ECI Cost Estimating Practice
ECI Cost Estimating Practice
Cost Estimating
Value Enhancement Practice
March 2000
JH/AK/13-Aug-01
CONTENTS
1 INTRODUCTION 1
2 ESTIMATING APPROACH 1
2.1 Scope Structure/Definitions 1
2.2 Project Execution Strategy 2
2.3 Estimating Strategy 2
2.4 Information Requirements for Different Types of Estimate 3
2.5 Accuracy 5
2.6 Cost Structures 6
2.7 Review and Approval 7
3 ESTIMATING PROCESSES 7
3.1 Formats/databases 7
3.2 Checklists 9
3.3 Location Factors/ Performance Efficiencies 9
3.4 Commercial Considerations 9
3.5 Escalation Forecasting 11
3.6 Allowances 11
3.7 Uncertainty/Risk Assessment 12
4 RETROSPECTIVE REVIEW 13
4.1 Company Culture 14
4.2 Cost Estimate Structure 14
4.3 Cost Reporting 15
4.4 Change Control 15
4.5 Post Project Review 15
6 GLOSSARY OF TERMS 19
JH/AK/13-Aug-01
1 Introduction
The purpose of this document is provide guidance as to what is considered to be best
practice in the compilation of cost estimates over the complete project lifecycle. It will give
an insight into the estimating process, guidance as to what should be included in an
estimate, point to some of the valuable estimating aids that exist, and show how estimates
should be presented. This document also considers the importance of a retrospective review
in improving the estimating process, as well as the skills and competencies required by the
estimator.
Given that definition, the difference between a good estimate and a bad one can be
considerable. An estimate, by definition, is never “correct.” However, good quality
estimates are those that are neither conservatively high (due to excessive contingencies,
allowances, etc) nor optimistically too low (due to lack of proper scope definition or
unrealistic targets).
Companies that consistently produce poor quality estimates will either win no new business
or consistently overrun and erode profit margins.
A better way to gauge the quality of an estimate is by the magnitude of under-run or over-
run, an approach that is only really valid if there are no scope variations or identified
changes during project execution. That in itself is a rare event.
An estimate can only be considered to be good if: -
• each item is priced objectively.
• the project out-turn is within the calculated (estimated/assumed/implicit) spread.
On this basis, estimate quality can only be judged against some independent measure, the
best being that of the completed project on which the estimate was based.
2 Estimating Approach
-1-
package list or set of deliverables, referred to as a work breakdown structure, a cost
classification listing, referred to as a cost breakdown structure, and organisational
responsibility assignment, referred to as the organisational breakdown structure. Whilst the
level of detail required for estimate production may vary, depending upon the type of
estimate involved, the following principal points should be considered:
• Scope and its boundaries.
• Schedule.
• Exclusions.
It is recommended that an initial (top down) order of cost and preliminary assessment of
risk be allocated to each cost item/work package. This makes it possible to identify the
significant cost drivers/risks within the project and enables the estimator to prioritise the
work.
The project execution strategy should be documented, possibly in the form of a strategy
matrix. This will list all items within the defined scope and assign a responsible party for
each phase of execution, from initial basic/process design through to commissioning and
start up.
Whilst developing the project execution strategy consideration should be given to the
information and time available for preparation of the estimate. This will help to determine
the most appropriate techniques to be adopted and whether in-house estimating is
sufficient. If not, external support will be required. The estimating strategy should also be
detailed in a methodology document.
-2-
2.4 Information Requirements for Different Types of Estimate
The type of cost estimate produced is determined by the following factors: -
• Purpose / use (e.g. for feasibility, sanction, control, tender).
• Project stage / phase (e.g. pre-sanction, pre-construction, construction).
• Level and quality of project scope available (e.g. preliminary details, outline or
detailed design).
• Level and quality of data available (e.g. historical, project specific, quotations).
• Time for preparation of the estimate (limiting the estimating methodology and the
quality that can be brought to bear).
• Resource level and availability.
• Risk culture of the company.
Estimating techniques vary according to the nature of the estimate required, the level of
engineering data and the time available in which to produce the estimate. However, in order
to produce any estimate, a number of individual constituent parts need to be present:-
• Project scope
• Project specification
• Quantity
• Rates (cost data)
• Project schedule data
• Local factors / performance efficiency data
• Escalation norms
In all categories data might be actual - prepared specifically for the estimate, historical - in
the form of data from a previous project of a similar nature, or judgmental - based upon
knowledge and experience.
