Earned Schedule overcomes known Schedule Variance and Schedule Performance Index shortcomings, which exhibit unclear trends towards the end of a project.
What Is Wrong with Schedule Variance / Schedule Performance Index?
Schedule Variance (SV) is criticised for its use of monetary units to express variances in time, while the Schedule Performance Index (SPI) is known to have unreliable behaviour from the middle to late stages of project progress. The figure below displays the cumulative increase of the Planned Value (PV) and Earned Value (EV) of an example project, which is used to illustrate SV and SPI flaws.
Time is Money
SV measured at the current review period, i.e., at actual time (AT), is shown as the vertical difference between the PV and EV and is expressed in dollars. However, it would be better if this variance were expressed in time since it measures the difference between planned work and earned work in time. Earned Schedule (ES) overcomes this limitation by projecting EV onto the PV curve.
Time Is Quirky
SV and SPI exhibit quirky and unreliable behaviour towards the latter stages as the project progresses. This is because the vertical distance between PV and EV narrows towards the end of the project and ultimately will be zero at the project finish.
This observation is explained by the EV formula, which is equal to PC * BAC. The percentage of the project completion (PC) varies between 0% (at the start), and 100% (at the finish) and hence EV varies between 0 and BAC, just like the PV curve.
This decreasing deviation between the PV and EV leads to SV = 0 and SPI = 1 at the project finish, indicating a perfect on-time project schedule, even when the project is unacceptably late. This unreliable trend has been described by Lipke (2003) in his paper “Schedule is different.” Further studies show this quirky behaviour starts from 50-to-60% project completion onwards.
As an alternative to the EVM time metrics, ES is an extension of EV and is calculated using the current EV value and the known PV line. Consequently, no extra input parameters are needed per review period to estimate the project’s ES value.
ES is the translation of EV expressed in monetary units into a metric expressed in a time unit. Its value is obtained by projecting the EV at the actual time AT onto the PV curve. Values of ES lower than AT denote a project delay while ES values larger than AT indicate a project ahead of schedule.
Earned Schedule Performance
Similar to the time performance measures of an EVM system, the ES metric is used as input for measuring and assessing the time performance of a project in progress, and results in two performance measures summarised as,
- Schedule Variance, SV(t), is a time-based alternative for the SV measure and expresses the time variance in a time unit rather than a monetary unit.
- Schedule Performance Index, SPI(t), is similar to SPI, a time-phased performance index expressed as a percentage of the expected baseline performance.
Referencing the Microsoft Project Demonstration Project contained in ‘Determine & Action Project Variance’ are the resulting SV(t) and SPI(t) charts which show the relative improvements in actionable insight when compared to the EVM SV and SPI metrics.
The advantage of the SV(t) and SPI(t) measures lies in their reliability to measure the correct time performance from project start to project finish. While the SPI is known to exhibit a quirky and unreliable trend towards seemingly perfect performance, even for projects ending dramatically late, the SPI(t) does not show this misleading trend and is therefore much more reliable.
If you would like to know more about leveraging data-driven actionable insights for your project schedule, then feel free to contact me on firstname.lastname@example.org.