CHAPTER 4
CREATE ESTIMATES
This chapter covers the Create Estimates stage. The objective of Create Estimates is to take the planning documents and other inputs and then use estimating techniques to create the project estimate for activity resources, activity durations, and costs. Note that this chapter will focus on the different types of estimating techniques and not on industry-specific tools or models.
This chapter includes the following sections:
4.1 Introduction
With the project estimating approach now determined and documented, estimates of resources, duration, and costs can be determined. As shown in Figure 4-1, the project estimating approach, along with other inputs, is used to create a completed estimate and supporting detail regarding the basis of estimates. There are several different types of techniques that can be used to create estimates.
4.2 Inputs
In order to execute the Create Estimates stage, it is important to have the proper inputs available. Estimates, by their nature, are approximations. Therefore, the confidence levels of the estimates are determined by the quality and detail of available information and the stakeholder expectations.
4.2.1 Project Estimating Approach
The project estimating approach document, created in the Prepare to Estimate stage, outlines the overall approach, techniques used, project information assumptions, constraints, and other important information needed to create the estimates.
4.2.2 Estimating Information
Estimating techniques require information about the current project and historical data from past projects and industry benchmarks. Document analysis, planned versus actual, risk registers, and lessons learned from other completed projects are invaluable for providing insight that can be applied to the Create Estimates stage.
Current project documents provide critical information for the Create Estimates stage, such as: activity list, activity attributes, scope baseline, project schedule, human resource plan, resource calendar, risk register, enterprise environmental factors, and organizational process assets.
4.2.3 Estimators
During the Prepare to Estimate stage, the people needed to create the estimates were identified. Estimates are best produced by the people performing the work, as they will be the most likely to produce the most accurate estimates for their work and they are motivated to achieve what they have proposed.
It is often the case that two experienced people may have different levels of personal productivity and, therefore, provide different estimates of the time they would need, to complete identical activities. For example, Frederick Brooks in The Mythical Man-Month quoted studies of programmers by Sackman, Erikson, and Grant that found: “…ratios between best and worst performances averaged about 10:1 on productivity measures…” [9]. This example leads to the conclusion that individuals should produce their own project estimates. In the absence of the actual human resource assigned to the activity, someone who is very familiar with the work being performed should provide the needed estimate. This estimate can be adjusted by the project manager to allow for resource differences.
4.2.4 Enterprise Environmental Factors and Organizational Process Assets Influences
Enterprise environmental factors that can impact the project should be considered in the Create Estimates stage. In addition, any existing organizational process assets, such as estimating tools, techniques, procedures, or models should be identified and used. Estimating databases, productivity metrics, and/or published commercial information may be available which could influence or assist in the Create Estimates stage.
4.3 Activities
The PMBOK® Guide—Fourth Edition offers many tools and techniques for those processes that produce estimates. This standard is primarily concerned with techniques that are considered to be good practice on most projects, most of the time. Tools are often industry, organization, or project-type specific and are not covered in this standard. Estimating techniques can be divided into three major categories: analogous, parametric, and bottom-up as described in Sections 4.3.1 through 4.3.3.
Most of these techniques for creating an estimate for activity duration, activity resources, and costs are relatively simple and require very few details about the project, if the model is best suited for making quick and rough estimates for the project during the very early stages. Techniques that produce more accurate estimates require more detailed and more voluminous information regarding the project and take more time to develop.
Figure 4-2 highlights each technique in relation to the known project information using the WBS as the base of reference.
- Analogous techniques, also known as top-down estimating, are used when very little information is available about the project, or the new project is very similar to a previous project or the estimators have great experience with what is going to be estimated. This category of technique results in a total project estimate and is the technique of choice for early estimates where detailed information is not available.
- Parametric techniques are useful as additional information becomes available, and the details of level one and level two of the project WBS can be estimated. Additionally, analogous and parametric models are also used to make an estimate of the total duration of the project, as well as an estimate of the cost and resource utilization.
- Bottom-up techniques are applied as the estimating tool of choice when the detailed project data becomes available. Using this technique, the expenditure of every resource of every component of the project is estimated as a prelude to rolling up these estimates to the intermediate levels and to the total project. This technique will result in a transparent and structured estimate for the project, which can be tracked and managed.
