ProjectFlightDeck

Concept: Some project topologies degrade the reliability of ES metrics early in a project’s timeline. ES Longest Path (ES-LP) alleviates the difficulty.

Chart 1

Practice: As described in the previous post, the impact of ES-LP on schedule management is still being explored. At ProjectFlightDeck, we have found that ES-LP improves metrics such as the Schedule Performance Index for time (SPIt).

The LP SPIt is derived from the ES(L) which is, in turn, based on the LP duration forecast. As described in the previous post, the ES(L) is defined as follows:

Following the pattern set by the standard SPIt, the LP SPIt is defined in this way:  

LP SPIt. Like standard SPIt, the LP SPIt is a measure of schedule performance efficiency. But, the LP SPIt is based on the ES(L) which is, in turn, derived from the LP EACt. Because the LP-ES avoids problems associated with false and misleading scenarios, the LP SPIt offers a more reliable reading on schedule performance.

As shown in Chart 1, the sample project used previously illustrates the difference between LP SPIt and standard SPIt. Not only is the LP SPIt more stable than the standard SPIt, it is generally lower, implying a less optimistic view of schedule performance.

The threshold values used for SPIt further clarify the difference between the two metrics. As you may recall, threshold values are used to calibrate the efficiency of schedule performance. Table 1 summarizes the thresholds:

For both LP SPIt and standard SPIt, if the SPIt is below .8, we say that schedule performance is very poor and assign a red status. When the SPIt is above .8 but below .9, performance is poor, and we apply a yellow label. If the SPIt is between .9 and 1.0, the performance is allocated a green status. In the green zone, SPIt of 1.0 (rarely seen in practice), indicates good performance, exactly aligned with expectations. Otherwise, the performance is considered fair.

Like the fuel efficiency of a car, the SPIt can go above the nominal rating. Although SPIt greater than 1.0 is often viewed positively, it can indicate problems. So, the red-yellow-green labels are also applied to SPIt greater than 1.0.

If the SPIt is greater than 1.0 but less than 1.1, performance is very good, and a green status is awarded. If the SPIt is greater than 1.1 but less than 1.2, the performance rating is remarkably high, and a yellow label is assigned. For SPIt greater than 1.2, performance is implausibly high, and the red status applies.

Table 2 shows what happens when we apply threshold values to the example.

Table 2

The abrupt swing in SPIt from red to green is indicative of lower stability in the standard SPIt values. The absence of “green” periods from the LP SPIt values indicates that schedule performance is worse than it appears from the standard SPIt.

Given that LP SPIt benefits from improved reliability, it is an ideal choice for managing highly parallel schedules or schedules where metric accuracy is of paramount importance.

Concept: Some project topologies degrade the reliability of ES metrics early in a project’s timeline. ES Longest Path (ES-LP) alleviates the difficulty.

Chart 1

Practice: The impact of ES-LP on schedule management is still being explored. At ProjectFlightDeck, we have found that ES-LP improves more than just duration forecasts. It also improves standard ES metrics such as the Earned Schedule measurement and the Schedule Performance Index for time (SPIt).

The ES-LP version of the ES metric is the ES(L)—the amount of time earned given the Longest Path. The Longest Path SPIt (LP SPIt) is a measure of schedule performance efficiency given the Longest Path.

ES(L) Burndown. As discussed in a previous post, the ES Burndown Chart is an effective way to depict the ES metric.  Like the standard ES Burndown, the ES(L) Burndown shows whether or not time is being used as planned on the project. Unlike the ES Burndown, the  ES(L) metric is not based on the standard ES forecast. A bit of math clarifies the difference.

Let’s start with ES. A simple derivation exposes the relationship between ES and the standard ES forecast, the Estimate at Completion for time (EACt):


From this equation, it follows that:

ES(L) , on the other hand, is defined as follows:  

In both cases, the metric is calculated as the Planned Duration times a performance factor. For ES, the performance factor is the ratio between the Actual Time and the standard forecast. For ES(L) the performance factor is the ratio between the Actual Time and the Longest Path forecast.

Unsurprisingly, the different performance factors produce different results (see Chart 1). Also unsurprisingly, the ES(L) Burndown varies from  the Expected Burndown more than the ES Burndown does. The reason is that ES-LP factors out shorter paths through the schedule. By focusing on the longest forecast path in each period, the ES(L) Burndown paints a less optimistic picture than the other Burndowns.

The ES(L) Burndown is easy to understand, making it an ideal communication tool. If the Burndown line is above the Expected Burndown line, the project is late. If the Burndown line is below the Expected Burndown line, the project is early.

ES(L) Burndown has an additional advantage: improved reliability. As discussed in previous posts, ES-LP serializes the schedule and avoids the pitfalls of false and misleading scenarios that can affect the accuracy of the forecast. The ES(L) Burndown, therefore, gives a more accurate reading of schedule performance, especially in highly parallel schedules.

The next post describes the impact of ES-LP on the SPIt.