Earned Value Management
Forecasting time

Controlling a project is key to the success or failure of the project. Earned Value Management (EVM) is a well-known technique to control the time and cost performance of a project and to predict the final project duration and cost. It is an easy tool to generate early warning signals to timely detect project problems or to exploit project opportunities. An overview of the EVM metrics is given in "Earned Value Management: An overview" and the formulas are summarized in "Earned Value Management: The EVM formulary".

In "Earned Value Management: Forecasting project outcome", it has been shown that the general forecasting formula for predicting the final duration, known as the Expected At Completion - Time (EAC(t)), is equal to:

EAC(t) = AT + PDWR

with
AT: The actual time moment (i.e. today)
PDWR: Planned Duration of Work Remaining 

The way the PDWR is calculated depends on the performance factor PF, which refers to the assumption made about the expected performance of the future work, as follows:

• PF = 1: Future performance is expected to follow the baseline schedule.
• PF = SPI or SPI(t): Future performance is expected to follow the current time performance.
• PF = SCI or SCI(t): Future performance is expected to follow the current time and cost performance.

with SCI the Schedule Cost Index (SCI = SPI * CPI and SCI(t) = SPI(t) * CPI). 

In this article, more detailed information is given on the time forecasting formulas and techniques. More precisely, three different EVM time forecasting techniques are discussed as follows:

• Planned value method
• Earned duration method
• Earned schedule method

Table 1 gives an overview of the three forecasting methods and the three different versions. Each version differs in the calculation of the PDWR and the performance factor. The nine different EAC(t) formulas are explained in the remainder of this article and illustrated on project data summarized in table 2.

Table 1: Overview of EVM time forecasts (3 versions x 3 performance factors)

Planned value method 

The planned value method calculates the PDWR somewhat differently, resulting in three different EAC(t) calculations which do not follow the general EAC(t) = AT + PDWR formula. 

The first version of the EAC(t) formula is based on two new concepts, known as the planned value rate PV_{rate} and the Time Variance TV, as follows: 

PV_{rate} = BAC / PD
Time Variance TV = SV / PV_{rate}
Version 1: EAC(t)_{PV} = PD - TV 

The second and third version re-calculates the planned duration PD by taking the SPI and SCI = SPI * CPI into account, as follows: 

Version 2: EAC(t)_{PV} = PD / SPI
Version 3: EAC(t)_{PV} = PD / SCI 

Earned duration method 

This forecasting method relies on the Earned Duration metric, which is equal to ED = AT * SPI. The final project duration using the earned duration method is calculated as 

EAC(t)_{ED} = AT + (max(PD, AT) - ED) / PF 

The use of max(PD, AT) can be explained as follows:

• Project progress is still early: if AT < PD then PD is used
• Project progress is already late: if AT > PD then AT is used

The three versions of this prediction technique only differ in their performance factor PF which can be equal to 1, SPI or SPI * CPI. 

Earned schedule method 

This forecasting method relies on the Earned Schedule metric (see "Measuring Time: Earned value or earned schedule?") and calculates the final duration prediction as: 

EAC(t)_{ES} = AT + (PD - ES) / PF 

Consequently, the three versions of this prediction technique only differ in their performance factor PF which can be equal to 1, SPI(t) or SPI(t) * CPI. 

Project example 

In table 2, all forecasting methods have been calculated based on fictitious project data with a planned duration PD = 9 weeks, a project finish with two weeks delay and a total budget BAC = € 150, as discussed in "Earned Value Management: Reliable time performance measurement". The PV_{rate} is equal to € 150 / 9 = € 16.67 per week. The SPI, SPI(t) and CPI values are also given to allow the reader to calculate the EAC(t) predictions.

Table 2: Overview of EVM time forecasts (example)

It should be noted that the earned schedule methods provide duration forecasts which lie closer to the real duration (11 weeks) than the other methods. However, this table only serves as an example table, and therefore, results cannot be generalized. More information on the accuracy of forecasting methods can be found in "Predicting project performance: Evaluating the forecasting accuracy".