The traditional approach for estimating cost or schedule starts with the "best estimates" for the Most Likely durations and costs for tasks. This is the standard approach for PERT based projects, for approaches suggested by the PMBOK or PMI style of project management.
The steps usually go like this:
- Build a network of the project’s activities
- Determine the “best” estimate of the duration for each activity in the network
- Compare the activities’ “best” estimates to find the critical path
- Sum all the “best estimate” durations of activities on the critical path
- Define this sum of tasks to be “best” estimate of the project’s schedule duration
- This will almost always be optimistically wrong (remember the discussion about Anchoring)
- Or it is pessimistically wrong
- Either way – it is wrong
The problem with the term "best" is there is no standard definition.
For each activity, the "best" estimate is ...
- The “most likely” duration – the mode of the distribution of durations? (Mode is the number that appears most often)
- It’s 50th percentile duration – the median of the distribution? (Median is the number in the middle of all the numbers)
- It’s expected duration – the mean of the distribution? (Mean is the average of all the numbers)
These definitions lead to values that are almost always different from each other Rolling up the “best” estimate of completion is almost never one of these.
Durations are probability samples not single point values
We know this because ...
- “Best” estimate is not the only possible estimate, so other estimates must be considered “worse”
- Common use of the phrase “most likely duration” assumes that other possible durations are “less likely”
- “Mean,” “median,” and “mode” are statistical terms characteristic of probability distributions
This implies activity distributions have probability distributions. These are random variables drawn from the probability distribution function (pdf). The “Actual” project duration is an uncertain quality that can be modeled as a sum of random variables. The PDF may be known or unknown.
These durations are "educated guesses," but rarely do they have known underlying probability distributions.
