Using Earned

Value
A Project Manager's Guide

ALAN WEBB
\I

GOWER

63Alan Webb 2003

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise
without the prior permission of the publisher.
Published by
Gower Publishing Limited
Gower House
Croft Road
Aldershot
Hants GUI1 3HR
England
Gower Publishing Company
Suite 420
101 Cherry Street
Burlington VT 05401-4405 USA
Alan Webb has asserted his rights under the Copyright, Designs and Patents Act 1988 to be identified as
the author of this work.
British Library Cataloguing in Publication Data
Webb, Alan
Using Earned Value: A Project Manager's Guide
1. Project management. I. Title
658.4'04
ISBN 0 566 08533 X
US Library of Cataloguing in Publication Data
Webb, Alan
Using Earned Value : A Project Manager's Guide I Alan
Webb.
p. cm.
1. Cost control. 2. Program budgeting. 3. Project management.
I. Title.
HD 47.33.W43 2003
658.15'52ac21

Typeset in 9 point Stone Serif by IML Typographers, Birkenhead, Merseyside and printed in
Great Britain by MPG Ltd, Bodmin.

Contents
List of Figures
List of Tables
Foreword
Acknowledgement
Preface
1. Earned Value -What and Why

2. Origins and History

3. Terms and Methodology

4. Work Breakdown Structures

5. Plans and Budgets
6. Data Structures and Reporting Relationships

7. Managing With Earned Value Data

8. Software for Earned Value Methods
9. Implementing Earned Value Methods

Appendix 1: Terms Used Within Earned Value Methods

Appendix 2: Earned Value Management Systems Criteria
Appendix 3: Formulae Used for Earned Value Calculations
Index

vii
ix
xi
xiii
XV

List of Figures
Adding the value generated as time passes gives a greater insight into the
project than simply comparing the planned and actual values. The worth in
financial terms associated with the value generated is termed the 'earned value'
The range of application of earned value principles as seen in the USA
Benefits associated with earned value methods as seen in a British survey
An example of an earned value progress report from Prestige PC dating from
1990
The relationship between the plan and the costs over time
Planned and actual expenditure for a project at the mid-point of its life
Adding an assessment of progress gives further meaning to the project
position
The earned value is estimated from the actual achievement based on the
original costs
Earned value assessment for a project that is running ahead of schedule
Cost and schedule variances
Schedule slippage can be established from the earned value and the planned
S-curve
The Cost and Schedule Performance Indices
Calculating the end conditions from the BCWS, BCWP, ACWP, CPI and SPI
The initial view of the project before work commences
The position of the project at the end of Month 3
Predicted position at project completion based on earned value measurements
and formulae
Assessing schedule performance from earned value cost measurements can be
improved if the time dimension is put back into the calculation
Top-level work breakdown structure for a housebuilding project
Expanded work breakdown structure for a housebuilding project
Expanded work breakdown structure with allocation of work
Partial work breakdown structure
Alternative work breakdown structures
Hybrid work breakdown structure showing the different levels of interest
and how they are summarized upwards
Work breakdown structure used on the Concorde project
The project cycle of activities
Generating an initial plan for both activities and costs
Build-up and allocation of contingency to various budgets as used on some
UK projects
Allocation of contingencies and budgets from DODI 7000.2

viii

Using Earned Value

Organization structure for C/SCSC reporting according to the US Department
of Defense in DODI 7000.2
Simple expansion of the Cost Accounts in the RAM can lead to a complex
relationship between the basic data structure and the project plan
Functional work breakdown structure based on a matrix
Basic task logic is derived from the work breakdown structure
Detailed task logic is derived from the basic logic by breaking down the
major tasks into minor tasks or activities
Satisfying both customer and company requirements may need a software
package that can handle multiple WBS codes
Organization and information relationships for earned value performance
measurement
Plotting the Cost and Schedule Variances against time can show a trend, if there
is one
Relationship between BCWS, BCWP, ACWP and the cost and schedule
performance indices
Alternative project positions indicated by the movement of the CPI and SPI
Cost and Schedule Performance Indices plotted against each other on
a quadrant diagram
The behaviour of the performance indices
Slip diagram techniques can be used to forecast the end conditions for both
time and cost by tracking variations in the EAC and the estimated completion
date at each reporting point
A project status report for a work package
The effect of front-end loading can disguise a potential cost overrun during
the early part of the project
The 'rubber baseline' - continuous revisions to the budget timing can also
disguise a potential cost overrun
Options for revising the project cost baseline to reflect the current situation
The basic input screen for Cobra
Four different forecasting methods have been used to estimate the final cost
(EAC) of work package No. 1.0.01 1410 01
Cost Performance Report (CPR) generated by Cobra
Tracking of budget and actual costs over time in Cobra
Earned value report from Primavera Project Planner P3

61
62
66
69
69
71
71
79
80
81
81
83

88
90
93
93
95
102
103
104
105
107

List of Tables
3.1
3.2
3.3
3.4
3.5
4.1

Expenditure plan
Position at the end of Month 3
BCWP calculation at Month 3
Earned value calculations using the additional formulae
Example of the possible error in calculating the Estimated Time to
Completion (ETTC)
Chapter numbers for the main work blocks within the Concorde work
breakdown structure

29
29
30
34
35
48

Foreword
Earned value is coming of age. From its distant roots in work measurement at the dawn of the
industrial revolution, earned value management has gone through many evolutions. Once
the sole domain of the United States Department of Defense, earned value has now been
adapted for use by commercial enterprises and governments around the world. This
development parallels the path of project management, also undergoing its own evolution
on the international stage. Once earned value was seen properly in the light of integrated
project management and as the responsibility of the project manager, great strides were
made in recognizing its core principles, thus making it even more adaptable to a broader
range of projects.
Much of the criticism that has been levied on earned value in the past was due to an
inflexible and dogmatic approach that was applied in the early days; however, recent
developments in international standards and tireless work by its advocates have shown the
world that the key principles are simply good project management techniques. The basic
principles of managing projects with earned value are relatively easy to understand, if
approached with common sense and the knowledge that earned value can and should be
adapted to different environments and business cultures.
Earned value can sometimes be thought of incorrectly as the myriad of variance
calculations and specialized terminology, but the mathematics is only one part of the
equation. As Alan describes so well in this book, earned value enhances good project
management through the development of integrated baseline planning and project control
techniques. His discussion of alternative work breakdown structures will be of particular
value to those commercial entities not bound by stricter government standards.
Alan also notes that perhaps one of the harder parts of using earned value is gaining
acceptance from everyone from within the organization. But for those who have lived
through the process, the insights gained from earned value management will forever change
one's perspective on project management. Once a project manager has used earned value to
manage a project, he will usually always want to manage with earned value.
While this book has been written primarily for project managers within the UK, it is my
sincerest hope that this book will be a valuable resource and reference guide for project
managers around the world, as earned value takes its rightful place within the project
management community.
Eleanor Haupt
President, Project Management Institute's College of PerformanceManagement
January 2003

Acknowledgement
Thanks are due to Eleanor Haupt, President of the Project Management Institute's College of
Performance Measurement, for her invaluable advice and assistance with the preparation of
this text. She took on the laborious job of reading the entire manuscript in detail and
provided considerable insights into the evolution of earned value methods and current US
practice that would have been almost impossible from this side of the Atlantic.

Alan Webb

Preface
I first became involved with earned value performance measurement back in the late 1970s,
when I worked on a project to develop a major airborne weapons system that the United
Kingdom was developing jointly with the United States. If it had not been for the US
involvement, I have no doubt that earned value would never have been used and I would
never have gained that initial experience. As things turned out, the US withdrew their
support and a joint project became a purely national affair. With US funding taken out of
the equation, things became very tight and it was time for a reappraisal; one of the first
things that went was the earned value approach. No one mourned its demise, most were
pleased to see it go, as it clearly cost a lot in terms of clerical effort for the information it
produced.
The years passed and by the mid-1980s I was in the position of Senior Project Manager
with a different company but still involved with airborne weapons development. Faced with
new responsibilities in an increasingly harsh commercial climate, the whole question of the
value being created by the project teams under my control grew in significance. It resulted in
a number of personal initiatives including the introduction of a thorough value engineering
programme and the use of earned value performance measurement. Despite the earlier
experience, the ideas behind earned value management always seemed like the commonsense approach, but the problem was making it work without creating a complex and costly
bureaucracy. At that time there were no books on the subject in the UK; what knowledge I
had was from memory of the earlier application plus a sheet of formulae, and this turned out
to be a big advantage. It freed me from all the conventional wisdom associated with the
prescribed approach; instead it meant thinking the whole thing out in a way that would suit
our method of working. Experience with the previous organization involving other
computerized management systems also taught me that if systems were not designed to be
integrated across all their interfaces at the start then getting them to work together sometime
later would be a nightmare. As much as possible, interfaces had to be reduced to a minimum
and data structures had to be both simple and universal. It also became clear that only a
computerized approach would work with the limited staff available, so a search began for a
suitable package. We were fortunate in that at the start of 1989 the first integrated project
planning packages appeared that could perform earned value calculations. Implementation
was generally trouble-free - the biggest point of contention was use of the new activity
coding system which seemed alien to the older team members. Within two months of
starting we were producing earned value reports on all new projects, although some projects
at the mid-point of their lives were never converted to the new methods.
In many respects the systems that were introduced would not have conformed to the
prescribed view of how earned value methods should be applied: there was no formalized
Responsibility Assignment Matrix and the Work Breakdown Structure was of a completely
new type but it worked and did everything I required. Senior management soon started
taking note of the monthly reports and, taken together with the other initiatives, it raised

xvi

Using E a r n e d Value

awareness of cost and progress issues among all the project staff in a way that I believe no
other approach would have done.
In the period since the early 1990s awareness of earned value methods has certainly
grown and it has been greatly aided by the appearance of more project planning packages
that contain earned value features. Nevertheless, very little has appeared in print about
earned value in the United Kingdom; what has been written has generally appeared in the
form of magazine articles which may be informative but do not say enough to provide a
complete view. The only books that deal with the subject have been written in the USA with
a specifically American market in mind. Unfortunately, they tend to endorse a procedure
that is heavily influenced by a prescriptive, US government-inspired,approach to managing
project contractors. As a result, anyone reading them might tend to believe that this is the
only way that earned value can be used. In fact, nothing can be further from the truth; the
earned value method can be used on projects of most types whether they are handled
through contractors or are a purely internal affair. The method can be implemented as little
more than an extension of the accounting process, or it can form part of a highly structured
and formalized management control system, or it can be used in ways that lie between these
two extremes. The important point is to recognize the project situation for what it is and use
the earned value approach, not as a prescriptive process that has to be followed to the letter,
but as an adaptable method and a valuable source of project data that provides as much
information as you need, but no more.
In writing this book I hope that the reader will gain a greater insight into the whole
process of earned value management whilst remaining free of the specifically American
methodology that some will see as containing aspects that are unnecessary and inappropriate
to their situation. However, I make no apology for the many references to American project
history, practical experience, procedures and terms that appear in the pages that follow. It
must be remembered that earned value management is an entirely US-inspired technique; all
the original writings and specifications came from the US as do the commonly used terms
and definitions. No proper discussion is possible without reference to these original works
and the way they were implemented. They have shaped much of what followed and their
influence still remains with us to this day.
Earned value principles of performance measurement are beginning to gain in
popularity in the UK but it must be said that many organizations would be resistant to
adopting the full US disciplines. That may be no bad thing as the United Kingdom is a nation
that is known for taking the best of what it sees in foreign culture and practice, adapting it to
suit itself and finally adopting it as its own. And so it may be with earned value techniques:
this book may be the first that will be an original British contribution to this important
subject and perhaps more will follow from around the world as the technique becomes
widespread and new ways of using it are devised.

Alan Webb
Horsham, West Sussex, December 2002

CHAPTER

1 Earned Value - What and

There cannot be many who practise in the field of project management who haven't brushed
against earned value performance measurement at some point in their careers. For some it
will be one of those techniques that was studied as part of a project management course but
that was as far as it went, while for others it will be part of their normal project activities. One
suspects that the latter group are still in the minority despite the fact that earned value
methods have been around for about forty years. The obvious question must be why is it that
a technique that has some real advantages for project managers has been largely ignored for
so long, even though many in the profession are fully aware of it? Of course the answers are
never simple but it cannot be denied that the technique gained a tarnished reputation from
its early history and the way that it was first implemented in the United States. The popular
view was that it was complex, bureaucratic, costly, peppered with alien terms and acronyms
and something that one just wouldn't employ unless forced to do so. Not surprisingly, very
little use was made of it outside the USA during its first thirty years but, since 1990, a more
enlightened approach has developed along with suitable software tools that have made the
technique appealing to a much wider audience.
The earned value principle is not difficult to understand - it comes from a basic concept
that goes back to industrial engineering and accounting procedures that were around well
before the discipline of project management arrived on the scene. Prior to the introduction
of earned value methods, project managers were used to measuring the performance of their
projects by reference to Gantt charts and Critical Path Analyses for the scheduling aspect,
and the difference between the planned expenditure and the actual costs to see how the
money was going. From a time dating back to the 1950s, it was realized that this was not a
very satisfactory way of managing projects as there was always the problem of reconciling
these two different measures of project progress. Furthermore, some highly influential
customer organizations were embarrassed by cost overruns that never seemed to be predicted
until it was too late to do anything but swallow hard and pay up. The answer they came up
with was perfectly simple: make both a detailed plan and a detailed valuation of all the work
in the project before you start, then, as the project progresses, make a note at each reporting
point of 1) how much value should have been achieved according the plan, 2) how much
value has been created according to the work done and 3) how much money has actually
been spent. These values are shown in Figure 1.1. Those three numbers form the basis of all
earned value methods; with a few simple mathematical ratios one can quickly judge the state
of progress in terms of both the cost and the schedule. Anyone introduced to this idea for the
first time would probably use terms such as 'obvious' or 'elementary' to describe such a basic
concept, so it seems all the more baffling that it should not have received a more enthusiastic
response. Some of the reasons have already been hinted at but like many good ideas there can
be a wide gulf between theory and practice. Although simple in concept, the practice was

Using Earned Value

much more complex because it was most often applied to projects that were very complex in
themselves.

measurements
Adding the value generated as time passes gives a greater insight into the project than
simply comparing the planned and actual values. The worth in financial terms associated with the value
generated is termed the 'earned value'

Figure 1.1

Earned value methods had something of an unfortunate start that has never helped the
process gain wide acceptance. In those early days, earned value was not seen as a
performance measurement system in its own right to be used and adapted as required;
instead it was introduced as part of a much larger and highly prescriptive government
initiative. In the first incidence, this was the US Department of Defense's PERTICost system
which proved to be a significant failure; that experience alone coloured the next thirty years'
contracting practice. Next it was used more successfully on the Minuteman programme and
finally it was incorporated in a US DoD project management specification that detailed the
Cost/Schedule Control Systems Criteria. This important document made a provision for
earned value measurement and reporting but it was wrapped up in a logical but bureaucratic
management process that was designed to meet the particular needs of the DoD. The fact was
that few people made the distinction between the earned value accounting and measurement principles and the total system in which it was contained. As the principal market for
earned value methods was among the contractors to the DoD and the only documents that
explained the procedure were the government specifications, the perception spread that this
was the 'right', 'official' or 'only' way that earned value could be applied. It is perhaps not
surprising that the rest of the world viewed it as a quirk of US business practice and took little
notice.
It might just have remained that way if it had not been for the PC revolution of the 1990s
when a new breed of project planning software burst onto the market. A break from the
stand-alone systems of earlier generations, the new software allowed integration with other
company systems and also contained earned value calculations. These systems were not

Earned Value - W h a t a n d W h y

based on government requirements but instead recognized that earned value performance
measurement is actually a process that springs from the integration of a planning system
with an accounting system. Freed from the US government-inspired approach to total project
control, project managers using suitable project planning software packages could begin to
use earned value performance measurement as a valuable project control tool but in ways
that suited themselves. It was at this point that earned value finally 'came of age'.

What sort of projects suit earned value methods?

Any project with a structured plan of work, a cost structure and a suitable data-gathering
system can make use of earned value methods but it would be unreasonable to suggest that
the approach is equally applicable to all types of project. In general, projects that are well
suited to earned value methods are those which have most or all of the following
characteristics:
a clearly defined objective
a clearly perceived route to the goal
work taking place over an extended period
a high labour content
tasks of a creative nature
a formalized management structure
cost and time limitations.
An ideal application is engineering development work and it was on projects of that type that
the first use of earned value was made.
Without a clearly defined objective no one can say what route the project might take or
where it might end up; it may at some point focus on an objective, in which case effort will
be concentrated upon it or it may lose sight of any real objective and be terminated. Projects
of this type cannot be controlled in any formal sense; what plans are made are likely to be
short-term or subject to constant variation and no proper performance measurement is
possible even though data may be gathered about expenditure and work.
All performance measurement relies on the idea of a clearly perceived plan that takes the
project from inception to the attainment of the goal. Some projects cannot be planned in
detail because the route contains many uncertainties that can only be resolved at the time
they are confronted. This can occur with projects that contain a large research element where
the outcome of any experiments cannot be known in advance and the outcome may
determine what new lines of enquiry are pursued. It can also apply in an opportunistic
situation where the project strategy might include taking advantage of what opportunities
arise, for example, where sales to third parties, or the involvement of outside parties, is
involved. Without a stable plan in which all can have confidence, earned value performance
measurement is not suitable.
Performance measurement demands a planning, a monitoring and a data-gathering
process. In practice, data gathering and processing take time; in large companies or large
projects, the time-scales can run into weeks or even months before accurate data becomes
available. Projects looking to use earned value methods for control and forecasting should
take note of the actual reporting time-scales; if the projects are of short duration with respect

Using E a r n e d Value

to the reporting cycle, the performance measurement data may be of little use for actual
control although it may have some historical and statistical value. With short-duration
projects, more direct control techniques may be more suitable.
Earned value was conceived as a method of valuing work done. Although purchased
items or materials are not excluded from the earned value approach, this aspect was not the
principal focus. If a project is to consist primarily of purchases with very little labour (for
example, a project that consists of the purchase of a suite of standard equipment amounting
to 90 per cent of the project value, with the only labour element being the installation and
commissioning) then earned value methods would have little real impact as a control
mechanism. In a case such as this, control would be much better exercised through the
pricing agreement with the supplier, which might include discounts, incentives or liquidated
damages for late delivery.
The first applications of earned value methods were on major defence engineering
development projects. As such, these projects contain a high degree of innovation involving
problems which demand a creative solution. These kind of projects are among the most risky
and are prone to schedule and cost overruns. Earned value methods were devised to deal with
such problems, in particular, to generate some objective measures of progress in a somewhat
unstable project situation. Whereas there is no absolute reason why earned value methods
cannot be used on routine projects with little creative demand, the question becomes: is it
really worth doing? In situations that are largely repeats of earlier well-understood projects,
in which there is little risk at the start, or are mostly repetitive production-type operations,
then earned value methods are not likely to be of great benefit. Control might be better
exercised through simple progress recording and fixed price contracting arrangements which
should be easy to agree if things are well known at the start.
Without a formalized management structure, there is not much point in attempting
performance measurement unless one is only interested in statistics. Earned value management implies not only a well-defined plan against which performance can be measured, but
that someone is going to take responsibility for implementing the plan, take note of the
performance measurement results and carry out whatever actions are indicated. If no proper
management structure exists or the relationships are only vaguely defined, it may prove
impossible to obtain objective information about progress, as well as getting anyone to be
responsible for achievement. It may be possible to implement an earned value system of sorts
if the planning and data-gatheringaspects exist, but the results will have little practical value
in terms of the influence they exert over progress.
Another name for earned value management is 'integrated cost and schedule control',
because it brings together a way of measuring achievement against both time and cost goals.
If a project has neither of these limitations within the control horizon or the cost limit
cannot be tied to any particular work or outcome, then earned value methods are not
appropriate. Examples of projects of this type are found in long-term research and
development, for example, finding a cure for certain types of cancer: it could take five years
or it might take ten -and nobody knows.
The case for implementing earned value methods is no different from the case for
implementing any other management technique: it is quite simply an economic one. There
is always a cost associated with obtaining any information on a project and the issue is one of
deciding whether the costs are going to bring the required benefits. In fact, it is not possible
to answer that question as the decision must be made before the project starts and no one can
say what extra costs might be incurred on the project or what overruns might occur if earned

E a r n e d Value - W h a t a n d W h y

value is not employed. So it becomes a question of what type of project is most suitable, and
the answer is given above: if a project does not fit a substantial part of this profile then other
methods might be more suitable, and some suggestions about what to use have been made.

How sophisticated do we have to be?

Earned value methods were first embodied in some highly detailed US government
specifications that appeared in the 1960s. Since then, earned value has been applied across a
range of industries and in different ways; Figure 1.2 shows a contemporary US view. Some of
the terms shown in the diagram may be unfamiliar and more explanation is given in later
chapters but the overall view is clear. Figure 1.2 makes an important point: far from being a
prescriptive process that can only be done in one way, earned value management should be
adapted to needs as determined by the scale of the project and the organization undertaking
it. At the left-hand end, we have small organizations and comparatively simple projects, in
this case the core earned value accounting principles and simple reports will be sufficient.At
the extreme right-hand end, we have very large projects that are the province of the US
government departments such as the Department of Defense and here compliance will be
required with a national standard for earned value management, incorporating 32 criteria
that define the process. Whereas this situation is not directly applicable to the United
Kingdom, as no UK government department demands compliance with a US national
standard, the principle remains the same. Simpler projects can use a simplified approach,
while more costly and complex projects need a more rigorous procedure.
Commercial or Defence

Where

small
companies

US Government
Organic

Larger companies

Major contractors
to the US DoD

DoD non-major

DoD major

When

As required

Corporate policy

Contracts
(>I2 months)
>$6.3m

contracts
> $73, TDT&E
>$315m Prod

Reports

Streamlined,
simple

Tailored to needs

Simplified
C/SSR

Detailed
CPR

Method

Core EV principles Tailored applications

Full compliance with

ANSllEIA-748-1999
All $values Fiscal Year 2000

Figure 1.2 The range of application of earned value principles as seen i n the USA. Note that only the
largest defence contract would be viewed as requiringthe mll rigours of ANSIIEIA-748-1998 criteria
compliance (Source: Eleanor Haupt)

What can we expect from earned value?

When earned value was first introduced there is no doubt that the sponsors were looking for
a much better insight into the progress of their projects from both a cost and a schedule

Using E a r n e d V a l u e

standpoint than they had before. In particular they did not want any nasty surprises from
contractors making sudden demands for more money and increased time without any clear
warning that the project situation was deteriorating. That situation has not altered; it is as
important today as it was forty years ago to have a clear view of how well a project is doing
and where it is heading. Its importance appears in a number of ways:
Early warning of a deteriorating situation creates an opportunity to do something about it
before it is too late.
Accurate forecasting allows better decisions to be made about the course of the project.
Accurate forecasting allows better decisions to be made about matters outside the project
which may be influenced by the progress of the project.
An open and verifiable view of progress improves sponsor confidence.
These are all good reasons why earned value performance measurement is an important
project management technique but there can be additional benefits. Earned value methods
demand effective planning, costing and monitoring systems; the emphasis placed on these
aspects can improve overall project management through the discipline they bring.
Furthermore, management using earned value techniques requires a proper system of
controls with the appropriate allocation of responsibility for achievement.
In 1994, a survey was carried out in the United Kingdom among users of earned
value methods; one question was what benefits have users experienced through the
implementation of this approach? The response is shown in Figure 1.3. The most striking
feature is the broad range of benefits that are claimed and the high incidence among the
respondents for some of the most important features. Not surprisingly, all claimed to have
seen much better integration of costs and plans but over 80 per cent also saw improved cost
forecasting and earlier sight of problems, both of which are important benefits in their own
right. Two-thirds noted better financial control and, most importantly, better overall
management awareness of the project situation. As to actually making a reduction in the
overall project cost through the use of earned value, only one-third claimed to have seen any
evidence of this but this is a difficult issue as it is not possible to say what costs might have
arisen if earned value methods had not been used.
Better integration of costs & plans
Better cost forecasts
Earlier sights of problems
Better end-date forecasts
Improved financial control
Better management awareness
Better staffing forecasts
Better communications
Actual cost reductions
Subcontractor improvements

20

40

60

80

Percentage response
Figure 1.3

Benefits associated with earned value methods as seen in a British survey

100

E a r n e d V a l u e - W h a t a n d Why

Earned value methods come at a price; there is no doubt that introducing them in a project
environment can be a significant complication compared to running a project without them.
They can call for changes to operating practices that go well beyond producing a few new
reports as they can demand much greater discipline within the planning'process, much
greater emphasis on objective reporting, improved integration between the planning
systems and the accounting systems, and they might even require new software to be
installed. Inevitably there can be problems when a new approach is introduced, particularly
when it can demand changes such as those mentioned, but the benefits are well worth
having. It is a significant fact that of the companies surveyed, none was contemplating
abandoning the practice; once the initial problems had been overcome, the experience
encouraged some of the organizations to make more widespread use of the technique.

CHAPTER

2 Origins and History

The fundamental ideas that have come to form the earned value approach to project control
stem from two distinct lines of thought: those that originated with industrial engineers and
those that come from project managers. Industrial engineers wishing to measure the
performance of production lines realized, perhaps as far back as the nineteenth century, that
three measures were needed to establish how well and cost-effectively the manufacturing
process was working. They created the concept of a 'cost standard'; that is, the amount of cost
or value t o be attributed to a single unit of output. This is a very important measure as it is
usually based on a detailed study of the labour content, charge rates, overhead structure,
material content and their associated costs. Such figures are vital when it comes to both
product pricing and the expected profitability of the production process. To assess the
efficiency of the production line in operation, they counted the number of items actually
made, compared it with the planned or expected number of items to be made and finally,
from the material accounts, the time bookings and the current overhead rates, they worked
out the actual costs incurred. Applying the cost standard to the actual and planned output
figures generated standard costs that could be compared with the actual costs. Three figures
were generated:
the planned output at standard cost rates
the actual output at standard cost rates
the actual cost incurred.
Comparing these figures led to the ideas of:
process efficiency- actual output at standard cost compared to planned output at standard
cost
cost efficiency-actual output at standard cost compared to actual incurred cost.
If either of these measures showed a poor position then the profitability of the process was in
doubt and something needed to be done. The actual output at standard cost is what has come
to be known as the 'earned value', that is, what has actually been produced at the cost you
expected to pay.

A government initiative
Until the advent of Critical Path Methods (CPMs) in the late 1950s, projects were viewed in a
very different light from production processes; they were unique undertakings for which no
standards could be established, and certainly not from long-term observation and
measurement. Critical path methods (CPM) and PERT (Program Evaluation and Review

Origins and History

technique) methods were the first systematic attempt to plan projects in detail and control
progress through rational analysis and decision making. Planning charts of the Gantt type
had been used for at least fifty years before that time but the analytical approach was lacking.
CPM and PERT relied on high-speed computers which were just making their entrance into
business operations; with hindsight perhaps too much faith was placed in just what these
methods were really capable of doing with the technology then available. It was realized at
the time the PERT system was invented that if cost rates could be added to the hours and
durations that are attached to project activities, a powerful system would exist for predicting
and, ultimately, controlling the costs of projects. The US Department of Defense (DoD) and
the National Aeronautics and Space Administration (NASA), as the largest instigators of
projects involving the highest technical risks, saw the immediate application of the idea; in
1962 they sought to impose the system on their major contractors under the title PERTICost.
This approach contained the 'earned value' concept as now it was possible to draw an
analogy with the industrial engineer's approach. One could see the plan in terms of its
anticipated costs and the accounting system should provide the actual expenditure, all that
was needed was a way of valuing the work done and, in theory, this should be a relatively
simple process: 'A comparison of the actual costs accumulated to date and the contract
estimate for the work performed to date will show whether the work is being performed at a
cost which is greater or less than planned.jl
However, the DoD's application was very heavy-handed, as it specified the total costing
system that was to be used throughout the project. The result was a complete rejection by an
industry that had already made large investments in its accounting systems and was not
about to change to a new and untried method, especially as the required software had barely
been developed. By 1965 PERTICost was dead and so too was PERT as industry found them
too demanding and by and large preferred the simpler CPM which did not require the threeestimate method inherent in PERT.
All this happened at a time when some very high-profile US aircraft development
projects, notably the C-5A Galaxy freighter and the F-111 bomber, were experiencing major
cost and schedule overruns that seemed to come out of the blue. There was a suspicion that
the true picture had been deliberately obscured until it was too late to do anything about it,
which was a major embarrassment to the DoD civil servants in charge of procuring these
systems. The DoD was clearly open to criticism if it had to keep going back to Congress to ask
for more money on projects that should have been reasonably straightforward if properly
conceived at the start. Concurrent with these two programmes was the Minuteman antimissile system development project and for this a special reporting system had been set up.
This was called the Minuteman Contractor Performance Measurement system; Devised by
Brigadier General Samuel Phillips, it embodied the earned value principle of performance
assessment against a fixed 'baseline' plan but without any of the complex systems and
planning requirements of PERTICost, and it appeared to be more successful. In particular, it
did not require the detailed activity costings that were a feature of the network-based PERT
approach but relied on higher-level work packages that were derived from another important
data set: the Work Breakdown Structure. Realizing a serious mistake had been made in the
handling of the introduction of the new management processes, the DoD and NASA decided
that a less rigid approach would be more appropriate; in particular it would allow industry to
choose its own management methods and software tools providing it met a series of criteria
that were deemed to be good management practice. These were finally embodied in a US
Department of Defense Instruction DODI 7000.2 of 22 December 1967 called 'Performance

10

Using Earned Value

Measurement for Selected Acquisitions'. The title gives the clue to what this instruction was
really intended to do, in particular it was about performance measurement (that is, visibility of
progress in plain terms), selected, implying that this was not to be a blanket procedure applied
to all DoD programmes, and acquisitions as this was to be used on acquisition programmes
(that is, projects that were to create new products to be acquired by the DoD). Projects such as
long-term R&D were excluded. The title does not suggest that project performance would
actually be improved by the use of this instruction although that clearly must have been a
hope. One suspects that what was uppermost in the minds of the DoD managers was:
visibility of what they were getting for what they were paying
measurement rather than guesswork
no more embarrassment due to contractors concealing the real position of the project.
DODI 7000.2 set out what were to be known as the Cost/Schedule Control Systems Criteria
(C/SCSC),2 popularly called the 'C-Spec'. In many ways it was a seminal work in the
evolution of project management as it set a precedent for much of what has followed, so
much so that its methodology has acquired a stature that is above its real place among some
practitioners who have perhaps forgotten or never knew its original purpose. Earned value
reporting was a part of this standard but it included another important feature: data
structuring. The structure it set out is not, however, essential for earned value measurement
but it has become inextricably linked to it in the eyes of some advocates, and possibly the
project management profession as a whole, which has in turn obscured the real position.
DODI 7000.2 was not a lengthy document, the core of it consisted of 35 simply stated
criteria such as:
The contractor's management control systems will include policies, procedures and
methods which are designed to ensure that they will accomplish the following:
a) Organization
1) Define all authorized work and related resources to meet the requirements of the
contract using the framework of the CWBS (Contract Work Breakdown Structure)
2) Etc.
b) Planning and Budgeting
1) Schedule all authorized work in a manner which describes the sequence of work and
identifies the significant task interdependencies required to meet the development,
production and delivery requirements of the contract
2) Identify physical products, milestones, technical performance goals or other indicators
that will be used to measure output
3) Etc.
c) Accounting
1) Record direct costs on an applied or other acceptable basis in a formal system that is
controlled by the general books of account.
2) Summarize direct costs from the cost accounts into the WBS without allocation of a
single cost account to two or more WBS elements.
3) Etc.

O r i g i n s a n d History

11

d) Analysis
1) Identify at the cost account level on a monthly basis using data from, or reconcilable
with, the accounting system:
a) Budgeted cost for work scheduled and budgeted cost for work performed
b) Budgeted cost for work performed and applied (actual where appropriate) direct
cost for the same work
c) E ~ c . ~
The above extracts give an idea of the standard; although much of it can be considered as a
sensible approach to managing a project, it was both prescriptive and demanded a great deal
of bureaucracy. In addition to the basic criteria there was a requirement for highly detailed
reporting in precisely defined formats (DODI 7000.10: C/SCSC Reports). As if that was not
enough, the US DoD developed hundreds of detailed questions about how a company
functioned and sent audit teams to ensure contractor compliance. The teams sometimes
adopted an over-zealous attitude that did not lend itself to the general acceptance of the
methods on the part of industry, particularly as many of the questions were considered to be
commercially sensitive, marginal to the central issue of managerial control and really none
of the DoD's business. The scene was thus set for an uneasy relationship between the
government and its contractors. In practice, the implementation of DODI 7000.2 was slow
and by 1972 only 36 contractors had been certified as compliant; it was generally regarded by
industry as both DoD interference and costly bureaucracy.

