact on the cost of complex projects from project risks of all types, including traditional cost-type risks and the indirect but often substantial impact from risks usually thought of as affecting project schedules. Integrated cost-schedule risk anlaysis helps us determine how likely the project will go over budget with the current plan, how much contingency reserve is required to achieve a desired level of certainty, and which risks are most important so the project manager can mitigate them and achieve a better result. Integrated Cost-Schedule Risk Analysis provides solutions for these and other challenges. This book follows on from David Hulett's highly-praised Practical Schedule Risk Analysis. It focuses on the way that schedule risk can generate cost risk, and how to handle this relationship. It also applies the Risk Driver Method to the analysis so that you can clearly and transparently identify the key risks, rather than just the most risky cost line items. With detailed worked examples and over 70 illustrations, Integrated Cost-Schedule Risk Analysis offers the definitive guide to this critically important aspect of project management from surely the world's leading commentator.

CHAPTER 1
Introduction:
Why Conduct Cost Risk Analysis?

Introductiosn

Key areas covered in this chapter include:

• some relevant experience of cost overruns from experience

• the nature of cost estimates including the components that make up most estimates

• why typical contingency estimates do not provide for risk on the subject project

• how corporate culture biases the estimates of cost and contingency reserve

• project risks are defined and discussed in general

• why project risk tends to cause overruns (as opposed to underruns or on-budget performance).

Project cost estimates are statements made by knowledgeable people such as engineers or estimators expressing their views about how much a completed project will cost. However, because events in the future are never completely known, it is wise to investigate the conditions and events that may make these projects cost more or less than the official estimate.

This is not a criticism of estimating, which is a prerequisite of cost risk analysis. Rather it is an explicit recognition of a conceptual difference between trying to estimate a single value for project cost (estimating) and performing a risk analysis of that single value. The risk analysis explicitly assesses the risks to that estimate and provides a range of possible costs with their probability as well as the risks that cause the difference between them and the estimate.

Using quantitative methods of project cost risk analysis (and its integral companion schedule risk analysis) provides the means to examine the impact of individual risks on the overall project cost. Cost risk analysis is the way to determine the likelihood of finishing the project on budget as well as to estimate the requisite contingency reserve needed to provide the desired amount of certainty about achieving the cost plus reserve. Cost risk can also identify the most important risks to project cost so risk mitigation can be effective.

Experience shows us that cost estimating on projects is rarely successful because cost overruns routinely occur. One study of public transportation infrastructure projects found that 9 out of 10 projects had overrun their initial estimates and that overruns of 50 to 100 percent were common. “We make the following observations regarding the distribution of inaccuracies of construction cost estimates. Costs are underestimated in almost 9 out of 10 projects. For a randomly selected project, the likelihood of actual costs being larger than estimated costs is 86 percent. The likelihood of actual costs being lower than or equal to estimated costs is 14 percent” (Flyvbjerg et al. 2002).

For IT projects, an industry study by the Standish Group found that average cost overrun was 43 percent and that 71 percent of projects were over budget, over time and under scope. Spectacular examples of cost overrun include the Sydney Opera House with 1,400 percent, and the Concorde supersonic airplane development with 1,100 percent. The cost overrun of Boston’s Big Dig or Central Artery/Tunnel Project was 275 percent. The cost overrun for the Channel Tunnel between the UK and France was 80 percent for Construction costs alone. Each of these projects also experienced increases in financing costs (Wikipedia 2009).

Edward Yourdon’s book Death March cites the generally accepted assessment of IT projects’ performance that the average IT project is likely to be 6 to 12 months late and 50 percent to 100 percent over budget (Yourdon 1997).

With this experience fresh in all project managers’ minds, and the minds of the public, is there a way to find out what these projects will really cost? Is there some way to identify risks to the project plan and its cost so we can mitigate them to achieve a better result? How can cost risk analysis help the project management community to avoid the worst of the cost overruns? In describing cost risk and its extension, integrated cost – schedule risk analysis, this book provides answers to these questions.

Project Cost Risk Analysis Purpose and Summary

Project cost risk analysis examines the risks to the project and specifies how they may affect the project schedule and costs. (Schedule risk analysis performs the same functions for the project schedule. See Hulett 2009). The analysis of cost risk can be conducted in the project’s conceptual development phase as soon as there is a notional budget, and should be continued periodically throughout project execution as the estimate is refined and more risks are identified and quantified.

