While helping a recent client understand the potential risks associated with their project schedules -as they were tired of projects blowing past their milestone dates - I found it useful to walk them through some of the concepts in David Hulett's "Practical Schedule Risk Analysis".
The key point of David's work is to make risk explicit in your schedules by thinking in terms of ranges of dates and costs for a project rather than a single point dates and costs Single point estimaties assume that each and every scheduled activity will be completed on time and to cost a quality expectations; this is an ideal that is never realised in the real world and has major risk implications for project delivery .
A simple way to reduce the schedule risk associated with this type of estimation is to use three point estimates (optimistic, expected, pessimistic) as opposed to single point estimates.
Using multiple estimates at the task level of a schedule allows you to account for differing degrees on confidence you may have in particular activities and their associated timings. These types of estimates can then be used within a Monte Carlo simulation tool to develop a probability based model of expected date and cost ranges.
In the example above there is 50% probablity that the project will finish on the 24th of August but a 90% probability it will complete by the 30th of August.
Schedule assurance, in particular schedule risk analysis allows you to make uncertainty explicit in your schedule and use that to identify key delivery risk factors and develop mitigation to help reduce those risks.