System problems
During the 1960s, the majority of project cost on DoD-type projects was attributable to
labour and, using time-booking systems that were common in all companies, planned,
earned and actual labour cost could be relatively simply calculated providing the time
records could be related directly to the plan. The same was not true of the material aspects of
the project. Whereas practically all companies with a procurement system can produce two
basic figures - the budgeted cost for material purchases and the actual cost of materials
purchased - few can produce an 'earned' value of materials at any point in the process. This is
because the value of materials can vary with time due to a host of factors such as the effects of
inflation, the effects of different batch sizes, price changes in the market, etc., but the DODI
instruction demanded that all these variations be accounted for on a continuous basis
through a series of tracking points. Things got even more difficult when materials were
bought in large quantities which could be used on a variety of development projects as well
as on routine production work. Needless to say, material accounting under earned value
methods was a painful process: 'firms had to create new or substantially modified material
accounting systems to comply with the criteria'. That might have been all to the good but
experience has proved 'there is very little relationship between what is required to provide for
an efficient contractor procurement system and the satisfaction of those criteria which
impact material^'.^ In other words, operating an earned value type of material accounting
system does little or nothing for creating an efficient purchasing arrangement, it can simply
lead to heavy and expensive systems. However, one must remember that the DoD had been
stung by embarrassing, unforeseen overruns; in the Cold War climate of the 1960s)with its
heavy emphasis on defence, it was willing to pay the high costs of the bureaucracy.

12

Using E a r n e d V a l u e

Despite the difficulties, contracting companies began using the approach and this led to
two important developments. First, software companies started to produce package systems
that would support the CISCSC requirements, do the earned value calculations and generate
the highly detailed reports. Second, a body of knowledge began to accumulate about the
performance of projects that were run under earned value conditions. Unsurprisingly,
commercial industry that may have become aware of the C/SCSC approach had no
inclination at all to follow the lead; some progressive firms adopted earned value accounting
principles for reasons of their own but without the inherent bureaucracy contained in the
criteria. In 1975 the US Department of Energy adopted a broadly similar approach with a
CISCSC contracting arrangement for its major projects but beyond that no other
government departments in the USA or elsewhere took much interest in earned value until
the 1990s. It is significant that the UK Ministry of Defence has been well aware of the
methods but has never imposed a CISCSC contractual arrangement on its contractors,
although it now advocates the use of earned value management principles where
appropriate.

The consultant's view

Inevitably, a complex system like CJSCSC had its critics: senior people in both industry and
government argued that if the DoD really wanted to save money on its projects it should
abandon the C/SCSC process. The DoD took this criticism seriously; 15 years after the
introduction of DODI 7000.2 it commissioned consultants Arthur D. Little to conduct an
independent enquiry into its usefulness and whether it was worth the additional costs it was
clearly having to pay. Arthur D. Little reported in 1984, having taken evidence from both the
government side and industry. The report concluded that both sides agreed that using
CISCSC was effective as a project control mechanism and the benefits outweighed the costs.
However, there were sharply differing views from industry and government about the details
of how it was implemented. The main, and well-voiced, criticisms from industry were that 1)
the prescribed approach required far too much detailed reporting and variance analysis,
much of which was of too low a level to be of use for overall control, and 2) the requirement
to maintain the baseline position, when work-around plans were what was really needed,
was too restrictive for effective control. The government criticisms were much less cohesive
and to some extent at odds with each other, indicating (surprisingly, as it was a governmentinspired scheme) that they had lost sight of the original objectives while failing to appreciate
what was practical and achievable. The study reported: 'In summary, effective internal
management does not of itself require detailed reporting to the government. However,
reporting to the government should be expected as a reality of doing business with the
government, particularly when the work entails cost risk to the government. The
government need for information is not always taken into account by contractor^.'^ This
statement reveals the whole position and purpose of the CISCSC approach; if you simply
substitute the word 'sponsor' for 'government' you can apply this more generally. In short,
the CISCSC process is in place primarily for the benefit of a sponsor who is paying a
contractor, or group of contractors, to undertake a project on his behalf, who is also bearing
both the cost and schedule risks - and expects to be kept well informed. It also makes a tacit
admission that if it were not for the requirements of the government, industry would run its
projects more cheaply and simply. Other complaints covered the unnatural and complex

Origins and History

13

approach required to handling material costs, the competence and narrow-mindedness of

personnel on the audit teams and the unrealistically short time-scales that were allowed for
the development of a realistic baseline plan. As both sides felt there were sufficient
advantages to retain the general approach, Arthur D. Little's recommendations contained
nothing fundamental and were largely confined to administrative matters covering
organization, training and some points of operation.

A new direction for the 21s t century

From a US government perspective, the C/SCSC method must be considered a success, even if
the old delays and overruns persist. The imposition of the criteria improved contractor
awareness of good project management practice in a way that might not have come about by
pure competitive pressure. However, there has been a comparative failure, until recently, to
get earned value methods into the commercial world. This situation has been greatly
influenced by 1) its somewhat tarnished reputation as costly and bureaucratic, which
stemmed from the early days and the DoD's approach, and 2) doubts about its applicability
in a commercial environment that is increasingly looking to fixed price contracting and the
imposition of liquidated damages as a way of containing cost and schedule overruns.
By the early 1990s the DoD, which had been so instrumental in formulating and
promoting the earned value approach with its strict compliance criteria, began to realize that
its acquisition processes had placed too much emphasis on adherence to fixed plans and
reporting in detail against them rather than emphasizing doing things in the most efficient
way. Although it was getting plenty of data about earned value, what was not so clear was that
what it was buying represented good value for the money it was paying. Even more pointedly,
many items purchased for military use were found on examination to be little different, if not
identical, to items that were available to commercial industry at a fraction of the cost. In
1996, the DoD issued new directives in the DOD 5000 series that completely changed the
emphasis from one of detailed control at every level to a total system acquisition process that
made compliance with user needs, effective integration of complex systems and best value
products, the principal programme drivers. Gone was the old adversarial approach between
government and industry, to be replaced with the formation of 'Integrated Product Teams'
and the beginning of a new partnership.
A revolution in the approach to earned value came in 1993 when Gary Christle from the
Office of the Under Secretary of Defense set out a plan for reform in five key areas:
Change the emphasis from the government to the contractor. Compliance with CISCSC
should be based upon how the contractor actually manages rather than an imposed view.
Rather than automatic reviews and audits imposed by contract awards, C/SCSC reviews
would only be required if there was poor discipline or faulty certification.
Reduce the amount of surveillance required by both the government and contractor.
Put earned value in its proper context as an integrating tool for cost, schedule, and
technical management; this changed the emphasis from reporting on earned value to
managing with earned value.
Limit and tailor reporting to what can and will be effectively used.
Ensure comprehensive planning and common understanding by both parties through the
implementation of a baseline review after contract award.

14

Using Earned V a l u e

Several other significant changes occurred. Ownership of the policy for earned value was
transferred from the comptroller's office to the acquisition policy office, thus signalling that
the project managers, not the accountants, owned the process and the policy. A policy
directive was soon issued that changed the focus of earned value reviews from auditing the
processes (the old CISCSC compliance reviews) to integrated baseline reviews that were
conducted jointly by the government and contractor. The DoD underscored its
determination to ensure industry ownership of the process by asking five industrial
associations to propose industry-standard guidance to replace the 35 original criteria.
In 1996, industry responded with a standard that reduced the original 35 criteria to 32
(see Appendix 2). Overall, it was broadly similar; with the exception of three criteria which
were deleted, there is general equivalence on the remaining 32. However, certain important
changes were incorporated which recognized that the needs of commercial industry were
somewhat different to those of government. In particular, there were two very significant
changes: 1) the Work Breakdown Structure was no longer required to be a product-based one
expressed in terms of contract deliverables, but a structure that was compatible with the
management methods in use, and 2) the material cost reporting system was to be one that
accorded with normal industry practice. Overall, the emphasis across the criteria was
changed from strict adherence to procedure to one of using the most suitable and
managerially efficient methods in the circumstances. In December 1996, the 32 industry
criteria formally replaced the old 35 with the DoD 5000.2R regulations and industry was left
to find processes that were suitable to managing the job effectively. The 32 criteria were later
incorporated in an Earned Value Management commercial standard ANSIIEIA-748-1998
which was approved in May 1998 and this was adopted in 1999 as the official DoD approach,
completing the demise of the old ways. The official announcement stated:
The guidelines in this document are purposely high level and goal oriented as they are
intended to state the qualities and operational considerations of an integrated management system using earned value analysis methods without mandating detail system
characteristics. Different companies must have the flexibility to establish and apply a
management system that suits their management style and business environment. The
system must, first and foremost, meet company needs and good business practice^.^
This swept away the supporting checklist of questions which had become detailed criteria in
their own right and left the original authors' intent standing in the 32 criteria. Gone were the
days of CISCSC compliances reviews and audits with excessive demands for compliance with
the criteria. By stating that different companies need to define for themselves management
systems that suit their management style and business environments, the DoD had recognized
after thirty years a fundamental truth about project management as a process. Rigid
procedures, however logical and well intended, are no guarantee of efficiency, in fact they can
be quite the opposite. To be efficient, firms must know how to choose from the available range
of systems and techniques those methods that will be effective in the particular circumstances
of the project and apply them in a way that makes a real contribution to project success. This
means a flexible or adaptive approach that always keeps the ultimate goal of the project firmly
in mind and tailors everything to that end. Of course, many contractors who had developed
systems that were both CISCSC compliant and managerially useful continued to use them in
the new environment; many have used earned value methods as a primary component of their
business management methodology and find it effective.

Origins a n d History

15

The C/SCSC legacy

Earned value, as a method of calculation of project performance, is an accounting principle;
it actually relies on comparatively little project information, but earned value was only one
aspect of the criteria that were set out in DODI 7000.2 as it specified a total approach to
project management. Important features of DODI 7000.2 such as the Work Breakdown
Structure and the Cost Accounts are not essential to performing earned value calculations (if
one is only interested in accounting data), but a good planning system and a compatible
cost-reporting system are essential and they must be in place and working. This point was not
actually overlooked but it was de-emphasized due to the DoD's failure when it tried to
impose the use of the half-developed PERTICost system, an approach which made those
requirements very clear. There was so much trepidation over this point that the planning
system, which is really fundamental to effective earned value methods, was left out of the
principal data structure; there was no requirement to use a PERT-like networking system and
some contractors chose not use networks. Instead, it placed emphasis on 1) data structuring
that was based on product and work visibility, 2) accountability for achievement, and 3)
highly detailed reporting. Inevitably, this approach has found its way into the special
software packages that have been developed. For the most part, these were created to serve
the contractors to the DoD and DOEas these were the only markets and thus followed the
very prescriptive format. The legacy of DODI 7000.2 is still with us in the form of the
specialist earned value software tools and the 'conventional wisdom' of how earned value
should be applied.

UK experience

'

In the United Kingdom, very little use was made of earned value methods during the 1970s
and 1980s due to both ignorance of the method and lack of suitable software. Some firms did
adopt the principles using systems they created themselves; unless a firm happened to be a
contractor to the US DoD there was simply no reason to adopt the US approach. It has
already been pointed out that the earned value principle is more dependent on a planning
system and a cost-reporting system than it is on work breakdown structures (WBS) even
though that feature first got earned value off the ground. Throughout the 1980s, there was an
explosion in the development of project management planning packages due, in part, to the
development of personal computers. These new packages were much more versatile than the
old cumbersome systems and allowed the integration of planning data with other systems
such as time bookings and accounts. Once this had been achieved, elementary earned value
calculations became possible and by the early to mid-1990s many of the popular planning
packages offered earned value information as part of the report suite, although the report
formats were not compliant with the DoD requirements. Figure 2.1 gives an example of a
report from 1990 showing basic earned value data in a project status report. In some
packages, however, the use of the basic terms was not quite the same as in the original
definitions and derived formulae; this has unfortunately led to some confusion and possible
mistrust of the information,derived. Some early systems, it must be said, also contained basic
errors in the calculations.
By the mid-1990s, there were still very few companies in the UK using earned value
methods but a small number were surveyed by a group of students working for this author.'

Using Earned Value

16

PRESTIGE PC

:
1
: 12Apr90
TIME NOW
: 31MaflO
FORECAST END : Omec90
REQUIRED END : 31Aug90

EARNED VALUE -VARIANCES & INDICES

PAGE NUMBER
-..-.------------------------.
RUN DATE
REPORT ACO3

PROJECT: PROJECT DEFINITION STUDY

-----.--.-

VERSION 00
DRAWING OFFICE: PHASE 1

-----..-----------------..-...---------.----..--..----...--...---.----------------------------ACTIVITY

ACTUAL

EARNED

BUDGET

SCHEDULE

COST

SCHEDULE

COST

IDENT.
DESCRIPTION
TO DATE
VALUE
SCHEDULED
VARIANCE VARIANCE PERF. INDEX PERF. INDEX
--------------------------------------.----..--..---12310
D. 0. MANAGEMENT
12320
D. 0. -SCHEME CONTROL PANEL
12321
0. 0. -SCHEME LOOM LAYOUT
12322

.OO

.OO

1490.40

-1490.40

.OO

.OO

247.77

275.30

275.30

.OO

27.53

1.OO

1.11

1624.27

1569.21

1569.21

.OO

-55.06

1.OO

.97

853.43

853.43

853.43

.OO

.OO

1.00

1.OO

00.0

00.0

00.0

.OO

.OO

D. 0. -SCHEME NEW CASING

12327
D. 0. SCHEME FINAL DESIGN

"*TOTALS***
4157.03
4074.44
8070.07
-3995.63
-82.59
.50
------------------..----..----------.--.--.--.-------------------------------------------------

.98

An example of an earned value progress report from Prestige PC dating from 1990. Note the
cost and schedule performance indices

Figure 2.1

The survey found that the reasons for introducing earned value methods were rather various
and some commented that they had found some difficulty in getting things to work. All
indicated that they had achieved much better integration between costing and planning
than before but few felt that they had achieved any overall cost reductions on their projects
due to earned value being implemented. However, every respondent felt that once the
system was working there were benefits and none was contemplating abandoning the
approach.
Around the world, some government departments started to realize the worth of the
earned value method and began mandating its use for their contracts. One notable example
was the Australian defence department which is now demanding earned value reporting,
causing some British suppliers of defence equipment to formally implement the process. The
advent of the Internet has also had its effect and there is now a major website devoted to
earned value methods which contains much useful information: <www.acq.osd.mil/pm/>.It
also shows that a sizeable community of earned value practitioners has grown up, although
the correspondence on the notice board tends to indicate that it is mostly contained within
the USA. The correspondence also tends to indicate that there is 1) still some confusion
surrounding the terms and calculations, 2) continuing doubt about the applicability of
earned value to specific situations, particularly fixed price contracting, 3) difficulty with the
minutiae of dealing with the reporting requirements in a full earned value management
(EVh4)-compliantenvironment and 4) difficulty with the use of certain software packages.

Origins a n d History

17

The problems outlined above are not surprising, given both the diverse nature of projects
and the inevitable complexity associated with a system that ties together cost and schedule
aspects of a project that might previously have been treated separately. However, progress is
being made: the old prescriptive methods have been replaced with a more flexible approach
that will surely have a much broader appeal to project managers in a wide range of industries,
who might never have considered it before. As more projects embrace the methods and more
knowledge accumulates, new ways will be found to use earned value methods that will,
hopefully, mean that project managers will be better equipped than ever to ensure the
successful outcome of their projects.

Notes
1. Office of the Secretary of Defense and National Aeronautics and Space Administration Uune 1962)
DoD and NASA Guide PERT Cost Systems Design, p. 17.
2. DODI 7000.2 (1967) Performance Measurement for Selected Applications, Department of Defense,
Washington DC, 22 December.
3. Ibid., Section 3: Criteria.
4. Ibid.
5. Arthur D. Little Company (1984) Survey Relating to the Implementation of Costbchedule Cont~olSystems
Criteria within the Department of Defense and Industry - Phase 11, Department of Defense, Washington
DC, 15 August.
6. ANSI website: <www.acq.osd.milz.
7. Webb, A. J. (1995) 'Integrated Cost and Schedule Control, a survey of UK experience', Enginemr'ng
ManagementJournal,Vol. 5, No. 3, June.

CHAPTER

3 Terms and Methodology

All projects consist of a set of activities that lead to the achievement of the project goal. How
those activities are performed, who does the work and when they are done should be defined
in the project plan. However, there is no absolute requirement for certain types of projects to
have a plan -this may be because there is no particular need to complete the tasks in any set
order or by any given date, or because later events will be dictated entirely by the results of
earlier ones whose outcomes cannot be known with certainty in advance. Where no plan
exists, earned value methods are not appropriate and they will not work.
If the project can be seen clearly in terms of 1) the work to be done, 2) the value
associated with the work and 3) the order and duration of events, it is possible to generate a
time-phased plan and a time-phased statement of the value to be created or the costs to be
incurred. At the planning stage, the value to be created and the costs to be incurred can be
treated as the same thing; only by spending resources, in terms of cost, is anything of value
created. Project plans are typically expressed in the form of a network or a bar (Gantt) chart.
Most projects have a start-up phase where a small number of resources are needed, a period of
significant activity where the maximum resources are used and finally a winding-down phase
as resources are shed and the project concludes. When the cumulative costs associated with
these activities are plotted on the basis of time an S-shaped curve will result. The steepness of
the slope of the curve represents the level of expenditure; as this is greatest when the project
is at peak of activity, the slope of the curve is steepest on or about its centre. Taken together,
the planned activities and the costs appear as shown in Figure 3.1

The concept of the earned value

In the period prior to earned value methods, two simple measures of cost on projects were
taken: the planned costs and the actual costs. Consider a project with a four-year lifespan
with costs that follow an S-curve as shown in Figure 3.1 and budgeted to cost 80m. Now
suppose at the end of the second year the cumulative cost is planned to be E40m, providing
all the work is done according to plan and at the expected costs. Let us further suppose that at
the end of the second year we know from the project accounts that 30m has actually been
spent. The picture of this situation would appear as in Figure 3.2.
The question we might now ask ourselves is how well is this project actually doing at the
half-way point? On the face of it we might say it is not doing very well, as actual spend is well
behind the plan - it could be that the project is behind schedule and we might have cause to
worry. On further reflection we might decide that it could be doing very nicely and be
completed for a cost that is well below the budget, in which case we might have cause to
cheer. Alternatively, we might conclude that it could be either or both these possible
scenarios, in which case we really don't know where we are. In fact, it is impossible to tell
what the position of the project is with respect to either cost or schedule from these two

--

Terms and Methodology

,,
I

I\

II
I

Start-up

Peak activity

LIA

,I,

1-

Wind-down

rl-

I ,

19

I
I

1
;
=
I

Project plan

II

The project plan consists

of a series of time-phased activities
Each activity has a cost

I
I
I

I
I
I

I
I
I

I
F
I

Cumulative Cost

Project cost

Cumulative project costs

typically form an 'S-curve'

Time
The project plan and the project cost are related in terms of time

Figure 3.1

The relationship between the plan and the costs over time

figures alone. To establish the true position we need two more facts: what have we actually
accomplished and what is it worth at the values that we originally put on it when the plans
were made? For this we need to go back to the original plan and make an assessment of what
has actually been achieved in terms of what was expected when we drew up the plan. If an
assessment is made of the actual progress with the activities in the plan, it could appear as in
Figure 3.3 where the shading of the bars has been used to represent the degree of completion
of each activity at the end of Year 2.
We might now choose to make an assessment of the worth to the project of the
achievement so far. If we examine the percentage achievements of all the activities that
should have started by the end of Year 2, we might conclude that just over 31 per cent of the
total project plan has been achieved. (This can be seen by the relative areas of the shaded to
the unshaded portions of the activity bars, on the assumption that the rate of expenditure on
all the activities is similar.) Taking 31 per cent of E80m gives a worth to the project of the
work completed at the end of Year 2 as E25m. This is the earned value at the project's midpoint.
Now that we have this additional piece of information we can make a precise judgement
about the position of the project at our reporting point. The position is shown in Figure 3.4.
Looking at Figure 3.4 indicates that the project is actually in a poor position as not only is
actual expenditure below that planned but the earned value is even worse. This project is

20

Using E a r n e d V a l u e
I

k
I
I
I
I

Start-up

Peak activity
Report date

AII
I

r~-

I I
I
I
I
I
I

I
I
I
I

-I

I,

I
I
I
I
I
I
I
I

Wind-down
r~-

I
I
I
I
I
I
I

7
1

Project plan

Planned cumulative cost

f80m 4

Project cost

Planned cost
E40m 4
E30m 4

Year 1

1
1
-

Actual cost

Year 3

Year4

Time

Planned and actual expenditures for a project at the project mid-point

Figure 3.2

Planned and actual expenditure for a project at the mid-point of its life

thus well behind schedule and costing more than was expected for what has actually been
achieved. As a project manager, one would certainly not be happy with a situation like that.
Supposing however, our assessment of progress actually looks like that shown in Figure
3.5; here we can see that the shaded area equates to about 56 per cent of the total work
content at the reporting date. Fifty-six per cent of 80m is E45m; plotting this point on the
cost curves shows that in this case the project is actually doing very well. Not only is work
running slightly ahead of schedule but E45m worth of work has been completed for an actual
spend of only E30m. If this continues, this project could expect to finish early and at a
substantially reduced cost compared to the original estimate.
Calculating the earned value at a reporting point and plotting it against the planned and
actual cost curves allows us to see the precise position of the project in terms of both its costs
and its progress. This comparatively simple process is the basis of all earned value methods.
The important points to note are the fixed relationship between the plan and the costs and
the ability to make an accurate assessment of progress. In practice, earned values are
calculated regularly throughout the project, not just at the mid-point.
The distinction between the sum that has been spent and the value that has been created
or 'earned', as opposed to being planned or scheduled, is the distinctive feature of the

Terms and Methodology

21

presents achievement

f40m 4
f30m 4

Year 3

Year 4

Time

Achievement against the plan at the project mid-point

Figure 3.3

Adding an assessment of progress gives further meaning to the project position

approach and leads to the often used titles of 'Earned value costing' or 'Earned value
management'. It should however be remembered that the earned value method for project
cost reporting was only one feature of the total CISCSC approach to project control
embodied in DoD 7000.2 and implied in the new commercial standard ANSIIEIA-748-1998.
The terms 'Earned value' and 'C/SCSCt or 'C-spec' are sometimes used interchangeably but
this is mong. Earned value cost measurement can be applied without incorporating many of
the requirements of ANSIIEIA-748-1998.

Earned value terminology

DODI 7000.2 defined a series of terms that have become synonymous with the earned value
method of performance estimation. Although they are now common currency, widely used
and well understood, there has been a body of opinion that would like to see them changed
and other terms used. The original definitions and their associated acronyms are perfectly
clear and easy to understand, hence the argument for a change is a weak one; however it does
seem to be gaining ground in both the United Kingdom and the USA. The principal motive
seems to be to disassociate the earned value principles from its US government origins which

22

Using Earned Value

The value of the progress

made in terms of the
original planned costs

Adding the earned value to the cost curves at the project mid-point
Figure 3.4

The earned value is estimated from the actual achievement based on the original costs

are clearly identified with the standard terminology. As the standard terms are still' widely
used and are likely to remain in use for the foreseeable future, they will be used here.
It will be realized from the above figures that three measures of cost are required at any
reporting point. The standard terms for these three values are:
Budgeted Cost for Work Scheduled (BCWS) (Planned value) -This is the sum of all the
planned costs in the project, or any given part of the project, up to the reporting date. (This
is the 640m sum in Figure 3.4.)
Budgeted Cost for Work Performed (BCWP) (Earned value) - This is the cost of all the
progress achieved on the project, or part of the project, up to the reporting date and
expressed in terms of the planned costs originally set out in the initial estimate; it is also
called the 'Earned Value' as it represents what has been earned, not simply what has been
spent (this is the 25m sum in Figure 3.4). (Chapter 7 gives more details of how the BCWP
can be calculated.)
Actual Cost of Work Performed (ACWP) (Actual cost)-This is the total of all expenditure
on the project, or part of the project, up to the reporting date; it is the sum of what has

Terms and Methodology

23

The value of the progress

f45m 4
f40m 4
f30m 4
56% x f 8Om = E45m
Year 1

Year 2

Year 3

Year4

Time

Adding the earned value to the cost curves at the project mid-point
Figure 3.5

Earned value assessment for a project that is running ahead of schedule

actually been spent irrespective of what has been planned or achieved. (This is the 30m
sum in Figure 3.4.)
Once the earned value has been plotted, the difference between it and the other two values
can be subdivided into that part due to variations in the cost of the work done, that is, the
'Cost Variance', and that part due to work being done at a different time from that scheduled,
that is, the 'Schedule Variance'. These variances are shown in Figure 3.6 for the case in Figure
3.4.
From Figure 3.6 it will be seen that:
Cost Variance (CV) = BCWP - ACWP, the numerical difference between the earned value
and the actual cost at the reporting point

Schedule Variance (SV) = BCWP - BCWS, the numerical difference between the earned
value and the planned expenditure at the reporting point.
Whenever the cost or schedule variances are negative, the project is in a poor position; if the

24

Using Earned Value

Cumulative expenditure
fm I
Planned cost at completion E80m

80
7c

Planned spend

Report date

At the end of Year 2

Planned value

40

/I

Planned (BCWS) f40m

Cost variance

Spent (ACWP)

f30m

Earned (BCWP) f25m

I
Figure 3.6

Year 1

Year2

Year 3

Year 4

Cost and schedule variances

variances are positive then the project is doing better than planned. The separation of cost
performance versus schedule peformance is a key benefit of using earned value methods.
With projects that do not use earned value methods, accountants take the difference
between the planned cost and the actual spend (that is, BCWS - ACWP) to mean the 'Cost
Variance' as there is no other variance figure available to them. This is a point that always
needs to be remembered.
Cost and schedule variances may be calculated on either cumulative data or periodic
(typically monthly) data. These variances should be calculated at the lowest level of detail
established within the project and progessively summed through the various levels of the
project. This allows the project manager to see where the cost or schedule performance
problems are occurring and to take appropriate actions.
Although the Schedule Variance is a calculable number, it may be of less real value than
the Cost Variance. However, the Schedule Variance can be turned into a measure of schedule
progress by reference to the S-curve as shown in Figure 3.7.

Earned value calculations

Calculation of the Cost and Schedule Variances are the simplest of the earned value
calculations but further numerical information can be derived which may be even more
helpful. Two useful index numbers can be calculated that give an instant measure of
performance against both the cost plan and the schedule. They are defined as:

Cost Performance Index - (CPI) (Cost efficiency) The ratio of the value created to the
amount spent at a point in time on the project
BCWP
CPI = ACWP

Terms and Methodology

25

Cumulative expenditure

Planned cost at completion E8Om

Planned spend

Report date

Planned value

40

At the end of Year 2

/ Planned (BCWS) E4Om

Spent (ACWP)
25

:I

Earned value

/w

A//"r

E30m

Earned IBCWP) E25m

Schedule slippage 5 months

Year 1
Figure 3.7

Year 2

Year 3

Year 4

Time

Schedule slippage can be established from the earned value and the planned S-curve

Schedule Performance Index (SPI) (Schedule efficiency) -The ratio of the earned value
created to the amount of value planned to be created at a point in time on the project
BCWP
SPI = BCWS
These ratios are shown in Figure 3.8.
Index values greater than one (1) indicate performance either in cost or schedule terms
that is better than planned; values lower than one indicate a worse position. The CPI is,
perhaps, the more useful of the two, it shows the real worth that is being created by the
project, thus a CPI value of 0.85 indicates that for every pound spent, only 85 pence worth of
value is being created on the basis of the original budget. The SPI is somewhat more suspect
as a measure of progress as it is using money as an analogue of time, which may not be strictly
true. Nevertheless, both these performance indices give a valuable clue as to performance to
date and what the future may hold.
Over many years of running projects under earned value conditions, it has been evident
that trends, once established, tend to remain in force until the end of the project. More
depressingly, it has been noticed that if trends do change it is rarely for the better: the
situation is far more likely to get worse. Two formulae can be derived for estimating both the
expected cost at completion and the expected completion date on the assumption that the
trends seen up to the reporting point continue until the end of the project.
For the cost at completion at a reporting point, the formula is made up of two parts: the
cost already expended plus the estimate of futurecost, assuming the trends seen to date continue:
Estimated Cost At Completion (EAC)-The estimated final cost of the project is given by:
EAC = ACWP + BAC - BCWP
CPI
Where: BAC is the Budgeted Cost At Completion

Using Earned Value

26

Cumulative expenditure

At the end of Year 2

Planned (BCWS) f40m

Spent (ACWP)

Planned cost at completion f80m

Earned (BCWP)
Planned spend
Report date
Schedule Performance Index =
Planned value

40

Cost Performance Index =

(
Figure 3.8

h.

Actual cost

30

a
m
= 0.625
E40m

Year 1

sm

= 0.833

II

Year 2

Year 3

Year4

Time

The Cost and Schedule Performance Indices

It should be noted that extensive research carried out on hundreds of projects within the
USA shows this EAC forecasting method tends to be optimistic; it depends on the same level
of cost efficiency to continue to project completion but this rarely happens. More
sophisticated formlate are considered later.
For the overall duration of the project the formula also consists of two parts, the time
elapsed up to the reporting date plus the estimated additional time to complete the project assuming
the trends seen to date continue.
Estimated Time To Completion (ElTC)-The estimated overall duration of the project is
given by:
E

~ = ATE
C + OD - (ATE x SPI)

SPI

Where: BAC is the Budgeted Cost At Completion

ATE is the Actual Time Expended
OD is the Original Duration
The calculations using these expressions are shown graphically in Figure 3.9.
It will be seen that a straight line is drawn from the actual spend value to the predicted
end conditions. This is because these expressions say nothing about the shape of the cost
curve, they are simply point functions and simple linear relationships. In practice the cost
curve is likely to be some form of S-curve, but exactly what shape it will take we cannot say
from these simple formulae.
Both expressions 3.1 and 3.2 can be reduced algebraically:
Taking the expression for the EAC
To simplify the working, let:

Terms a n d Methodology

27

Cumulative expenditure Em

........................................................................................................................................

.,..,7.,
Planned cost at completion E80m BAC

50 40

Planned.value

Year 5

Figure 3.9

Year 6

Time

Calculating the end conditions from the BCWS, BCWP, ACWP, CPI and SPI

A C W=A
BAC = B
BCWP = P
CPI = C
B-P
now: EAC = A + C

Looking back at expression (3.1):

EAC=A+=
C

= A+

Bringing to a common denominator:

PA+BA-PA
P

and from expression (3.4)

B BAC
EAC=-=C CPI
In simple terms, the estimated cost at completion is the original budgeted cost at completion
divided by the CPI. This simple formula is often quoted without reference to the more
fundamental expression from which it derived.

28

Using E a r n e d Value

Looking at the ElTC, the expression is made up of two parts, ATE being the actual time
elapsed and the additional part being an estimate of future time based on the current SPI.
Let ATE = T
OD=D
SPI = S
BCWS = W

Looking back at expression (3.6)and bringing to a common denominator:

TS+(D-(TxS))
E'ITC =S
S
EnC=TS+D-TS
S

(3.7)

= D=OD
S

SPI

This algebraic simplification indicates that the E'ITC is simply the original planned duration
divided by the SPI. However, there are circumstances where this could be misleading; even
so, it is the formula that is often quoted. It should be noted that this formula is not normally
used within the USA. A more reliable predictor of schedule performance and the end-date can
be achieved through analysis of the project plan.
More formulae and performance indices have been devised but the terms, variances and
formulae quoted above are probably the most useful. The following worked example will
show how they are used.

Case example: Estimate of the outcome of a project at the

mid-point of its life
A project set to last seven months contains the eight major tasks with costs and timings
shown in Figure 3.10.

