Nov 13, 2020

The purpose and importance of a contingency analysis

  • Article
  • contingency analysis
  • project management

Cost contingency analysis is an aspect of cost estimation that is often misunderstood. Analysis methods such as predetermined percentages as a function of estimate maturity or study phase, deterministic calculation, expert judgment and probabilistic models can be used, but no matter which one is used, the proposed contingency amount is often understated and fails to stand up to third-party scrutiny.

  1. Why perform a contingency analysis?

    A contingency analysis is meant to address the following objectives:

    • Define the central estimate or project cost inclusive of contingencies, that is, the point representing an equal probability of overestimation and underestimation (also known as Pmean).
    • Define the accuracy of the estimate, measured from the central estimate, between predetermined points such as P10 to P90 (i.e., an 80% level of confidence).
    • Determine the level of confidence or the percentage of simulations that fall within the target precision measured from the central estimate (from Pmean).

    To produce this type of analysis, Monte Carlo simulation is typically used with a deterministic or probabilistic model. The preceding strategy is best suited to projects nearing full funding approval but could also be used in the pre-feasibility study.

    Cost contingency and risk allowance: What’s the difference?

    One common error of cost contingency analysis is failure to create a separate project risk register, and the attempt to generate a catch-all amount covering design developments, potential scope changes and materialized risks. Since contingency analysis and risk analysis are different, the two exercises should be done separately.

    “Contingencies” are “known unknowns,” that is, additional sums that may have to be committed in the framework of the project, but on which we do not have sufficient information. Project risks, on the other hand, are future events or uncertainties whose exact outcome is unknown and which could have an impact on cost and schedule. In other words, contingencies do not include potential scope changes or force majeure events, and in theory, the contingency reserve is meant to be spent in full during project execution.

    ISO 31000 defines risk as “the effect of uncertainty on objectives.” Here, uncertainty includes events that may or may not occur and uncertainties caused by lack of information or ambiguity. This covers both risks and opportunities. In this context, risk is different from contingencies because, theoretically, if we had enough time and resources to arrive at an accurate definition, the contingency reserve could approach zero, whereas the risks of a project may be mitigated, but not entirely ruled out. The probable consequences of risks involve uncertain future events that are, by definition, abnormal and outside the boundaries of planning and budget estimation.

    What contingency does not cover

    Here are some examples of risks that would not be covered by a contingency reserve, but which must appear in an official risk log: change in the nominal output of a facility, potential revenue losses due to delays, project acceleration, labour disruptions and closure in the event of a COVID‑19 lockdown.

    Another noteworthy pitfall in contingency analysis is the failure to appropriately segregate less well‑defined aspects of the estimate and an analysis that is too optimistic in its ranges.

    With the Monte Carlo method, the simulation also presents inherent characteristics that influence results. The estimator and project team should strive to perform contingency modelling independently of any preconceived notion of the results, and, at all costs, avoid narrowing ranges as a means of reducing the contingency amount or selecting a contingency amount corresponding to a probability lower than Pmean in order to lower the overall estimate value.

    Increasing or decreasing the contingency amount has no impact on project accuracy but simply changes the probability of underrun and the project’s risk profile.

    Conclusion

    A structured contingency analysis should effectively segregate elements that have lower engineering or cost maturity and identify project aspects involving a higher degree of complexity from those that are well defined. Variables assigned to contingency terms should generate sufficiently broad ranges that adequately cover potential growth in quantities, pricing and construction hours as a function of maturity level and site complexity.

    From the feasibility study stage onwards, the contingency analysis should allow the central estimate to be defined and to present contingency amounts as a probability of cost underrun; it should also be able to define the level of confidence of meeting the target accuracy.

    Lastly, the project should include a commitment to organizing an official risk assessment workshop, establishing a risk log and conducting a risk analysis before finalizing the estimate.

    Ideally, the estimate plan produced shortly after project kick-off would define the difference between cost and schedule contingencies and risks, along with the methodology to be used to determine the recommended cost contingency and residual risk amounts.

This content is for general information purposes only. All rights reserved ©BBA

Let’s
think it further
together
Contact us