What is an estimate at completion?
An estimate at completion, usually shortened to EAC, is the figure a team expects the final cost of a project to reach. It is compared with the budget at completion, BAC, which is the total the plan allows. When the EAC is above the BAC, the project is forecast to overrun.
An EAC is a forecast, not a fact. It changes every time the team learns something new about the work, the prices or the dates, so the useful question is never only what the number is but what it is based on. See the glossary entry for EAC and BAC.
Teams use an EAC in three ways: to report the expected outcome to management, to decide whether scope or money needs to change, and to check whether the budget still makes sense at all. Each use needs the same discipline: a stated method, a date, and a named person who owns the figure. A number without those three is an opinion.
What are the standard EAC formulas?
Textbook earned value management, often called EVM, builds the EAC from three measures. Actual cost, AC, is what the work has cost so far. Earned value, EV, is the planned value of the work actually done. The cost performance index, CPI, is EV divided by AC, and the schedule performance index, SPI, is EV divided by the planned value, PV. Three formulas are common, and each assumes something different about the work that is still to come.
| Method | Formula | What it assumes |
|---|---|---|
| Remaining work at the budgeted rate | EAC = AC + (BAC − EV) | The overrun so far is a one off, and the rest of the work goes to plan. |
| Remaining work at the current cost rate | EAC = BAC / CPI | The cost rate so far continues to the end. |
| Remaining work at cost and schedule rates | EAC = AC + (BAC − EV) / (CPI × SPI) | Both cost and schedule slippage continue to the end. |
| Reperline spend forecast, per budget item | Forecast spend = invoiced + committed | Only money already invoiced or committed. Work not yet ordered is not in it. |
| Reperline forecast end, per milestone | Forecast end = start + elapsed / progress share | The pace so far continues. Needs progress and a start in the past. |
How do the formulas differ in a worked example?
Take round, hypothetical numbers. The budget at completion is 100,000. The actual cost so far is 40,000, and the earned value of the work done is 30,000, so the work done is worth 30% of the budget but has cost 40%.
The cost performance index is 30,000 divided by 40,000, which is 0.75. At the current cost rate the EAC is 100,000 divided by 0.75, about 133,333. At the budgeted rate it is 40,000 plus 70,000, which is 110,000. The two answers differ by about 23,000. That gap is the point of the exercise: the team has to decide which assumption it believes, and why.
What is the S curve, and why is it background here?
An S curve is the cumulative plan of a project drawn against time. Spend and progress usually start slowly, speed up in the middle and level off at the end, so the line looks like a letter S. The planned value at any date is read from this curve, and comparing the curve with the spend and progress actually reached shows whether the project is early or late and whether money is ahead of or behind the work.
The S curve is a picture of the plan, not a law of nature. Its shape comes from how the budget and the schedule were phased, so a curve is only as good as that phasing. Reperline draws one on its overview: the cumulative plan of the milestones against their actual progress, month by month, with key milestones marked on the plan line. It is a progress curve, not a cost curve, and it is built from milestone weights and planned or actual dates. It shows where the project stands against its plan, but it does not turn that gap into an EAC. The S curve is useful for reading a plan.
Does Reperline compute earned value, CPI or SPI?
No. Reperline does not compute earned value, CPI or SPI, and it does not produce an EAC in the textbook sense. Nothing in the product turns percent complete into a value of work done, so there is no CPI and no SPI, and no formula that uses them.
What it computes instead are two forecasts, and both use only entered facts:
- Spend per budget item: what is invoiced plus what is committed. A commitment counts on the day it is agreed, not when the invoice arrives. An item whose forecast is more than 5% above its plan raises a signal.
- End date per milestone: a linear forecast. The start date plus the time elapsed, divided by the share of work done. A milestone at 40% after twenty days is forecast to need fifty days in total. Without progress there is no forecast.
Which figure should you read first, and what does it leave out?
Read the spend forecast as the money already promised. Take an item with a plan of 100,000, 40,000 invoiced and 75,000 committed. Its forecast spend is 115,000, which is 15% above plan. The item over budget signal fires, even though only 40,000 has been invoiced, because the commitments already tie the money up.
The forecast does not include work nobody has ordered yet, so it can sit below the final cost of the item. It is a warning built on promises, not a final bill. The second check is spend ahead of progress: invoiced money as a share of the plan, divided by the progress of the phase. Above 1.2 the signal fires. If 40% of the plan is invoiced while the phase is 25% done, the ratio is 1.6.
How does Reperline handle the forecast?
Reperline stores the plan, the facts and the commitments, and it applies its rules every day. Each figure shows its source: who entered it, when, and what changed. Where a fact is missing, the forecast is not filled in. A milestone with no progress has no forecast end date, and the system shows it as missing data rather than assuming it is on time.
If you need a textbook EAC built on earned value, with CPI and SPI measured from a work breakdown, Reperline is not that tool. It is built to answer two questions every day: where the project stands, and what it will look like in a month, using the budget and the schedule you already have. The budget module and the schedule module show the figures in detail, and the signals module explains which of them raise a warning. Plans and prices are on the pricing page.
When the two views disagree, the disagreement is the useful part. If the spend forecast of an item looks fine while its phase is behind on progress, the gap is a question for the site manager: either the work is ahead of the money, or the money is ahead of the work. The answer sits in a fact that somebody has to enter, and the system keeps showing the gap until that fact arrives.
For the wider picture of how plan, actual and forecast fit together, read construction cost control: plan, actual and forecast.