Earned Value Management Applied to Small Projects Basic Model

June 17, 2007 | Author: PM Hut | Filed under: Cost Management, Risk Management, Scheduling

Earned Value Management Applied to Small Projects Basic Model (Part 1)
By João Almeida

Commonly EVM has seen a very good acceptance in projects of very big dimension (usually called multi-million dollar projects) and unfortunately it’s also been a general practice to disregard the use of EVM in small projects (up to 4 Full Time Equivalent in a 6 to 12 month period) due to the benefit perceived of using it doesn’t compensate the effort/cost associated. This is particularly true in the IT area.

The aim of this article is to demonstrate that the usage of EVM for small projects is feasible and the benefit provided by EVM largely compensates the effort/cost of applying it when using a practical approach. This paper will describe the usage of EVM in those situations and will finish with a case study.

Project Life-Cycle

The project life-cycle for small project is of great importance, since it enables the analysis of the EVM values at a higher level providing more meaningful information and more stable indicators. Because the project life-cycle phases groups’ tasks of the same nature, this provides a very good basis for considering each phase the best level in the WBS to perform analysis of the indicators and generating cost and schedule predictions.

Progress Report

To track progress I use the 50/50 rule. This is a fast and easy way of tracking progress if the tasks don’t exceed a reporting period (usually one week). The advantage of this method is that team members only have to report hours and task closure (the start of the task is implied by entering the first set of ours) not interfering in the quick pace of small projects.

If the task does exceed several reporting periods than I recommend using an apportioned relationship to other discrete work packages or a combination of percent-complete estimates with milestones used as gates.

Decision Making

For small projects or big ones one of the main issues is that when looking only at the accounting variance (PV-AC) this doesn’t reflect the health of the project and problems are not identified early, forcing the project team to develop workaround plans to handle situations that could have been avoided.

So the main question is how to identify potential problems and when to act.

When analyzing the EVM indexes one might acknowledge degradation on the performance of the project or maybe a trend that could lead to a decrease of the performance of the project. But the main question remains: is it time to act? If so what to do?

The model that I use associates activation limits to the EVM indexes. For example if CPI or SPI:

CPI/SPI Action Table

For better control graphical markers were assigned to the specific index thresholds (bright green, green, yellow and red).

In each review period (usually one week) the indexes at the phase level must be analyzed and eventual problems should be looked at following a couple of rules:

If this deviation is a risk trigger or expected as part of a response to a risk activate the proper response (could be do nothing)

If the deviation is not explained by the identified risks than search for the cause (In small teams this is usually done quite fast due to the reduced number of communication channels therefore lower complexity – analysing CV and SV of the lower level tasks) and identify the risks that could happen due to the identified cause. Next plan for a proper response.

Risk Identification and Response

As much as possible during risk identification the risk thresholds should use the EVM indexes to enable association of actions to specific values of the indexes.

Predicting Future Project Outcome

To predict future project performance Estimated Cost At Completion was calculated in it’s optimistic form, ECAC Likely and EDAC were also used (If the progress is bigger than 15% otherwise the previsions should be equal to the baseline):

Estimated Cost At Completion Optimistic (ECAC Optimistic): Planned Cost/CPI. This assumes that the resources productivity will be constant.
Estimated Cost At Completion Likely (ECAC Likely): AC+(Total Planned Cost - EV)/(CPIxSPI). If SPI between 0 and 1. This assumes that if a task is late than the productivity will be affected and is equal to CPI x SPI for the remaining work (Planned Cost-EV). Otherwise ECAC Likely=ECAC Optimistic.
To predict future project duration Estimated Duration At Completion (EDAC) as used and is equal to Planned Duration/SPI.

When a task is recovering (a task is recovering when it was supposed to have finished but it is still being executed) EV is going to progress towards PV and PV is always the same causing the progression of SPI towards 1 (the baseline is extending until the task finishes). The following figure illustrates the idea:

EV PV SPI Chart

The SPI value is expressed in percentage in the right Y axe.

This example illustrates one of the biggest concerns in EVM since SPI in this cases moves way from the real trend of the schedule and if we want to predict future performance based on the classic SPI the predictions will each time be better since SPI is always approaching 1 (that is perfect schedule performance).

In this case we propose that SPI is calculated using the classical way for all the cases except this one. During the recovery period we want to calculate an “Adjusted SPI” that provides a correct prediction therefore: EDAC=Baseline Duration/Adjustedt SPI

Considering that the SPI for a task that is finished is:SPI=Baseline Duration/Actual Duration

In the case of a recovering task our best estimate for the Actual Duration is EDAC and instead of using Baseline Duration we must use (Actual Date-Baseline Start) so SPI does not maintain the same behaviour of approaching 1. Therefore: SPI=(Actual Date-Baseline Start)/EDAC

Replacing EDAC we get: Adjusted SPI=(Baseline Duration/(Actual Date-Baseline Start))*SPI

The next graph compares the evolution of SPI with adjusted SPI.

SPI vs Adjusted SPI

With the Adjusted SPI we can now calculate with confidence the future project performance and get a clear view of the status of the project. This approach has the advantage of being very easy to calculate using any of the project management software packages or the simplest spreadsheet.

João Almeida is currently working for Microsoft Portugal as an Engagement Manager for the Public sector. His personal blog can be found at http://projmblog.blogspot.com/.

Share this article:

Related Articles

This entry was posted on Sunday, June 17th, 2007 and is filed under Cost Management, Risk Management, Scheduling. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

No comments yet.

feel free to leave a comment

Comment Guidelines: Basic XHTML is allowed (a href, strong, em, code). All line breaks and paragraphs are automatically generated. Off-topic or inappropriate comments will be edited or deleted. Email addresses will never be published. Keep it PG-13 people!

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

All fields marked with " * " are required.

The Stevens Enterprise Project Management Master's Program

Project Management on the Far Side of Complexity

How-to Guide: Change in Project Management

How Much Project Management Is needed in a Federated Organization?

Become a Proactive Project Manager

Project Management - Communicating Successfully

Talk Like a Project Manager

How to Project Manage a Pyramid

The Art of Project Management: Scale

Project Management Foundations - Understanding the Maturity of Your Project Office

Why Are Companies “Good” but Very Rarely “Great” at Project Management?

Earned Value Management Applied to Small Projects Basic Model

feel free to leave a comment