Concept: When does a project need recovery? If you wait for a breach of TSPI’s threshold (1.10), it’s too late. If indicators are noodling along, not too great, not too horrible, it might be too early. Raising the alarm would hurt your credibility.
To decide, you need more information. Earned Schedule provides it.
Rather than start with the math, let’s start with its application. The math will come later.
Practice: Here’s an example that illustrates how uncertainty arises. And, then, what to do about it.
Decision Dilemma
Say that at month 6 of a 10 month project, the Schedule Performance Index for time looks like this:
The project got off to a slow start, but it improved quickly. The SPIt is still climbing! At over 96%, it looks pretty good.
How does the Estimate at Complete for time look?
Given that the Planned Duration is 10 months, the nominal EACt is very close to it, hovering just above the plan. The variance is still converging, but the nominal lies well within the current bounds.
Maybe, the project is not really in trouble, and there’s no justification for initiating recovery.
Like EACt, To Complete Schedule Performance Index looks to the future. What performance level is required to finish on time? Here are the results.
A late start with no EV throws off the first measurement. The first TSPI is already over the threshold. Given the false start, that’s not a good reason to initiate recovery.
Once EV begins to flow, TSPI drops away from the threshold and continues to ride just below it. Again, recovery is not clearly signalled.
Decision Support
In cases like the example, basic ES metrics can provide ambiguous signals. There are, however, more advanced ES metrics that help you make a decision.
One of the measurements is what Walt Lipke calls Period (or Window) of Opportunity. Here’s Walt’s description of the metric:
The Window of Opportunity is not the total remaining time in the schedule. Instead, it is the time from the status date to a TSPI threshold breach. The difference is important.
Often, we think that the time available for recovery is the remainder of the schedule. In the example, there are four months or about 120 days remaining.
That time box is an illusion. Walt’s research shows that once TSPI breaches the threshold value of 1.10, the project is unrecoverable. So, in reality, the only time left is between now and when the breach will occur.
The current level of performance marks the pace at which schedule is earned, which, in turn, determines how much room is left to recover.
The room is expressed as a fraction of the remaining time. If all the remaining time is available, the fraction is, obviously, 1.000, and it’s likely there will be no breach. If none of the remaining time is available, the fraction is, again obviously, 0.000, and it’s likely that there will be a breach.
Most often, the fraction is between 1.000 and 0.000. There is room to recover, but what’s important is how much.
Here’s the historical window of opportunity for the example.
In this case, the decision is less about the trend, which is down but not a sharp drop. Rather, it's about the end point. At the status date, the window has shrunk to .132.
The recovery window is not 122 days, it’s only 16 days. At the current level of performance, the project will be unrecoverable in less than a month.
With so little time to recover, it would be prudent to take action now. But, it’s possible that you might still want to hold off the decision. Recovery, after all, remains possible.
There’s more information available from Earned Schedule to help you decide. That's for next month.
PS I haven’t forgotten the math. But, this post is already long enough. I’ll save the math for later. |
