Appendix 3.4.5 - Project Cost Estimate and Contingency The Development of a Project Estimate

An estimate is developed by considering the scope of a given project and estimating the quantities of material and resources needed to successfully complete the project within a given schedule.

Any estimate carries risk. The allocation of allowances, escalation and contingency within an estimate and the assignment of an accuracy range to that estimate is a means by which a bidder endeavours to identify and manage the risks associated with any estimate.

  • Allowances

Allowances cover incremental resources (for example, hours and money) included in estimates to cover expected but undefined requirements for individual accounts or sub-accounts. They cover design allowance for engineered equipment, bulk material take-off allowance, overbuy allowances, unrecoverable shipping damage allowance, provisional allowances for poorly defined items and freight allowance (equipment and materials). There are two main types of allowances, assumed (based on the bidders’ perception of the project requirements) and validated or historical (based on the bidders’ estimating database).

  • Escalation

Escalation is a provision in actual or estimated costs for an increase in the costs of equipment, material, and labour from a set point in time and is due to a continuing price change over time until the completion of the project. Escalation does not cover hyper-escalation, that is escalation which is outside what is expected from published indices, Hyper-escalation should be covered by contingency and allocated based on the perceived risk.

  • Contingency

A bidder will typically include three main types of contingency in an estimate, estimate contingency, event contingency and management Reserve.

Estimate contingency is defined as a special monetary provision in the project budget to cover uncertainties or unforeseeable elements of time/cost in the estimate associated with the normal execution of a project, for example, labour rates and design development. Estimate contingency is calculated using a risk model with input from a knowledgeable team.

Event contingency is defined as a monetary provision in the project budget to cover the costs associated with the occurrence of one or more specific risks, for example incurring liquidated damages or impacts from severe weather or hyper-escalation.

Management reserve is a further contingency included based on the bidder’s management perception of the overall likelihood of the project cost and associated risks. What Contingency is not meant to cover

Contingency is not meant to replace the development of an accurate estimate commensurate with the stage of the project and the associated definition at that stage.

It is not meant to cover project scope change for example a change in pipeline throughput or terminal storage volume.

It does not cover for design allowance which should form part of the normal project estimate basis.

Contingency does not cover for management reserve or profit. These areas will also be discussed. Development of Allowances, Escalation and Contingency
  • Pipeline Materials

Most of the material qualities can be relatively easily quantified following the FEL process, the number and size of valves will be set, the location and specification of pig-traps will be defined. The associated allowances will be set based on historical data and escalation will be set based on the appropriate published indices. These do not cover the full estimate risks. The supply price that is the price at the time of purchase from the supplier is still likely to be subject to change as this often cannot be fixed until some months after the bid has been made to the developer. The risks associated with this will need to be assessed and appropriate contingency allocated.

  • Other Materials

Other materials are likely to be subject to more significant quantity variations. For example, the allowances for weight coating will cover some repair and damage and additional usage as part of the overbuy allowance. However, contingency may also be included in the estimate to allow for potential local rerouting which might be required to solve problem and undefined ground issues.

  • Construction Labour

The construction manpower estimate has many more variables. It starts with an assessment of the volume of work to perform, how many welds, how much ditch to dig etc.

Following assessing the volume of work the construction schedule is developed to meet the requirements of the bid, as described in the March chart section of the “The Road”. Resourcing by activity is then developed to achieve the required speed of production.

In generating the construction estimate many assumptions will have to be made for example how easy the soil is to dig, how much of the soil can be reused in the ditch, whether the ditch stand up without batter or stepping, and how well the ROW will stand up to multiple heavy traffic movements. All of these will be captured in the estimate basis.

An assumption is made of construction labour productivity and equipment availability rate. The weather in the construction season is reviewed and the impact on progress evaluated. Many more risks are also inherent in this estimate. (A review of the risk register will demonstrate the issues confronting the bidder.)

All elements of the buildup of the construction labour in the estimate will be reviewed and appropriate allowances, escalation and contingency included and defined in the bidder’s estimate basis. The determination of these figures can be complicated since, for example, the productivity/quality of the construction labour will not just influence the number of hours and therefore the number of people required to execute a project, it will also influence the loss and damage of materials due to poor installation or handling.

  • General

As the various areas of the estimate are developed the variability and risk in each is different. However the bidder’s estimate cannot assume that all the potential problems associated with the construction will occur on the same job, his bid price would not be competitive. Similarly it would be unwise to assume that no mishaps will occur either. A Monte Carlo analysis, or similar statistical analyses, will determine the overall level of contingency that will be required to bid a project at a level of risk that is acceptable to the bidder. What is the Estimate Range?