Project task durations, sequences and costs

Figure 3.10

The initial view of the project before work commences

30

Using E a r n e d V a l u e

To calculate the BCWP, each of the percentage completion figures must be multiplied by the
original budgets; this is set out in Table 3.3.
Table 3.3

BCWP calculation at Month 3

Task No.

% Complete

Budget f

BCWP f

100
45
10
15
20

10000
12000
2500
6000
6000

10000
5400
250
900
1200

49

36500

17750

1
2
3
4
5

Total BCWP

Applying the formulae gives the variances, performance indices and an estimate of the
project end conditions.

THE VARIANCES
The Cost Variance (CV) is: BCWP - ACWP = 17750 - 24300 =-66550
The Schedule Variance (SV) is: BCWP - BCWS = 17750 - 23500 = -65750

THE PERFORMANCE INDICES

The Cost Performance Index (CPI) =
- .iZZ50= 0.730
ACWP - 24300
The Schedule Performance Index (SPI) =

BCWS - 23500

= 0.755

THE ESTIMATES OF THE PROJECTEND CONDITIONS

The Estimated cost at completion (EAC)
EAC = ACWP + BAC - BCWP = 24300 + E50500-E1 7750 = 69163
CPI
0.730
The Estimated time to completion (ETTC)
= ATE + OD - (ATE x SPI)

SPI

-3

7 - (3 x 0.755) = 9.272 months from start.

0.755

Figure 3.12 shows the position graphically and it illustrates both the potential overspend and
slippage that will occur on the basis of the trends to date. It should be noted that the

Terms a n d Methodology

31

predicted values only indicate the end position; the path in terms of cost versus time by
which it gets there is not predicted, although it might be assumed that the path will have a
general shape similar to the BCWS curve.

Figure 3.12

9
1
0
Time (months)

Predicted position at project completion based on earned value measurement and

formulae

Additional formulae
The formulae used in the above example are probably the most useful, but more values can
be calculated if required. Given below is a selection of formulae that are sometimes used to
provide additional insight:

i
I

32

Using E a r n e d Value

Variance at completion - (VAC)

VAC = BAC - EAC
Schedule variance as a percentage of the scheduled achievement - (SV%)
SV%= BCWP - BCWS
BCWS

100

(3.10)

Cost variance as a percentage of the earned value - (CV%)

CV% = BCWP-ACWP XI00
BCWP

(3.11)

% of project schedule to be achieved at the report point

--BCWSx 100
- BAC
% complete at the report point

--BCWPx 100
- BAC

% spent at the report point

--ACWPx 100

- BAC

Other estimates of cost at completion - EAC

Besides the basic linear formula for the EAC given as expression (3.1), two empirical formulae
are sometimes used that take into account the combining effect of both cost and schedule
changes. This is saying two things. First, some costs are time-related irrespective of the task
content (for example, service activities such as managing the project), when the schedule
begins to slip costs will tend to rise as a result of the lengthening of the project and second, if
the project is perceived as slipping, additional resources may be deployed, possibly at
premium rates, in order to attempt to recover the position thus incurring extra cost.
EAC, = ACWP +

BAC - BCWP
(0.5 x CPI) + (0.5 x SPI)

EAC, = ACWP + BAC - BCWP

CPI x SPI
Formula (3.16) is sometimes quoted as the 'pessimistic' estimate, while others consider it to
be the 'most likely' value. Empirical evidence from actual projects would suggest that it is a
good predictor of the final outcome and it is often used. It should always be remembered that
the most accurate EAC can be calculated by traditional estimating methods through a
thorough assessment of hours for remaining work, revised bills of materials and additional
project expenses.

Terms and Methodology

33

The 'To-complete Performance Index' for budgeted cost - (TCPI(BAC))

TCPI(BAC) = BAC - BCWP
BAC - ACWP
This formula is used to indicate the level of cost performance that will be necessary to
complete the project within budget from the report point.
The 'To-complete Performance Index' for schedule - (TCPI(0D))
TCPI(OD) = BAC-BCWP
BAC - BCWS
This formula is used to indicate the level of schedule performance required to finish on time
from the report date. In any practical situation, both the To-complete performance indices
should be compared with the actual observed performance indices; if there is a significant
difference, particularly where much improved performance is required, then the realism of
these values must be in doubt.
Table 3.4 shows the calculated values for the additional formulae. Note how much larger
the Estimated Cost at Completion is using formula (3.16). This is undoubtedly the reason
why it is often dubbed the pessimistic estimate, but why it might also be the most realistic
one.

Difficulties with the predictive formulae

Despite the straightforward nature of the predictive formulae, they can lead to difficulties
under some circumstances. In some cases the simplified formulae (3.5) and (3.8) are quoted
without reference to the original, more fundamental, expressions from which they derive.
Whereas the algebra may indicate that this is acceptable, it has a hidden pitfall that
particularly affects the calculation of the E'lTC. Expression (3.8) states that dividing the OD
by the SPI is a correct way to calculate the estimated time to completion; the algebra seems to
indicate it, but that conclusion would be misleading. This may not appear obvious, given the
working shown, but the explanation lies partly in a quirk of algebra and partly in the nature
of what is being done. In essence, cost and time cannot be treated in the same way; when no
work is done, costs stand still, but this is not true of time, which goes on whatever the work
situation may be. Furthermore, the cost that a project incurs is the sum of all the costs in the
project; this is not true of time because the time that a project takes is governed by the time
taken by just a few activities, the Critical ones. Because of the additive nature of the costs and
the fact that the CPI is calculated from the cumulative costs, the EAC expression holds good
in its reduced form.
If we look at the SPI, calculated from BCWP/BCWS, we find that it has some rather
unusual properties because it is using cost as the analogue of time, which is not strictly true; it
works well until the original planned duration is exceeded, after which it changes its
character. The schedule aspect of 'Cost/Schedule Control Systems Criteria' has been a
problem and there was a debate in the early days as to whether or not the term 'schedule'
should be used in the title.

34

Using Earned V a l u e
Table 3.4

Earned value calculations using the additionalformulae

Earned value calculations for the case example
Given

BCWS =
BCWP =
ACWP =
BAC =
CPI
=
SPI
=
EAC =

23500
17750
24300
50500
0.73
0.755
69136

Variance at Completion

VAC = 50500 69136 = f18136

Scheduled Variance %
SVO/o= 7550- 23500 x 100 =-24.5%

23500

Cost Variance %

% of Schedule

~~~~~

-% complete

100= 46.5%

50500 .lo0 = 36.1%

Estimated cost at completion,

EAC, = 24300+l

- 68907

(.5x.73 + .5x .755)-

Estimated cost at completion,

EAC, = 24300 +

(.73+ .755)

= 83721

To-complete performanceindex (cost)

To-complete performance index (schedule)

TCPI (OD) = l5O5O0-

77501

(50500 - 23500)

= 1.21

Terms and Methodology

35

The problem is best illustrated with the followingworked example of a project with two
major activities with costs and times as shown in Table 3.5.
Table 3.5

Example of the possible error in calculating the Estimated Time to Completion (EITC)

I
The Plan (BCWS)
Planned Cost, Cumulative f
Activity A
Activity B
Cum. Total(W)

1000

2000
1000
3000

2000
4000

The Actual Costs Incurred (ACWP)

Activity A
1000
Activity B
Cum. Total (A)
1000

2000
1000
3000

2000
4000

3000
5000

4000
6000

The Earned Value (BCWP)

Activity A
Activity B
Cum. Total (P)

1000

2000
500
2500

1000
3000

1500
3500

2000
4000

1 .O
1 .O

0.833
0.833

0.75
0.75

0.70
0.875

0.667
1 .OOO

4000
3.0

4802
3.6

5333
4.0

5714
3.4

6000
3.0

Performance Indices
CPI (C=P/A)
SPI (S=P/W)
Estimates at Completion
EAC (B/C)
ElTC (D/S)

1000

Period
2

1000

Although the EAC shows a steadily increasing prediction in line with the actual costs, there is
clearly a problem with calculating the ETTC using the simplified formula as the predicted
duration starts to diminish after the planned duration (three periods) has been exceeded.
Furthermore, there is an apparent improvement in the SPI from 0.75 at Period 3 to 0.875 at
Period 4, even though there is no actual improvement in the rate of schedule progress on
activity B which is the only activity current between 3 and 4. The same applies between
Periods 4 and 5. This anomaly stems directly from using cost relationships to determine
progress through time; in fact, once the planned duration of any activity or project has been
exceeded the SPI becomes a measure ofpercentage completion, not schedule progress.
That does not fully explain the reason for the diminishing value for the ETTC and to find
it one has to look back at the original expression (3.2). The problem stems from the fact that
the algebra says nothing about the relative values of the terms in the expression. Once the
T x S) in formula (3.6) will eventually
duration (D) is exceeded, the whole expression

@-+)

become a negative value; this is obviously a nonsense as it has the effect of subtracting a sum
from the elapsed time and time never goes backwards!
However it can be corrected if it is noted that

36

Using E a r n e d Value

EnC=T+D-(T~S)
S
can be used for all cases where the planned duration is not exceeded, that is, T < D. Once the
originally estimated duration D is exceeded, T > Dl the formula is changed to:

that is, the elapsed time is substituted for the original duration.
This can be simplified to
T
ETTC = S

(3.20)

If this change is made and we look at the situation at Period 4, we now get an E'l'TC of 4 /
0.875 = 4.57 periods; an improvement on the original and obviously incorrect prediction of
3.4 periods. However, the result could be further improved if the calculation is based solely
on the performance of the activity or activities that actually govern the schedule progress of
the project. In this case, the reduced form cannot be used and the expression should be
modified to:
E'ITC=T+ D1- (T, xS,)

s1

For T, < Dl
orDl=T,ForTl>Dl
Dl T, S, are the values relevant to the critical activity.
In this example, Activity B is clearly critical to progress so it may be decided to generate the
ETTC at, say Period 3, using the factors that come from Activity B alone. In this case it will be
necessary to calculate the SPI for Activity B at period 3 and it is 1000 / 2000 = 0.5.
Substituting in formula (3.19)

Using the same formula at Period 4 also gives the correct prediction. In fact, the same
procedure of dividing the formula into its component parts could have been applied to the
EAC to generate a more precise estimate although this is less important than it is with the
ETTC.
The reader is thus left with the choice of using either the simplified formulae if you are
happy with the CPI and SPI generated by the aggregate of all costs in the project, or the
fundamental formulae if you wish to base predictions on the performance of those current
activities which are seen to be critical.
Problems with the SPI result from its use of cost as an analogue of time; however, it is
possible to put the time dimension back into the assessment of schedule performance if one
has the original BCWS curve plotted over time. By projecting back onto this curve from the

Terms and Methodology

37

BCWP value from the reporting point, it is possible to measure the achieved progess in terms
of time from the start of the project. When this figure is divided by the elapsed time, a
measure of schedule performance is derived that is based on the time dimension (horizontal
axis) rather than the cost dimension (vertical axis). This method of assessing schedule
performance does not suffer from the SPI value always tending towards 1.0 irrespective of
actual progress. Figure 3.13 shows the principle. However, it should always be remembered
that the best method of determining schedule progress and forecasting the future is to look at
the detail of the project plan and see just what is happening on the critical path.

Cumulative

Schedule performance:
Progress achieved
Elapsed time

Time

Putting the time dimension back into the

assessment of schedule performance
Assessing schedule performance from earned value cost measurements can be improved
if the time dimension is put back into the calculation. This removes inherent problems that arise from
using cost as an analogue of time and eliminates problems associated with the behaviour of the SPI as
conventionallycalculated.

Figure 3.1 3

Conclusion
The above formulae and worked examples show the basics of all earned value measurements
and calculations. It should be noted that many of the features that have become
synonymous with the concept of 'earned value management' are absent. None of these
calculations makes any mention of a work breakdown structure, a set of control accounts or
the organization structure, things which are fundamental features of the CJSCSC approach
and which many still consider to the 'correct' method when using earned value
management. In fact, none of these things are necessary in order to perform earned value
performance measurement at the overall project level. What is needed, however, is a well-

38

Using Earned Value

thought out and properly costed plan, an accurate and timely cost reporting system and a
formal method of assessing progress; all of these features are implied in the worked example.
In many real projects, a suitable project management software package is also necessary.
From a practical viewpoint, a suitable work breakdown structure is necessary unless the
project is very simple and contains few activities. The WBS provides a degree of organization
to the plan that makes life simpler from the point of view of data collection and interpretation
of the results and outputs.
The simple fact is that earned value performance measurement is an accounting
principle, it is not a managerial process at all. The managerial aspect comes from how one
sets up the project to obtain the data and what one does with the results that arise.
Earned value was introduced back in the 1960s as part of a much larger and very detailed
project management approach and it has not broken free from that heritage. As the worked
example shows, earned value performance measurement is something that can be done quite
independently of the managerial and work breakdown arrangement whilst it is equally
possible to create an organization structure and WBS that follows the CISCSC approach
without doing earned value performance measurement. The two aspects can be treated
independently, they are not inextricably linked. The important point is that project
managers who wish to use earned value methods do not have to adopt the whole CISCSC
implied methodology; they must learn to choose what parts are most applicable to their
project situation and tailor their approach accordingly.

CHAPTER

4 Work Breakdown Structures

With the advent of formalized project management methodologies in the USA in the 1950s
came the idea of the Work Breakdown Structure (WBS). This structure was devised to aid
Department of Defense civil servants to account for the money that had been spent by the
project against the items that had been delivered; the organization of project data in a
standard form was made a contractual requirement. Similar accounting structures must have
existed in many organizations long before the DoD requirements were written, as it has
always been necessary to know how and where money is being spent if control is to be
exercised over a business. However, the new formality associated with the work breakdown
structure has become firmly woven into the process of project management.

What is a work breakdown structure?

The work breakdown structure (WBS) is, quite simply, a method of identifying and
classifying the work content of the project in a rational and easily understood way. Its
purpose is to define discrete quantities of work so that:
They can be uniquely identified for what they are.
They can be seen for their contribution to the total project.
They can be monitored and controlled from a time, cost and content standpoint.
Responsibility for achievement and performance can be allocated.
Meaningful historic data can be obtained at the end of the project.
All of the above are important from an overall project management standpoint, so creating
an appropriate WBS is an essential first step in the handling of any complex project. The WBS
is not the same thing as a simple task list although this could specify all the work in the
project. Nor is it the same thing as a project plan -which could also be a method for defining
all the work in the project. A WBS only specifies work to be done; it does not specify the order
in which that work will be carried out as that is a property of the project plan. However, in an
integrated project control environment, both the WBS and the plan will need to interface in
a harmonious way.
The principal feature of a WBS is a rational arrangement of the tasks that define the work at
the lowest level suitable for control; in addition, the arrangement must allow the lower levels to
be summarized upwards to broader areas of related activity or responsibility. Some form of
coding system that relates to the levels is usually employed for reference purposes. A number of
different coding arrangements are possible but they tend to follow one particular aspect of the
project such as the product structure, the organization structure or the plan of work, as the
chosen aspect is considered particularly important in terms of both reporting and control.
Work breakdown structures used in the past have tended to be product-basedand this was

40

Using Earned Value

encouraged by the US DoD; its contracts define a WBS as a 'product-oriented family tree'. This
arrangement helped the DoD's own project managers to account for what their money was
being spent on; that is, they could relate expenditure to items of procured hardware or discrete
services. The WBS that is imposed under this arrangement is called the 'contract work
breakdown structure' as it is specifically designed for reporting against the contract statementof-work which is normally written in terms of the hardware or other specified services to be
delivered. The idea that a WBS should be 'product oriented' has become so embedded in the
conventional wisdom of earned value performance measurement and the general principles
of project management that some project managers believe it to be either the 'official', 'right'
or the 'only' form of WBS. A work breakdown structure has been desribed as:

... a product-oriented task hierarchy of all the work to be performed to accomplish the
project contractual objectives. The products may be software, hardware, documents, tests,
reports, support services or other quantified elements of the objectives ... Use of the term
'product oriented' does not mean ... that the WBS should follow a structural
decomposition of the functions of the delivered product unless this is appropriate.
This somewhat ambiguous definition is clearly aimed at a project situation with a contract
between a sponsor and a contractor for the work. It is thus designed to conform with a
contract that sets out the project in terms of deliverables: components, finished goods,
documents and services, for which both prices and time-scales can be determined. This is very
much how the US Department of Defense operated, expecting a considerable amount of time
and effort to be spent at the start of the project to plan and cost all the activities in detail before
contracts were signed. Nevertheless the DoD did expect the WBS to follow the principal
physical breakdown of the product because that is how it wanted costs to be reported. Besides
the overall project situation, it was interested in the position on each of the major elements of
the end product. This is particularlyimportant with a complex product; for example, a combat
'plane is composed of a set of 'systems' such as airframe, radar, engines, guns, etc, each of
which is a development project in its own right and which all have to be brought together in
the final product. Difficulties with any of these system components could impinge directly on
the conclusion of the total project. Cost reporting in that format also allowed the DoD
planners to assemble historic data that could be used to estimate and compare the costs
associated with developing new items of defence equipment. By knowing the costs and timescales associated with developing all the various components, new projects could be seen and
estimated more clearly at the concept development stage; this clearly placed the DoD in a
stronger position when it came to negotiating contracts for new systems.
From the DoD's point of view, this whole approach made considerable sense. Whatever
problems operating this way caused for industry, the DoD was not concerned; when the DoD
tried telling industry how to manage with PERTICost it was dealt a sharp rebuff. As a result
industrial managers were left to figure out a way to conform with its particular form of WBS
and the reporting requirements it imposed under CISCSC.

The concept of the work breakdown structure

The WBS is a logical way of describing all the work in a project without considering the
order of events. If, for example, we consider a project for someone to build a house and

Work B r e a k d o w n S t r u c t u r e s

41

garden to their own requirements, they might decide that the project consists of a set of
major tasks:
Administration - obtaining the land and the necessary permissions plus dealing with all
the contractors
Architecture- getting a set of plans drawn
Structure - engaging a builder to dig the foundations and put up the walls and roof
Internal systems -engaging contractors for the internal works
Landscape and garden - engaging a gardener for the external landscaping.
This simple division recognizes the distinctive nature of each of these broad areas while
encompassing all the work to be undertaken; at the very highest level a WBS could look as
shown in Figure 4.1

House and Garden

~~~~~
& garden

Figure 4.1

Top-level work breakdown structure for a housebuildingproject

The top-level WBS serves only to divide the work into identifiable and fairly discrete tasks of
different but internally related character. To use the WBS in a meaningful way both for
planning and control requires a further breakdown into more detailed aspects that are clearly
identifiable as being distinctive. An additional breakdown has been done in Figure 4.2.
At this point it is possible to see all the work in the project in a way that would allow us to
1) plan the project as we can clearly see that some aspects have to be completed before others
can start, 2) estimate at a broad level the likely cost of the project, providing we have some
data on building costs, and 3) divide the work among the various contractors in a way that
suits our purposes and is convenient to the skills available. If, for example, the various aspects
are allocated to the contractors in the way that the sponsor has decided that the work should
be done, the WBS will look as shown in Figure 4.3.
Note that the allocation of work does not necessarily follow the major divisions of the
WBS, although there is some broad grouping. For reasons known to the sponsor, particular
tasks have been allocated to particular contractors where there is either a specific preference
or one contractor has a skill that is particularly valued - the specially fitted kitchen, for
example. This situation is not uncommon, particularly with complex projects.
If necessary, this WBS could be broken down still further into aspects such as labour,
material purchases, and services. Thus the work block given as 'foundations' could be further
reduced to 1) hire of an excavator, 2) excavation of the foundations, 3) supply of ready-mixed
concrete, 4) pouring and levelling of concrete, and 5) disposal of the soil (see Figure 4.4).
Each of these is a separately identifiable purchase or task and is the lowest realistic level to
which this work block could be reduced. In practice the whole task might only be about three
days' work. For anyone managing the building of a house for their own use, such a detailed

Using Earned Value

42

I
Administration

House and Garden

Architecture

Internal systems

Landscape
& garden

Hi
Gas boiler

permission

regulations

-4

Fmnt drive

Bills of Materials

D
D
1 I-GiiqD b

planning permit

Site supervision

Water & C. heat

Lawns

Kitchen units

Borders & plants

Plaster work

& curtains

Figure 4.2

Expanded work breakdown structure for a housebuilding project

breakdown would almost certainly not be worthwhile, although the builder might find it
worthwhile if he has given a fixed price quote. However, for a developer building an estate of
perhaps 100 houses, work estimation at this level would be sensible because of the costs
involved; the three days of foundation work on one house turns into 300 days: more than a
man-year's work.

Creating a work breakdown structure

For a work breakdown structure to be effective in a practical project situation, four criteria
should be met. The WBS must:
1 Impose a rationality upon the project data -there should be a recognized and understood
logic to the structure.
2 Contain a hierarchical coding structure - lower levels of data must be capable of being
summarized upwards with no loss of accuracy or identity.
3 Compliment existing data gathering systems - the structure must not demand any greater
precision, from the data with which it must interface, than can be made available.
4 Conform to the requirements placed upon it - it must allow the generation of appropriate
data in a form that is useful.

Work B r e a k d o w n S t r u c t u r e s

43

House and Garden

Architecture

~ruTre

11

Internal systems

el

Pi
pzzizq
Land purchase

permission

P
Kitc en units

Work Allocation

r-l
H
Plaster work

Figure 4.3

Wall decorations

Expanded work breakdown structure with allocation of work

Point 1 is self-evident: without it there would be little point to creating a WBS although the
structure of the logic can be different depending on requirements. As we have seen from the
house-building example, even if a WBS appears logical from one viewpoint it may appear less
so when viewed in a different way - by the contractor, for example.
Point 2 follows directly from the logic that is created by observing the rule in Point 1. A
work breakdown numbering system is not strictly necessary as work package descriptions or
titles could be used, providing one remembers the logic of the structure, but if a
computerized or integrated project management system is in use a rational numbering
system is essential.
Decimal-based numbering systems are by far the most common and are easily developed
and understood. Taking the house-building project a step further, a partial expansion of the
WBS is shown in Figure 4.4 with a decimal-based numbering system included. Summing
upwards through the various levels of the coding structure will produce aggregate results at
successively higher levels; thus the total cost of the Internal Systems - 1.4 is the sum of all the
work blocks starting with 1.4, that is, 1.4.1, 1.4.2, 1.4.3, 1.4 etc. However, this WBS
arrangement cannot be used to simply determine the value of the work done, for example, by
Acme Buildings, the building contractor. The way in which the digits are used represents
both the levels of the WBS and the logic of its structure. What each level is taken to represent
will depend on how the WBS is to be used and what information is expected to be
represented; this has significant implications for the design of the WBS.

44

Using Earned Value

permission

regulations

management

House and Garden

1.o

H
Bills of
Materials
1.2.2

Foundations
1.3.1

Gas boiler
1.4.1

- Hire excavator

Electrics & TV
1.4.2

1.3.1.1

- Dig trenches

H
4
Assist with
planning permit
1.2.2

1.3.1.2

- Supply concrete

Site
supervision
1.2.1

1.3.1.2

- Water & C. heat

1.4.3

1
Lawns
1.5.2

Kitchen units

Pour & level

1.3.1.4
Wall decorations
1.4.6
Carpets &
curtains
1.4.7

I
I

Figure 4.4

Partial work breakdown structure- note expansion of 1.3.1

Points 3 and 4 have practical implications; if they are not met, the whole process of
project control could be made difficult, either from a data systems viewpoint or a contractual
one. Point 3 states that the WBS coding must fit with the data-gathering systems with which
it must interact. With a project like personal house-building, this point may not be significant as the incoming data are all likely to be under direct human control in the form of
conversations, letters, invoices, contracts etc, and the information can be extracted and
interpreted according to need. With automated systems that are typical of current industry
this situation is quite different; costs, times and other information relating to tasks are
gathered on computerized systems and normally conform to strict rules as to format and
content.
There is no point in defining a WBS coding system that demands levels of subdivision of
data that is greater than the basic level at which data is collected by the normal company
procedures. For example, if one wishes to collect costs automatically against tasks in the
WBS, it would be pointless devising a six-digit code for detailed tasks if the company's timesheet system (the principal system through which information about what work is being
done, by whom and how much) can only cope with five digits. Often these informationgathering systems are fundamental to the company's accounting procedures and, unless the

Work B r e a k d o w n S t r u c t u r e s

45

project is very big and of great importance, they are unlikely to be changed. Any project WBS
that aims to interact directly with existing company systems must conform to the limitations
they impose. In some cases this can cause problems, particularly if a company is in the
position of a contractor to a sponsoring company that demands reporting against a WBS that
the sponsor has devised for reasons of their own.
Point 4 relates directly to the uses that are expected from the data generated by reporting
against the WBS. In the case of the US DoD, it was concerned to gather data about the
progress and costs associated with the specific items of deliverable hardware, hence its
insistence on the WBS being 'product oriented'. This was the basis of the WBS logic and all
the input data was required to support this overall requirement.
If the funding sponsor for a project, be it the US DoD or any other organization,
demands cost reporting in a product-oriented format then clearly the WBS must, in one
form, conform with that requirement. If, in the house-building example, the principal
requirement is to gather and accumulate costs against each of the principal contractors then
the WBS format shown in Figures 4.1 to 4.4 would not be the most convenient one.
The product-oriented format is not the only structure possible; when devising the WBS it
is for the project organization to consider how it intends to control the project, what data it
wishes to use as input to the control process and what data it expects to generate as a
meaningful output. A wrong choice of WBS at the start of the project could lead to reporting
difficultieswhen the project gets into full swing.
Many forms of WBS can be devised; each project is likely to be unique, if only in a small
way, but they generally conform to one of four basic types:
product based
organization based
task based
hybrid structures embodying two or three of the above.
Figure 4.5 shows the first three alternatives.
The product or physical structure, Figure 4.5a, gives a costing system that relates costs to
the physical elements of the product. Here, when costs are collected, they will show what
parts of the product have been worked on and how much has been spent on designing
or making each element; what it may not easily tell is who (that is, what department) has
been doing the work. In practice, a totally product-based WBS may not be possible as aspects
of the project may have an all-embracing character and may involve a service rather than a
product; managing the project is one example, while compiling the project accounts is
another.
With an organizational structure, illustrated in Figure 4.5b, work is defined according to
the department within the company that does the work. When costs are collected through
the time-sheet or bookings system, cost reports will show which departments or divisions
in the company have been working on the project and how much has been spent in those
areas.
The task or functionally based structure, shown in Figure 4.5c, relates costs to the tasks
that are expected to be performed. Cost reporting in this arrangement shows what work
people have been performing in terms of the tasks they normally carry out. This system
might seem to offer less than the other two in terms of control but it has much more to
recommend it than either of the others, in the context of integrated earned value per-

46

Using Earned V a l u e

a) Product-basedWork Breakdown Structure

Product

Q
Unit

Sub-system

b) Organization-basedWork Breakdown Structure

Company

Evision

c) Task-based Work Breakdown Structure

Project

Figure 4.5

I
I

Work area

Alternative work breakdown structures

formance measurement. The reason is that in order to assess progress it is necessary to refer to
the project plan and this is normally structured around the activities or tasks that are
expected to be performed. Because the other structures lay their emphasis on either 1) what
part of the product is being worked on or 2) who does the work, it is often difficult to
construct a satisfactory plan in those formats that fully describes the planner's intentions.
Planners think in terms of tasks rather than components or departments, particularly where
the work involves bringing together components and where the work is shared across
departments.
Hybrid structures can also be constructed that embody elements of two or more of these
structures, an example is shown in Figure 4.6.
Each of the levels in the structure gives the information of interest that will be
accumulated at the given level. This will be significant for project management and is an
important factor in the design of the structure. In practice, such an arrangement might prove
difficult to integrate with both the planning system and the cost collection system as it has,

Work Breakdown Structures

47

a
m3n4m
Project

Overall Project Level

First Level of Interest

(Where is money spent?)

1 Product Design & 2

Manufacture

Assembly and

Second Level of lnterest

(On what are we spending?)

Assembly

purchases

management

Purchases

Planning

Work breakdown structure

(What are we doing?)

Fourth Level of Interest
(Who is doing it?)

