How is earned schedule different from schedule variance?

Classic schedule variance measures the lag in dollars; earned schedule measures it in time. Both start from the same fact, that less work has been earned than planned, but they express it differently. Schedule variance says six hundred thousand dollars of planned work is outstanding. Earned schedule converts that into when: the work done by the month-eight data date was actually planned to be finished around month seven and a half, so the project is roughly half a month behind in real time. For anyone building a recovery plan, half a month is a far more useful number than six hundred thousand dollars.

How do you calculate earned schedule?

Find the point on the baseline where the planned value equals the earned value you have now, and read off its time. On our example, earned value is five point four million. Walk back along the planned-value curve to where it last read five point four million, which happens at about month seven and a half; that time is the earned schedule. Then the time-based schedule variance is earned schedule minus the actual time, 7.5 minus 8, so about negative half a month, and the time-based index SPI(t) is 7.5 divided by 8, about 0.93.

Why does earned schedule forecast the finish date better?

Because it does not heal itself at the end. Classic SPI mechanically returns to 1.0 at completion, since all planned value eventually gets earned, so near the finish it tells you a late project is on time. Earned schedule has no such flaw: it keeps reading the true time lag right up to the last day, so a project that is a month behind still reads a month behind in month fifteen of sixteen. That makes SPI(t) a sound basis for projecting the actual completion date, where classic SPI collapses into uselessness exactly when the finish date matters most.

Construction Glossary · Earned Value

EarnedSchedule

The fix for earned value's oldest flaw. Earned schedule reads the lag in weeks instead of dollars, and keeps telling the truth even in the final month, where classic SPI gives up.

Definition

Earned schedule (ES) is the time-based version of the earned-value schedule reading. It is the point in time at which the work done so far was planned to have been earned. Schedule variance in time is ES minus actual time (SV(t) = ES - AT) and the time-based index is SPI(t) = ES / AT, which, unlike classic SPI, stays honest right to the finish.

Reading the lag in time

The $5.4M earned by the month-8 data date was planned to be reached back at month 7.5. The horizontal distance between those two times, not the vertical dollar gap, is the schedule variance in time.

Why it beats classic SPI

Classic SPI heals itself

Because all planned value is eventually earned, SPI returns to exactly 1.0 at completion, even on a project that finished months late. In the final stretch it reports a late job as on time.

SPI(t) stays honest

Measured along the time axis, SPI(t) keeps reading the true lag right up to the last day, which is exactly why it is the sounder basis for forecasting the real completion date.

How POD reads schedule performance

POD computes the schedule performance index from the earned value and baseline planned value it already holds, and pairs it with the float and critical path it reads from imported P6 and MS Project files. The dollar reading and the program sit side by side, so the schedule story stays anchored to the contractor's own program rather than a recomputed one.

Earned schedule & SPI guide

Frequently asked questions

The schedule reading that holds to the end

All of it, cost and schedule, rests on an honest earned value, and earned value rests on an honest percent complete.

Percent Complete Back to the Glossary