In summary, the process of performing cost risk analysis using quantitative methods, particularly Monte Carlo simulation, can be described as follows:

• Quantitative cost risk analysis methods require gathering risk data, usually from in-depth interviews about project risks. Interviewees are carefully selected and interviews may be intense.

• A Monte Carlo simulation approach is used to develop the possible costs of the particular project plan under consideration. Monte Carlo simulation is internationally recognized as best practice.

• We strip the contingency reserve that has been included “below the line” or embedded in line item estimates from the estimate and start from line item estimates that are without contingency reserve or padding.

• Using simulations we then estimate a contingency reserve that is explicitly based on the risks of the specific projects. This contingency is appropriate for the plan that is in place and is based on the risks identified and quantified. Hence the accuracy of the result depends on the accuracy of information developed during the risk analysis as well as the underlying estimate.

• The estimation of possible project costs enables management or other stakeholders to adopt a budget that will provide their desired level of certainty (given the plan and the risks that have been identified, quantified and analyzed). Management or other stakeholders determine their desired level of certainty.

• The risks that contribute to the contingency so estimated can be identified and prioritized using various quantitative measures. The list of high-risk line items or high priority risks to the project cost forms the basis of risk mitigation efforts. In this way the quantitative risk analysis leads to a better project plan because it drives toward early mitigation of the important risks.

Project Cost Estimates

Project managers, sponsors or owners expect to be told how much a project will cost when it is completed. To gather project cost information they turn to their estimators or to an engineering company that is engaged to assist in the early stages of developing the project plan, including developing its schedule and cost parameters, to agree on a cost number. Those experts, in their turn, will have developed the work breakdown structure (WBS) for the project and populated the cost line items with estimates based on material take-offs for the specified equipment, the estimated cost of equipment, raw materials and labor, as well as many other project cost elements.

Estimators often have available to them cost results from prior projects to use in developing the estimates of cost for a current project. Estimators may have experience to bring to bear, often taking data from prior projects and adjusting them for project size and complexity or new technology, to be applicable to the project they are estimating.

In some cases, particularly on aerospace or defense projects, estimators have done statistical or parametric analysis of the relationships of cost to cost-driving project elements such as weight, power, capacity, function points or other quantity measures deemed to be relevant to indicating project cost. Technology readiness is sometimes a cost factor as well. Parametric estimating, using a database of past projects costs and cost-driving elements, involves computing regression parameters that can be used to extrapolate the cost of the current project based on its design parameters for those same cost-driving elements.

Cost estimators also use information from the market to determine costs. Equipment suppliers are asked to indicate (1) whether they are interested in supplying the equipment for the project and (2) at what price the services or equipment would be supplied. Information from the suppliers, which is not yet a formal fixed-price bid, will be used in developing cost estimates for the equipment. For other project elements such as labor and bulk material the estimators may conduct a market survey, although most large companies will have that information continually under review by a separate department or a standing subscription to a service that monitors labor or material markets.

Ultimately, when procurements are made at a fixed price and when contracts are signed for project management, fabrication, construction, installation and commissioning, the estimator has some pretty hard numbers to use in the estimate. Of course, fixed price or lump-sum turnkey may not mean that there is no longer any cost risk, because of external compensable factors and changes to the original plan, but those contracts are good bases for the cost estimate.

Ultimately the estimates of individual line items are combined to generate a bottom line number that represents what the project will cost based on myriad assumptions and calculations, sources of information and adjustments. In many projects there is an element of new work that cannot be estimated based on historical precedent, however, which means that detailed cost estimates will be built up from engineering information that should describe the new work accurately.

If the project is large and expensive, estimates and worksheets may be scrutinized and checked by third parties against technical specifications and data from market information.

Still, consumers of these estimates need to understand that the estimates calculated this way are based on specific assumptions that may turn out to be wrong, and hence are subject to uncertainty that usually delays the project schedule and often increases the cost of the project above the estimates, even with contingency added. Some of those assumptions are that specific risks will not occur, yet experience shows that risk occur on every project.

Contingency Reserves

Project cost estimators recognize the uncertainty in their estimates and almost always add a contingency reserve to the estimate below the line to make provision for the risks that might occur. In fact project stakeholders require the estimate and addition of a contingency reserve of cost before they take the cost estimate seriously.1 If this contingency reserve were correctly calibrated to cover the cost uncertainty in the estimate, then the sum of the estimates and the contingency reserve for risk should be enough to provide an estimated target that is a safe for planning and budgeting purposes.

