Every iteration or release of a project requires some form of estimation. Schedule estimates generally start as ballpark numbers that need to be adjusted during the project. As factors stabilize, initial estimates can be more sure. You need to have stability in who the team consists of, their allocation, the completion of estimated scope, and control over additional scope that is not estimated. In addition, the average size of future scope items need to be close to the average size of historical scope items, which we will call “ave. hist. size.” The following facts make estimation very easy in a stable project environment.
- Work days equals Number of scope items of ave. hist. size (s) divided by Throughput
- Throughput equals Number of historical items that the same team fixed (a) divided by Number of work days required for all items
Combining the above, you have the formula, d = s / (a / w).
Say your team fixed 150 items (a) over 100 work days (d), and you have 30 items in scope. In addition, 8 of these items are twice the average size, or equivalent to 16 items of ave. hist. size (s). Therefore your total scope (s) is 30 – 8 + 16, or 38
Plug these numbers into the formula above, and you have d = 38s / (150a / 100d) = 38s X 100d / 150a. Remembering your Algebra, you know that because “s” and “a” are the same unit, these cancel each other. Therefore, 38 X 100d / 150 = 3800d / 150 = 25.33d. You can confidently expect to finish development of the 30 items within 26 work days plus a small buffer.
Notice that the scope items are not estimated individually, but as a group. This is the key to both simpler and more accurate estimation.
Tags: estimation, team capacity, committed scope, throughput, resource allocation, stable project environment, staying on schedule