Design
Engineer

Top Level

Project

1 .O

~~~l
1

Production design 8
Manufacture

1.1

Level 2

Part No. 4

1.1.4

Level 3

Scheming

1.1.4.1

Level 4

Stress Engineer

1.1.4..1.2

Stress

Hybrid work breakdown structure showing the different levels of interest and how they are
summarized upwards. As the lowest level in this structure is a skill (stress engineer, design engineer,
etc.): integrating it with the plan and reporting system may be difficult

Figure 4.6

Concorde Work Breakdown

Defenses GBnBrales

SUD Documentation

BAC Work Control Summaries

Major Tasks 01

rrtn rrln rrtn rrtn

m@mB

T
E1
Minor Task 01

\
Prototype
Pre-production
Production

Authorizations on Line Departments

Figure 4.7

Work breakdown structure used on the Concorde project

48

Using Earned V a l u e

at its lowest level, the responsible person or required skill. However, integration with the
plan would be possible at the third level, that is, the work being done.
Besides providing a rational structure to the project data that will aid both reporting,
interpretation and progress measurement, the WBS can form a bridge between the planning
system and the company's systems for cost and time reporting, providing it is properly
constructed. This is very important in the context of earned value methods, particularly if an
integrated approach is required. Work breakdown structures are best devised and
implemented at the start of a project.
An example of a WBS that was actually used on a major project is given in Figure 4.7 and
Table 4.1. Although this was constructed over 30 years ago, it is basically a task-based
structure rather than a product-based one and has, at its lowest level, minor tasks that relate
directly to the plan; it is thus a rather more modem approach than the traditional work
breakdown structures that have often been advocated.
Table 4.1 Chapter numbers for the main work blocks
within the Concorde work breakdown structure. Chapter
9 relates to flight tests which in turn is broken down firstly
by the three design standards and then into major and
minor tasks as shown in Figure 4.7.
Chapter No

Subject
Design
Systems Tests
StructuresTests
Wind Tunnel Tests
Mock-ups and Space Models
Simulator
Equipment Design and Development
Tool Design
Tool Manufacture
Aircraft Manufacture
Flight Tests
Not used
Ground Equipment
Spares
Technical Publications
PERT
Liaison and Transport
Translation and Enquiries
Refurbishing
Reserve
Material and Equipment Stocks
Post C of A Development Preparation

Work breakdown structures need care in their construction; a poorly conceived structure
will:

Work B r e a k d o w n S t r u c t u r e s

49

prove difficult to integrate with existing data systems and reporting structures, and
fail to provide information in a form in which it can be most useful.
The way in which the WBS relates to other data structures in the project is discussed in
Chapter

CHAPTER

5Plans and Budgets

All earned value techniques rely on a well-conceived project plan. The project work must be
clearly perceived at the start and well understood in terms of both what is to be done and the
order it should be done. Sometimes these criteria cannot be met; the reasons could be:
There is an extensive amount of experimentation involved.
The work is highly innovative with many uncertainties.
The strategy involves taking advantage of whatever opportunities may arise.
External forces are likely to dictate the course of the project.
With situations such as those given above, it is very difficult to plan any project with
confidence that the plan will remain firm and workable over anything but the shortest
horizon. Projects with these characteristics tend to be hostages to fortune; they could make
very rapid progress if things fall into place but equally they could be subject to delays,
overruns and major changes of direction and may even end in cancellation. This is not to say
that such projects are not worth starting: some may make a significant breakthrough that
could result in a major competitive advantage. What must be recognized is that projects
which cannot be planned with reasonable certainty contain significant risks. Projects of this
type might be described as 'entrepreneurial' or 'radical' in character; earned value methods
were not designed to handle such projects.
Earned value performance measurement may be a project accounting technique but its
origins lay in a contractual arrangement between a sponsor and a contracting organization.
In particular, the sponsor wished to have assurance that what has been contracted for will be
carried out in line with an approved schedule of activities and within an approved budget. If
variances arose, the sponsor wanted to be able to see where those variances were occurring
and be able to estimate what the overall effect on the project was likely to be. So, earned value
methods started as a form of assurance that projects were understood in sufficient depth,
planned in sufficient detail and controlled in such a way that a contract could be agreed
between both parties that had a reasonable chance of being fulfilled in line with the original
expectations. A contractual arrangement is not, however, a requirement for earned value
methods as they could be used equally well on an in-house or private venture project. What
is important is that the project is understood and planned to a degree that would be suitable
for a contract to exist. However, the existence of a contract is not a guarantee that a project
will be planned to the degree that it deserves and this fact undoubtedly accounts for some of
the well-publicized project disappointments.

Planning and the project cycle

The principal steps in the project life cycle are given in Figure 5.1; for convenience, the cycle
has been divided into five major blocks. The execution and control blocks are so interlinked

Plans a n d Budgets

51

that they could be considered as one but, as they represent different functions, they can be
treated separately.

lnitiate
ldentify objectives
Understand constraints
ldentify stakeholders
Develop strategy

Plan
ldentify work content
Define work scope
Create schedule of work
Generate budgets
Create organization
Define procedures
A

Execute
Authorize work
Perform defined work
Implement procedures
Implement changes and
corrective actions

,
<

Control
Monitor performance
Control changes
Understand variances
Initiate corrective actions
Forecast the future

Complete
Delivery, handover
Settlement of claims and
fees
Disposal of redundant
assets
Post-project review
Figure 5.1

The project cycle of activities

All projects start with some form of initiation phase where the ideas for the project are first
discussed. Projects are started for a host of reasons but they all have one thing in common:
the sponsor or initiator expects to gain some form of benefit from a successful conclusion.
These benefits may be seen as the objectives of the projects; if they are not achieved, the
project will be worthless. However, the realities of the world in which the project exists can
have a major bearing on the form of the project, which could include such things as the
likelihood of achieving the objectives and the constraints that are placed on what can be
done. The process of initiation should conclude with the development of the project
strategy, that is, the overriding set of decisions that will shape all that follows. Strategic

52

Using Earned Value

matters could involve fundamental decisions regarding what will be done in-house and what
will be outsourced, what risk-sharingarrangements will be put in place, what partners to seek
and what organization will be set up. In some cases these may be simple decisions because
the project is small and straightforward or because there is a successful precedent for doing
things a particular way, but in other cases these could be major decisions that have a
fundamental bearing on what project structure is created and how the project is conducted.
Planning for the project must be carried out within the strategic framework. For some
organizations, such as contractors, who are told what is expected of them, project strategy
may not be a major issue. Even so, contractors are often advised only of the requirements:
how they achieve them is up to them, but to convince the sponsor they are the right
contractor to employ, they may have to produce a credible plan.
Planning involves identifying the work to be done. This can rarely be done in detail at
the start as not enough information will be available. However, a start has to be made and the
most useful starting-point is to rationalize the work into identifiable and discrete major
blocks of work. The process of doing this has already been described in Chapter 4, on work
breakdown structures. A work breakdown structure is not a plan as it does not define the
order of working; earned value methods imply the control of both cost and schedule and the
schedule aspect is absent from the WBS.
For the purposes of control, all planned activities require a 'scope of work' to be defined this is quite simply a statement of what is included in the activity and, by exclusion, what is
not. At the highest level of the project, the scope of work could be fundamental to what is
included in the contract between the parties, while at much lower levels it could be simply a
definition of an activity. Nevertheless, without a clearly understood scope of work no one
can be sure what is expected from any activity.
A definition of work to be done through a host of activities will not form a plan unless it
shows how the work is to be done, in particular, the order in which the activities must be
performed. Unless it consists of a set of unrelated activities, no project can be performed in
random order; with most real projects there is often a very clear basic order that is evident
from the nature of the project and it cannot be avoided. Nevertheless, there are usually
choices about the way individual activities are performed and how they relate to other
dependent activities. Positioning activities in relation to one another and placing them in
their correct position in time is the essence of the planning process, it is fundamental to
effective project management and to the use of earned value methods. To be useful, the
planning must be done in sufficient detail that discrete activities can be identified at a level
that can 1) be used for direct control, and 2) be related to other activities so as to accurately
define the whole programme of work. There are some obvious difficulties with this that relate
directly to our own abilities to perceive accurately all that is necessary and all that will occur.
The larger and more complex the project and the greater the span of time over which it is
conducted, the more difficult this becomes. This problem is recognized in the concept of
'rolling wave planning' which expects activities in the near term to be planned in detail
while activities that recede further into the future are planned in progressively broader
blocks.
All activities require resources of one sort or another - these could be materials, labour,
services or facilities. The notional allocation of resources to the activities shown in the plan is
often termed 'resource scheduling'. It may be the case that allocating resources to the plan as
originally set out will show resource demands that cannot be met. This will call for a revision
of the plan to conform with the resource constraints. When the schedule of activities is made

Plans a n d

Budgets

53

compatible with the available resources the project will have a viable plan, assuming there
are no unforeseen elements that could make it impossible to complete some activities on
time. All resources have a cost, whether it is direct payments for materials, services and
facilities or charges for labour and the associated overheads. When resources have been
scheduled against activities in the plan, costs can be estimated for each activity. Summing all
the activity costs in the plan will lead to an overall project cost. Projects contain a degree of
uncertainty that is inevitable in anything that attempts to look into the future and do
something that has not been done before. It is therefore normal to include a contingency for
costs associated with unforeseen work and events that may occur but are not included in the
plan. The addition of a contingency to a basic identified cost for an activity creates a 'budget'
- a sum of money which it is expected will cover the cost of the activity when the work is
performed. Summing all the budgets plus any overall contingency will generate the initial
project budget; this is often termed the 'baseline' position, that is, the initial starting-point
from which all progress will be measured. This initial planning process is illustrated in
Figure 5.2.
Define
overall scope

Rationalize
work

1
1

Overall project

D1
Major work block

Major work block

Plan tasks

Assign
resources

Add
contingencies
and combine

The initial planning process

Figure 5.2

Generating an initial plan for both activitiesand costs

54

Using Earned Value

For contemporary projects of any size, the planning and budgeting process will be done
using a software package that contains network drawing and analysis, resource scheduling
and activity costing. Compared to the position of 15 years ago, things are very much easier
for the project planner and a great deal of drudgery has been removed; nevertheless the
amount of thought and attention to detail has not been lessened. If anything, modern
methods require much greater discipline as the element of discretion does not exist in
integrated, automated systems. More information on current packages is given in Chapter 8.
The work content of the project, the plan of work, the schedule of resources and the set
of budgets form the basic set of information that any project needs if it is to go forward in an
orderly way, but information is not the only thing a project requires. All projects must have
an organization structure and a set of rules and procedures by which it must operate.
Decisions about these important aspects are required at the planning stage - it is too late to
leave it until the project gets going.

Baselines
The project baseline is an essential feature of any project performance measurement system;
if the project started from a position that was known to be incomplete or unclear then
attempts to measure, as opposed to simply observe, progress are always going to be suspect.
Baseline plans take time to compile in the detail required - for some very large projects,
periods in excess of one year might be required with a great deal of thought and discussion
with all the parties. It is at this stage that projects attempting to use earned value, particularly
in the United Kingdom, have gone wrong; quite simply, far too little time has been spent on
developing the baseline plan in the degree of detail required. This in turn has arisen from the
practice of putting work out to competitive tender with, perhaps, a 12-week period in which
to respond with the fully costed plan and then awarding the contract on the basis of the
lowest price. If the work is complex there may not have been sufficient time to develop the
plan fully; when the work starts the deficiencies in the plan become apparent and changes
must be made which then makes reporting difficult against the original plan. This can be
made worse if the contract is let on a fixed-price basis; however, in this situation the
contractor may not be willing to provide anything more than BCWP data.
Earned value methods were developed in the 1960s when cost-plus-fee contracting was
the normal arrangement for large defence procurement contracts, as this reflected the
inherent uncertainty in this type of work. The US Department of Defense recognized that the
time taken to submit a tender might be too short to develop an accurate baseline plan so it
allowed a period after contract award where further work on the plan could be done before
the baseline was finally established. It was accepted by the DoD following the Arthur D. Little
study, that with large projects 18 months might be needed before a baseline programme
could be agreed. This fact alone could mean that significant changes could be introduced
into the plan submitted at tender award in the light of more detailed study. It also tends to
emphasize the point that earned value methods were intended for use in a cost-plus-fee
contractual arrangement. Both these points have been overlooked, or were never
understood, when some projects have attempted to use earned value methods; inevitably
they have run into reporting and measurement difficulties.

Plans and Budgets

55

Budgets and contingencies

All plans and budgets for projects are, to a some extent, estimates of what will happen.
Estimates imply a degree of uncertainty, they are not a guarantee and they could be wrong. It
is beyond the scope of this text to go into the reasons why estimates turn out to be wrong or
the many methods that have been devised for generating estimates for project work but
techniques include:
subjective estimating, based on experience and intuition
synthetic estimating, based on a build-up of known measurable elements (synthetics)
comparative estimating, based on a rational comparison with previous projects
parametric estimating, based on formulae and key project characteristics (parameters).
Whichever methods are used there will still be an element of uncertainty, and the greater the
degree of novelty or innovation the greater the uncertainty. Resolving issues of uncertainty
that arise in the course of the project generally involves expending more effort and resources
and they could also require more time than was originally conceived. It is therefore normal
to add a contingency to cover costs, and possibly time, that may be needed to address
uncertain issues that could occur.
Company pressures towards underestimation can be considerable. Whether the project
is an in-house one or the subject of competitive tendering, there may appear to be
advantages in a low initial estimate. All project managers know that a low-cost figure is more
likely to gain board-level approval and, in general, is more likely to win in a competitive
situation. Another view that is often taken is that if a low initial budget is set, a lower overall
project cost will result than if a higher and more reasonable figure had been fmed. Some
managers adopt a hard line on estimates, saying they should all be based on success first time
and that contingencies should never be included as once they are known people will use
them whether needed or not. This attitude may stem from:
Wishful thinking: 'If I say it will happen, it will happen.'
A mistrust of human nature, possibly based on experience or an attitude to life in general.
Neither of these show an enlightened attitude, as contingencies represent a reserve to cover
the unknown and unforeseen; to refuse to admit contingencies is to assume perfect
knowledge of the future.
Of course, there could be good reasons for tendering with a low price but if it is done in
the full knowledge that costs may rise and a loss may be incurred, it can be a rational decision
aimed at generating business. This is quite a different situation to a low price being generated
through internal company pressure that simply distorts judgement.
Some organizations adopt a more reasoned view of contingencies and include a structured
build-up within the project budget; Figure 5.3 illustrates this principle. Individual task budgets
are set at a level that, in the view of the estimator and the manager responsible for the task, is
reasonable for the defined work. This will be the budget issued to the personnel required to
perform the task. Several minor tasks can come together to form a major task or work package
and, by either a rule-of-thumb or a risk assessment, a contingency is added to the budget to
cover underestimates due to unforeseen difficulties in all the minor tasks. The work package
manager may, in theory, control the use of contingency to accommodate any overruns.

56

Using Earned Value

Company budget

Initial estimate,
defined work

Reserve for
underestimate
of defined work

unforeseen tasks
and underestimate
of difficulties
Reserve for
unforeseen problems,
management effects
and new requirements

Managed by
Task Operative

Figure 5.3

Work package
Manaaer

Project
Manager

Board of
Directors
Proiect

Build-up and allocation of contingency to various budgets as used on some UK projects

All work package budgets, including their contingencies, can be summed to give a total
budget; to this can be added an overall project contingency which is under the direct control
of the project manager and can be released to the work package managers as needed. In the
case of an internal project, the company directors can determine the company budget by
adding another contingency which can only be released on application to the board by the
project manager. With externally sponsored projects, the sponsors can set their own
contingency fund to cover contractor cost overruns or additions to the project plan that they
may authorize. Whether it is an internally funded or an externally sponsored project, there is
no requirement to divulge the size of this contingency to the project manager - it may be
wise to keep this matter quiet.
Just how easy it is to 'manage' a contingency fund must be questionable; once a task hits a
significant problem and starts overspending, money may have to be spent until the problem
is solved, whatever the size of the contingency. Despite this drawback there are merits in this
approach:
It recognizes that the future contains unknowns and if problems arise they are likely to
cause project costs to rise.
Provision is made in the overall company plans for increases in project cost.
Allocation of the contingency fund to specific tasks gives early warning of potential
overspends.
The above arrangement of contingencies, or something similar, is widely used in project
work and DODI 7000.2 sought to formalize a slightly different arrangement with specific
terms. Illustrated in Figure 5.4, these terms have become synonymous with the earned value

Plans and Budgets

8%.

57

Possible expenditure based on risks and unknowns

Budget at Completion (BAC)

Performance Measurement
Baseline (PMB)
Budget

f
Expected expenditure
Planned expenditure
based on known work
Build-up of contingencies

Sum of all priced

Control Accounts
or
Work Package
Budgets

Time

Allocation of contingencies and budgets from DODI 7000.2. This approach is implied in
ANSIIEIA-748-1998and is thus the accepted US practice

Figure 5.4

approach in the USA but they are not so familiar in the UK and, if the work is being done on
an internal project where there is no external sponsor, they are not entirely appropriate.
The position is shown in Figure 5.4; unlike the British practice from the early days of
CISCSC until the EVMS standard of today, US practice typically holds contingencies at the
project level. This contingency is known as management reserve, and is held and controlled
by the project manager for allocation to additional work effort that is still within the scope of
the contract. Management reserve should not be allowed to cover or mask cost overruns. The
undistributed budget is simply a holding account where the budget for recent contract
changes may be held until detailed planning can be accomplished. As the work and budgets
are planned in detail, the allocated costs are removed from this holding account. The
combination of the existing baseline plus undistributed budget is known as the Performance
Measurement Baseline (PMB). An important distinction between management reserve and
undistributed buddget is that the latter is for known work, whilst management reseve is held
for unknown work. Thus, management reserve is excluded from the PMB. However, the
Budget at Completion (BAC) includes the management reserve and is thus the current
estimate of the likely overall cost based on an assessment of risks. As the project progresses,
some risks will materialize and demand additional expenditure, hence the PMB will move
towards the BAC.
Even more confusing can be the question of profits and how to deal with them. This is not
an issue if the project is a purely internal affair but it can be important when the project is
being undertaken by a contractor. It need not be an issue if work is procured on a simple timeand-materials basis where the appropriate profit is included in the agreed rates. In a cost-plusfee situation, however, profit is taken by adding a sum to the actual expenditure; the amount
will be governed by whatever formula is agreed between the sponsor and the contractor. In

58

Using E a r n e d Value

the USA, profits are always excluded from performance budgets. This element must be taken
into account when the sponsor fixes his final budget; however, the amount of profit paid to
the contractor will be set by the actual out-turn costs, which could be more or less than the full
budget amount. Furthermore, the profit actually claimed can be influenced by any incentive
included in the contract that awards a higher percentage profit if the task is completed at a
cost below the planned budget or caps the profit if the cost budget is exceeded. One must be
very careful when it comes to incentives, particularly when the performance of the deliverable
article or system is also a factor in the profit calculation; it may not be possible to calculate an
incentive profit payment (or reduction) until some way into the project and after many
activities are complete and paid for. It is also perfectly possible that some activities, materials
or services can be procured at fixed prices with profits included.
Earned value, it should be remembered, is an accounting convention for performance
measurement; as such it is a project management tool, it is not a profit-and-loss accounting
system for a project as a business venture. For this reason it is up to the project sponsor and
contractors to agree the budgeting and costing arrangement they intend to use for earned
value performance measurement. If profits cannot be simply assessed it would be better to
leave them out of the earned value process and perform all the measurements and
calculations on the basis of costs alone. Separate provisions will then have to be made in the
contractors and the sponsor's project accounts, both for claiming and paying profits.
At the end of the planning and costing process a plan defining all the activities and their
associated budgets will be generated for Board or customer approval, but it is at the Board's
approval stage that things can start to go wrong. The plan may look reasonable and the
project desirable, but in the eyes of the Board the costs are seen to be too great. Perhaps there
is a fear that the competition may be offering something cheaper, that the customer may
reject the proposal or that there is not enough money in the company budget for the given
year. Sometimes it can come from the customer who may feel that only one contractor has
the proper capability to do the required work but the tendered costs are too high. All too
often senior management, either on the company or the customer side, concludes that the
estimate itself must be wrong. Clearly, they reason, the planners and managers have put in
excessive contingencies, made pessimistic assumptions and this time there will not be the
problems that occurred on the last project. As most of the estimates represent the judgement
of the individuals, they can usually be 'persuaded' by zealous Board members to modify that
judgement to something more acceptable. After downward revisions, contingencies are
usually the next thing to go; they may be seen as a luxury provision for something that may
never happen and if they are included they will be spent whether needed or not. Finally,
arbitrary cuts may be applied to bring the budget down to an 'acceptable' level. This
depressing story will be only too familiar to many project managers.
Of course, estimates should always be scrutinized and challenged if they seem unreasonable, but one should be careful that scrutiny combined with wishful thinking does not lead
to distorted judgement. History shows that the 'savings' made on paper are likely to be
illusory and will disappear as the project proceeds. Few development projects are completed
within their budgets and most overrun their targets, sometimes by quite massive amounts.
This may not necessarily imply a lack of control but simply the impossibility of foreseeing all
that will happen in the future and the problems that may be encountered. Project managers
should guard against applying wishful thinking and undue pressure for low estimates in an
attempt to get their project selected and should be wary if they observe the process in other
more senior individuals. An overly optimistic view at the beginning combined with budgets
f

Plans a n d Budgets

59

below the real needs of the project can sow the seeds from which really thorny problems can
grow later on.
A budget can be considered as a formalized statement of the amount of money
provisioned for some activity or purchase. It should, at least at the start, be seen by all parties
as adequate to cover the intended work. Budgets should be allocated to members of the
project staff charged with directing activities for which they are responsible. Budget
allocation may be hierarchical with work package, or control account, budgets being held at
a section-head level while lower-level task budgets are delegated to reporting staff who are
more directly involved with the work. When a budget is authorized by the project manager,
the budget holder is free to spend the sums on the tasks for which he or she has responsibility
and are contained in the approved project plan. With a properly controlled project, it should
not be possible to start spending on activities that have not been authorized.

CHAPTER

6 Data structures and Reportiy

Relationships

The previous chapters have dealt with the information and calculation aspects of earned
value methods without reference to the organization structure in which they must operate
and the other information systems with which they must interact. To understand how
eamed value methods were intended to operate within a company we need to return to
the original US Department of Defense's CostISchedule Control Systems Criteria (CISCSC)
specification, what it had to say has generated a legacy that is still with us.

The fundamental data structure

Besides the work breakdown structure (WBS) discussed in Chapter 4, the implementation
guide to the DoD's CostISchedule Control Systems Criteria (CISCSC) procedure went on to
define the 'organization' as a structure that reflects the way the contractor has organized his
people for both responsibility and reporting. When this structure is laid out as a hierarchical
arrangement it is often and quite sensibly referred to as the 'Organization Breakdown
Structure' (OBS) although these words were not actually used in the original document.
Furthermore, the CISCSC implementation guide sets out a fixed relationship between the
OBS and the WBS; where these two structures meet - that is, where a package of identifiable
work can be associated with a responsible part of the organization - a control account is
created (this was formerly known as a 'cost account' and is sometimes still referred to as such,
particularly with some software packages). This relationship is shown in Figure 6.1; the array
of control accounts is called the Responsibility Assignment Matrix (RAM). This &xed
relationship was fundamental to the way in which the CISCSC procedure was intended to
work, and is mandatory when some specialist eamed value software packages are used, again
reflecting how some aspects of earned value have become dominated by this initial US
approach.
The significance of the control account as the principal means of cost collection has not
really been appreciated in the UK and the term is little used; much more stress is laid on the
WBS for cost collection. As for the project schedule, this was normally contained within a
network but, due to the failure of PERTICost, the DoD did not insist on the use of networks
for project planning and some contractors avoided using them. However, given the
abundance of cheap and effective project management software, that approach is now
antiquated, as all contemporary software makes use of a network as the principal data model
for the project even though a Gantt chart may be the preferred form of display.

Data S t r u c t u r e s and Reporting Relationships

Where the lowest
level of the WBS meets
the lowest level of the OBS
a Control Account is created

61

+-

Organisation Breakdown Structure

Senior

Middle

Work Breakdown Structure

Sub
Assemblv

Junior

Organization structure for C/SCSC reportingaccording to the US Department of Defense

in DODI 7000.2.

Figure 6.1

Whereas the relationship shown in Figure 6.1 is easy to comprehend at a high level and thus
reasonably easy to construct, things can get a lot more complicated when the schedule or
project plan must be included in the data structure. The US DoD sidestepped this issue,
possibly because it recognized the difficulties, and simply left it to individual contractors to
find a way through these complications.

Increased complexity
It must be recognized that it is a major complication for any project control system to use
earned value methods compared to a reporting system that does not use them. The reason for
this is that to use earned value methods, both the timing of events and the degree of progress
must be input into the project data system in a coordinated manner. However, this precise
coordination is not an absolute requirement for a project reporting system. For example, it is
quite simple to devise an elementary work breakdown structure, perhaps dividing the project
into a few major work blocks - admin, design, manufacturing, etc. - and asking all staff to
book their time and record their purchases against these blocks. At any reporting point, costs
can be accumulated for each block so that actual cost-to-date against total budget can be
seen, but there is no measure of value created. Likewise, it may be relatively simple to draw a
project plan in Gantt chart or network form and use it both to indicate the work to be done
and mark it up with progress as it is observed. This is an uncoordinated arrangement as the
plan and the WBS need not bear a rigid relationship to one another other than that they both
relate to the same project - this is the simplest system of all.
With the uncoordinated arrangement, staff in, for example, the manufacturing area may
book their time to the manufacturing work block for the given project but what specific work

62

Using Earned Value

they have been doing is not recorded. For small, short-duration or relatively simple projects
this may be all that would be needed; the general overview given by the cost report and the
marked-up schedule should give the project manager the visibility necessary for high-level
control. Day-to-day direction, on the other hand, requires much closer control than would
be possible through the time-booking and accounting system which, if it operates on a
monthly cycle, might not produce data until six weeks after some of the work is done.
Despite the relative simplicity of this reporting arrangement, it need not imply a lack of
overall control as formalized change requests and approved budget revisions can still be
applied to ensure that unapproved work does not get into the plan and approved changes are
formally recognized. Many organizations operate in this way satisfactorily; until the advent
of earned value methods the uncoordinated approach was the norm.
When companies first began applying earned value methods, little thought was given to
the relationship between the plan and the WBS and, in consequence, the technique proved
awkward to apply; the reason is given in Figure 6.2. In theory, each control account should

WBS

The theory: each control account translates into a work package plan.
OBS

The reality: the work package plans may be intimately interlinked.

Figure 6.2 Simple expansion of the Control Accounts in the RAM can lead to a complex relationship
between the basic data structure and the project plan

D a t a S t r u c t u r e s a n d ~ e ~ o r t Ri enl a~t i o n s h i p s

63

translate into a major work package which can then be planned for its detailed tasks. As one
aspect of the control account is the organization structure, this tended to generate major
work packages that related to skills, jobs or departments. However, in many tasks, people
from different departments were expected to work together or in a strict sequence so the plan
actually contained a complex series of links between each of the major work packages. This
underlying complexity is not apparent in the basic WBSIOBSIRAM structure. This, in itself, is
not a problem until things start to change; when this occurs changes in one area can
permeate through the plan causing changes throughout. This inevitably has implications for
the reporting baseline as it no longer reflects the intention of the project.
This difficulty was ignored by the DoD who simply demanded that industry find a way to
make their imposed idea of a fixed relationship between the WBS, OBS and RAM work in the
context of an earned value reporting system. Difficult though it may have been, this was not
impossible as the project plan was in the hands of the planners whose job it was to monitor the
situation and make an assessment of progress, whilst the costs were in the domain of the
accountants who can normally provide the up-to-date expenditure. In the early days these two
sets of data were not necessarily coordinated and they certainly were not integrated, as no
suitable software existed to do the integration. Things got really difficult when the plan existed
in several different forms; this can happen when some part of the project cannot be
satisfactorily handled by a network and has to be dealt with another way. An example of this
can occur if the project includes a major manufacturing effort that requires items to be made in
varying standards and in large numbers (for example, weapons development projects). In this
case the whole manufacturing plan might be handled by a Manufacturing Resource Planning
(MRP) production control type of system and separated from the network-based plan.
Despite the problems, by some means, progress and value of work done was agreed
between the planners and the accountants and a measure of performance derived. This was
set against the baseline plan and used for reporting to the customer. A consequence of this
was the advent of two plans: a customer, baseline plan for reporting purposes, and a current
or internal plan that reflected the latest thinking and to which people were expected to work.
Despite all the advances in project management methodology, with a project that is beset
with frequent changes and is also the subject of a formalized contractual arrangement between
the parties, it is difficult to avoid ending up with two sets of plans. A look at the earned value
website note-board will show that more than thirty years after EV was implemented this is still
a subject of debate (see <www.acq.osd.mil/pm/>). It probably never will be fully resolved, as
there is a fundamental incompatibility between the notion of reporting against a plan which
reflects the original thinking and asking people to work to the latest instructions which will
inevitably be based on knowledge gained since the original plan was created. Some project
managers have actually tried to run projects for long periods without allowing any changes or
updates to the plan, even when deviations occur. This can work if the project generally
proceeds according to the plan and deviations are minor, but if there are serious deviations,
failure to update the plan with the latest thinking simply leads to confusion and lack of
direction. Other companies have formalized the two-plan arrangement by openly admitting to
a 'customer programme' for reporting to the sponsor while having a 'company programme'
which reflects the current plan to which staff are working. Often the customer programme has
both time and cost contingencies included which are not in the company programme (either
as a kind of incentive to the team or a deliberate management reserve), although it is generally
assumed that these will be consumed in the course of the project. The inevitable double bookkeeping involved in this approach leads to a lot of bureaucracy. Whether such a method is used

64

Using Earned V a l u e

is likely to depend on 1) the attitude of the sponsor regarding the degree to which he demands
reporting against a fixed, immutable baseline plan, and 2) how well perceived at the outset is
the plan and the likely level of changes.

Integrated systems
Although earned value systems are now advocated as a method for general project control at
all levels, it should be remembered that it was originally advocated by an external sponsor,
the DoD, who wished to have both overall project information and detail where it was
needed about general progress in terms of cost and schedule. In short, the sponsor wanted a
clear and unequivocal view of exactly what the overall cost and schedule progress was in
terms that could not be disguised by the contractor. The sponsor was not, however,
interested in exercising day-to-day control. But, by imposing a series of operating criteria
upon the contractor, which were considered to be good management practice, the DoD
hoped to ensure that proper day-to-day control was applied, though not necessarily through
the earned value aspect alone.
The demands of reporting against a baseline plan combined with the lack of suitable
software to integrate both the plan and the costs meant that, in the early days, earned value
reporting was done largely for the benefit of the sponsor; in the case of the DoD, it was a
service for which it was willing to pay the extra costs. Day-to-day control was carried on by
other means but both the project (program) manager and the sponsor were interested in the
earned value reports, as they clearly showed the overall project position and highlighted
areas where things were not going to plan. It must be said, however, that the difficult areas
were often clearly perceived and may well have been openly discussed well before the earned
value reports came out! Nevertheless the overall trend information was useful to both parties.
A principal reason why the DoD imposed earned value methods on its contractors was
for the visibility of cost and schedule progress that EV generated; this may well have been
more important than any improvements in performance that came through the managerial
demands of the C/SCSC. This visibility is very useful when a multitude of defence projects are
proceeding simultaneously with each making different levels of progress. Being able to
forecast the likely progress of a variety of projects over a period of several years ahead is
hugely valuable when it comes to planning future actions and budgeting for cash in future
years; furthermore it can avoid the embarrassment of having to go back to the government
for more funds due to unforeseen situations arising that could, with more foresight, have
been anticipated. For example, if clear slippages are being foreseen for the in-service date of a
new fighter aircraft, this has far more implications than for just the new fighter project alone.
Existing aircraft will have to be maintained in service for a longer period involving additional
and as yet unbudgeted maintenance costs, which could be large if the planes are nearing the