Often the estimators have approved institutional contingency-sizing rules, usually percentages applied to specific components of the estimate. Sometimes percentage contingencies are even applied to numbers that already contain contingencies. Contingency amounts following these rules are historically based and can help to determine a believable total cost for the finished project. However:

• The process of developing a contingency reserve is complex and based on data that may not be correct for this project.

• There may be risks affecting this project that have not been encountered before and therefore are not incorporated in the reserve calculation.

The typical contingency reserve appears to provide for the average or expected (in the statistical sense, the mean value) level of uncertainty whereas the stakeholders typically want to be protected against more than the average level of risk. Often estimates have contingency costs or padding built into or allocated (spread) to the specific line items in addition to the contingency below the line. If this occurs, the estimator may be able to reconstruct contingency-free estimates that were generated and can be used for quantitative risk analysis. Contingency-free estimates are needed for a cost risk analysis.

Do Contingency Reserves Provide for Project Risk?

Contingency estimates that rely on percentages determined by engineering practice or history may not be appropriate for managing risk in the project.

Overall project risk is determined by the individual risks that affect the project, how likely those risks are to occur on the project, and what impact those risks may have on the project if they occur. Deciphering these currents and cross-currents without a formal risk analysis is a daunting task. It is often not performed in estimating the contingency reserve that accompanies the estimate, in favor of applying historical percentages and factors.

In addition to not knowing the extent of risk on the project, the estimator generally does not know the degree of risk aversion of the customer on this project. The degree of protection of the cost objective (degree of certainty) required by management or the customer is influenced by many factors including how the project fits within or advances the organization’s strategic objectives; the financial situation of the performing and customer organizations; the inherent degree of risk aversion of the organizations; the risk-taking behavior of the industry, and other factors.

It would be difficult for the estimator to gauge all of these factors simultaneously in sizing the contingency reserve of cost. How is the estimator to know the risks without examining them? Even if armed with a risk register that has risks prioritized using qualitative risk analysis methods (APM 2004; Hopkinson et al. 2008; PMI 2008) the estimator will not know how they fit together to affect overall project cost risk. The estimator generally falls back on institutionally fixed percentages to calculate cost contingency reserves. The estimates of contingency reserve generally provide for an average project contingency even if the stakeholders would prefer to have more protection from overrun risk.

The contingency reserve percentages used by the estimator and the estimating engineering organization may, but rarely do, provide for risk on a specific project. Since the project owners understand that the contingency is somewhat arbitrary (at least as it pertains to the specific project) the contingency is susceptible to adjustment by the owner. On some projects, the project owners have been known to persuade the estimator or engineering company to reduce the contingency reserve to serve a political purpose – reducing the apparent cost of the project – for the consumption of management, owners or other stakeholders.

In one instance the estimate of cost (before contingency) increased from the first to the second estimate, and the owner forced the engineering company to reduce the contingency reserve by a like amount so that the total estimate including contingency stayed the same in the two estimates. Although this example is extreme, owners’ disregard for the integrity of the contingency reserve is not uncommon and may be based on their lack of respect for the process by which the contingency is estimated. Wanting to please the owner and gain future estimating business, and realizing that the institutional contingency reserve percentages are subject to interpretation, the estimators at the engineering company may bow to owner pressure and revise the contingency reserve estimate. (Revise, in this context, means, in most cases, reduce it to whatever the customer will accept.)

There has to be a better approach to determining the cost contingency that provide both for the specific risks that may affect cost and the customer’s degree of risk aversion. This approach is quantitative cost risk analysis. In fact, since uncertainty in time influences uncertainty in cost, we need to apply both quantitative integrated cost and schedule risk analysis.

Why Do So Many Projects Overrun their Cost Estimates?

In many cases, project managers know that other projects have overrun their budgets, but they may feel that (1) this project will be secure from overruns because they are good project managers, (2) if the project overruns its budget the causes of those overruns will be evident to all and will not be judged to be their fault, or (3) since all projects overrun there will be no personal consequence to them. There is a fair amount of self-deception in the project management world about both cost estimates and schedules.

The main reasons for costs to overrun the estimates, assuming that the cost estimators are competent, relate to the corporate culture...