Concept: The time available for recovery is expressed by the Window of Opportunity. That’s not just the remainder of the schedule. It’s the point at which the To-Complete Schedule Performance Index (TSPI) will hit 1.10—at that point, the project becomes unrecoverable. And, that is often much sooner than the project’s end date.
What must be achieved in that fraction? TSPI gives the performance for the remainder of the timeline. But, as just pointed out, that’s often longer than the recovery window itself. So, what’s required exclusively in the Window?
Walt Lipke answers the question. He’s developed math that calculates the efficiency required for each period in the Window—the “Improvement Profile”.
Successful recovery depends on achieving the efficiencies in the Profile.
Practice: As for other recovery metrics, practice with the Improvement Profile does not hinge on understanding the underlying math. This is a case, moreover, in which Walt explains the math. So, I’ll move directly to its implementation. FYI, I’ve included an outline of the math in the Appendix.
The example offered in the February post has background numbers. They’re repeated here.
Part way through a 10-month project, the Schedule Performance Index for time looks like this:
The project got off to a slow start, but it improved significantly. The SPIt is currently well above 90%, and it’s still climbing! That looks pretty good. It certainly doesn’t cry out for recovery.
How does the Estimate at Complete for time look?
Given that the Planned Duration is 10 months, the estimates are not too bad. The variance is converging, and the nominal amount is just slightly above plan.
Maybe, the project is not really in trouble, and there’s no reason to initiate recovery.
What does the To Complete Schedule Performance Index tell us?
The late start and its nil SPIt throw off the first measurement. The other measurements seem to be hovering below the 1.1 threshold. Again, there’s no clear signal to start recovery.
So, what do you do?
Get more information.
The Window of Opportunity described in the February post helps.
Based on current performance, the window is .132. That’s a narrow window—so narrow, that it would be reasonable to initiate recovery.
Still, recovery is a dramatic step, and a window remains open. Maybe, there’s other information to help us decide.
One such piece of information is how the recovery needs to proceed. With that, we can assess whether the required efficiencies are realistic.
The Improvement Profile gives us the data. It looks like this.
As the graph makes clear, the efficiency improvements are striking. Almost all of them exceed the maximum efficiency attained thus far.
Furthermore, after a sharp initial increase, efficiency has leveled off--well below the desired level for the future.
That’s more evidence to start recovery.
And, yet, it might be argued that the desired recovery *rate* appears to be lower than the rate achieved through the first three periods.
(Mathematically, the initial rate is “N/A” because it starts from 0. The slope of the first curve, however, suggests that the project is capable of improving efficiency quickly.)
It’s a thin thread, but there’s reason to hold the decision in abeyance.
That is, until next month. Then, a final recovery metric will conclude the story.
** **
**Appendix: Improvement Profile Math**
Walt derives the Improvement Profile through a clever combination of geometry and algebra. He uses geometry to create base equations. He then uses algebra to isolate the key parameter.
TSPI is used to set the overall efficiency required to complete on schedule. A certain level of efficiency is then required within the Window to avoid hitting the 1.1 threshold.
That level of efficiency is called the “recovery efficiency” or SPItr—the schedule performance index for time required for recovery.
Here’s the equation Walt deduces:
SPItr= [(TSPI - SPIt) x ((Estimated Duration – Actual Time)) / (Estimated Duration – Actual Time – (Improvement Period/2))] + SPIt.
The recovery efficiency in the improvement periods equals the overall efficiency required to complete on time (TSPI) less the efficiency currently achieved (SPIt).
That amount is scaled to the time available for improvement. Finally, the increment of efficiency required is added to the current efficiency.
Walt’s math goes on to allocate a portion of the required improvement to each individual period in the Window. That gives us an Improvement Profile with efficiency broken out by period.
To view Walt’s derivation of the key equation, see the Derivation SPItr tab of his Prediction Calculator. For the allocation to individual periods, expose the formulas used in the Improvement Profile section of the Calculator & Graphs tab of the same spreadsheet.
I realize that the math can appear intimidating. Happily, its implementation does not require understanding the derivation.
An analogy explains why I’m sanguine in making such a claim.
Meteorologists routinely use kilograms in their calculations. I dare say that most of them do so without a deep understanding of quantum mechanics.
Say what?
The kilogram was recently re-defined by International System of Units (SI). To do so, they used Planck’s constant. That number originated in quantum mechanics over a hundred years ago. Scientists with a deep understanding of the math figured out how to use the constant (plus a “Kibble balance”) to define the mass of a kilogram.
So, meteorologists no longer depend on a lump of metal produced in the 19^{th} century. They have a more accurate, reliable, and scalable measure to use in their calculations. They do so without necessarily understanding the physics behind it.
For more on this analogy, see the article here, |