ends of their design lives and supporting programmes such as pilot training might have to be
put back. Knowledge based on reliable forecasts is of great value when it comes to decision
making; decisions such as those associated with retaining older aircraft in service might have
to be taken well in advance of the conclusion of the main project, as you simply cannot
afford to wait until the end to see what happens. With earned value reporting and the
associated forecasting, the DoD could practise a form of 'portfolio management' of its many
projects with greater ease and certainty than before. By having a better view of what is likely
to happen in the longer term across a range of projects, the DoD could make better decisions

Data S t r u c t u r e s a n d Reporting Relationships

65

about what actions to take in the short term about individual projects and others that might
be affected by their progress. The ability to forecast ahead for a group of interacting or related
projects is possibly one of the greatest advantages that comes from using earned value
methods, yet it is an argument that is rarely put forward.
Whether the DoD really believed that by introducing earned value methods it would get
its projects delivered any quicker or cheaper than before is not known by this author,
although it must have been a hope. Most experienced managers know that the majority of
problems on advanced projects are technical in origin - if a project hits a seriobs technical
problem in a vital area, often little can be done except continue working on it at whatever cost
until the problem is cured or the requirement is reduced. The DoD's experience with projects
run under earned value conditions has proved this point as it has been observed that once a
project starts to slip, that initial rate of slippage is never reduced and frequently gets worse.
Since those early days, project management software tools have developed considerably
and the kind of integrated systems that were not around through the 1970s and the early
1980s are now here in abundance. Software specifically designed for both planning, costing
and earned value reporting is available but its arrival throws into focus another problem.
In terms of both cost reporting and project planning, an uncoordinated approach was
commonly used as 1) it was simpler to operate, and 2) integrated software to make use of
coordination was not available. Because of this, cost and schedule reporting could operate in
a somewhat undisciplined way that made life easier for both the planners and the accountants; any difficulties that arose could be resolved by discussion and agreement. The advent of
integrated software means that a more disciplined approach must be adopted as the elements
of discretion and adaptation are absent with mechanized systems.
The simplest and most direct method of updating the project plan is from the data that is
captured by the time-bookings and purchasing systems as they usually supply the most
accurate and up-to-date information about where time and money are being spent but, as has
been said, it is often coded according to the WBS rather than the plan. Some contemporary
software will allow WBS codes to be attached to activities in the project network but they can
only be used as a method of summarizing costs associated with related groups of activities;
they cannot be used as a means of directly updating any activity because several activities can
all have the same WBS code and the software has no way of distinguishing between them.
Earned value methods demand a coordinated approach to both cost and schedule
reporting; using an integrated software package (that is, one that contains both planning and
earned value features) demands that an unambiguous way is found for linking the elements
in the WBS with each task in the plan. Both the WBS and the project plan represent two sets
of structured knowledge but any system that forges a link between two sets of data demands a
discipline of its own that must be recognized and understood. If that discipline is built in at
the outset, then the process of control will be much easier; try to put it in when the project is
underway and things may become awkward. By creating a WBS that allows the plan to be
coded directly from it, a way can be found to structure both the network and the WBS so that
they are precisely related and mutually compatible. The most convenient way of doing this is
to use the WBS coding as the key code for activity numbering in the network. This
fundamental decision will affect the whole project's data structure and has to be taken at the
start. Furthermore, it imposes a discipline of thought upon the project planners that they
may not have been used to previously.
For complete cost control, any WBS must satisfy the criteria that the work packages are
both mutually exclusive and collectively exhaustive; that is to say no work should appear

66

Using Earned Value

twice and all work must be covered. This principle must not be violated and it must apply
equally to the plan. In this respect a matrix type of WBS is particularly useful as it lends itself
to generating all the required packages and can also be used to draw the top-level network
directly from the matrix. Figure 6.3 shows part of a WBS based on a matrix that was used
successfully for managing a development project. As such it should be considered as an
example rather than a prescriptive approach as no two projects are alike and the conditions
on one may not translate conveniently to another.
Responsible Department ----)
Major Tasks

..-5 .-P . ? :

Project Phase

.-m

? - % E =

Chapter within Phase

2
Purchasing

Design

g " o s

, ,$
1

.-=

:.

3
Manufacture
m

.-

U)

5 .g

2
a,
a
3

m
.G

2 G

V)

=
S
a

An 'X' indicates a controllable

work package at the major task level

Figure6.3

Functional work breakdown structure based on a matrix

The most significant feature of Figure 6.3 is that the WBS is expanded by including the
principal tasks that the staff normally perform as well as the main subdivisions of the project
and the responsible sections of the organization. This feature takes the WBS one step further
towards the project plan than is normally the case. The principal tasks that any department
of a company can perform are termed major tasks and they are the lowest level in the WBS. A
major task is thus the point at which the project plan and the WBS meet. To define the major
tasks, a careful look will be needed at what each department actually does as discrete,

Data S t r u c t u r e s and Reporting Relationships

67

identifiable and controllable functions. The arrangement of these major tasks, on the basis of
the order in which they are performed, will be the starting-point for the project plan.
In Figure 6.3, a formatted sheet has been used to create the WBS. Along the top the major
responsible departments have been defined: Design, Drawing Office, Purchasing, etc. Within
that definition, the main functions they perform have been specified; for example, for the
Design department the main functions are:
1 supervising

2 modelling
3 scheming
4 designing
5 support to manufacture and test
6 specification definition
7 defect investigation
8 meetings and liaison.
When this approach has been put into practice, six, seven or eight main functions have been
found perfectly sufficient to define the tasks that a department performs to a level at which
adequate plans can be created.
At the left-hand side of the sheet, the main elements of the work to be undertaken can be
inserted; these are subdivided into two: Phase and Chapter; alternatively, the main work
elements can be subdivided according to the product structure if cost reporting is required in
that format. Inserting a cross in a box in the matrix will indicate the identification of a package
of work that must be performed and one worthy of control in its own right. This arrangement
produces a task-based WBS as the tasks that each department is expected to perform are at its
lowest level. A Work Package Number can be derived directly from the position of the cross in
the matrix. In this example it is a four-digit number made up from the elements:
1st Digit The Project Phase - a major subdivision of work within the project as a whole
(for example, Phase 1: prototype design and construction, Phase 2: Customer
demonstrations and full-scale development, and so on)

2nd Digit The Chapter - a major sub-division of work within a project phase (for
example Chapter 3, construction of a prototype in Phase 1).
3rd Digit The Responsible Department Code - the area of the company in which the
particular skill is to be found (for example, Responsible Department 2 is the Drawing
Office). Work Area Code 9 is reserved for bought-out materials and other items of
expenditure that do not consume labour but incur costs.
4th Digit The Major Task Code - the particular activity that is required to be done (e.g.
Major Task 5 in work area 1 is Design support to manufacture and test). Major tasks 9 and 0
are left free in all cases and can be used to specify any special tasks that are not easily
defined by the pre-designated numbers. There is also a work area code 0 which has free
fields that can be used to specify any other miscellaneous activities that do not easily fall
into any of the other categories.

With this arrangement, a maximum of ten subdivisions can be made within the chapter,

68

Using Earned Value

phase, department or major task. If more are required either the number of digits must be
increased or a letter can be substituted, which will allow up to 26 subdivisions.
Once the WBS has been established, all the work packages in the project will have been
identified. The project planning engineer, in conjunction with the project manager, must use
judgement in selecting an item to be the subject of a work package in its own right. For
example, a major task in the design area may be 'meetings and liaison'; in a large project with
regular liaison meetings with the sponsor or other contractors, this may be a significant item of
both time and cost and worthy of special control. In a smaller project, even though there will
be meetings, these may be considered as part of the normal design process and not singled out.
The greater the number of identified work packages, the greater will be the detail contained in
the plan and the subsequent reporting, but this must be set against the extra complexity and
the need to ensure that times and costs are correctly booked to a multiplicity of task codes.
A basic network should be constructed using the work packages or major tasks to lay out
the main logic and establish overall time-scales. The major tasks themselves will be too large
and general for both precise logic definition and project control, hence they must be broken
down into minor tasks or activities. These activities will specify a definable amount of work
to the same level as would be expected in a normal network but they must all emanate from
and be contained within their own work package; activities which cross work package
boundaries are not allowed. The activities are numbered by adding an extra digit or digits
onto the work package number from which they arise. Some planning engineers may see this
approach as robbing them of some of the freedom of thought that they had previously
enjoyed, and resistance to using it may be encountered. However, this resistance must be
overcome; the point to make is that planners should no longer think of planning for
activities alone but planning for the total project control process, which includes the costing
and the reporting systems.
The major advantage of this approach is that just one number is used throughout; there is
never any need to transpose data from one coding system (the WBS) to another (the network
codes) in order to update the network with progress or generate the costs associated with work
packages. It means the network can be updated with progress directly from the data generated
by the time-booking and accounting systems as the codes in both systems are identical. There
is no need to employ 'data capture sheets' which require staff to record their working hours
against both the cost code (work package number) and the activity number. In practice, when
data capture sheets are used either staff don't bother to fill them in or when they do, half the
time they are filled in incorrectly as the precise matching that is needed between the cost codes
and the activity numbers is not achieved. The benefits of using a single code number
throughout cannot be overstressed; an arrangement that involves independently maintaining
two sets of data in parallel whilst keeping a strict relationship between the two but using
different codes can be one of the biggest sources of errors, confusion and wasted effort,
particularly if the project is large and the plan is subject to frequent changes.
Figure 6.4 shows a sample of the basic logic in precedence form drawn from the matrix of
Figure 6.3, while Figure 6.5 shows the basic logic's expansion to form the detailed network.
Figure 6.5 gives further information as the activity boxes contain details of the grade of
labour to be used along with the estimated hours and the activity duration in days.
In the example in Figure 6.5 a standard system of labour coding has been used; this is
recommended when you wish to plan company-wide across a range of projects all calling on
the same resources. In Figure 6.5, the labour coding system, made up of two letters and a
digit, is one that was actually used on aerospace development projects. The two letters

Data S t r u c t u r e s a n d Reporting Relationships

I Desian

sup&ision

69

I
H

WPN 1311

Prototype
layouts

-4

WPN

prototype
detail design

Prototype
purchase
estimating

WPN 1314

Prototype
modellingQ
wpp1

iQi

4 prototype

layouts

Prototype
manufacture
planning

detail drawing
WPN 1323

WPN 1322

WPN 1342

Basic task logic is derived from the work breakdown structure

Figure 6.4

Act. No. 131101

Design,
superv~sion
Lab.
TMI

Act. No. 131301

Prototype body
layout
Lab.
TE3

Hrs
60

Dur'n
10

**

Hrs Duration
1000 500

*m-Act. No. 131101

layout

Activity number

Act. No. 131302

Release mechanism
layout

Act. No. 131401

Holdback unit
deta~ldesign *

Activity title

Lab.
TE4

Hrs
120

Dur'n
20

Lab.
TE5

Hrs
40

Dur'n
8

Labour category Estimated labour Duation in working

hours
days

Detailed task logic is derived from the basic logic by breaking down the major tasks into
minor tasks or activities

Figure 6.5

indicate the job. The first letter indicates the discipline - T = Technical (Design Engineering),
D = Draughting etc.; the second letter indicates the role within that discipline - M = Manager,
E = Engineer, etc. The digit at the end indicates the hourly charge rate for costing purposes.
Eight standard rates were used into which all chargeable staff were allocated, rate 1 being the
highest hourly rate and rate 8 being the lowest.
Thus activity No. 131101 (Design Supervision of the Prototype) requires 1000 hours of
technical management (TM) time over a 500-day period at charge rate 1. With the project
plan expressed in this form, it can be loaded into a project management software system.

70

Using Earned Value

The arrangement described above was used successfully; the use of a single coding
system for both the WBS and the network plan resulted in remarkably trouble-free operation
when compared to the problems that have been encountered when two different coding
systems have been in use simultaneously. Figure 2.1 in Chapter 2 is taken from a project run
under this arrangement.

SYSTEMS INVOLVING MULTIPLE CODES

Despite the clear advantages of using a single code as the basis of both the work breakdown
structure and the planning there are occasions where using two codes cannot be avoided.
This can happen where a company providing products or services to a larger project has a
WBS imposed on it by its customer who demands reporting on cost and progress in a
prescribed format. From the point of view of a prime contractor or a customer organization,
using a common WBS and reporting format across all supplying contractors makes good
sense, particularly when it comes to integrating data from a variety of sources, but it can
create fundamental problems for the supplying contractors. The customer organization may
demand a product-based WBS for valid reasons of its own, ignoring the practical advantages
of a task-based WBS at the supplier level; however, a more problematic issue could be the
number format.
A customer could devise a work breakdown coding that is eight characters long
including both letters and numbers, but what happens if the cost-collection and timerecording system in the supplier company can only accommodate six characters? Clearly an
issue like this could go to the heart of the company's accounting system and could be
fundamental to the design of all its standard paperwork (for example, time-sheets, invoices,
purchase requisitions, and so on) and its data input formats; it could be very difficult to
change. The answer lies in the appropriate choice of project management software. Many
packages allow multiple work breakdo- structures and here one WBS can be used for cost
and time collection and thus can be made compatible with the existing company systems
while another can be used for reporting to the customer. In this case the WBS code for
reporting to the customer acts only as a sort-code which allows data to be reorganized in the
required way but plays no part in the data-gathering process. In this case many activities can
carry the same customer WBS code if that is required. Furthermore, should there be some
errors in the allocation of customer WBS codes to activities, they will not fundamentally
affect the operation of the reporting system even if they do produce some small errors in the
reported position. The system is summarized in Figure 6.6.

Organizational aspects
Figure 6.7 shows an example of the data structure and organizational relationships plus the
principal data flows necessary for an earned value performance measurement system to work.
In this case the WBS forms the bridge between the organization structure that defines who is
responsible for each task and the network which holds the plan of what is to be done and
when. A suitable project management software package or suite of packages is essential to
perform all the necessary calculations. Besides the normal start and finish dates, criticalities,
etc., it will generate:

Data Structures and Reportinq Relationships

71

PROJECT SYSTEM

Satisfying both customer and company requirements may need a software package that
can handle multiple WBS codes, one serving the company systems requiremens while sorting processed
data to suit the customer

Figure 6.6

Responsible section managers

Progress

Cost structure

Network/plan

Proiect manageement software package

Figure 6.7

Work breakdown structure

I/
I

Organization and information relationships for earned value performance measurement

72

Using E a r n e d Value

the costed schedules of work which form both the budgets and the instructions regarding
what each section manager is expected to perform, and
the performance reports.
The budgets must be agreed with each section manager before the system is put into
operation as no manager is going to be held responsible for budgets to which he has not
agreed. Budgets and targets can be set but they have no meaning unless those with authority
take responsibility for achievement. Section managers who are sufficiently close to the task
and have enough authority must be appointed to take responsibility for performance against
their budgets. It is these managers who will receive the cost performance reports and will be
expected to take action according to whatever is indicated.

COST BREAKDOWN
In addition to the WBS and the OBS another term, the 'Cost Breakdown Structure' (CBS),
has come into use but this was not a feature of the original DoD specification and the term is
not officially recognized in the USA. The CBS has been defined as 'a system for subdividing a
project into a) hardware elements and sub-elements, b) functions and sub-functions and c)
cost categories to provide more effective management and control of the project'.' This is a
somewhat ambiguous definition as these divisions might seem either mutually exclusive or
overlapping and it leaves it open to the individual as to what this structure is and how it may
be used; consequently there are a number of possible interpretations. One obvious and useful
interpretation is to attach the values of all the cost accounts to their respective WBS elements
and this will result in a costed work breakdown structure. Whether or not this can be deemed
to be a cost breakdown structure in its own right is open to debate but it does fulfil the
requirement of part a) of the above definition and it does have the merit of being directly
related to another fundamental data structure. Another possibility is to take all the work
packages and subdivide them according to cost categories such as material, labour and
overheads, then recombine all the individual costs into the overall cost categories so that one
can see total labour, total materials, etc. Whether this can be deemed to be a cost breakdown
structure is equally open to debate as it is not a structure in its own right but something that
results from other data sets; what importance is attached to such information and what use is
made of it rather depends on the project. However, in the context of integrated project
planning and control, a more relevant structure might be the cost structure of the
organization that is performing the project.
Computerized systems that are capable of planning projects can include resource
requirements. When these are allied to the appropriate charge rates, it is possible to put costs
on all the activities in the project. A fully costed project is fundamental to earned value
performance measurement and the application of the process relies on a computerized
system. Thus, from an operational standpoint, the most appropriate cost breakdown
structure is that associated with the charge-rates and overheads of the organization
performing the project. It is this interpretation that is shown in Figure 6.7.

Notes
1.

Flemming, Q. W. (1988) Cost/Schedule Control Systems Criteria - The Management Guide to C/SCSC,
Chicago: Probus Publishing Co., p. 506.

CHAPTER

7 Managing With Earned Value

Data

The use of earned value measurement techniques can produce data that gives a valuable
insight into the workings of the project, but will be of little value unless project managers act
upon the information in a way that preserves the project's objectives. However, what earned
value can reveal may also be highly sensitive if it shows a worsening situation, and here the
project manager needs to use a degee of discretion

Measuring project progress

In Chapter 5, Figure 5.1 shows the principal elements in the project cycle and it will be
realized that using earned value methods has a significant effect on the approach to
planning. EV's other major impact is the contribution it makes to the control process. Once
the earned value performance measurement system is in operation, each responsible section
manager will be required to participate in the control process through: 1) the work that they
authorize, 2) their assessment of the amount of progress, and 3) the controlling actions they
take. Generating an accurate value for the earned value or BCWP has often been considered
as one of the more contentious parts of the whole earned value management process. The
generally accepted criterion is that earned value reporting should be both objective and
verifiable but sometimes a subjective or an arbitrary element in this must be accepted.
However, as the project progresses more activities are completed; the BCWP for a completed
activity becomes the Original Budget (OB) and doubt about its value is removed. The greater
the numbers of activities that are completed, the less the potential error within the overall
project BCWP, as that figure contains a continuously reducing proportion of activities in
progress; in effect, progress with the project tends to reduce the errors in the estimate of the
overall BCWP. A number of methods for estimating the BCWP are possible; each has its
drawbacks, but all are worthy of consideration for different circumstances. The methods can
be broadly divided into four categories: subjective assessments, objective measured
assessments, rules of thumb and indirect assessment.

SUBJECTIVEASSESSMENTS
These methods rely on the experienced judgement of individuals close to the work. Examples
of this approach are:
a) At intervals that relate to the reporting calendar, either weekly, fortnightly or once a
month, the section heads are asked to estimate the percentage completion of each of the
activities, for which they are responsible, that are currently in progress.

74

Using Earned Value

b) At intervals that relate to the reporting calendar, the responsible section heads are asked
to estimate the amount of time they expect to elapse before each of their current activities
is complete; that is, they are asked to predict the remaining duration.
Providing a degree of objective honesty is applied, both these methods can work well and are
suitable for activities that span perhaps three to five reporting intervals. They tend to be used
where there is no directly measurable unit of output, but there are clear indicators of progress.
For example, the output of test reports from a test programme gives a measure of progress but
some aspects of the test programme may be time dependent and the test reports may be
clustered towards the end.
With method a), the BCWP at the reporting point for any activity is given by the
formula:
BCWP = OB x % complete
where OB is the original budget for the activity
The sum of the BCWP values for all open and completed activities is the overall BCWP for the
project at the reporting date. This method was used in the example given in Chapter 3 (see
p. 27).
With method b), using the predicted remaining duration, one needs to assume a
uniform rate of spend in each activity. This may not be a valid assumption in all cases but it is
reasonably accurate if the activities are small enough. The BCWP at the reporting point for
any activity is given by the formula:
BCWP =

OD x BCWS
ATE + predicted remaining duration

where OD = Original duration

ATE = Actual time elapsed
The predicted duration is found by adding the expected remaining duration to the time
elapsed since the activity first started; the BCWS is found from the current cost schedule as
generated by the computer or by a simple calculation based on a uniform spend rate. Should
the original duration be exceeded, the actual time elapsed is substituted for the OD in the
above formula to become:

BCWP =

ATE x BCWS
ATE + predicted remaining duration

If one was to ask the section heads to provide both figures (and there could be good reasons
for doing so), you can arrive at two different estimates of the BCWP. For example, a task is
set to last seven weeks with a budget of 7000; if at week 4 the section head estimates the task
to be 60 per cent complete and will require another four weeks to complete, what is the
BCWP?

M a n a g i n g With Earned Value Data

75

By method a) the BCWP is 7000 x 0.60 = 4200

By method b) the BCWP is

4000
= 3500
4+4

It is up to the individual which of these figures is more believable and there is nothing to say
that the supervisor is wrong in his assessment of the situation; method b implies a uniform
rate of spend and that may not be the case. Providing that the responsible managers can be
trusted to make a reasoned assessment, method a can work well. It might seem easy to
hoodwink the system, especially in the early stages of each activity, but as time passes and
the assessed progress is not, in fact, being made, it soon begins to show and the manager
cannot disguise it.

OBSERVED MEASURED ASSESSMENTS

These methods rely on some form of scoring based on achieving defined milestones or
delivering set quantities. Examples include:
c) At each reporting point, an amount of earned value is credited to each current activity
according to an agreed method of scoring based on a milestone or an achievement target.
d) At each reporting point, an amount of earned value is credited to each current activity
according to the number of work units completed or delivered.
Method c is useful where there is some easily measurable unit that indicates progress or
where there are clearly visible milestones that will show progress when they are achieved. For
example, in the drawing office, there is a package of work that covers the output of a set of
drawings for a certain item. Inspection of the scope of work and discussions with the designer
have resulted in the section leader in the drawing office making an estimate of 80 drawings in
the final set and a budget of 100000 for the work. Because of the way the drawings are
created, some drawings may be started but not completed until later drawings are done,
when some minor details can be finalized. Drawings may not be issued until checked, which
may need to be done in batches of related drawings; this can make the rate of issue of
drawings somewhat 'back end heavy' as it may be well after the mid-point of the planned
duration before the first completed drawings emerge, even though more than half the work
may be done. In a case like this, the idea of a uniform rate of output indicating the rate of
progress just cannot be applied; here the rule for assessing the BCWP might be: credit 10%
(10 000) of budgeted cost when the activity starts, give another 50% (50000) when 25%
(20) drawings have been issued, give another 25% (25 000) when 75% (60) drawings have
been issued and give the final 15% (15 000) when all the work is complete. This pattern
reflects the way that output does not indicate work done until nearly the end of the task; with
the formula given above, 60% of the task value will be earned when 25% of the required
items have been delivered. An arrangement like this is likely to give an accurate assessment of
the earned value at the various stages, but there is a drawback. There may be no general rule
that can be applied - each case must be treated individually and a unique formula devised.
Furthermore, that formula then needs to be incorporated into the earned value process and
correctly applied as the project progresses. Software projects have attracted special attention;
metrics involving the number of completed source lines of code (SLOC1s)have been used as

76

Using Earned Value

well as a reverse measurement based on the reported numbers of defects per 1000 SLOC's
(this value falls in line with a commonly observed pattern as the task approaches completion). The various output achievement targets can be considered as 'milestones' and the
same general comments apply to any milestone-based measurement of work accomplished.
The drawing office example shows'a situation where work and output are not simply
related because of the creative, interactive front-end activities that inhibit initial output, but
there are other cases where this is not so, here earned value can be related directly to
measured output. Method d can be used and examples could be: 1) any repetitive production
process, provided the initial set-up is treated as a separate activity, and 2) construction
activities such as laying bricks or pouring concrete. It only requires a suitable cost value to be
established for the measured unit of output, for example, /finished item, /1000 bricks laid,
/100m3 of concrete poured, to count the output and compute the earned value. As with
method c, a separate unit of value must be computed for each case but there is none of the
complication of a special formula.

RULES OF THUMB
These are simple rules for crediting earned value that make no particular pretence to detail
accuracy, but, bearing in mind that errors in the combined overall earned value tend to
reduce as the project progresses, they are adequate for control. Examples are:
e) Credit 20% of the total activity budget, OBI as the BCWP when the activity is recorded as
having started, credit the remaining 80% when it is recorded as complete.
f) Credit 50% of the total activity budget, OBI as the BCWP when the activity is recorded as

having started, credit the remaining 50% when it is recorded as complete.

Simple rules such as this are particularly appropriate where activities are of a duration that is
typically between one and three reporting periods long and there are many simultaneous
activities. The advantage of this approach is that it can be computerized so the earned values
can be calculated directly from progress information, regarding activities started and
finished, that is entered into the planning system as part of the reporting cycle. If planned
activities can be kept relatively short in duration there may be little to be gained by being any
more sophisticated.

INDIRECT ASSESSMENT
These methods apply when the earned value is determined by some outside factor. Two cases
typically arise: 1) when the activity is time dependent, or 2) when earned value is related to
the progress of some other activities. The first case is typical of an ongoing service to the
project as a whole, the most obvious example being managing the project: as long as the
project continues the management team has to be in place. The second case is typical of a
discontinuous service to some particular aspect of the project such as inspection. For
example, inspectors are only required when there is something to inspect, usually when
there is some aspect of a deliverable system that is 1) nearing completion, 2) being
incorporated into something larger, or 3) ready to deliver; when this occurs is entirely
dependent on progress with the deliverable system. Methods used in these cases are:

M a n a g i n g With Earned Value Data

77

g) Calculate the earned value based on the elapsed time at some predetermined rate of
accrual based on a planned level of effort.
h) Calculate the earned value by apportioning an amount from an overall budget based on
progress with the controlling activity.
Method g is known as the 'level of effort' approach and is applied to general service activities.
If costs are uniformly spread over the activity duration, earned value is credited according to
elapsed time; in this case the BCWP is always equal to the BCWS and there is no schedule
variance. Actual expenditure may not be uniform and could also be more or less than
planned so there can be a cost variance. Levels of effort can be planned to take into account
known changes in effort; in that case, expenditure will not be uniform.
Method h is often called the 'apportioned effort' approach. Quite simply, an amount of
value from the overall budget is directly credited to the activity based on a fixed relationship
with the value earned on a controlling activity to which it is directly related.
It will be clear that there is no single and generally applicable method of assessing the earned
value of any activity in a project, particularly if it is a complex project involving a mix of
different skills and tasks. Project managers charged with running projects under earned value
conditions need to choose the most appropriate methods based on:
the required degree of accuracy
the nature of the particular activity
the practicality and cost of detailed measurement
the software in use.
It is likely that a mixture of methods will be used in any real project situation.

Dealing with materials and purchases

When earned value was first implemented in the USA, accounting for materials and
purchases proved to be a troublesome aspect (the background to this has already been given
in Chapter 2). Much of the difficulty stemmed from an unnatural and largely unnecessary
requirement to separate materials out and continuously account for them when they are
really part of a bulk quantity that could be used on a variety of projects and might only be
allocated to a particular project at the point of issue. Putting aside that issue, the diverse
nature of purchased items can mean that accounting for them can still be a problem that
might need a variety of solutions.
Externally purchased items can be divided into a number of broad categories for which
different methods of performance assessment might be applicable. Some suggestions, according to the type of purchase involved, are listed below:
Externally purchased services -This can include anything that is subcontracted out that
is not the subject of an earned value reporting arrangement in its own right. Services might
be purchased on an item-by-item basis with budget and time-scale for each item, they
might be provided on a continuous basis and thus accounted for on a level-of-effort basis,

78

Using E a r n e d Value

or they might be provided on an as-and-when basis as part of some general contract when
an estimated percentage of total value might be appropriate. There is no single approach to
assessing the BCWP that can be recommended as each of these is a different case so any of
the above methods might be suitable. Actual costs for work performed are usually credited
on receipt or payment of the invoices depending on the policy in force.
Externally purchased items of equipment and/or tools - Providing the items are bought
specifically for the project, the BCWP values are credited, according to the original price
estimates, at the time of delivery or acceptance. ACWP values are credited on receipt of
payment of the invoice. However, there may be more than one invoice because there
might be a down payment with the order or a series of stage payments, particularly if
the equipment is large or expensive. If the stage payment is significant and tied to an
observable milestone, the relevant percentage of the total purchase value could be credited
to the BCWP at the time of payment.
Externally hired facilities -These are similar to services but could include such things as
test houses which are hired rather than purchased. These are normally paid for on the basis
of the time they are used; BCWP values can be credited on the basis of the percentage of
the planned period of hire. The ACWP can be credited on receipt or payment of the
invoices.
Externally purchased, project-specific materials - Single delivery items can be treated in
the same way as equipment, this can apply to such things as special forgings, castings or
mouldings. With bulk materials such as building sand, that might be delivered over an
extended period or supplied on a call-off basis, the BCWP can be credited as a percentage of
the total order quantity and the AWCP credited on receipt of the invoices.
Externally purchased, non-project-specific materials - These can be among the most
difficult of items to assess. They can include a host of things such as general stock material
(sheet, bar, plate, etc.), paints, finishes, sealants, adhesives and fasteners. With construction projects, these items are less common as most materials are specially bought for
the project in quantities specified by the quantity surveyor from the architect's drawings.
With development engineering projects, this can be more difficult; at the start of the
project, there may be no detail design and, at the early stage, the designers might have a
wide choice of materials from which they can choose to fabricate the parts. Which
materials and processes are chosen only emerge as the design evolves.
It is difficult to give a firm recommendation for estimating the BCWP as the situations
can be diverse; possibly the most useful might be to use an estimated per cent complete
figure for the total manufacturing work package as provided by the manufacturing
supervisor, or an apportioned effort approach might be applicable if the manufacturing
activities have been planned in sufficient detail.
Other non-project-specific materials might be involved that do not relate to
manufactured aspects but to other parts of the project such as the budget for project
stationery and office materials, if a separate budget is created for this aspect. In this case
BCWP values can be credited on a level-of-effort basis tied to the complete project timescale.
ACWP values can be credited on the basis of issues from stores, providing the stores
accounting system can recognize the work-package, control account or activity numbers
used by the project monitoring and earned value calculation system. This might not be a

M a n a g i n g With Earned Value Data

79

simple matter as manufacturing may be accounted for in a different way from project staff,
particularly if the company does routine production work in parallel with development
manufacturing using the same facilities.
Experience in the USA has shown that regardless of the type of earned value management
used for material, it is important to try to match the BCWP and ACWP values within the
same reporting period. Otherwise, reporting the BCWP in one month and the actual costs in
another month could result in misleading cost variances. In order to achieve this match-up
of earned value and actual costs, some companies have chosen to use 'estimated actuals'
against the earned value.

Performance measurement
The simplest indicators of project cost and schedule performance are the Cost Variance (CV)
and the Schedule Variance (SV). These are elementary measures of the difference between
planned and actual spend at a reporting point. They can be recorded at each reporting point
and plotted as a time series as the project progresses; an example is shown in Figure 7.1. Cost
and schedule variances can be recorded for any work package, control account or the project
as a whole; sometimes warning and action limits can be placed on these figures indicating
that some action should be taken to contain spending. Variances can be calculated on the
values for the individual periods or on cumulative costs from the start of the project; the
latter figure is probably more useful.
Cumulative f k

Variance f k

BCWS BCWP ACWP

CV

SV

24.0

19.0

21.3

-2.3

-5.0

57.5

55.3

57.9

-2.6

-2.2

108.0 102.4

122.5

-20.1

-5.6

147.0 135.7

145.5

10
0

-10

-9.8 -1 1.3
-20

201.5 196.2 214.5

-18.3

-5.3
Action

265.0

-30

Figure 7.1 Plotting the Cost and Schedule Variances against time can show a trend, if there is one.
Action and warning limits are also shown, the fact that the cost variance passed the warning limit at
Period 3 may account for the jump in performance in Period 4, but the more likely explanation is that
actual cost bookings have been running ahead of credited earned value in Period 3, possibly because
something has been paid for in advance of delivery (for example a down-payment for tooling associated
with the manufacture of a major item) which is then corrected in Period 4

The variances can be useful for measuring the absolute cost differences at a reporting point,
but they are not the best indicators of overall performance. The two most useful measures of
project performance obtainable from the earned value method are the Cost Performance
Index (CPI) and the Schedule Performance Index (SPI). Useful though they are, they say

Using Earned Value

80

nothing about that other equally, if not more, important aspect of project performance technical performance; this aspect should never be forgotten when looking at any earned
value measurements. CPI and SPI values greater than one indicate performance either in cost
or schedule terms that is better than planned, values lower than one indicate a worse
position. The CPI is, perhaps, the more useful of the two, it shows the real worth that is being
created by the project, thus a CPI value of 0.85 indicates that for every pound spent, only 85
pence worth of value is being created on the basis of the original budget. The behaviour of
