This paper addresses the subject of ‘schedule risk’ and how it may be determined in a valid and defensible manner. In doing so, it addresses an underlying management objective to have the best estimate of the statistical distribution that best represents the behaviour of the Total Schedule Time, as a stochastic variable. Having this information, allows management to properly assess the risk associated with the schedule and the impact thereof on other objectives, particularly project cost.
Schedule risk is represented by a probability distribution about its Mode, for the Total Schedule Time. Likewise, the risk associated with each activity comprising the schedule should be represented in the same way, given that the Total Schedule Time is a function of its component activities.
As an organised program of activities, a schedule may be a simple series of activities, a set of parallel activities or a complex combination of both, ie a network of activities. Because the latter is almost always the case in practice, risk analysis needs to follow an orderly process, one step at a time, from start to finish through the schedule.
Given details of a network, the process will be to determine (by estimation or assumption), in sequence:
- a probability distribution of time for each activity comprising the network;
- the extent of correlation between activities;
- the critical path of the network;
- the most likely value for the Total Schedule Time (Mode of the critical path);
- the probability distribution for the total time of the critical path; and
- the Mean of the distribution and its Range for the specified Confidence Limits.