Cost estimating techniques generally fall into one of two categories:
Top Down / Parametric - used when project scope / design information is minimal.
This type of estimate is usually prepared using, for example, major plant items and factors,
cost per unit of function / use, and cost per gross area or volume etc.
-3-
Initial estimates for preliminary assessment purposes will usually be prepared with little
more data than knowledge of the size and capacity of the plant to be built, probable location
and completion date. They will often be based upon ballpark estimating techniques such as
cost per tonne of required product and rely heavily on historical data.
Slightly more detailed estimates generally have a better scope definition for major work
packages, but may rely on the use of factors for peripheral items such as off-sites, utilities
and the balance of plant.
As the need for more detailed estimates arises (e.g. for project sanction or budget
estimating), the degree of engineering input will be increased. Preliminary design data and
drawings will be produced improving scope and boundary definition and enabling initial
parametric data to be replaced.
For fully detailed estimates (e.g. for control purposes or for fixed price tendering), the
engineering input will increase again, enabling detailed specifications, drawings and bills of
quantities to be produced and prices sought from potential suppliers and contractors.
A project programme should be produced to a level of detail dependent upon the level of
estimate. As a minimum the programme should identify the high-level activities and
allocate them to the project schedule, thereby introducing time factors and key dates into
the estimating process.
In all cases the estimating process must ensure that the estimate covers the required scope
of work and no more, and is good practice to ensure a clear audit trail is visible.
-4-
Information Quality
It is good practice to use a checklist to assess the quality of information used in compilation
of the estimate.
In recent years a number of measures have been developed to assess the completeness of
engineering data used in the compilation of estimates. One example is the Project
Definition Rating Index (PDRI), developed by the CII, which allows the status of design
information to be rated and a score obtained. It should however be noted that until the basic
design concepts have been fixed, the rating would be subject to change.
2.5 Accuracy
The accuracy of the estimate should improve throughout the course of the project as scope
and specification definition become firmer and as costs progress from estimated to actual.
Figure 1 shows a typical distribution:
Conceptual
FrontEnd Study
DetailDesign
Estim ate Accuracy
-5-
The accuracy of an estimate will generally improve with the time spent on its preparation.
Figure 2 below indicates the broad relationship between estimate accuracy and time.
Cost of
Producing
the Estim ate
Increasing A ccuracy
Figure 2: Increase in estimate accuracy and cost of preparation (typical)
Cost estimates are prepared for a variety of reasons, and it is therefore important to organise
them in such a way as to satisfy the various uses to which they will be put. Whilst they may
not be prepared specifically for use in implementing a project they usually form an integral
part of the project control process. It is therefore desirable that the estimate structure be
developed as early as possible and be used consistently throughout the life of the project.
Cost estimators should be aware that the commercial success of the project will be
significantly influenced by the estimating process.
-6-
2.7 Review and Approval
The requirement to review and approve a cost estimate within an organisation should be
determined by an internal company control procedure. The procedure will cover factors
such as estimate type and level of company exposure. Other factors such as estimator
experience and company familiarity with the type of work may also be addressed.
Before a review is started, and as part of the estimating process, cross checks should be
made against benchmarks and other existing data to assess the level of the estimate. This
will put the estimate into context and provide justification in the review and approval
process.
The review should not look simply at the monetary values, but consider the process
followed in the preparation of the estimate. For large projects the estimating process should
be detailed in a methodology document forming part of the estimate back up.
The approval process needs to win acceptance from all contributors such as project, design
and construction teams, and estimators as well as management. To achieve this buy-in it is
good practice to hold interim reviews before final completion of the estimate. Decisions on
assumptions can be shared, and the estimate can be allowed to evolve, producing fewer
surprises and minimising errors.
3 Estimating Processes
-7-
Estimate Framework
Estimates may be structured in a number of ways, but typically they will either reflect the
wbs or the cbs.
Quantity
The quantity of each item.
Units of Measurement
The unit by which the item has been quantified, i.e. hours, linear metres, tonnes etc.
Unit Rate
This is the monetary cost for a unit of the estimate item. It should be at the same base date
as the other rates included in the estimate and have a visible currency base.
Basis of Estimate
Indication of the source of the rate utilised, location of any working papers/ material take-
off, information on the quality of the estimate etc. (allowance, quotation etc. for the audit
trail). In addition,
This provides the basic building blocks of any estimate. It is important, therefore, that any
estimate sheet is logical in its structure and presentation and clear in its content to assist in
the scope checking and verification process. It is essential that a clear audit route is visible
for both the basis of scope, calculations and rates.