these two indices can be plotted as the project proceeds; they give a good indication of the
real progress and what the future may hold. Figure 7.2 shows a typical situation in
terms of the cumulative ACWP, BCWP and BCWS; the indices have been derived and are
plotted. Using the indices gives a clearer view of the position than reading the relative
ACWP, BCWP and BCWS values. Figure 7.3 shows the pattern of the CPI and SPI under
various conditions.
Cumulative costs f x 10 000

Cumulative

MONTH ACWP BCWP BCWS

1

.75

.75

CPI

SPI

.75

1.6

1.4

1.6

,875

,875

2.8

2.4

2.7

,857

.888

4.7

4.3

5.1

,914

.843

7.4

6.6

7.8

,892

,846

10.1

8.9

10.9

.881

316

12.8

11.1

13.6

,867

,816

14.4

12.7

15.4

,882

.825

Performance indices
1.3

0
1

Months from start

1.2

rformance above plan

1.1

nths from start

1 .o
0.9
0.8

erformance below plan

0.7

Relationship between BCWS, BCWP, ACWP and the cost and schedule performance
indices. After the fluctuations in the CPI and SPI during the first 4 periods, notice how they settle down
to steadier values; this is typical due to the aggregating effect of past data contained within each
successive calculation

Figure 7.2

82

Using E a r n e d Value

The behaviour of the performance indices

Figure 7.5 shows a project in which the original planned BCWS values are given in the top
bar against each task number, beneath that is given the actual spread of the BCWP and the
ACWP values as they appeared at the time. This project had an initial duration of seven
periods but the result was that it actually took ten periods. The cumulative total BCWS,
BCWP and ACWP values are given as well as the CPI and SPI values that result at each
period's end. The graph at the bottom shows how these values vary. It will be noticed that
there are considerable swings in the values of both these indices in the first few reporting
periods but as more tasks are completed, and their results are fixed, an aggregating effect
occurs and the CPI value settles down to a more accurate and stable level. The behaviour of
the SPI has already been discussed in Chapter 3, it too tends to settle but once the original
duration (in this example, seven periods) is exceeded it tends towards a value of one,
irrespective of the actual schedule performance. The aggregating effect of completed
activities means that the further the project progresses towards its conclusion, the more the
SPI tends to become a measure of percentage completion rather than a measure of schedule
progress. It would have given a better indication of schedule progress for longer if, at some
point, the project had been replanned on the basis of an expected longer duration.
If these early swings show high adverse values they can have an alarming effect on staff
morale, especially if people are inclined to see them as a direct reflection on their work. Some
organizations have thus made it a policy not to openly report CPI and SPI values before three
or four reporting periods have passed, or in other cases they have applied an exponential
smoothing1 process to the values in order to damp out variations.
Experience in the USA over the last thirty years of running projects under earned value
conditions has revealed some interesting conclusions. In particular, the CPI has consistently
shown itself to be a good predictor of overall cost performance. It has been noticed that its
value tends to stabilize by about the 20 per cent point in the overall project duration; beyond
that its value tends not to vary by more than 10 per cent until the project end. In the majority
of cases, variation will be towards a worsening situation rather than an improvement. Part of
the explanation for this observation lies in the mathematics of the aggregating effect of the
completed activities as demonstrated in the example above. Just as importantly, however, it
tends to confirm the well perceived fact that trends in overall performance, once they are
established, tend to continue until the end of a project. It might indicate that management's
ability to significantly influence matters, once project momentum builds up, is rather more
limited than some would wish.

Reporting for earned value measurement

There can be little doubt that obtaining accurate reporting data is one of the most problematic
aspects of earned value management and the difficulties often come down to aspects of the
internal systems and accounting practices of the firms engaged on the project. By and large,
earned value methods were successful in the USA because the companies using EV tended to be
large organizations that had regular dealings with the US government, knew the imposed
requirements of working on government contracts and were thus prepared to make the
investment in appropriate systems, procedures and software. This was not simply a one-way
process: the data was also useful for project management and the accounts department.

M a n a a i n a With Earned Value Data

Cum.ACWP

15

50

CPI

0.667

1.0

SPI

0.50

110

185

270

365

425

460

480

83

485

0.863 0.784 0.833 0.835 0.812 0.826 0.823 0.824

0.714 0.730 0.659 0.725 0.824 0.863 0.950 0.988

1.00

mber

The behaviour of the performance indices - these can show initial swings that tend to be
damped out by the passage of time and project progress. Note the SPI always has a value of 1 .Oat the
project's end irrespective of actual schedule peformance

Figure 7.5

84

Using Earned Value

Organizations that undertake significant government contracts tend to have good time
recording, purchase accounting, bookkeeping and costing systems in place as part of their
normal operating procedures, because the organizations are largely static and have a mostly
permanent staff base. This situation is not so true of organizations that are more transient in
nature which come together only for the purpose of the project and make use of large
numbers of contractors on a more casual basis, possibly on sites remote from the main
premises. This can be typical of the construction industry where staff working on the project
may not be required to fill in time-sheets, progress is assessed using quantity surveying
methods and invoicing is based on material purchase accounts and the numbers on the
payroll. Whereas it is possible to assess overall progress and establish the overall cost at any
time, it may be very difficult to perform any lower level of earned value analysis; it may be
impossible to get any detailed correlation between BCWP and ACWP at anything other than
the highest level. This should not however imply that such projects are necessarily out of
control, in fact very detailed day-to-day control can be exercised through the site supervisors
and there are plenty of examples of construction projects that are completed on time and
within budget. Part of that success may, however, come from the practice of awarding
contracts for work on the basis of fixed prices with milestone payments, bonuses for early
completion and using liquidated damages for late delivery. The difference in this approach is
that control is not exercised through the reporting and accounting systems (which is what
earned value is really about); it is done by other means, which need be no less effective.
Reporting systems must be set up to gather the cost and progress data. The time-booking
system will usually prove convenient for gathering labour cost data and the purchasing or
accounts departments will have data on orders, purchases and payments. The relationship to
the master plan must be quite clear and the WBS matrix and activity numbering system will
help. A dedicated software system with the earned value performance measurement features
must be acquired and project management staff must become familiar with its use and accept
the disciplines it brings with it; some changes to company procedures may be necessary.
Work must proceed through a system of task authorizations; unauthorized work must not be
allowed to start. Here the project management software can help as the better-quality
packages allow the generation of customized reports and this feature can be used to create
authorizations at the appropriate time directly from the plan.
The generation of regular performance reports is vital: responsible managers must see
how they are doing at regular and well-defined points. It must be made clear that progress
assessments will be required for each work package and responsible managers must submit
their reports on time. Resistance may be encountered here but knowing that senior
management will be interested in the results can be a great incentive. Project management
staff responsible for operating the system must set up a regular routine for themselves and
must ensure that authorizations go out on time and that managers know when progress
assessments are needed.
Subcontractors can be one of the bigger problems when it comes to timely reporting.
This can be made more difficult if there are imposed reporting deadlines set by the sponsor
who also demands the most up-to-date information and there is a chain involving a prime
contractor and subcontractors. The problem that can arise is that many organizations
operate on a monthly accounting cycle, that is, data is gathered during the month then
processed at the month's end to produce a complete set of project accounts a week or a
fortnight later. This data might then have to be fed into the project management software
that will produce the earned value reports - and a few more days will go by. It could be three

M a n a g i n g With E a r n e d Value D a t a

85

weeks after the month's end before the earned value reports are available; some of the data
contained in it could thus relate to activities undertaken anything up to seven weeks earlier,
clearly too late to be of use in direct control. Subcontractor performance data also needs to be
collected and incorporated in the reports, so even more time is required as they may have to
go through a similar process; by the time the sponsor sees the reports, more than a month
has elapsed.
Some organizations, particularly those in the USA connected with advanced technology
aerospace development projects, have recently adopted a weekly earned value reporting
cycle. It should be remembered that these are also companies which have made large
investments in their reporting and accounting software systems and are in a position to
impose the required disciplines on the project staff. They are claiming considerable success in
maintaining project schedules by this approach; whether this is really due to the effect of
earned value measurements or simply that senior management are taking a more detailed
interest and becoming more actively involved in solving pending problems one cannot say.
There may also be something of an incentive for the project team members through knowing
their performance is being actively monitored. There are certainly advantages in adopting
tight managerial control particularly when deadlines are important, providing it is not seen
as interference or 'micro-management'; how much of this should come through frequent
earned value measuement will depend on the required investment in reporting systems and
the ease with which the necessary disciplines can be imposed. A balance needs to be struck to
ensure that reponsible staff don't get deflected into spending large amounts of time tending
to the information system rather than directing the task in hand.
Things can get more complex if subcontractors work on different accounting cycles or
admit to publishing their accounts one month in arrears. There is no simple solution to this
situation if it arises and it has been a subject of ongoing debate among earned value
practitioners; one solution that has been advocated is to use 'flash' data, that is, information
supplied as the most up-to-date estimate of progress and actual cost at the reporting point.
This is sent to the prime contractor for incorporation to meet the reporting deadline. When
accurate data becomes available, some time later, this is substituted for the flash data in the
performance measurement system. It must be said that this is not an ideal situation as it leads
to doubling the amount of work and is also prone to introducing errors. The answer
ultimately lies in all parties agreeing a realistic reporting cycle time and then deciding on just
how much more worth can really be attached to having data available at an earlier time. The
answer may well be that there is not very much. As has been implied, earned value methods
are not the most appropriate for day-to-day control: direct contact with the work, progress
meetings and action lists are more effective. Cost and schedule variances may jump from
month to month but the overall performance indices will tend to become more stable as the
project nears its end, so up-to-the-minute earned value reporting tends to progressively
reduce in value as a control mechanism, although not from a measurement viewpoint, in the
later stages.
Another problem sometimes encountered is that of the reporting units. It was very clear
from the start of earned value assessment in the USA that all reports would be in cost terms:
dollars in the first case and no allowance was made for any other unit. Cost is a common
denominator when it comes to measuring the worth of the various components of the
project, such as labour, services, materials and purchases. Some organizations, because they
have found difficulty in obtaining accurate or timely information from their accounting
systems, have chosen to use a different unit of measurement, typically the man-week or man-

86

Using Earned Value

hour. Whereas it is mathematically possible, given a suitable time-booking system, to

establish BCWS, BCWP and ACWP values in terms of man-weeks of effort and do the
appropriate calculations, this can only be applied to the direct labour element of the project.
Few projects consist of nothing but labour; in some projects it may be only a third of the cost.
The problem comes in combining values determined in man-weeks with values for
purchased items which can only be given as costs. The answer is clear: if an organization
wishes to use earned value methods for complete project control then it needs to make the
appropriate investment in its data gathering and accounting systems so that all data can be
reduced to costs.

Earned value reports

What reports are produced and the formats used can vary according to the requirements of
the various parties involved in the project. When earned value methods were first applied,
the whole process was embedded in a very prescriptive series of US DoD procedures. These
included a number of precisely defined report formats of which the most significant were the
Cost Performance Report (CPR) and the Cost/Schedule Status Report (CJSSR).The CPR was
designed for complex, high-cost contracts and was defined by five formats: 1) earned value
data by WBS element, 2) earned value data by functional category (design, manufacturing,
test, facilities hire, etc.), 3) changes to the baseline, 4) actual and projected staffing levels, and
5) variance analysis explanations. The C/SSR is a simpler version of the CPR, used for less
complex, less costly contracts below the CPR threshold and only contains two formats earned value data by WBS element and variance analysis explanations. With the C/SSR, the
planned value and earned value can be assessed at any level deemed logical by the project
manager whereas the CPR demanded a more rigorous assessment at the control account and
work package level. The principal data contained in the C/SSR are:
At project level
1 Contractor and contract details
2 Programme details
3 Report period number
4 Original target cost
5 Negotiated contract changes
6 Current target cost (4 + 5)
7 Estimated cost of authorized, unpriced work
8 Current budget base (6 + 7)
At activity level
9 Activity number
10 Cumulative BCWS
11 Cumulative BCWP
12 Cumulative ACWP
13 Current schedule variance (Culmulative BCWP - BCWS)
14 Current cost variance (Cumulative BCWP - ACWP)
15 Budgeted cost at completion
16 Estimate at completion
17 Variance (16-15)

M a n a g i n g With E a r n e d Value D a t a

87

Overall
18 General and administrative costs in categories 9 to 17
19 Overall work package totals in categories 9 to 17.
This produces a suitably comprehensive report on the earned value position of any work
package or activity. Standard report formats were also specified and these have been followed
by the specialist earned value software packages. An example of a CPR document generated
by the Cobra package is shown in Figure 8.3 in Chapter 8. It contains an additional split of
data between the current period and the cumulative BCWS, BCWP, ACWP and associated
variances when compared to the simpler CISSR; it also shows the general format used for
reporting on US government contracts.
Although the CISSR format is a very comprehensive one, it requires the specialist earned
value performance measurement software to produce it for anything other than the simplest
projects. In many cases this will not be necessary, particularly if one is not committed to the
full rigours of a US government approach. Simpler reporting formats are used that are
generally restricted to variance reports and performance reports.

VARIANCE REPORT
At project level
1 Programme details
2 Report period number
3 Work package (control account) number
At activity level
4 Activity number
5 Cumulative BCWS
6 Cumulative BCWP
7 Cumulative ACWP
8 Current schedule variance
9 Current cost variance
10 Budgeted cost at completion
11 Latest revised estimate
Overall
12 Overall work package totals in categories 4 to 11.

PERFORMANCE REPORT
At project level
1 Programme details
2 Report period number
3 Work package (control account) number
At activity level
4 Activity number
5 Cumulative BCWS
6 Cumulative BCWP
7 Cumulative ACWP
8 Current schedule variance

88

Using E a r n e d Value

9 Current cost variance

10 Schedule performance index
11 Cost performance index
Overall
12 Overall work package totals in categories 4 to 11.

An example of a typical variance report is shown in Figure 8.5 in Chapter 8 (p. 107) and a
typical performance report is shown in Figure 2.1 in Chapter 2 (p. 16). Both these reports
were generated using the earned value features of standard project management planning
packages. They do not conform to the requirements of a CPR or CISSR but they are adequate
for normal management control purposes.
Graphical representationsof the performance indices and the variances can be helpful in
visualizing how the project has been performing since its start. Examples of plots of variances
and indices have been given in Figures 7.1, 7.2, 7.3 and 7.5. Brave project managers may even
choose to plot variations in the EAC and the estimated completion date. Figure 7.6 shows
how this can be done.
Estimated completion date

Estimated cost at completion

Estimated values

based on <
trends in

Report date

Actual cost of
work performed

Plotting end condition values as the project progresses allows

a final estimate to be made based on the trends in these values

Figure 7.6

Slip diagram techniques can be used to forecast the end conditions for both time and cost
by tracking variations in the EAC and the estimated completion date at each reporting point. Projecting
forwards can give an estimate of the final position well in advance of a formalized assessment

Project managers using diagrams such as those shown in Figure 7.6 would be well
advised not make the results generally available or well known. If slippage in the programme
comes to be seen as the accepted way that things are progressing, it could demotivate the
team and turn an estimate into a prophecy that fulfills itself because everyone expects it.

Interpreting the project position

Despite the detailed costs, mathematics and forecasts, it should always be remembered that
earned value calculations only provide a very limited view of a project - they have nothing to

M a n a g i n g With E a r n e d Value D a t a

89

say about what is actually going on within the project. When it comes to understanding
what these reports mean, you must look into the workings of the project; earned value might
indicate there are slippages and overruns but gives no clue as to the cause or what might be
done about it. All the normal project progress information must still be retained within the
overall project control system, including written explanations of deviations from plan,
forecasts of future situations and problems outstanding that need to be resolved. An example
of a status report is given in Figure 7.7. It shows some earned value data but it gives a great
deal more information about what is actually happening in the work package during the
reporting period; this may be more valuable to the project manager than pure numerical
data. One should always remember that earned value reports are an addition to the normal
progress reports, not a substitute for them.
Perhaps two of the most significant questions are who should see the earned value
reports and how should they be viewed? This is not an idle question as information
contained within them, particularly if it shows predictions of a cost or a schedule overrun,
could be seen as potentially explosive. There is no question that some senior managers,
either through zeal, ignorance, pressure upon them or a belief about how people should be
managed, have a great deal of difficulty in coping with estimates that show a project slip or a
cost overrun, particularly if they are in a fixed price or a liquidated damages contract
situation for which they may be ultimately responsible. Putting undue pressure on the staff
to improve performance or using threats can be counter-productive, it can simply lead to
people attempting to cover up or misreport the situation to appear more favourable.
Alternatively, it could result in people reverting to indifferent or uncommitted behaviour as
they reason that the senior management really do not understand the situation or have no
creative solutions to the problem, so why should anyone pay anything other than lip-service
to them? Slippages and overruns can come about due to a variety of reasons and they do not
necessarily have to be anyone's fault. For example: 1) some external aspect of the project's
circumstances could change, over which the project might have no control, 2) an
unexpected failure could occur on a test that requires a lengthy investigation programme
that might even lead to major redesign, or 3) a supplying contractor could suddenly run into
difficulties.Each of these is a potential cause of a delay and an overrun; in fact, I experienced
all of these during my project management work - none could have been envisaged at the
planning stage.
All forecasts and estimates, if they are made through reasoned assessment, should be
taken seriously as they can contain valuable information about a developing situation that
should be an aid to planning the way forward. It must always be recognized, however, that all
forecast information is potentially sensitive if it shows any variance from the plan; where
slippages are indicated they could become a self-fulfilling prophecy if the forecasts come to
be common knowledge, particularly if they appear 'official'. The other important point is
that forecasts made using the earned value formulae are essentially straight-line projections
based on the performance ratios to date. They contain no information about developing
future situations and make no allowance for the unforeseen occurring at any point in the
future. For this reason, earned value data and the associated forecasts should be distributed
only to those with a genuine need to know and they should always be accompanied by a
reasoned explanation of the project situation, including those many aspects that are simply
not addressed by purely numerical information.

90

Using Earned V a l u e

SAT/
SFACTWY
AT PRESENT

SINCE LAST REVIYM\I

Figure 7.7 A project status report for a work package - note the earned value data along with other
highly relevant information about the progress that is actually being made

M a n a g i n g With E a r n e d Value D a t a

91

Revisions to the baseline

Earned value performance measurement was founded on the idea that the project could be
seen in sufficient detail for an accurate plan to be constructed that would show all the major
blocks of work and the proper time phasing until the very end. For certain types of project,
for example, routine construction, road building or ship building, this can hold good in the
general sense even though there may be minor variations to the plan. Other types of project,
particularly those with a high degree of technical innovation, for example, weapons system
development and some major software systems projects, can have very different features.
These projects are characterized by changes and surprises; the unexpected is the one thing
that can be expected and there are frequent revisions to whatever plans have been made. In
certain respects, one could say these projects are perhaps the least suitable for earned value
methods, yet it was on just such projects that the whole approach was started. It was the
desire to place certainty and foresight into projects that were inherently uncertain that was a
principal motivator. If all projects were of the routine and highly predictable type, nobody
would ever have started down the performance measurement route - it simply would not
have been worth it.
When earned value methods began, the US DoD laid great emphasis on the development
of a complete and detailed plan for every phase of its projects as early as possible. To think
that an initial plan, even after a full review, would remain stable over the duration of the
project in the face of inherent uncertainties that must have existed was an unrealistic
expectation and recognized as such. The DoD did, however, improve the chances of stability
by allowing the plan to contain contingenciesthat could be used to overcome difficulties and
maintain the plan without too many deviations and thus uphold the overall integrity of the
baseline. In particular, it allowed the provision of a management reserve within the budget at
completion that could be distributed to existing or new activities that were within the overall
framework, if circumstances demanded. An approach similar to this can be recommended in
any project situation where there is an element of uncertainty.
The development of a baseline plan in which all can have confidence at the start is
essential to the effective use of earned value methods. This was recognized in the USA by the
adoption in 1994 of the Integrated Baseline Review process in which the plan could be
thoroughly scrutinized before it was acccepted as the base position from which all
performance would be measured. Integrated Baseline Reviews take place after contract award.
A particular aspect of the review process is a thorough assessment of the risk areas in the
project plan. It demands that risks are identified and assessed in the areas of 1) technical
performance, 2) adequacy of the budget provision, 3) overall time-scale, 4) availability and
suitability of resources, and 5) appropriateness of the management process. This is a very
comprehensive review designed to ensure that a thorough understanding of all the project
issues exists on both sides. The important point to note is the detail of the planning and the
thoroughness of the evaluation before the full performance measurement process is
implemented. This approach has not tended to be mirrored by British practice, which lays
great emphasis on competitive tendering, often with relatively short periods in which to
generate the full proposal and submit it. Contracts are then frequently let on a fixed-price
basis; there is no proper review and risk assessment but the project proceeds through a series
of change orders that are agreed between the sponsor and the contractors to cover
developing situations that were not envisaged in the original submission.
In any effective project control system, the handling of changes is a significant element.

92

Using E a r n e d V a l u e

No project can be properly managed if uncontrolled changes can creep in; if that happens it
can lead to a chaotic situation and a lack of overall direction. However, changes cannot
always be avoided; they may be forced upon the project by events and they may be essential
if the project is to move forward in a proper manner. There is simply no benefit in trying to
direct people to report against an old plan that simply does not reflect what is being done or
what is intended, particularly if planned work is no longer needed and new and unplanned
tasks must be performed instead. To fail to admit changes in the face of obviously altered
circumstances or new knowledge can be as harmful as uncontrolled change. All projects need
a proper change control procedure to:
formally acknowledge change that has come about due to deviations from existing plans
or altered circumstances,
assess proposals for change that arise from new knowledge about the project situation
ensure that approved changes are properly incorporated in the project plan and acted
upon.
As changes are accepted and incorporated, the original baseline plan will be increasingly less
relevant as a statement of the way ahead. It can lead to the development of two plans: a
baseline for reporting purposes and a working plan for day-to-day control. This is not ideal
and the more the two plans deviate the more difficult the whole reporting situation can
become. It must therefore be recognized that baseline plans are likely to have a limited
operational life and there will come a time when all the changes must be rationalized into a
new baseline. This is true of any project containing inherent uncertainty.
Besides changes, baselines for both budgets and time-scales can be subject to other
problems, some of which are the result of management actions.

FRONT-END LOADED BASELINES

Front-end loaded baselines can occur when work in the near-term is perceived clearly and
can be estimated precisely but longer-term work is less well defined. This lack of definition
may exist for perfectly valid reasons as some of the long-term work may be dependent upon
the results of the near-term work. There can be a temptation to put in realistic budgets for
near-term work while underpricing longer-term work in the expectation that contract
changes or budget amendments will be negotiated in the intervening period to cover any
overspends that may arise later on. Unscrupulous contractors may also be able to exploit lack
of knowledge on the part of the sponsor, particularly in highly specialized areas of
technology to cover a low initial contract budget. The effect may not become apparent until
well into the project when actual costs suddenly begin to diverge from the baseline plan and
there is flood of change requests.

RUBBER BASELINES
This problem was particularly prevalent in the time before there were integrated project
planning and costing software packages. In the early days, because these two aspects were
separate, it was possible to shift the budgets for work without actually shifting the
appropriate activities in time. Although this practice was prohibited under formalized earned
value contractual arrangements, some contractors became quite clever at bringing budgets

M a n a g i n g With Earned Value Data

Cum.
Cost

- -X

Final position
0

Actual costs

\c'

Effect of 'front-end loading'

may
until not
wellbinto thee project
c

'

/-xTarget
'

Project baseline
expenditure curve

Time
Figure 7.8

The effect of front-end loading can disguise a potential cost overrun during the early part

of the project

forward without the corresponding movements in the planned work. The effect was similar
to front-end loading by a slow process of subtly altering the baseline at each reporting period,
thus disguising the potential for a cost overrun.

BASELINE REVISIONS
The fact that most projects are subject to variations, from whatever cause, was recognized
from the time that earned value methods were first devised. It was also recognized that
continuing to measure progress against an out-of-date plan would, in the end, lead to a totally

;/>

Cum. I

Actual expe diture

! TX

Rubber baseline
to successive budget
shifts to the left
\

Time

The 'rubber baseline' -continuous revisions to the budget timing can also disguise a
potential cost overrun

Figure 7.9

94

Using E a r n e d Value

false view. Customer organizations might simply have to swallow hard, sweep up all the
changes, for better or worse, and create a new baseline position at a time when it is deemed
that the current baseline plan no longer gives a true representation of the intended work.
The DoD's approach to this problem was to allow two conditions: 'reprogramming' and
'replanning'. The essential difference is that reprogramming involves allocating additional
funds to the baseline in excess of the contract value and possibly schedule changes that
exceeds the contract schedule dates, while replanning allows the contractor to move the
activities in time without altering their budgets. A contract change is not required if the work
content has not changed. This is a purely US approach but it does have the merit that it stops
changes to the plan becoming the subject of a contract change where they don't involve a
change in the project budget. However, this condition might not apply if the contractual
arrangement involves some form of delivery incentive or liquidated damages; if this is the
case, a change in the schedule, irrespectiveof the budget, would have to be accompanied by a
contract change. Whether or not one would choose to use the distinction between
'replanning' and 'reprogramming' is a matter that needs to be agreed between the parties
when the control procedure is set up.
Reprogramming always involves a change to the baseline, and four conditions were
identified:
1 Eliminate the schedule variances - With this method the total budgeted cost for work
scheduled (BCWS) is made equal to the budgeted cost for work performed (BCWP) and a
new baseline is projected forward from the current BCWP. This will result in a baseline
with the same overall time-scale but more compressed activities, this additional priority
may result in an increase in the baseline cost at completion which must be accepted in the
revised plan. Such an approach could be used where the end date for the project is
absolutely fixed, for example, the opening of a new stadium for the 2008 Olympic Games:
once the date is announced it cannot be allowed to slip.

2 Eliminate the cost variances - Here the BCWP is made equal to the current actual cost of
work performed (ACWP) and the new baseline is projected forward from a new budget
position based on the current BCWS plus the ACWP less the BCWP at the point of
implementation of the new baseline. This creates a baseline above the ACWP line, thus the
schedule variance will continue to be negative even though the cost variances disappear.
3 Eliminate both the schedule and cost variances -This is a complete reprogramming of

the whole project with a new baseline projected forward from the current ACWP, at which
point all variances are eliminated.
4 Preserve all variances - In some cases, the existing variances are small enough to be

tolerable and no adjustment to existing data is made.

Which of these conditions is applied depends on the circumstances and the degree of
reprogramming involved. Option 1 might be applicable in some circumstances, particularly
if there is known to be some slack in the programme, if activities can be 'crashed' or the end
date is absolutely fixed. Option 2 is something of a mathematical trick to ensure that the
schedule variances remain negative even though additional funds have been allocated to
speed up the completion of the project. In practice this awkward and somewhat artificial
method is rarely used; the justification for such an approach would have to be political rather

M a n a g i n g With E a r n e d Value D a t a

95

than practical. Option 3 is a chance to inject new realism into the project plan free of any
artificial constraints; it must be considered the best option unless there are overriding
political considerations to do otherwise. The graphical interpretation of the first three of
these options is given in Figure 7.10.

Option 1 Eliminate the schedule variances

Revision date
EAC
Cost variance
added to BCWS

Option 2 Eliminate the cost variances

Revision date

BCWS (Old baseline)

Option 3 Reprogramming from the current actual position

Options for revising the project cost baseline to reflect the current situation. Option 3 is
probably the most realistic, unles there are good reasons to adopt Option 1; Option 2 is not
recommended (source: Flemming, Q. W., The Management Guide to C/SCSC)

Figure 7.10

96

Using E a r n e d Value

Once a new baseline plan has been generated, reporting against the original baseline
becomes irrelevant and the whole performance measurement process starts from a new
position. In the case of Option 3, the CPI and SPI values revert to 1.0. In practice, issuing a
new baseline programme is done only when the original plan has become so out-of-line with
the current situation that the relationship between the two is no longer meaningful.
Nevertheless, plans issued to the project team should always reflect the latest thinking and as
soon as any changes from the baseline are introduced, errors in the earned value calculations
begin to appear although initially they may be small enough to ignore.
Very often the generation of a new baseline requires considerable clerical and planning
effort as it will be normal to assess not only the current position but the entire future plan.
Few projects of any magnitude are completed without the creation of revised baselines at
some point in their lives; as much as possible this process should be restricted to the few
occasions that are really necessary but, when a new baseline is created, the changes should be
as comprehensive and far-sighted as possible.
Earned value methods are an aid to the project manager, they are not an end in
themselves. The data they produce can be of great use for the insight they can give into the
progress that is being made and what the future may hold. However, EV does demand
reporting against a fixed plan; sometimes events dictate that the plan is not going to be met
and something needs to change. Performance measurement may be very useful, but when
rapid changes are needed to keep the project on track they have to be made irrespectiveof the
reporting requirements. The dead hand of the past should never be allowed to rule the future.

Note
1 Exponential smoothing was invented by R. G . Brown in the 1950s as an aid in making short-term
forecasts, principally for the next time period ahead of the latest figure. Brown's formula is a form of
mathematical smoothing to establish the trend in a series of more erratic data. As well as forecasting, it
can be used for reporting purposes; the formula is:

The latest smoothed value =the previous smoothed value + a proportion of the difference
between the latest actual value and the previous smoothed value.
In mathematical terms this is given as:
ST=ST-, + a(VT-ST-1)
where
ST=Smoothed value at time T (that is, at latest time)
ST-, = Smoothed value at time T -1 (that is, smoothed value at the previous period)
VT=Actual observed value at time T (that is, latest observation)
a = Smoothing constant
This formula is sometimes revised for easier working to give:

The weights attached to each observed value in the series of values that make up any smoothed value,
ST,form an exponential series as each decreases by a fixed fraction (1- a) over the previous, with the
greatest weight being attached to the most recent observation.
For reporting purposes, values smoothed by the exponential process are used, as data that exhibits
considerable variation may be subject to misinterpretation if too much notice is taken of the local
fluctuations. The degree of sensitivity in the resulting smoothed values to changes in the most
recently observed data is controlled by the factor a. If a is set to 1, the new smoothed value will be

M a n a g i n g W i t h Earned Value Data

97

equal to the latest observation and there will be no smoothing. If a is set to 0, all variation from the
initial position is removed and the smoothed figure remains constant at the initial value. In practice,
setting a between 0.1 and 0.2 is found to give the most appropriate smoothed figures both for
reporting and forecasting purposes.

CHAPTER

8 s o m a r e for Earned value

Methods

In many aspects of contemporary project management, software systems have become an

essential part of the process and earned value is no different. It could be argued that the advent
of high-speed computing in the mid-1950s created project management as we know it today;
methods such as critical path analysis would have been quite impractical without it.
Unfortunately, however, the theorists ran ahead of the practitioners. Although critical path
and PERT methods were fine in theory, in the early days the practice was often made difficult
by decidedly less than adequate software. So problematic was it that some project managers
actually abandoned drawing and analysing networks during the 1970s and reverted to handdrawn Gantt charts, task lists and weekly progress meetings. However, the PC revolution that
started in the mid-1980s has reversed that position and a host of reliable and affordable project
management software is now available that can perform a wide variety of tasks. Unless the
project is very small, most project managers have access to some form of planning software.
It has already been said that the original take-up of earned value methods in the USA was
slow and that lack of suitable software must have been a contributory factor, particularly
after the failure of PERT/Cost, which embodied earned value principles. However, the lack of
suitable software was not the only problem as in some cases, the data structure and mode of
operation did not support an earned value approach.

A cautionary tale
From the late 1970s to mid-1980s, I was involved in the management of an Anglo-US airweapon development project; because of its high value, the British prime contractor was
obliged to install earned value methods when the USA decided to join the project, after it had
been underway for some years. The plan was not contained in any single document as it
existed in a variety of forms - networks, supplies schedules and trials lists which bore a
relationship to one another -while the work breakdown structure, used for cost collection,
had only a loose relationship to the various aspects of the plan. When this was coupled with
a project involving nine major subcontractors, all developing different aspects involving
thousands of components to be made within a very complex and interlocked plan, the
problems become obvious. No possibility existed for automatically assessing the progress of
any specific activity and tying it to anything but the broadest work blocks in the cost
collection system. In practice, to comply with the US earned value requirement was
immensely time-consuming: everything had to be done manually and the result was of
dubious accuracy. When the USA pulled out of the project, the first thing the British
company did was to abandon earned value measurement - it had all been too troublesome
for what little of value was produced.

S o f t w a r e f o r Earned Value M e t h o d s

99

The reason for dispensing with EV methods did not lie with principles of EV itself; the
problem was that when the project started it was an all-British affair for which earned value
