Contingency planning is an essential part of project management simply because of the degree of risk and uncertainty involved in such an undertaking. It involves having some kind of safety net in case of unforeseen circumstances. Contingency is added to both the project budget and the project schedule in order to allow for unforeseen costs or additional work requirements. The degree of contingency to be added will depend on the degree of risk and uncertainty associated with the project. This can only be determined by carrying out a risk assessment. In general, there are two approaches to risk assessment. One is to assess the risk associated with every control package, and allocate contingency to each accordingly. The other is to allocate a single contingency figure for the entire project, and draw down contingency resources for individual control packages as and when needed. There are a number of advantages and disadvantages to both approaches, although the latter approach seems to be more widely used.

Project costs are to a large extent dependent on the amount of time it will take to complete the project, simply because this will be closely related to two of the major cost factors involved – man-hours and the total contribution to overheads. Any delays in completing the project are therefore likely to have a serious impact on cost. Perhaps the single biggest risk factor for any project is therefore the accuracy (or lack thereof) of the estimated task duration for each work package or control package. The estimates should of course be carried out by those responsible for doing the work and who (presumably) have the necessary knowledge and experience to come up with an accurate estimate. Historical data, if it exists, can also be used as a basis for estimating task duration. Ultimately, however, the accuracy of the estimate will depend the skill, knowledge and experience of the individual doing the estimating. It is essentially a best guess based on available information.

The degree of schedule contingency required will depend on the degree of certainty with which the original estimate has been made, and the degree of risk associated with any significant schedule slippage. If there is a penalty clause built into the contract for late completion, the degree of risk can be considered to be quite high. If the schedule itself is subject to a number of variables that cannot be accurately predicted, there is a more chance of it not being particularly accurate. Another issue that will affect how much schedule contingency is needed is the availability of additional resources that can be deployed in order to bring the project back on schedule, should things start to slip. If additional resources can be assigned to the project relatively easily, the degree of risk may not be so great. Alternatively, it may be possible to negotiate with the client to reduce the scope of the project in order to meet the project deadline.

The calculations required by the Project Evaluation and Review Technique (PERT) provide three time estimates for carrying out each work activity. The pessimistic estimate gives us a task duration based on a worst case scenario, while the optimistic estimate provides a task duration that assumes everything goes according to plan. Somewhere between the two lies the most likely estimate – essentially a best guess. The three figures thus derived can be used to find the expected task duration using a simple formula to calculate the average time the task would require if it were repeated a number of times over a given period. The PERT techniques are dealt with in more depth elsewhere, but they provide a means by which reasonably accurate task durations can be derived. They also give an indication of the range of values for task duration that might apply to a given work package. The numbers derived from a PERT analysis, together with the results of a risk analysis, can be used with a Monte Carlo simulation to analyse the likelihood of a range of possible outcomes. Monte Carlo methods are computational algorithms that repeatedly carry out the same calculations using pseudo-random numbers, and are often used to model the behaviour of systems in which there are a large number of variables (and therefore a high degree of uncertainty and risk). Many project management software packages now provide automated facilities for carrying out the necessary calculations. The risks associated with delays in the project schedule can thus be realistically assessed, and contingency time can be added to the schedule that is commensurate with the degree of risk involved.

Cost overruns have two primary causes. The first is due to the expenditure of more money than has actually been budgeted for the work planned, perhaps due to inaccurate estimates of the time required or poor task performance. The second is due to the addition of unplanned work to the project, either because something was left out of the work schedule at the planning stage, or because the scope of the project has been extended, or because technical issues have forced us to carry out the work in a different way. Budget contingency refers to a sum of money that is added to the initial project cost estimate to allow for such additional work. Remember that changes to scope will require an adjustment to the project baseline, both in terms of the budget and the schedule. If the project is being undertaken on behalf of a customer, any additional work requested by the customer that represents an extension to the project scope should be accompanied by a renegotiation of the contract.

It is important to manage contingency. It should not be seen as a way of compensating for poor task performance, or poor project management. It is in fact a reserve that should only be used to accommodate additional work, should it become necessary. If schedule or budget contingency (or both) are drawn down to bring a particular control package back on track, the project baseline should be amended accordingly. The action should be recorded, and will figure in the post-project appraisal process so that lessons for the future can be learned. Once the project is complete, any unused budget contingency is either included in the profits realised by the project, or is released for use elsewhere.