Cost forecasts for our major projects rely on the assessment of events that may, or may not, occur during construction. A project forecast will include financial evaluations for these uncertain ‘risk’ events.

Take a situation where a civil engineer believes that there is a risk of encountering difficult rocks when building a tunnel. From experience the engineer judges the situation is 40% likely to happen and will cost £10m to overcome if it does. A provision of £4m (40% multiplied by £10m) is added to the project cost forecast. There will be many similar risks on a large project and the valuation of each will be added to the overall forecast.

The probabilities are generally subjective – different specialists, experts and advisors can legitimately have different views on the possibility of future outcomes. In the example above another engineer might judge the probability as 20% rather than 40%, in which case the provision will be halved to £2m (20% of £10m). Prior to the start of construction it will be very difficult to decide which engineer is ‘right’. When the project is complete it will be very hard to draw conclusions; if it happens/does not happen both engineers can say they said it might/might not happen.

Effort is required to resolve ambiguity between expert judgements. This will involve understanding why there are differences between expert views and seeking additional information. It might be possible to have a more robust forecast by working through to a consensus. Often differences in viewpoint will remain, in which case transparency is the responsible approach so that decision makers can factor relevant ambiguity into their plans.

The differences in judgement can be exploited to create a more favourable picture of the project. A simple way to lower the overall cost forecast is to always take the most favourable assessment; in our example use the 20% and ignore the 40% viewpoint. Do this across all the risk items in the project cost forecast and it can be taken to a much lower point than if, for instance, the average judgement was used.

Generally there is more likely to be consensus on the impact costs in this type of situation; the engineering remedy is likely to be more tangible with established methods and known timescales and material costs. However, when assessing probabilities nobody can be definitively wrong or right unless they state 0% or 100%; although in the long run it might become apparent that some individuals and organisations are more skilful, or more honest.

‘It appears that there are enormous differences of opinion as to the probability of a failure with loss of vehicle and of human life. The estimates range from roughly 1 in 100 to 1 in 100,000. The higher figures come from the working engineers, and the very low figures from management.…… It would appear that, for whatever purpose, be it for internal or external consumption, the management of NASA exaggerates the reliability of its product, to the point of fantasy.’ - Rogers Commission report into the Challenger Space Shuttle Disaster

Big Projects, Bad Predictions

Vulnerable Probabilities