methods were never considered and appropriate data structures were never created. Costs
were collected and allocated to the work blocks in the work breakdown structure so project
cost against the WBS budget was always known. Progress, however, was a completely
different matter, as the network was used primarily as a way of planning future work from
whatever position the project found itself in; there was no reporting against a 'baseline'
position for anything but the very highest level. This situation stemmed from another
problem: the plan was extremely complex, with many thousands of activities. The customer
required it to be planned and costed in great detail so that a position could be fixed from
which all changes could be costed and agreed, a baseline position but based primarily on
cost; time was a secondary issue. This was the result of yet another problem: by the time the
prime contractor had set out a baseline position it could take three to four months to get this
agreed among all the contractors, who each had to give their input. When a baseline position
was finally agreed, it was already out of date as a working document. Furthermore, the project
was plagued with technical difficulties that needed work-around plans to be developed on
virtually a daily basis. For this reason the baseline plan was never viewed as something to
report against, as it was always out of date; managers looked upon the plan that was created
as a continuously updated view of what to do in the future and that became the principal task
of the planners. As tasks were completed they simply dropped off the plan and historic
schedule information about them was lost. None of this should imply that the project did
not make progress or that there was no control. For example, when some serious cost issues
faced the project, much of the test work was removed from the plan in order to keep within
budget, an indicator that very drastic controlling actions were taken. The project was
completed successfully as a purely British project and the weapon went into RAF service.
With hindsight one could say that the use or non-use of earned value methods would have
had no bearing on the outcome of this project, which was subject to some serious technical
problems that required a significant effort in cost and time to overcome. The requirement for
the weapon was a national priority; unless the problems had been viewed as intractable or
the strategic situation changed, the project was going to continue to its conclusion.
Lack of suitable software undoubtedly contributed to the three to four months it took to
create and agree the plan; huge numbers of activities had to be integrated across the nine
participants, using comparatively crude network analysis tools which were quite useless for
handling the massive and complex supply schedules created to deal with the large amounts
of hardware that had to be made, assembled and tested. This initial difficulty led to a position
where the concept of reporting against a baseline plan had little real credibility with a project
team which regarded plans as a transient view of a changing future, rather than an
established view from which all deviations could be measured. This attitude was reinforced
by the frequent changes of plan that resulted from the many technical and manufacturing
problems which hampered this development project throughout most of its life.
The problems were all made worse by the fact that the project planning packages of
the time could not translate the project network into a Gantt chart, as a result, no visual
representation of the project plan was possible, nor was any software available that could
perform earned value calculations. It is thus not surprising that, under the circumstances,
earned'value methods proved virtually impossible to implement in a satisfactory way and
were dispensed with as soon as the requirement was no longer mandatory.

100

Using Earned Value

Contemporary software
The above tale relates to a time about 17 years ago when the tools available today just did not
exist - this lack of suitable software undoubtedly contributed to an already difficult situation.
Today, many of the problems can be resolved with current tools but certain basic issues still
remain. Earned value methods have always relied on the fundamental idea that a project can
be seen from start to finish with reasonable clarity at the beginning. If this condition is not
met because of too many unknowns or later events in the project are to be determined by the
outcome of earlier events, such as in a pure research project, then earned value methods are
not the most appropriate. Earned value methods are best suited to projects of the more
routine type or those that are well understood at the outset. Remember, earned value
concepts started from standard costing applied to stable industrial processes, not unstable
projects. If a large amount of experimental and developmental work is expected, then a great
deal of initial planning and feasibility study work needs to be done to establish a baseline
plan to which all can agree before the main project starts. The scale of this task should not
be underestimated if the project is complex.
Many of the contemporary project planning and management software packages
contain:
network analysis
bar chart drawing
allocation of resources to activities
allocation of cost rates to resource codes
work breakdown structuring
project costing and budgeting
resource scheduling
progress reporting
report generation
earned value calculations.
Because of the very comprehensive nature of these packages, they can do much more than
simply hold the project plan, analyse it for its properties and generate Gantt charts. The
ability to add resources and cost means the planning system can now become part of the
estimating and budget compilation process, which in earlier times was viewed as a separate
process. The ability to generate costed schedules and hold reported progress means the
potential for earned value analysis is contained within the basic operational structure,
providing a link can be created with the accounting system. This was the essence of the
PERTICost method of the early 1960s which failed due, among other things, to inadequate
software, but whose basic concept is now achievable thirty years later. Earned value
calculation is a relatively recent innovation that has only become general during the last ten
years. An example of a contemporary project planning system is Primavera Project Planner
which contains earned value analysis features as part of the package.
Not all packages contain full earned value features that include the comprehensive
reporting required by the US DoD, but some vendors offer add-on modules which enhance
the basic system. An example is Artemis Cost View, a full cost management system which can
be run in conjunction with Artemis Project View. Artemis Project View is the direct descendent
of Prestige PC from K&H used in Figure 2.1 (see p. 16); when this project planning software

S o f t w a r e for Earned Value M e t h o d s

101

was acquired by Artemis Systems from K&H in the early 1990s the earned value features were
removed, but the usefulness of this facility has resulted in earned value analysis being
reintroduced in the latest version. Artemis Cost View is the direct descendant of another K&H
product, I/CSCS.
Unlike the integral earned value features in a package like Primavera Project Planner,
specialist add-on packages can be run in stand-alone mode by importing data from other
sources as well as running in conjunction with their sister products. Cobra is an example of
an add-on or stand-alone package; it is designed to run in conjunction with Open Plan
although Open Plan does have some earned value features. Besides Open Plan, Cobra can link
directly with MicrosoflProject and Primavera Project Planner, to import schedule data; it can
also import data from other planning packages through the creation of transaction files.
Because the original customers for this specialist software were mostly contractors to either
the US Departments of Defense or Energy, the output report formats are similar and reflect
their intended compliance with the US DoD data item specifications for reporting. Those
project planning and management packages with integral earned value features do not tend
to follow the DoD formats, even though they may do many of the standard earned value
calculations, as they are intended for more general application and would not meet the strict
budgeting and reporting requirements of the US earned value purists.

Specialist software for cost and schedule control

Taking Cobra, as an example, a brief description will enable you to see how the basic
principles are applied in a current product. Two basic data sets are required: the work
breakdown structure (WBS), which defines the work to be done, and the organization
breakdown structure (OBS), which defines the structure of the company or companies that
will do it. This structure stems directly from the original DODI 7000.2 requirement and
reflects the intended market for this software, as well as illustrating how those original
decisions taken over thirty years ago are still influencing current practice. Each element of
the WBS must be identified with a single element in the OBS (that is, one unique block of
work is the responsibility of one particular department) and the two taken together represent
a 'control account' or 'cost account'. Within each cost account, lower-level elements of work
can be defined and these are termed 'work packages'. Thus the basic screen which appears at
the start and shown in Figure 8.1 is a split view which shows, in the upper half, the cost
accounts defined by the combination of the WBS and OBS codes and, in the lower half, the
work packages associated with the selected cost account. Once the basic data has been
organized into these structures, more detailed information about each work package can be
entered.
Cobra provides a complete costing facility, but, before any work can be costed, the items
that incur cost have to be established along with their cost rates and appropriate overheads.
The split-screen format is used to define, in the upper half, the budget elements - engineers,
technicians, etc. - and in the lower half, the calculations that are used to determine the cost
figure. Costs can be planned, budgeted and collected at either the cost-account or workpackage level, hence planning can be done at a higher level for blocks of work that will be
broken into more detail at some later point. However, earned value calculations can only be
performed at the work-package level, hence it is this level that reporting must be done if it is
wished to use this feature; this, in turn, may have implications for the organization's

102

Using E a r n e d Value

The basic input screen for Cobra. Note the requirement to define both the WBS and OBS in
the upper part of the screen - these are the cost accounts. Selecting a cost account will reveal the
associated work packages in the lower part. This clearly reflects the intended compliance with DODI
7000.2 requirements

Figure 8.1

accounting system. Work packages need both a value content comprised of either a labour
element given in time units (hours) and/or a pure cost (materials and purchases), plus start
and finish dates. To this limited extent, Cobra contains the time-phased plan but no logic is
included to link the time relationship of one work package with any other, hence the plan
cannot be analysed for its properties nor will a change in any particular work package's dates
reflect itself on any future or linked packages.
As most projects will be planned using a conventional project planning package, Cobra
incorporates an 'Integration Wizard'. This feature connects directly to either Open Plan,
Primavera Project Planner or MicrosoflProject to pick up schedule date information. In addition
the Integration Wizard can pick up more data: specifically work breakdown structures,
resource definitions and charge rates from the planning software if the fields are defined to it.
Time-phased budgets are obtained from the activities to which resources have been assigned;
activity status information can also be obtained directly from the planning tool.
Budgeted Costs for Work Scheduled are calculated automatically from the defined work
contents and overheads, but when it comes to calculating the earned value (BCWP), there is a
choice of methods (% complete, 50% at start, 50% at finish, etc), and each work package
must have one method specified. Entering a date for 'time now' will cause the BCWSs to be
calculated. Giving each work package a status of either 'opened' or 'complete' will allow the
calculation of earned values based on the chosen method. To complete the calculation,
Actual Costs of Work Performed need to be obtained from the accounting system and
entered either manually or by computerized means. This requires the accounting system to
use the same coding structure as Cobra. Unlike most other planning systems, Cobra is able to
make a forecast of where things might end if the trends to date are continued. Using the
BCWP, BCWS and ACWP figures, nine alternative methods are available depending on

S o f t w a r e f o r Earned Value M e t h o d s

103

which groups of figures (such as current value, three-month average or cumulative to date)
are used to calculate the performance indices. Figure 8.2 gives an example of a forecast report.

1.O1.O1 1410 01 Fuselage

........

EAC2

EAC4

Four different forecasting methods have been used to estimate the final cost (EAC) of work
package No. 1.01.01 1410 01. There is considerable difference in the results

Figure 8.2

Figure 8.3 shows a Cost Performance Report (CPR) from Cobra which precisely follows the US
DOE format. Graphical representations of project progress are also available as shown in
Figure 8.4. A large selection of alternative reports is provided and it is possible to create
reports to your own requirements.
Beside earned value reporting these specialist packages offer many other featuressuch as:
trend analysis and forecasting
report consolidation to increasingly high levels
multiple project combination reporting
departmental efficiency calculation
tracking of schedule and budget revisions
proposal pricing
currency cost conversion.
For large projects involving extensive accounting and cost reporting requirements, these
specialist packages should be considered.

1 .01.03 Flight Control

1.02.02 Flight Simulation
1.03.01 Experiment Design
1.04.03 Test Prototype
1.05.02 Project Management
Gen. and Admin.

4,650
8,505
9.110
10,656

10,656

4,650
8,505
9,481

-37

13.778

Cost Performance Report (CPR) generated by Cobra. The format follows that defined o n C-spec DODI 7000.10:Contract Cost Performance, Funds
Status and Cost/Schedule Status Report

Figure 8.3

S o f t w a r e for Earned Value M e t h o d s

105

1st Year Earned Value Report

425000 7
400000

375000

Figure 8.4

Time Now

Tracking of budget and actual costs over time in Cobra

General project planning packages

Whereas stand-alone or specialist earned value packages have tended to conform to the US
requirements, earned value calculations are also possible with the better quality project
planning packages. Primavera Project Planner P3 has already been mentioned and is a typical
example as it can perform certain earned value calculations and generate some earned value
reports without any of the rigid requirements of the more specialist packages. However, the
calculations and the report formats are much more limited, but if there is no requirement to
conform to DoD-type reporting, they should certainly be considered as they offer a
completely integrated approach within one piece of software.
Earned value methods always demanded a time-phased schedule that shows where and
when the money is to be spent. From the earliest days of computerized project planning it
was realized that adding resource quantities and charge rates to the activities in the plan
would generate a time-phased cost schedule and this feature was included. In pure
computational terms, adding information to each activity regarding the money actually
spent, the amount of progress achieved and performing earned value calculations is
comparatively simple; the worked example given in Chapter 3 (see p. 27) shows how
straightforward the basic calculations are. However, it was not until the late 1980s that this
addition started to become available in project planning packages; by the mid-1990s many

106

Using Earned Value

packages included an earned value element. This delay probably stems from the fact that
throughout the 1970s and 1980s earned value methods were largely confined to US defence
and energy contracts and the very special data structure and reporting formats that were
demanded could not easily be made compatible with a general purpose planning package. As
interest in earned value had not spread outside the defence community, I can only speculate
that the software developers saw little point in including these features.
Planning packages exist to schedule tasks in time and generate criticalities, floats and
other information about each activity or group of activities. In all cases the key element of
data is the activity number as this is the unique identifier for all information about that
activity: the quantity of resources, the duration, the day it started, etc. To update the plan,
information must be entered at activity level. Remember that in the original DoD
specification the plan was excluded from the data set, which instead laid emphasis on the
work breakdown structure and the organization structure. This in turn promoted the idea of
cost reporting against work breakdown or cost (control) account codes rather than activity
numbers, which tended to mean that project plans could not be updated from the costing or
time-sheet systems, as reporting was in an incompatible format. If, instead, the activity
number is used as the key code for project reporting, earned value calculations become
possible within the planning system, but work breakdown or cost (control) account codes
become sort-codes rather than reportinglinput codes.
The use of a standard planning package to perform earned value calculations is
illustrated in Figure 8.5, which is a report prepared by Primavera ProjectPlannerP3. Notice that
progress and cost information is given for every activity and that progress must be reported
by indicating the percentage completion of each activity. Notice also that both resource
codes and cost account numbers are also shown but they are not unique to any activity; for
example, Resource code TESTMGR relates to activities A2DE1002, ATDE1002, M2AN0658,
MNAN0658 and MNAN2707. In order to establish the earned value information, both actual
cost and progress must be entered against each activity; as far as the individuals working on
the project are concerned, the activity number is the figure to be entered on the time-sheet,
not the cost account number. When this is done the data can be sorted by cost account
number or resource code. In Figure 8.5, this has been done by cost account number and it can
be seen that the complete position of cost account no. 4000 is given. Examination of the total
line shows that this account is running well ahead of schedule but at slightly increased cost
(CPI = 7442.8617545.86 = .986). If required, the total project can be summed for all its cost
accounts to give the overall project position. Besides the cost account numbers, the system
can assign work breakdown structure codes to activities and these work as sort-codes in the
same way as the cost account codes. Note the basic similarity of this report to the one
generated by Prestige PC shown in Figure 2.1 (see p. 16); these two systems are similar in
concept and, not surprisingly, there is a similarity in their treatment of earned value.
This system contains both work breakdown structure codes and cost account codes plus
the ability to provide total cost and basic earned value data for every cost account in the
system. This is the reason why systems like Cobra can interlink with systems like Primavera
Project Planner, transfer data, perform further earned value analyses and generate forecasts.
Although this system can perform simple earned value calculations it is clearly much less
sophisticated than a system like Cobra in terms of the reports that it can produce and its
ability to generate graphs and forecasts. It might, however, be all that is required to provide
the necessary project control, particularly if the project is commercial in nature rather than
being done to any government requirements.

TSNS Financial Institution

PRIMAVERA PROJECT PLANNER

REPORT DATE

EARNED VALUE REPORT QUANTITY

Active IS Development Projects

8MAR96 RUN NO. 727

START DATE lOJUL95 FIN DATE 9JAN97

1448

Earned Value Report for Programmers &Testers

DATA DATE 28AUG95 PAGE NO. 12
-----------------------------------------------------------------------------------------------------------------.
COST
ACCOUNT

PCT
RESOURCE ACTIVITY ID CMS

..............CUMULATIVE TO DATE ..................... ...............VARIANCE ...........................AT COMPLETION ..........

ACWP

BCWP

BCWS

COST

SCHEDULE

BUDGET

ESTIMATE

4000

N TESTMGR A2DE1002

100.0

24.00

24.00

.00

.OO

24.00

24.00

24.00

4000

N TESTMGR ATDElOM

100.0

24.00

24.00

.OO

.OO

24.00

24.00

24.00

4000

N TESTMGR M2AN0658

.O

.OO

.OO

.OO

.OO

.OO

96.00

96.00

4000

NTESTMGR MNAN0658

100.0

96.00

96.00

96.00

.00

00

96.00

96.00

4000

N TESTMGR MNAN0707

.O

.OO

.OO

7.38

.00

-7.38

8.00

152.00

4000

NTESTQA

A2AN0409

100.0

880.00

880.00

.00

.00

880.00

880.00

880.00

4000

N TESFQA

A2TE0527

100.0

72.00

72.00

.00

.00

72.00

72.00

72.00

4000

NTESTQA

A2TE0717

100.0

24.00

24.00

.OO

.00

24.00

24.00

24.00

4000

NTESTQA

A2TE1038

100.0

24.00

24.00

.00

.OO

24.00

24.00

24.00

4000

N TESTQA

ATAN0409

100.0

880.00

880.00

.OO

.00

880.00

880.00

880.00
72.00

4000

N TESTQA

ATTE0527

100.0

72.00

72.00

.OO

.OO

72.00

72.00

4000

N TESTQA

ATE0717

100.0

24.00

24.00

.OO

.00

24.00

24.00

24.00

4000

NTESTQA

ATE1038

100.0

24.00

24.00

.00

.00

24.00

24.00

24.00
128.00

4000

N TESTQA

C2TE2817

O
.

.OO

.OO

.00

O
.O

.OO

128.00

4000

N TESTQA

CSTE2817

.O

.OO

.OO

.OO

.00

.OO

64.00

64.00

4000

N TESTQA

RPTEI852

.O

.00

.OO

.OO

.00

.OO

144.00

144.00

-------------------. ---------------------------------------------------------------TEST'

TOTAL

38.5

7574.86

7442.86

1259.32

-132.00

6183.54

19320.00

19846.86

Figure 8.5 Earned value report from Primavera ProjectPlannerP3. This example shows t h e final sheet of a 12-sheet r e p o r t . Notice that there i s a large positive
s c h e d u l e variance indicating that this block of work i s running well ahead of s c h e d u l e , as can be seen from t h e large number of activities that are shown a s
100% complete a g a i n s t a B C W S of z e r o which indicates that they are not yet planned to start. What i s missing from t h i s report i s t h e r e a s o n for this situation

a
V)

108

Using Earned Value

Contact details for the software mentioned in this chapter are

given below
-

Product title

Vendor

Tel/Fax

Web site

Artemis
Cost View
Project View

lnternational
Corporation Ltd
261 Bath Road
Slough
Berks SL1 4DX

Tel
(44) 01 753 727100

www.artemispm.com

Artemis International
Solutions Corp.
4041 MacArthur Boulevard
Suite 260
Newport Beach
California 92660

Tel
(01) 800 477 6648

Microsoft Ltd
Microsoft Campus
Thames Valley Park
Reading
RC6 1 WG

Tel
(44)0870 60 10 100

Microsoft Project

OpenPlan
Professional
OpenPlan
Desktop
Cobra

Pimavera Project
Planner

Welcom UK
26/28 Church Road
Welwyn Garden City
Herts AL8 6PW

Fax
(44) 01 753 727099

Fax
(01) 949 660 7020

Fax
(44) 0870 6010 200
Tel
(44) 01 707 331 231

www.welcom.com

Fax
(44) 01 707 330187

Welcom
15995 N. Barkers Landing Rd
Suite 350
Houston
Texas TX 77079

Tel
(01) 281 558 0514

Primavera Systems Inc.

Second floor,
Commonwealth House
2 Chalkhill Road
London W6 8DW

Tel
(44) 020 8563 5500

Primavera Systems Inc.

Three Bala Plaza West
Suite 700
Bala Cynwyd
PA 19004

www.microsoft.com

Fax
(01) 281 584 7828

Fax
(44) 020 8563 5533
Tel
(01) 61 0 667 8600
Fax
(01) 61 0 667 7894

www.prirnavera.com

9 Implementing Earned Value

Methods

Earned value performance measurement is no different to any other tool, it needs to be used
properly if it is to be effective; use it wrongly and it may be more trouble than it is worth. To
make it work successfully in any project organization, two major categories of elements must
exist:
Systems elements
a time-based plan with a full costing
a work breakdown structure allied to the plan
a work authorization process
a data collection system
a method of assessing and reporting progress
a suitable accounting process
a report generating system
a change control procedure

Managerial elements
- an organization with a responsibility/work matrix or similar arrangement by which staff

are responsible and accountable for progress on their assigned work

- the right culture; most importantly acceptance by the project staff and the backing of

senior management.
An activity network forms the ideal medium for the time-based plan; it is well understood as

a project management technique and can be handled by any project planning package.
Suitable software is available at prices to suit all types of project; furthermore, most project
planning packages have the ability to accept resources and charge rates onto planned
activities and from there produce both resource loadings and costings. Comprehensive
systems will allow costs to be directly enterable to cover purchases of services and materials.
Whatever system is used, it must be capable of producing total costs against individual
activities that are phased to reflect the plan. Those costs should be capable of being
summarized upwards and they should have a fixed relationship to the work breakdown
structure, if a formalized WBS is being used.
Although using a work breakdown structure is not an absolute requirement in order to
perform earned value calculations, rationalizing the work into a recognizable structure does
help with both reporting and control and it is to be recommended. Work breakdown
structures tend to exist in one form or another in most companies and projects; they may not
be recognized as such but they often appear in the form of 'cost codes' that are used for time
and cost bookings; any rational coding system represents some form of work structuring.

110

Using Earned Value

What may be absent is the formality and type of structuring needed by earned value
methods. If rational or useful data structuring is not present then this issue needs to be
addressed; whatever structure is devised it must conform to the requirements of the project
and the limitations of the data gathering systems in use within the organization.
Work must proceed in an orderly way and the managerial process must includea way of
formally authorizing work to start. This need not be a major restriction on the project's
activities as ultimately all the work is (or should be) under the project manager's control, but
without a proper system of authorization, work could start in an out-of-sequence fashion
which could lead to nugatory work that may need to be altered or repeated later in the
project, or to budgetary problems if there are cash flow limitations.
The data collection system is absolutely fundamental to the efficient working of earned
value methods. All organizations that are commercial in nature have mechanisms for
gathering data about their costs but the issue is whether the data is sufficiently detailed and
timely. In most companies that are sited in one place, basic data about project costs can be
obtained through the time bookings and purchase accounting systems as these are
fundamental to commercial operations associated with projects. This data may be less easy to
compile in more transient project organizations, for which special arrangements might have
to be made.
The method of assessing progress for earned value measurement is likely to require a new
discipline that must be maintained, ideally it should be linked to the responsibility/authority
structure. Those responsible for the success or completion of specific tasks should be required
to report progress and be accountable for achievement.
Earned value methods were founded on the idea of cost as a common denominator that
linked all elements of a project, hence an accounting system that is accurate, timely and
capable of taking basic progress and other project data and turning it into costs associated
with defined activities or work packages is essential.
None of the above will mean anything if the information cannot be turned into
meaningful results that can reported for managerial action. The earned value method is a
highly prescriptive, mathematically based method of recording progress; it demands very
specific data to be reported. Although there is no absolute requirement for a computerized
system to do the work, in practice it would be difficult for anything but the most
straightforward projects. With projects involving thousands of activities, a computerized
approach is essential in today's environment where staff costs are comparatively high. Choice
of suitable software is a significant issue and some of the options are covered in Chapter 8.
Projects, particularly advanced technology or innovative ones, are inherently unstable.
It is part of the job of the project manager to bring stability and order; successful project
managers can do this with the right procedures, adequate resources and a competent team.
Uncontrolled changes lead to an unstable situation, hence they should, as far as is possible,
be restricted to the absolute minimum. When changes are necessary they should be properly
considered before being formally incorporated in the plan with all the required budget and
contractual changes.
The right type of organization structure needs to be created in which staff with the
appropriate authority are made responsible for progress and direction and are accountable
for what has been accomplished. If people in authority are unwilling to accept responsibility,
then any form of managerial control is going to fail. If this is the case, the problem could, in a
few instances, lie with particular individuals but in the majority of cases the problem is likely
to lie with the organization itself. The causes can be many but they could include:

I m p l e m e n t i n g Earned Value M e t h o d s

111

Actual authority over resources or actions is not matched to the implied responsibility.
There is no input to setting the objectives or task goals.
The objectives or task goals are perceived as unrealistic.
The objectives or task goals are unclear or poorly defined.
The plan, as set out, is considered unworkable.
Too many outside influences over which there is no control could determine the outcome.
Too much stigma or heavy penalties are associated with failure.
Whatever the cause, if evasion of responsibility is perceived it is likely to be symptomatic of
some other failure which needs to be properly diagnosed and addressed.
The earned value methods with the associated disciplines and systems must be accepted
by the project team, who will have to cooperate in its workings; if they do not, for whatever
reason, it will fail. Ultimately, the whole system must have backing at the most senior level if
it is to succeed; top management must endorse the procurement and installation of whatever
systems are required and they must be seen to take an active interest in the reports and
projections that come from it. If it is perceived that top management are not really interested
but are simply paying lip-service to fashion or customer wishes and see the whole process as a
distraction, then the system will gradually fall into disuse as the incentive to maintain the
required discipline is not there.
Any organization contemplating introducing earned value methods should consider all
the above points and take steps to see that all have been adequately addressed. If any are
neglected then installing and using earned value will be difficult, if not impossible; what
results are produced will be of doubtful value and are likely to be disregarded.

Personnel aspects associated with earned value methods

It is probably true to say that many if not most organizations that have implemented earned
value methods have not found the process either simple or straightforward. Difficulties can
arise both from technical aspects including the procedures, the software, the data gathering
or the management of the data set, and from the staff associated with the project.
The technical aspects have already been mentioned in earlier chapters and it comes
down to the individual organizations to chose systems and procedures appropriate to the
needs of their projects; for this a thorough examination of both the project and the
organizational requirements, plus a study of available products, is essential.
For many staff working in the development field, earned value performance measurement will come as something of a shock. There can be little doubt that, upon its
introduction, there will be resistance from some quarters, particularly from those members of
the project staff that have not been used to any form of performance monitoring. It may also
come from project managers who may feel that a technique that exposes potential cost or
schedule overruns to demanding or unsympathetic senior management at an early point in
the project could be damaging to their own prospects. This resistance is understandable,
particularly with complex or innovative work where there is always an element of the
unknown and people cannot be expected to meet and solve unknown problems within fixed
time and cost parameters. However reasonable this proposition may seem, the business
world, in general, is finding it increasingly difficult to take that view and there is a growing
insistence on the part of customer organizations for fixed price development contracts with

112

Using Earned V a l u e

closely defined technical objectives. The underlying reason for this is a desire to shift the
effects of cost, schedule and technical risks from the sponsor to the contractor. The
traditional response to uncertainty, in a fixed price situation, is to put in large contingencies,
both in time and cost, but this becomes more difficult in the context of competition. It
becomes imperative that industry should pay closer attention to both the initial estimating
and the control that is exercised while the project is underway. Departmental managers must
become more responsible for the value generated by their departments and performance
measurement is probably the only way it can be done.
Difficulties with staff can be expected: the three quotations are given below regarding
problems with the project staff are taken from a survey of earned value in the United
Kingdom in 1994:'
[We have not achieved] buy-in to the system from every-one. [There is] unwillingness to
change work practices, [a] not-invented-here syndrome [and people] don't want to be
accountable for [their] work.
Not all the staff have signed on to the concept. The system shows efficiency against
original budgets. In the early stages low efficiencies were achieved, due to inaccurate
budget levels but staff saw this as a reflection on their work.
Its implementation suffered from a lack of education and training support and there was
unwillingness to adopt required new working practices to go with the discipline which the
tool imposed.
It is important to ensure the full cooperation of all concerned. A 'hearts and minds'
campaign, together with education and training, will be necessary and it must be seen to be
effectively backed by senior management. Performance measurement must not be seen to be
a threat, although some will see it that way, as this contributor to the survey indicates:
Resistance was driven by the fear of change and the desire to preserve the status quo,
particularly by those who stood to lose control over the content and release of
information. Forms of resistance ranged from any perceived deficiency that could be
found, through constant questioning of the tool selection process and the need for and
applicability of the new project management system.
The use of earned value techniques opens up the details of the project for scrutiny in a way
that no other system does and that can lead to real insecurity. For those that have been used
to managing by throwing a 'cloak of secrecy' around the project or creating deliberately
vague or imprecise reports, earned value methods can be a real threat. One of the most
favoured techniques is to hide behind the slowness or imprecision in the company's
accounting system as a way of obscuring the true picture. Another is to keep revising the plan
so that all reports are essentially history that can't really be tied to anything concrete.
The majority of people want to do a good and valuable job; performance measurement
actually reveals the value of their efforts in a way that has not previously been so clear. It also
provides warning signals about things going wrong; usually something can be done about a
difficult situation, providing it is recognized early enough. These are positive benefits but
they need to be sold to the staff, commitment is all important. It must be pointed out that
every member of a company is a stakeholder in its future for it is the company's future that

I m p l e m e n t i n g Earned Value M e t h o d s

113

holds the key to their own prosperity. The continued presence of any company in the
marketplace is totally dependent on providing the customers with what they want at a price
they can afford and that implies giving value for money. The earned value concept is aimed
directly at the 'value for money' objective.

Getting started
If earned value performance measurement is to be implemented, the decision should, ideally,
be taken at the outset of the project and incorporated in the planning and control
arrangements. It can be implemented at a mid-point in a project but that can be difficult and
time-consuming. Managers submitting estimates must be persuaded to thoroughly scrutinize
the project plan and the technical specification to ensure a full understanding of what is
expected and its implications. Undue pressure for low estimates should not be applied.
Planning engineers should also ensure that adequate time is allowed for each activity.
Resource demands should be approved by senior management for it is they who will be
tasked with ensuring that resources in terms of staff, materials and money will be made
available when required. In simple terms, a new and more formalized arrangement will exist
between all the parties. Departmental managers must be made responsible for their budgets
and this must be clearly understood in their relationship with the project manager.
Reporting systems must be set up to gather the cost and progess data. The generation of
regular performance reports is vital; responsible managers must see how they are doing at
regular and well-defined points. It must be made clear that progress assessments will be
required for each work package and the responsible managers must submit their reports on
time. Project management staff responsible for operating the system must set up a regular
routine for themselves and must ensure that authorizations go out on time and that
managers know when progress assessments are needed.
Undoubtedly one of the most important aspects of implementation is the education and
training of the project staff. Earned value performance measurement is a methodology in its
own right; it is not an intuitive process. People can easily get confused and it has acquired a
history of its own with all the attendant baggage. The latter aspect includes a reputation for
being bureaucratic and generally applicable to the larger type of project that can absorb the
associated costs. Furthermore, a considerable US-inspired jargon has evolved with a large
number of unfamiliar terms and acronyms as well as some tightly defined budgeting and
reporting requirements. All of this can be very off-putting, implying a kind of built-in rigidity
that the conventional wisdom associated with earned value management does nothing to
dispel. It is not difficult to see how staff can question the need for procedures they may see as
marginal to the main thrust of getting the project tasks completed. As the quotations above
show, it is easy for staff to become both suspicious and fearful of the discipline and the
performance measurements aspect.
Proper education about the whole subject of earned value performance measurement
needs to be made a significant part of the implenlentation process. The reasons for this are
twofold: first, people must know what the special terms mean and what the procedures are in
order to provide the right information to make the whole system work; second, they must
know what the outputs are, how to understand them and how they can help in the
management of the project, so that unjustified fears and resistance can be replaced by a more
enlightened view. Few people engaged on project work will ever come to love earned value,

114

Using Earned Value

unless they happen to make a living out of it, but people can certainly learn to live with it,
accept it for what it is and appreciate the benefits it can bring. Two quotations from the
survey reinforce the message:
If the project team do not buy-in to the products of the system they tend to be discounted
and pro-active dealing with risks is reduced.
I would spend more time explaining to staff why we are using the system and the benefits
it has to us.

Perhaps the most important point of all about implementing and using earned value
performance measurement is to see it as a process that is essentially adaptable to individual
project circumstances. Its widespread adoption across the general field of project management
in the United Kingdom and elsewhere outside the USA was undoubtedly hampered by the
rigid methodology that was often seen to accompany it whenever any serious descriptions of
the process appeared. This, as we have seen, stemmed from its US Department of Defense
origins that tended to stamp an official 'correctness' on a particular way of working that many
saw as overly bureaucratic and inappropriate to their own project circumstances. The
specialist software tools tended to confirm this view. Many project managers were right to
reject earned value in the way it had been prescribed, but it is possible to take a more tailored
approach that chooses those aspects that are relevant while avoiding other aspects that are
marginal. The ending of the old CISCSC regime in the USA has perhaps signalled to the world
that there is a more flexible approach which can give all the benefits of performance
measurement that concentrates on results and usefulness rather than rigid procedures.
Other issues can arise from the introduction of earned value methods, in particular the
use to which the performance measurement results can be put. For example, problems can
arise for the individual team members if low performance indicators are viewed as a
reflection on their work. This interpretation could be used by unscrupulous managers as a
stick to wield over the team, but it may be completely unjustified and lead to a loss of
cooperation or a collapse in morale.
In addition, contractual arrangements can be put under strain, particularly where fixed
price contracts are involved. Earned value methods were devised in an era when cost-plus-fee
contracting was the norm; contractors were paid according to the bills they incurred thus