-8-
3.2 Checklists
The purpose of an estimating checklist is to ensure that no significant items have been
omitted. Different organisations will require certain items to be included for each type of
estimate. The degree of detail to which any checklist is developed should reflect the needs
of the organisations involved.
Best estimating practice dictates that an organisation should have a generic estimating
skeleton, which it can utilise as a checklist. This is likely to reflect the company code of
accounts, ensuring that historical cost data is retained in a standard structure. A generic
structure has the advantage of being more readily standardised than work breakdown
structures, which are usually based upon programme activities. However, in both cases the
use of checklists allows for the management of estimate preparation in a consistent manner
across an organisation.
Performance efficiencies for labour and location factors need to be applied to labour and
material for specific locations on plant (e.g. working at height). In the case of international
estimating, the labour rates and productivity must reflect the country in which the work is
to be undertaken. Whilst there are a number of published lists of location factors for
specific countries (e.g. Richardson International Construction Factors) it is important that
location factors used be applicable to the industry in question. The estimator may have to
gather information for the location in question.
It is likely, as with other parts of the estimate, that assumption made on exchange rates will
turn out to be wrong, for reasons that are beyond the control of the estimator. Uncertainties
surrounding these assumptions are often better managed through contractual processes and
should not adversely affect the perceived quality of the estimate.
-9-
It should be clearly understood who is to carry the exchange rate risk. Organisations may
operate in different currencies and be better placed to undertake risk mitigation exercises
such as currency hedging (taking out an option to buy currency at a later date at a fixed
rate), forward buying and taking out currency insurance. Whatever strategy is adopted, the
estimate should reflect it and the risk assessment should attempt to quantify the residual
risk.
A cash flow will illustrate the timing of payments both in and out of the company/project.
One key purpose of this profile will be to reveal the extent of any financial exposure. This
will help determine the level of financing required to execute the project.
The expenditure phasing, also known as ‘Value of Work Done’ (VOWD), shows the value
of work achieved with time. At the start of the project it will reflect the planned physical
progress of the work and should be consistent with the project schedule.
Cash flow and expenditure phasing will complement each other, with differences between
the two indicating the accrued value of a project.
When calculating any escalation provision for an estimate the expenditure phasing profile
should be used as this reflects the actual timing of the work being carried out.
Financing
Any provision required for project financing may be evident from a review of the two
profiles of cashflow and expenditure.
Where payments to a Contractor are made according to actual progress, the Contractors
financing costs will be small and particularly so where there is neutral funding. Neutral
funding is where an Owner makes payments to the Contractor in line with expenditure.
This makes the Owner totally responsible for the funding of a project.
If payments by an Owner are to be based upon milestones achieved, the Contractor may
require project financing. As part of the cost risk analysis, if carried out, consideration
should be given to the possibility of a milestone missing its schedule, thereby incurring
additional finance charges.
- 10 -
3.5 Escalation Forecasting
In order to estimate the out-turn cost of a project it is necessary to take into account the
potential impact of cost inflation over time. For example: -
• A significant time gap may exist between completion of an estimate and
commencement of the project. During the intervening period there may be general or
specific changes in cost due to market activity.
• A contract might include an escalation clause allowing the cost to change in line with
general increases in labour costs.
• When evaluating the economic value of a project, the full revenue earned and cost
incurred need to be incorporated.
It is therefore important that each estimate clearly identifies the period of time for which
the estimate is valid. Normally qualifying the estimate as 4Q99, mid 2000, current or real
(now) satisfies this requirement. When the estimate has taken escalation into account it will
be reflected by the timetable for which it is valid. Terms used to reflect this can be
‘escalated’, ‘inflated’, ‘MOD’ (money of the day), ‘outturn’ or ‘nominal’. Whatever terms
are used, they should be consistent across the company.
When calculating escalation it is important to: -
• Identify and isolate those items within the estimate that will be subject to inflation.
Establish whether vendor and/or subcontractor quotes are fixed. If not, establish
which elements are to be escalated, and when from.
• Escalation is rarely applied to the whole estimate. Where it is the estimate might be
raised to a value higher than necessary.
• Utilise published indices, industry or trade specific. Within larger companies there
may also be internal inflation factors applicable.
• Where standard factors are not applicable, be clear about assumptions made.
