Today, I released a minor version update to Statistical PERT® Normal Edition. Version 2.1 adds a new worksheet, SPERT Normal® (Mixed entry). This new worksheet offers estimators a very flexible way to enter uncertainties into a SPERT worksheet. Using this worksheet, estimators can very easily create three-point estimates with little effort using heuristics to generate the minimum and maximum extreme values for each uncertainty, and then estimators can selectively choose to change those auto-created values with row-level, exception-based values.
In this worksheet, estimators have three ways to create the necessary 3-point estimate that every Statistical PERT calculation requires. Firstly, two global heuristics are specified above the Minimum and Maximum column headings. The Minimum heuristic will reduce the value entered in the Most Likely column by a percentage value. Similarly, the Maximum heuristic will increase the value in the Most Likely column by a percentage value. These two heuristics will affect all entered uncertainties in this particular worksheet.
Estimators can, at the row level, override any globally-derived Minimum and Maximum point-estimates in one of two ways, One way to do that is specify a row-level heuristic by selecting a percentage value in either the Min % or Max % column. These percentages only affect a single row. They both either reduce or increase the value in the Most Likely column, just like the global heuristics do, but only at a single row level. The other way estimators can override a calculated minimum and/or maximum point-estimate is to manually enter a value in the Min point and/or Max point column. Manually entering a minimum and/or maximum point-estimate will always override any point-estimate value that is a calculated result.
Most Likely point-estimates must always be manually entered by an estimator; there is no built-in way to calculate a Most Likely point-estimate. But once a Most Likely point-estimate is entered, an estimator can either use a global or row-level heuristic to create the minimum and/or maximum point-estimates for each uncertainty, or estimators can continue to manually enter specific values for those point-estimates.
While this new worksheet may look a little more complex because of the extra columns, using it is still very simple. This simply makes it easier for an estimator to create exception-based point-estimates when dealing with a large number of uncertainties in the SPERT worksheet.
The example workbook includes examples of all three ways to create minimum and maximum point-estimates. The first three rows (ID 1, 2 and 3) have minimum and maximum point-estimates that are created from the global heuristics. For these three uncertainties, minimum point-estimates are 50% less than the most likely point-estimate, and maximum point-estimates are 100% greater than the most likely point-estimate.
ID 4, however, overrides the global heuristic with two row-level heuristics: the minimum point-estimate is only 10% less than the most likely outcome, and the maximum point-estimate is only 50% greater.
ID 5 uses a row-level heuristic to reduce the most likely outcome by 25%, but the maximum point-estimate (160) is manually entered and overrides the global heuristic for creating maximum point-estimates.
ID 6 specifies a value of 40 for the minimum point-estimate, which overrides the global heuristic for creating minimum point-estimates, and the maximum point-estimate of 200 is manually entered, too.
ID 7 specifies 80 for the minimum point-estimate, but uses a row-level heuristic to create a maximum point-estimate that is 75% greater than ID 7’s most likely value.
So, the new Mixed entry worksheet offers estimators the greatest amount of flexibility to create 3-point estimates in the easiest way possible. Download Version 2.1 now!