reporting the A C W was consistent with this practice. Where a fixed-price arrangement is in
operation, some contractors have refused to make actual cost data available as this is now
considered to be commercially confidential. The argument is that as the cost risk is now
being borne by the contractor there is no reason to disclose information that could, if the
contractor is seen to do the work significantly more cheaply than quoted, leave him open to
charges of making excess profits. The quotations below show some of the problems:
Some of the more senior staff did not believe that the reports and efficiency graphs should
be submitted to the client even though the company had signed a contract saying that it
would.
They [the contractors] perceive they are on a 'price' and will manage accordingly, Fence]
reluctance to reveal cost information.
Earned value methods belong to a time when it was normal for sponsors to bear the cost risks
associated with projects but current practice is increasingly shifting the risks onto the

I m p l e m e n t i n g E a r n e d Value M e t h o d s

115

contractor where these methods may not sit so easily. From the sponsor's point of view, in a
fixed price situation with no variations, the ACWP will always equal the claimed BCWP and
the cost performance index will be 1.0. as all the cost risks are transferred to the contractor.
From a contractor's viewpoint, there could well be merit in maintaining an earned value
measurement system for internal use as it can generate important information with respect
to progress against the requirements of the project and the anticipated profits. However,
much of this data might be viewed as commercially sensitive and not to be disclosed to the
sponsor, a view with which one has to sympathize given the shift in the risk aspect from the
sponsor to the contractor. Even in a fixed price contract situation, all sponsors have a right to
be informed of schedule progress as this could affect other related activities on the project.
Using fixed price contracts is a very direct form of cost control but it is one that is
exercised through the contracting process and the transference of risk rather than through
the demands of the reporting system and the accounting procedure. It must be recognized
that in this situation there is an element of conflict between the demands for openness on
the part of the sponsor and the equally valid requirements for confidentiality on the part of
contractors. Ultimately it is for the two parties to the contract to agree what information will
be made available; there may be no hard and fast rules about this - it may just come down to
what is acceptable in the circumstances.
None of this should imply that earned value performance measurement could not be
employed on a project that used fixed price contracts. Baseline plans can still be agreed,
variation orders can still be made out to cover changes, and contingencies can still be drawn
upon. Performance reports can still be required from the participants but aspects of the actual
costs incurred by the contractors may be hidden. Ultimately this may not be of great concern
to the sponsor; as far as he is concerned the actual costs are those contained on the invoices
from the contractors. Fixed price contracting simply increases the degree of confidence about
what the value of those invoices will be.

Perspective
Earned value principles have been around for over forty years but during the first thirty the
application was largely confined to the US defence and energy community. The advent of
affordable PC-based project planning software with earned value features has brought the
method to the project community on a much wider scale, but its application is still not a
general feature of project management in the way that network-based planning packages are
used.
Earned value was born in a highly disciplined environment with complex projects
procured under very exacting contract conditions by a sponsor with an army of
administrators to ensure compliance. Despite the fact that many of the projects to which the
method was applied were highly innovative and subject to frequent changes and delays, the
basic assumption was of a well perceived future with few changes against which actual
progress could be measured with precision. This is rarely the case hence one of the principal
assumptions has always had a somewhat shaky foundation. Acceptance of the need for
change is found in the procedures for baseline revision but the more frequently this has to be
invoked the less valid any inferences from earned value measurementsbecome.
This implies that earned value methods are more suited to projects with a well-perceived
plan at the outset. Where the project is one in which the later activities are dependent on the

116

Using Earned Value

outcome of earlier ones, the baseline plan cannot be seen with clarity. This fact was
recognized by the adoption, on large development projects, of a procedure for breaking
major projects into discrete phases, each of which would be costed and managed in its own
right; at the end of each phase there is an opportunity either to stop the project with no
further liabilities, or to continue. During one phase enough work would be done to ensure
that many of the technical risks were removed and the next phase could be planned with
reasonable detail; hence a decision to proceed could be taken with a high degree of
confidence in the outcome.
Other types of project may find earned value of less use because there are significant
difficulties in estimating the work content if much of it consists of research and
experimentation. Drug development is one such type of project where ethical and safety
issues come into play; ultimately these are far more important issues than cost and schedule
matters and can have a major bearing on the course and duration of the project.
Earned value assumes an ordered world of well-defined and measurable quantities, in
terms of both time and cost. However, the advent of risk analysis techniques over the last ten
years has shifted the emphasis to the uncertain and the variable. It may be a more
sophisticated view of the real world, where things are expressed in terms of probabilities, but
often it does not make things easier to manage. The probabilistic view certainly does not sit
easily with the concept of a fixed baseline, but project management has always been a
pragmatic discipline that can embrace potential conflicts and make the best of what is
available.
On the basis of the survey, the experience of companies that have implemented earned
value performance measurement is that of a qualified success, but a success nevertheless. The
majority of those surveyed said their experience in one area of activity would encourage
them to make it a company-wide or project-wide process. None was contemplating
abandoning the approach having got it to work, even if they had encountered problems.
Two basic problems emerged: first, that of staff commitment and second, the
appropriateness and detail generated in the reports. If staff do not fully understand and
subscribe to the system objectives, resistance will be met and exacerbated by lack of proper
education. The level and appropriateness of the reporting arrangement needs careful
thought. Too much detail can demand large amounts of clerical effort that may be seen as
costly. Large amounts of detail that are not easily summarized upwards can leave the picture
unclear and management may be left with no obvious choices of action as the messages are
too obscure. When that happens the worth of the entire process can be called into question
as management revert to other methods that seem to provide the information they want.
Whether those other methods really provide any better information is not really the issue,
what tends to be favoured in real project management situations is what is convenient,
available, understandable and is acceptable within the politics of the project. Earned value
specialists, take note!
Although earned value methods have many strengths and advantages for project
management, they also have certain weaknesses, some of which have already been indicated.
Perhaps the biggest of all is that earned value methods can measure current performance and
indicate future trends, but if those trends indicate trouble ahead, earned value gives no clue
as to what to do about it. Although earned value methods may indicate where the problem is,
they don't say what it is. Other techniques within project management are rather more
helpful. For example, Value Engineering contains a methodology that will lead to improved
value products, Risk Analysis contains such methods as decision theory which, if followed,

I m p l e m e n t i n g E a r n e d Value M e t h o d s

117

should lead to the best decisions, while Network Analysis can identify the critical path and
the subcritical activities and thus the overall duration. Earned value methods remain
completely silent when it comes to what actions to take because they contain no inherent
method of problem solving; cost and schedule difficulties are rarely the cause in themselves
but are symptoms of other more fundamental problems.
Earned value performance measurement should be viewed as a useful tool for the project
manager that complements all the other tools in the set but does not replace any of them. All
the other skills and disciplines, both technical and human, are still necessary but better
information should result in improved decision making and a better-informed team.
Together they should lead to better overall project performance, which is the holy grail.

Note
1.

Webb, Alan (1995) 'Integrated cost and schedule control

EngineeringManagementJournal,Vol. 5, No. 3, June.

- a survey of

UK experience', The

APPENDIX

1 Terms Used Within Earned

Value Methods

The terms given below are those that are commonly used in the discussion and use of earned
value methods within the general field of project management. Many were first coined in the
context of US government contracting practice and their original definitions reflected that
fact. However, some of the terms are now widely used in other non-US contexts in which
their original definitions are not so appropriate; the definitions below are for general
interpretation, explanation and guidance rather than the original definitions.
Actual Cost of Work Performed (ACWP) - The total actual expenditure for any activity,
purchase, project or part of a project at a point in time, the Actual Cost (AC).
ANSIIEIA-748-1998 - A US commercial standard for guidelines for implementing Earned
Value Management Systems on projects; it is now applied to US government contracts as well
as commercial enterprises.
Apportioned effort (AE) - A method of distributing earned value to activities whose progress
is governed by another activity, by taking a value that is directly proportional to the value
earned on the related activity.
AuthorizationTo Proceed (ATP) - A formal instruction to start work on a task.
Authorized work - The work for which authority to proceed has been given; in an earned
value situation this implies that a budget has been approved for it and it is contained within
the project plan.
Authorized, Unpriced Work (AUW) -The authorized work for which no definite budget has
been set. This is typically due to negotiations not being completed on contract changes.
Baseline plan - An agreed plan against which all changes will be recorded and all progress
and costs will be measured.
Budget - An amount of money authorized or agreed to cover a defined amount of work or
purchases.
Budget at Completion (BAC) - The sum of all planned project costs plus any contingency
for management reserve.

A p p e n d i x 1 : T e r m s Used W i t h i n E a r n e d V a l u e M e t h o d s

119

Budgeted Cost for Work Performed (BCWP) - The total planned cost associated with the
completed work or acquired purchase on any activity, purchase, project or part of a project at
a point in time: the Earned Value (EV).
Budgeted Cost for Work Scheduled (BCWS) - The total planned cost for any activity,
purchase, project or part of a project scheduled to be achieved by a given point in time: the
Planned Value (PV).
C-Spec - Abbreviation for the US government's CISCSC project control procedure. Now an
outdated term.
Comprehensive Estimated Cost at Completion (CEAC) - An all-embracing estimate of
costs at completion, often associated with a complete revision of the baseline programme.
Contingency (cost) - A sum of money set aside or provided to cover unforeseen expenditure.
Contingency (time) - An amount of time contained within a project plan to allow for a
schedule overrun.
Contract Budget Base (CBB) - In a contracted situation: the estimate of the overall contract
budget made up of the contract target cost plus an estimate of the authorized, but unpriced,
work.
Contract price - In a contracted situation: the price payable by the customer under the terms
of the contract for the properly delivered goods and services contained within the contract
scope.
Contract Target Cost (CTC) - In a contracted situation: the sum of the authorized and
priced work but excluding the estimated cost of any authorized, unpriced changes.
Contract Target Price (CTP) - In a contracted situation: the negotiated estimated cost (CTC)
plus profit or fee.
Contract Work Breakdown Structure (CWBS) - The work breakdown structure as
contractually agreed between the parties (often product based).
Contractor - One who performs work on behalf of another on the basis of a legally binding
agreement.
Control account (formerly Cost account) - A budget account associated with a work
breakdown structure element, at which lower-level tasks are gathered for the purposes of
managerial control and specifically assigned to one element within the organization
structure.
Control account manager - One who is responsible for the costs and progress associated
with a control account.

120

Using E a r n e d Value

Cost Performance Index (CPI) - The rate at which value is earned for the actual costs
incurred, it equates to the BCWP divided by the ACWP. Also called the cost efficiency.
Cost Performance Report (CPR) - A US DoD defined report format for reporting cost and
schedule progress data for contracts that require EVMS compliance.
Cost-plus-fee contract, Cost-plus contract - A contractual arrangement in which the
contractor is reimbursed for his actual incurred costs plus an additional fee for profit. Also
called a 'Cost reimbursable' contract.
CostISchedule Status Report (CISSR) - A US DoD-defined report format for reporting cost
and schedule progress data for contracts that do not meet the threshold for EVMS
compliance (a simpler report than the CPR).
Cost Variance (CV) - In the context of earned value reporting: the arithmetic difference
between the earned value (BCWP) and the actual cost (ACWP) on any activity, purchase,
project or part of a project at a point in time.
Not in the context of earned value reporting: the arithmetic difference between the
planned cost (BCWS) and the actual cost (ACWP) on any activity, purchase, project or part of
a project at a point in time.
CISCSC -Acronym for CostISchedule Control Systems Criteria, a US Department of Defense
procedure for project management incorporating earned value methods, used for assessing
and ensuring contractor compliance with a set methodology. Also known as C-Spec, no
longer in use.
DODI 7000.2 - A US Department of Defense Instruction (1967) for project management on
major acquisition projects incorporating the CostISchedule Control Systems Criteria,
superseded by DoD 5000.2R.
DOD 5000.2R - A US Department of Defense Instruction (1996))containing in Part I1Section
B the earned value management systems guidelines.
Earned Value - The value attached to that which has been achieved on any activity,
purchase, part of a project or project in terms of what was planned or expected at the start.
Also known as the Budgeted Cost for Work Performed (BCWP).
Earned Value Management Sysems (EVMS) - Project management procedure using earned
value principles and implied in US standard ANSIIEIA-748-1998.
Estimate At Completion (EAC) - The sum of all actual costs at a point in time plus all
estimated remaining costs to complete a project or part of a project. Also referred to as the
Estimated cost at completion.
Estimated cost at completion -Another term for the Estimate At Completion (EAC).

A p p e n d i x 1 : T e r m s Used W i t h i n E a r n e d V a l u e M e t h o d s

121

Estimated Time to Completion (ElTC) -An estimate, at a point in time, of the likely overall
duration of a project or part of a project.
Estimate to Completion (ETC) - An estimate, at a point in time, of the costs still to be
incurred to complete the project or part of a project.
Fixed price contract - A contractual arrangement where a contractor is paid a fixed sum for a
defined piece of work irrespectiveof his actual expenditure.
Incentive contract - A contractual arrangement where the contractor is rewarded with an
increased fee if performance above an agreed level is achieved.
Integrated Baseline Review (IBR) - A formal process in which both the sponsor and the
contractor examine a baseline plan to establish its completeness, realism and practicality.
Latest Revised Estimate (LRE) - The most recent estimate of the final cost of a project, or
part of a project, based on known work. Note, this term is sometimes used interchangeably
with the Estimate At Completion (EAC).
Level of Effort (LOE) - A method of costing work which is time dependent rather that
associated with any definitive output, for example, support services such as project
management or maintaining the project accounts. Value is accrued according to the elapsed
time and resources employed rather than observed progress.
Liquidated damages agreement - A contractual arrangemeent that provides for a final, nonnegotiable settlement for a loss due to some aspect, other than cost, of non-performanceof a
contract; it is often applied to lateness in delivery or product performance below that
stipulated in the agreed specification.
Management Reserve (MR) - A financial contingency under the direct control of senior
management outside the performance measurement baseline.
Original Budget (OB) -The original amount of money provisioned for an activity, purchase,
project or part of a project.
Original Duration (OD) - The original planned time over which an activity, project or part
of a project should take place.
Organization Structure, Organization Breakdown Structure (OBS) - The logical and
hierarchical arrangement of both function and authority of a company engaged on a project.
Over Target Baseline (OTB) - A baseline programme resulting from formal reprogramming,
where additional funds over the value of the contract are incorporated into the baseline plan.
Performance Measurement Baseline (PMB) -An agreed programme of work against which
progress will be measured which includes all known and planned work; it includes all costs
within the Budget at Completion less management reserve.

122

Using E a r n e d Value

Prime contractor - A contracting organization with overall responsibility for delivering a

project; normally this will involve coordination and control of a group of lower-level
contractors.
Project manager - One who is charged with the overall direction of a unique undertaking.
This term is sometimes synonymous with program manager or programme manager.
Responsibility Assignment Matix (RAM) - A matrix showing how control accounts for
tasks in the work breakdown structure are assigned to functions within the organization.
Schedule Performance Index (SPI) -The rate of progress against the original schedule with
respect to elapsed time (by reference to the earned value). It equates to the BCWP divided by
the BCWS at a point in time. Also called schedule efficiency.
Schedule Variance (SV) - The arithmetic difference between the earned value (BCWP) and
the planned cost (BCWS) on any activity, purchase, project or part of a project at a point in
time.
Sponsor - One who finances a project; he or she may also have a controlling interest.
Statement of Work, Scope of Work (SOW) - A definition of a product, group of products,
tasks or services to be procured under a contract; by definition it limits what is to be supplied
to that which is contained in the statement.
To-Complete Performance Index for budget (TCPI(BAC))- At a point in time: the rate of
generation of earned value required to complete the project within the budget at completion.
To-Complete Performance Index for schedule (TCPI(0D))- At a point in time: the rate of
schedule progress required to complete the project within the planned overall duration.
Undistibuted Budget (UB) - A holding account used to temporarily contain the budget for
recently negotiated contract charges until detailed planning can be accomplished.
Variance at Completion (VAC) - An estimate of the likely difference between the planned
cost and the actual cost of the project or part of a project when complete. The arithmetic
difference between the BAC and the EAC.
Work Breakdown Structure (WBS) - A logical, hierarchical arrangement of the work
elements within a project.
Work package -An identifiable, discrete and defined block of work.
Work package manager - One who is responsible for the conduct of the work contained in a
work package.

APPENDIX

2Earned Value Management

Systems Criteria

In 1996 a set of 32 criteria for the operation of earned value management systems within US
companies was produced. This was based on the 35 criteria set out in the DODI 7000.2
instruction for government defence contracts but was intended for more general use.
These criteria were later incorporated in an American national commercial standard
ANSIIEIA-748-1998 which provides general guidance for the operation of earned value
management systems. In 1999 this standard was adopted by the US Department of Defense
as the approach to be used for project monitoring and control.

Organization
1 Define the authorized work elements for the program. A work breakdown structure

(WBS), tailored for effective internal management control, is commonly used in this
process.
Identify the program organizational structure including the major subcontractors
responsible for accomplishing the authorized work, and define the organizational
elements in which work will be planned and controlled.
Provide for the integration of the company's planning, scheduling, budgeting, work
authorization and cost accumulation processes with each other, and as appropriate, the
program work breakdown structure and the program organizational structure.
Identify the company organization or function responsible for controlling overhead
(indirect costs).
Provide for integration of the program work breakdown structure and the program
organizational structure in a manner that permits cost and schedule performance
measurement by elements of either or both structures as needed.

Planning and Budgeting

6 Schedule the authorized work in a manner which describes the sequence of work and
identifies significant task interdependencies required to meet the requirements of the
program.
7 Identify physical products, milestones, technical performance goals, or other indicators
that will be used to measure progress.
8 Establish and maintain a time-phased budget baseline, at the control account level,
against which program performance can be measured. Budget for far-term efforts may be
held in higher-level accounts until an appropriate time for allocation at the control

124

Using E a r n e d Value

account level. Initial budgets established for performance measurement will be based on
either internal management goals or the external customer negotiated target cost,
including estimates for authorized but undefinitized work. On government contracts, if
an over-target baseline is used for performance measurement reporting purposes, prior
notification must be provided to the customer.
9 Establish budgets for authorized work with identification of significant cost elements
(labour, material, etc.) as needed for internal management and for control of
subcontractors.
10. To the extent it is practical to identify the authorized work in discrete work packages,
establish budgets for this work in terms of dollars, hours, or other measurable units.
Where the entire control account is not subdivided into work packages, identify the farterm effort in larger planning packages for budget and scheduling purposes.
11 Provide that the sum of all work package budgets plus planning package budgets within a
control account equals the control account budget.
12 Identify and control level of effort activity by time-phased budgets established for this
purpose. Only that effort which is unmeasurable or for which measurement is impractical
may be classified as level of effort.
13 Establish overhead budgets for each significant organizational component of the
company for expenses which will become indirect costs. Reflect in the program budgets,
at the appropriate level, the amounts in overhead pools that are planned to be allocated
to the program as indirect costs.
14 Identify management reserves and undistributed budget.
15 Provide that the program target cost goal is reconciled with the sum of all internal
program budgets and management reserves.

Accounting Considerations
16 Record direct costs in a manner consistent with the budgets in a formal system controlled

17

18

19
20
21

by the general books of account.

When a work breakdown structure is used, summarize direct costs from control accounts
into the work breakdown structure without allocation of a single control account to two
or more work breakdown structure elements.
Summarize direct costs from the control accounts into the contractor's organizational
elements without allocation of a single control account to two or more organizational
elements.
Record all indirect costs which will be allocated to the contract.
Identify unit costs, equivalent units costs, or lot costs when needed.
For EVMS (Earned Value Management System), the material accounting system will
provide for:
(1) Accurate cost accumulation and assignment of costs to control accounts in a manner
consistent with the budgets using recognized, acceptable, costing techniques.
(2) Cost performance measurement at the point in time most suitable for the category of
material involved, but no earlier than the time of progress payments or actual receipt
of material.
(3) Full accountability of all material purchased for the program including the residual
inventory.

Appendix 2 : Earned Value M a n a g e m e n t Systems Criteria

125

Analysis and Management Reports

22 At least on a monthly basis, generate the following information at the control account
and other levels as necessary for management control using actual cost data from, or
reconcilable with, the accounting system:
(1) Comparison of the amount of planned budget and the amount of budget earned for
work accomplished. This comparison provides the schedule variance.
(2) Comparison of the amount of the budget earned the actual (applied where
appropriate) direct costs for the same work. This comparison provides the cost
variance.
23 Identify, at least monthly, the significant differences between both planned and actual
schedule performance and planned and actual cost performance, and provide the reasons
for the variances in the detail needed by program management.
24 Identify budgeted and applied (or actual) indirect costs at the level and frequency needed
by management for effective control, along with the reasons for any significant variances.
25 Summarize the data elements and associated variances through the program organization
and/or work breakdown structure to support management needs and any customer
reporting specified in the contract.
26 Implement managerial actions taken as the result of earned value information.
27 Develop revised estimates of cost at completion based on performance to date, commitment values for material, and estimates of future conditions. Compare this information
with the performance measurement baseline to identify variances at completion important
to company management and any applicable customer reporting requirements including
statements of funding requirements.

Revisions and Data Maintenance

28 Incorporate authorized changes in a timely manner, recording the effects of such changes
in budgets and schedules. In the directed effort prior to negotiation of a change, base
such revisions on the amount estimated and budgeted to the program organizations.
29 Reconcile current budgets to prior budgets in terms of changes to the authorized work
and internal replanning in the detail needed by management for effective control.
30 Control retroactive changes to records pertaining to work performed that would change
previously reported amounts for actual costs, earned value, or budgets. Adjustments
should be made only for correction of errors, routine accounting adjustments, effects of
customer- or management-directed changes, or to improve the baseline integrity and
accuracy of performance measurement data.
31 Prevent revisions to the program budget except for authorized changes.
32 Document changes to the performance measurement baseline.

APPENDIX

3 Formulae Used for Earned

Value Calculations

Formulae used for Earned Value calculations

Cost Variance (CV)
CV = BCWP - ACWP
Schedule Variance (SV)
SV = BCWP - BCWS
Cost Performance Index (CPI) (Cost efficiency)
BCWP
CPI = ACWP
Schedule Performance Index (SPI) (Schedule efficiency)

Estimated Cost At Completion (EAC) The estimated final cost of the project.
EAC = ACWP + BAC&:CWP
Where: BAC is the Budgeted Cost At Completion
Alternative formulae for the Estimated Cost At Completion
EACl

BAC

=
+

BCWP

0.5 x CPI + 0.5 x SPI

EAC2 = ACWP + BAC - BCWP

CpI x SpI
Estimated Time To Completion (ETTC) The estimated overall duration of the project.
ETTC = ATE + OD - (ATE x SPI)
SPI

A p p e n d i x 3: F o r m u l a e Used f o r Earned Value C a l c u l a t i o n s

Where: BAC is the Budgeted Cost At Completion
ATE is the Actual Time Expended
OD is the Original Duration
Note: this formula becomes increasingly less reliable as the project approaches
completion and fails once the Original Duration is exceeded.
Variance at completion (VAC)
VAC = BAC - EAC
Schedule variance as a percentage of the scheduled achievement (SV%)
SV% = BCWP - BCWS ,100
BCWS
Cost variance as a percentage of the earned value (CVYo)
CV% = BCWP - ACWP ,100
BCWP
Percentage complete at the report point

--BCWP x 100
- BAC
Percentage spent at the report point

--ACWP x 100
- BAC
Percentage of project to be achieved at the report point

--BCWS x 100
- BAC
The 'To-complete Performance Index' for budgeted cost (TCPI(BAC))
BAC - BCWP
TCP1(BAC) = BAC - ACWP
The 'To-complete Performance Index' for estimated cost (TCPI(EAC))
TCPI(EAC) = EAC - BCWP
EAC - ACWP
The 'To-complete Performance Index' for schedule (TCPI(0D))
TCPI(0D) = BAC - BCWP
BAC - BCWS

127

Index

accounting system 9,15,62,68,70,84,85,

102,110,112,124
action limit 79
activity code (number) 65, 78,84,106
actual costs 8, 18,20,22, 58, 79, 106, 115, 125
actual cost of work performed (ACWP) 22,26,
30-35, 78, 79,80,86,87,102,107,115, 118
actual spends (expenditure) 1,2,9, 77, 79
actual time elapsedlexpended (ATE) 26,27,74,
127
ANSIIEIA 748-1998 (Standard) 5,14,21,57,
118,120,123
apportioned effort 77, 118
Artemis Cost View (software system) 100, 108
Artemis Project View (software system) 100,
110
Arthur D. Little (consultants) 12, 13, 17,54
authority 110,111
authorization 110,123
authoriztion to proceed (ATP) 118
authorized work 118,114,125
baseline plan 9,13,53,54,63,91-6,99, 115,
116,118,123
baseline review 13
Brown, R. G. 96
budget 53-9,62, 70, 72,77, 79,92,94, 100,
102,110,113,118,123-5
budget at completion (BAC) 26,27,32,33,57,
118,121,126
budgeted cost for work performed (BCWP) 22,
25,30,32,34,37,54, 74, 75,86,87,94, 102,
115,119
budgeted cost for work scheduled (BCWS) 22,
26,32,33,34, 74, 77, 79,82,86,94,97, 102,
107,119
C-Spec 10,22,119
C-5A Galaxy, Lockheed, aircraft 9
changes 91,92,110,125
change request 62

Christle, Gary 13
Cobra (software system) 87, 101, 102, 103, 106,
108
code (number)(system)67-70, 100, 102, 109
company programme 63
comprehensive estimate at completion (CEAC)
119
contingency 53,55,56,62,112,115,119
contract budget base 119
contract price 119
contract target price 119
contract target cost 119
contract work breakdown structure (CWBS) 10,
40,119
control account 60,62,78,79,86,87, 101,106,
119,123-5
cost (of project) 53,63,85
cost account l l , l 5 , 60, 101,106,119
cost breakdown structure 72
cost code 109
cost efficiency 8,24, 120
cost performance index (CPI) 24,26,32-5,
79-83, 96,120,126
cost performance report (CPR) 86-8,103,120
cost-plus-fee contract 54,57, 114
cost/schedule control systems criteria (C/SCSC)
2, 10, 11, 12, 13, 14,21,22,37,38,40,60,
64,114,120
cost/schedule status report (CISSR)86-8,120
cost standard 8
cost variance (CV) 22,23,30,32, 77, 79,85,87,
94,125-7
criteria (for implementation) 5,9, 13, 14, 64,
123
critical path (method) (analysis) 1,8,9,98, 117
customer 70, 71,125
customer programme 63
data collection system 110
data capture sheet 68
data structure 110

Index

Department of Defense, Australia 16

Department of Defense, US (DoD) 2,5,9,11,
15,40, 45,46,54,56,60,63,64,65,86,91,
93,100,114,123
Department of Energy US, (DOE)12,15,101
difficulties with the predictive formulae 33,
35-7
DOD 5000.2R (US instruction) 13,14,120
DODI 7000.2 (US instruction) 9-12, 15,10,
101,120,123
earned value management system (EVMS)14,
57,120
earned value (definition) 22, 120
earned value concept 1,2,19-22
earned value formulae/calculations 24-8,33-7,
79-83,126,127
efficiency (of production line) 8
equipment (purchase, hire and cost of) 78
estimated actual cost 79
estimated cost at completion (EAC) 25,26,27,
31,32,33, 34,35,88, 103, 120 125, 126
estimated time to completion (E'ITC)25,27,
32,34,35, 120
estimate to completion 121
exponential smoothing 82,96
F-111, General Dynamics, aircraft 9
facilities (hire and cost of) 78
fixed price contract 84,88,111-115,120
flash data 85
Fleming, Quentin W 72,95
front loaded baseline 92
functionally based work breakdown structure
45

129

labour code 68
latest revised estimate 81, 121
level of effort 77, 78, 121, 124
life cycle (of product) 50
liquidated damages 84,89,121
major task 66-8
management reserve 57,63,91,121
manufacture resource planning (MRP) system
63
materials 11, 53, 78,85, 102, 124, 125
material accounting 8,11,84
matrix (data structure) 6 6 4
measured assessment 75
micro management 85
Microsoft Project (software system) 101, 102,
108
milestones 75, 76, 78,84, 123
ministry of Defence, UK (MOD)12
minor task 68,69
minuteman contractor performance
measurement system 9
Minuteman, missile 2,9
most likely estimate 32
multiple codes 70
National Aeronautics and Space
Administration - NASA 9, 17
network (project plan) 9,18,60,61,63-70,99,
100,117
numbering system (of WBS) 43

Haupt, Eleanor xi, xiii, 5

hiring (of equipment, facilities and services) 78
hybrid work breakdown structure 45,46

objectives (of project) 111

Open Plan (software system) 91,102,108
original budget (OB) 74, 75, 76, 121
original duration (OD) 26-8,34,121,127
organization breakdown structure (OBS) 60,63,
101,106,121
organization structure 45, 54, 60, 70, 110, 119,
123,125
over target baseline 121, 124

IICSCS (software system) 101

incentive contract 58, 121
indirect assessment 76
industrial engineer 8,9
integrated baseline review 91,121
integrated cost and schedule control 4
integrated product teams 13
integrated systems xv, 64-70
invoice 78,84

payment 84
percentage complete 20,30, 75, 76, 127
percentage spent 127
performance measurement baseline (PMB) 57,
121,125
performance measurement 10,7946,125
performance report 113,115
personnel aspects 111-13
PERT 8,9,98

Gantt chart 1,9, 18, 61,98,99, 100

130

Index

PERTICost 2,9, 15,40,60,98, 100

pessimistic estimate 32,33
Phillips, Brig. Gen. Samuel 9
planned cost 19
planned output 8
planned spend 79
planned value (PV) 22
portfolio management 64
Primavera project planner (software system)
100-108
prime contractor 121
process efficiency8
production (line, process) 8
product oriented work breakdown structure 40,
45
profits (from project) 57,58
progress 64, 100, 106, 110
project manager (definition of) 122
project plan 40,50,61-70,110,112
purchasing (of materials and equipment) 77,
78,84,85, 102
quadrant diagram 81
risk (in project) 57,91, 114, 115, 116
replanning 94
report 70, 72,82,84,86,88, 101,106, 125
reporting system 61,63,65,68,85, 115
reprogramming 94
resistance (to EV methods) 70,84, 110, 112
resource(s) 52,72, 100,105,106
resource costs 53
resource schedule 52, 100
responsibility111
responsibilityassignment matrix (RAM) xv,60,
63
rolling wave planning 52
rules of thumb 76
S-curve 18, 19, 26
scope of work 52, 122
schedule performance index (SPI) 25,26,27,
28,30,32,33,34,35, 79,81, 82,83,96, 122,
126

schedule variance 22,23,32, 79,80,85,87,94,

107,122,125-7
s e ~ c e(purchase
s
and hire of) 76
software (for EV methods) xv, 2, 7, 15, 16,38,
54,56,60, 63,65,69, 70, 71,84, 85,87,92,
98-108,115
sponsor (of project) 45,50,58,64,65,84,85,
92,112,115,122
staff commitment 110-16
stage payment 78
standard cost 8
status information 102
status report 89,90
subjectiveassessment 73, 74
sub-contractor 84,85, 123
Survey of EV in the UK (1994) 6,15,116
task authorisation 84
technical performance 79
tender (for contract) 54,55
time booking (systemlsheet)5, 8, 11, 15,62,
68,84,106,101,110
to-complete performance index numbers 33,
34,122,127
tools (purchase of) 78
training (of staff) 113
uncoordinated approach (to plan and cost) 61,
65
undistributed budget 57, 122, 124
United States Government xvi, 5,13,82,87
United Kingdom Government 5
variance at completion 32, 122, 127
variance report 87
warning limit 79
work breakdown structure (WBS) xv, 9,15,38,
37,39-48,52,60-72,84,86,99,100, 101,
106,109,122-4
work package 55,56,67,68,72, 78,79,86,87,
89,90,102,103,107,122,124,125
work package manager 56,122

The Relationship Manager

The Next Generation of Project Management

Tony Davis and Richard Pharro

Traditionally, project managers have been allocated a project and their role
has been to deliver on time, to quality standards and within budget. With hindsight
the client only recognises what they really want once the project is delivered - and
there is often a gap between expectation and final product. The project
management role is now changing and the total impact on the business needs
to be addressed more effectively - enter the Relationship Manager.
The true role of the Relationship Manager is to act as an orchestral conductor:
to go to the client and demonstrate his understanding of the client's
short-, medium- and long-term objectives;
to translate this into a form which the project team can address;
to receive from the project team a specification of the work to be
undertaken, including plans, estimates and schedules, together
with detailed work and cost breakdown structures to check that
they have really understood what needs to be done and by when and only when the Relationship Manager is satisfied does he return
to the client to sell VALUE.
The Relationship Manager has been written to fill the gap between technical
and business aspects of successful project delivery. It provides practical
guidance on how to make this new way of working a reality and details
the skills and techniques necessary to make a success of the latest developments
in project management.

GOWER

The Bid Manager's Handbook

David Nickson
Winning significant business on the right terms is an increasingly complex,
challenging and time-consuming task, and a successful bid is a vital part of any
business offering its services or products to another. This book will help you to
enhance the probability of success in winning bids at the desired margins and to set
up and run effectively a bid management team. Aimed at two main groups of
readers (sales staff managing multi-disciplinary bid teams, and project and
technical managers who find themselves managing a bid to support a sales
campaign), The Bid Manager's Handbook provides an invaluable resource in the
battle to win new business.
Taking an extremely practical approach and using real-life examples,
David Nickson leads the reader through every stage of planning for, producing
and delivering a bid. By the end of the book you will:
know what needs to be done
know how to present the information to the prospective
client effectively
have gained the writing and editorial skills needed to put a sales
case across
have identified the skills that are needed to manage a bid
Crucially it also shows how to save time - the most important commodity
in any bid as it is always a scarce resource - without affecting quality. In short,
The Bid Manager's Handbook is the definitive guide to managing winning
bids effectively and efficiently.

GOWER

Project Management for

Successf uI Product Innovation
Alan Webb
This comprehensive book provides a complete guide to managing projects
involving the development of new products. It aims to give the practising
project manager an insight into the many processes that are involved
in handling one of the most complex of industrial activities.
The book is arranged in a logical sequence covering the development of
project management, project management structures, aspects of planning,
monitoring and control, economics and value management, design
management, intellectual property issues and production start-up. Particular
emphasis has been given to risk management which is recognized as both
a difficult subject and also one of growing importance to today's project
manager, especially in product innovation. A complete explanation is given
of the latest and most relevant techniques together with guidance as to where
and how they should be applied. Where software tools are available these
are named and, in some cases, brief descriptions are included; in all cases
contact details of the vendors are provided.

GOWER

GOWER

Gower Handbook
of Purchasing Management
Third Edition
Marc Day
Published in association with the Chartered Institute
of Purchasing and Supply
The revised third edition of the Gower Handbook of Purchasing Management views
procurement as standing on the boundary of the firm, looking outwards and scanning
the environment for new opportunities and threats. In this respect, as in many others,
the new edition is quite different from the previous two, reflecting the many changes
that have taken place for businesses over the years. In particular this edition has been
slimmed down and focused to assist the reader by working systematically outwards
using a purchasing lens to view the wider business world. The aim is to show the
potential contribution that purchasing can make as a driver for organizational
efficiency and business development. It is this latter requirement, the need for
purchasing to generate revenue, that has been identified as being ever more prominent
as a demand on purchasing directors' time and effort.
The book is now split into three sections. Part I lays the foundations for building
the organization of purchasing in a corporate environment. Part Il overlays
further applications on the foundations of purchasing organization. The assumption
is made that the purchasing activities of a firm are proactive in outlook, gathering
knowledge and measuring their current corporate purchasing performance, while
also looking to generate revenues for the business. Finally Part 111provides
case studies which bring to life some of the learning achieved through the
framework laid out in the previous parts.
Written by leading practitioners and academics, and published in association
with the Chartered Institute of Purchasing and Supply, this book is destined
to become a classic in the field.

GOWER