The range of an estimate is defined as the difference between the lowest and highest probable values of the estimate.

In single-point estimating, the estimator assigns a single cost value to the estimate. But picking a single point is equivalent to stating the project WILL cost this much and clearly does not take into account that this is an estimate with surrounding uncertainty. The single point tends to be the most likely cost in the estimator’s view, the probability of achieving this cost is not fully evaluated.

Three-point estimating allows for uncertainty around the estimated cost. To help establish the most likely value of the estimate many approaches can be used. One such approach is a risk based assessment using Monte Carlo techniques. It is normal to represent each area of the estimate as a triangular distribution.

In the example above 20 individual costs could be found for the cost of a commodity. However the estimator can idealize the cost by knowing just three points as follows


= $2,000
likely= $5,000
maximum= $11,000

Using a simulation and allowing the cost to vary between the high and low values in a random way described by the shape of the triangular distribution results in a total project cost distribution as shown in the diagram below. In this example the most likely cost (mean) or the 50/50 estimate P50 is $74.5 million. This contrasts with the base case estimate of $70.9 which was found by adding only the most likely figures together.

The above graphic represents the output of a real estimate the distribution is slightly squewed. For the purposes of the ongoing discussion this distribution will be represented by a smooth normal distribution as follows.

In a normal distribution without skew the mean, median and mode are aligned and have the same value, all equal the 50/50 or P50 probability.

A good estimate from a developer’s perspective should have equal probability of overrun and under-run (i.e., a 50% probability). This is a risk neutral approach, the assumption being that some projects will overrun while others will under-run and, in the long run, they will balance out.

The more conservative, risk-averse attitude used by companies that need to ensure each project returns a profit to their company (true for contracting organisations) normally specifies a probability of 80% or higher that the project will not overrun. This is a safer route but by specifying a high probability the required contingency (or contingency and management reserve) will increase and with it the project cost to the developer.

This results in a sub-optimal use of funds. Large contingencies on projects in the developer organisation’s project portfolio will sequester monies that could otherwise be put to productive use (e.g., funding additional projects, beefing up R&D, investing in product improvement, new equipment). This is a key reason why reduction of risk to the bidder by the provision of a good FEL and by equitable allocation of risk, as discussed in “The Road “, is beneficial to the developer. The excessive contingency is removed and the funds remain with the developer for his use. Contingency added to the bid by a bidder, due to poor project scope definition, becomes part of his bid and is lost to the developer.

Contingency is released or consumed by the project team as each of the risks is passed. It must be noted that the contingency which is determined in the development of the estimate is total required contingency. It does not reflect what is sometimes called "management reserve", a discretionary amount which is added to the estimate for possible scope changes or unknown future events which cannot be anticipated by the project team. Addition of this reserve increases in proportion to the lack of project definition and to the history the bidder has of the way in which the client manages change.

At the final management review of the estimate past project metrics are commonly used to gauge the result and to provide a reality check.

Some special risks also impact the assessment of the final project contingency. These include commercial terms of contract, for example, liquidated damages. Whilst these can play a part in a contract with well-developed conditions and FEL they are often applied without full consideration of the impact on schedule and, as such, when the bidder performs his risk analysis they are found to result in significant risk and a high probability of occurrence. In such cases the bidder adds the risk-based impact of these items to his final estimate expecting that they will be paid in full or in part. The developer has just unwittingly increased his cost for the project development. Estimate Accuracy

What does a stated estimate accuracy of 15% mean?

Any discussion of accuracy must be related to a specified confidence interval.

In the next figure the median/mean/mode cost is $200 million. The 80% confidence interval in this example (i.e., the confidence that the actual cost will fall within this range 80 times out of 100) corresponds to costs between $170 and $230 million. The difference between $200 million and $170 or $230 million is $30 million,  which is 15% of $200 million. Hence in this example the estimate of $200 million has a + or - 15% accuracy with 80 % confidence. How do we set Contingency?

Contingency is only meant to cover the project development as it has been described in the scope and basis of design, which at the current state of project definition cannot be accurately quantified, but which history and experience show will be necessary to achieve the given project scope.

There is a tendency for those not involved or unfamiliar with estimate development to view contingency as evidence that the estimator is inflating or "sand-bagging" the estimate to improve the chance of bringing in a successful project i.e. one that achieves its budgetary goals. In an effort to reduce the projected cost of a project, clients and those unfamiliar with the process often try to limit contingency to a fixed percentage of the base estimate or in some cases delete it entirely.

However, contingency forms an important and integral part of the estimate; it is not potential profit and as we will discuss later should be expected to be spent in the development of the project.