3.6 Allowances
When preparing an estimate, items may be classified into one of the two categories,
namely:-
1. Known Items – These estimates are for firm, identified scope, and the values may be
based upon measured or calculated quantities.
2. Unknown Items – These estimates are a provision or allowance for scope which
cannot be quantified but which is firmly believed to exist. The values may be based
upon historical precedent, ratios, factors or even personal judgement. The estimate
methodology should clearly state the reasons for and the basis of the assumptions
made. Sometimes this category is called ‘Known Unknowns’.
- 11 -
In addition to the specific items above, there is also a need to consider and if necessary
provide for the uncertainty inherent with all estimates. This final allowance is usually
called Contingency or Unallocated Provision; this, by definition, is an unspecific provision
for the whole estimate and generally covers for minor errors or omissions, as well as the
uncertainty associated with quantities, unit rates and productivity. It is customary to show
this as a percentage, and the actual amount will reflect the uncertainty and confidence in the
estimate. The level of contingency can be determined objectively by the estimator or can be
calculated using probabilistic methods which are described in more detail in the following
section.
Whichever method is used, the level of contingency should take into account the degree to
which the estimate is based upon known and unknown items. Contingency is not meant to
provide for the ‘unknown scope’, neither should allowances be introduced to cover for
uncertainty. Care must therefore be taken by the estimator to avoid any overlap of
provision, and a blurring of distinction between the different categories of allowance.
An estimate should not be treated by its customers as an exact value, but rather a set of
values within a range of possible outcomes. One of the last and most important roles of the
estimator is to ensure that the customer appreciates the range of potential outcomes.
One method by which a range of potential outcomes is calculated is the probabilistic risk
analysis. This requires the estimator to address the uncertainty for items within the estimate
such as quantities, unit rates, productivity ratios, levels of allowance, etc, by producing a
range of potential outcomes for each item. These are called distributions and can be both
above and below the initial values used in the estimate. Depending upon the level of
sophistication of the risk model, and the application used, it is possible to include
correlations between separate estimate items. This allows trends between similar or related
estimate items to be taken into account by the risk analysis.
The result of the risk analysis is a range of potential outcomes for the total estimate and is
best represented in graphical format either as a histogram or cumulative probability S curve
(as shown below). The S curve shows both the range of outcomes and their relative
probabilities of occurance. From the example shown, there is a 40% chance that the
outcome will be 125 or less; likewise, there is nearly 100% chance that the outcome will be
150 or less.
- 12 -
For the purposes of presenting an estimate to a customer, it is common practice to submit
the estimate at standard levels of probability such as P50, P10 or P90. A P50 value has a
50 % chance of being either overrun or underrun, whereas a P90 has a 90% chance of
coming in at that level or under.
Probabilistic Risk Analysis is a powerful tool when used correctly but it is not suitable for
all estimates, particularly where there is little detail available, such as in early conceptual
estimate. The VEP on Risk Management should be consulted for a fuller description of the
processes involved.
100.0
80.0
60.0
Increasing
confidence 40.0
of not
exceeding 20.0
the value
shown
0.0
75.0 100.0 125.0 150.0 175
Value
4 Retrospective Review
To reduce uncertainty cost estimators must focus on providing the most cost-effective level
of service in support of the project decision making process.
A key aspect of reducing uncertainty is to look back at what has been previously been
achieved, both internally within one’s own organisation and externally via benchmarking
and continuous improvement processes. However, it is important to recognise that this
process necessarily focuses on what could have been achieved whilst possibly ignoring
what should have been achieved – narrowing perspective and restricting extraordinary
performance.
- 13 -
• A change control procedure for identification, categorisation and evaluation of
changes to the original project scope on which the estimate was based.
• A record of significant events that impacted on out-turn cost.
• Access to those involved on the project to capture circumstances that may be difficult
to document.
Having captured issues at the project level a summary review of the whole portfolio of
projects is needed to determine significant trends. This then needs to be disseminated to all
concerned in order that systems and processes can be updated.
This process should be repeated frequently in order to determine whether improvements are
being achieved.
Views on successes and failures are collected through a series of interviews, sometimes
undertaken by an external independent facilitator. Care should also be taken when stripping
out exceptional costs from final cost analyses and derived cost ratios. Every project has
these costs to a greater or lesser degree and it would be unrealistic to exclude them from the
basis for judging accuracy and developing future estimates. Information on the degree of
variance, where it occurred and the reasons for it, provide valuable input for future risk
analysis studies.
When the work is completed and the final accounts settled they are recorded against the
relevant cost headings in order that final costs can be compared against the estimated costs
and a table of variances produced. This table should include the real reasons for
differences, developed by polling all involved in the project.
- 14 -
4.3 Cost Reporting
As each phase of the project progresses the actual costs involved are documented through a
translation process from estimated costs, through commitments in the form of purchase
orders, contracts or internal work instructions and eventually into invoices and receipts.
Records of each change are collated in a change register. This should include evaluation of
acceptable options, justification of the selected changes, cost and schedule impacts and
authorisation records.
This provides the basis for adjustments to the estimated final project cost as well as an
audit trail for retrospective review in line with the wbs/cbs.
The first opportunity arises on completion of the project and should concentrate on
feedback on the quality of cost estimating – how accurate was the forecast project out-turn
cost. Other cost indicators (actual cost per key cost driver, level of errors and omissions,
contingency use etc.) should also be developed at this stage. There is also an opportunity to
review the project delivery strategy to determine whether best value for money was
achieved from the chosen procurement route.
The second arises when there is a full picture with respect to plant reliability, maintenance
and operations costs. At this time there will also be feedback on the accuracy of the indices
used to forecast escalation, currency exchange rates, tender price index, product prices, etc.
- 15 -
Unfortunately, a common feature of many organisations is the tendency to finish one
project and move on to the next opportunity. Investing time and effort on closing out
should not be seen as delaying the onset of the next opportunity, rather as an opportunity to
enhance the project estimating process. Lessons learned, what went well and what did not,
increase the opportunity for continuous improvement.
Cost estimators should have a sound engineering background with relevant experience of
the industry they are working in.
- 16 -
Supporting Skills and Knowledge
1. Computer Operations
2. Measurements/Conversions/Statistics & Probability
3. Cost/Schedule Terminology/Basic Applications
4. Basic Business and Finance
5. Inflation and Discounting.
Project Management
1. Management Theory/Organisational Structures
2. Behavioural Science/Motivational Management
3. Integrated Project Control
4. Planning and Scheduling
5. Quality/Materials Management
6. Resource/Productivity Management
7. Contracts and Contract Administration
8. Social and Legal Issues in Management
9. Health, Safety and Environmental issues (HSE).
Economic Analysis
1. Value Analysis/Value Engineering
2. Depreciation
3. Comparative Economic Studies
4. Profitability
5. Life Cycle Costs
6. Time Value of Money/Engineering Economics
7. Forecasting
- 17 -
Increasingly estimators are earning due recognition and respect for the valuable
contribution that they make to project management. With this recognition however comes
responsibility and the need to ensure that estimators by their education, training and
experience deliver the highest standard of service possible. This is not just a one-off
requirement but should involve a long-term commitment to improvement and development.
Other routes to professional accreditation include Certified Cost Engineer (CCE) which is
accredited by the International Cost Engineering Council or through registration as
Incorporated, Chartered or European Engineer through bodies such as the Association of
Cost Engineers which are affiliated to the Engineering Council or the Federation of
National Engineering Associations.
Although a number of professional opportunities exist to the Cost Estimator, they are no
substitute for on the job training which will ensure that hands on experience is gained in the
field.
- 18 -
6.0 GLOSSARY OF TERMS
Value Enhancing Practices: - Those elements of project execution which offer the
greatest opportunity for performance enhancement on construction projects;
Probability Statement/Curve: - This is the output from the probabilistic Risk Modelling
process and represent the percentage confidence of achieving a range of monetary
outcomes, for instance 90% confidence in achieving an out turn value of £x or less.
Location parameters: - These are the factors that influence the estimated cost based upon
the location and manner in which the work is to be undertaken, they include working in
different countries together with performance efficiencies for (say) working at height.
Methodology Document: - This is a formal document which determines exactly how the
estimating process is to be undertaken, and defines the responsibilities for information
supply and estimate compilation.
Cost Breakdown Structure: - This is a cost classification system, which allows the total
cost to be divided into a series of cost accounts.
Work Breakdown Structure: - This is a cost classification that allows the collection of
costs against a generic work package list or set of deliverables.
Organisational Structure: - This is a cost classification structure that allows the collection
of costs against organisational responsibility assignment.
Derived Cost ratios: - These are metrics resulting from project analysis that can be utilised
to carry out sanity checks on estimates i.e. design to capital ratios.
- 19 -