4.3.1 Use of Analogous Techniques
Analogous techniques are the simplest form of estimating and are appropriate when, in the opinion of the project manager and the project management team, there is significant similarity between the proposed project and those projects that are contained in the historical database. Sometimes known as top-down estimating, analogous estimates are the least complex, and are less precise than other models.
Analogous techniques are usually used to develop early and rough estimates of the projects. Rough estimates are commonly known as order of magnitude, conceptual, or preliminary estimates. In a way, analogous models are always used in estimating, with the distinction that during the early stages, they estimate the total cost of the project, whereas when used in conjunction with the bottom-up estimate, they determine the resource utilization of individual elements.
An example of an analogous technique is a facility that is in the planning stages, which is identical to an existing senior care facility that was built with a final cost of US$4.5 million. For the purposes of the initial planning, a total project cost was determined to be between US$4 million and US$5 million. In this scenario, the buildings were considered to be identical. However, if the first facility had 344 rooms and the new facility was planned for 455 rooms, the analogous technique could be used to derive cost or duration using extrapolation. Assuming there is a direct correlation between the number of rooms and the cost, an average cost per room would be US$13,081 (US$4.5 million/344). Using this cost per room, for a 455-room facility would result in a preliminary estimate of US$5.95 million ($13,081 × 455).
Generally, analogous techniques are customized to industry sectors, using industry-specific historical data. For example, ratio estimating for a power plant might use a 30% ratio factor and construction might use a 50% ratio factor. Using historical data from the industry and the organization, the accuracy and reliability of the results will be substantially higher.
The most notable analogous techniques are ratio estimating, power-series estimating, range estimating, and three-point estimating.
.1 Ratio Estimating
This technique is interchangeably called equipment ratio or capacity factor. The premise of this technique is that there is a linear relationship between the cost of a project with one or more of the basic features of its deliverable. The basic deliverable features that will need to be quantified and used with this model are either physical attributes or performance characteristics. The ratios or factors may be derived from general industry data or from personal experience and enterprise-specific data.
An example of using generic data is a construction project, where the total cost of the project is estimated to be two times the cost of the material and embedded equipment.
Another example is when the cost for a high-level design of a system development project constitutes 30% of the total cost for the project. Secondly, the cost of human resources may constitute 50% of the cost for a construction project and 75% of the cost for a system and software development project.
.2 Power-Series Estimating
Ratio estimating presumes that there is a linear relationship between the capacity and cost, in that the ratio of two capacities is the same as the ratio of the cost of those capacities. The power series is an enhancement of the ratio estimating technique in that it presumes that the ratio of the cost of two projects is related to a value that is derived by raising the ratio of the capacities to a certain power. Likewise, the second part of this technique presumes that the ratio of the duration of the two projects is related to a value that is derived by raising the ratio of the costs to a certain power.
The graphical analogy of these two techniques is that ratio estimating can be depicted on a straight line on a linear scale, whereas power-series estimating can be depicted by a straight line on a semi-logarithm scale. In ratio estimating, the actual ratio or slope of the line varies depending upon the industry, enterprise, or project type. Likewise, in power-series estimating, the slope of its line (which is the power quotient) varies depending upon the project type.
An example of this technique is when a project manager has already built one experimental car at a cost of US$2.3 million over the course of an 18-month project. In the opinion of the project manager, the determining factor for the complexity of a car is its top speed, and the top speed of the recently built car is 214 mph. If the project manager is asked to build a car whose top speed is 250 mph, the estimate of the cost and duration for construction of the proposed car would be:
(1) Three quarter rule for the cost:
Cp = Ce (Sp/Se)0.75
Cp = 2.3 (250/214)0.75 = 2.3 × 1.123 = US$2.6 million
(2) Square root rule for the duration:
Tp = Te(Cp/Ce)0.5
Tp = 18 (2.58/2.3)0.5 = 18 × 1.06 = 19.1
Where: T = duration, C = Cost, S = Size (Capacity), and Subscripts e and p = existing and planned, respectively.
.3 Range Estimating
In order to increase the reliability and usability of the early estimates, the estimator should provide not just the most likely value but rather the full range of all possible values for the final cost of the project. One subtle feature of this technique is that the accuracy of the estimate is embedded in the estimate.
For example: the project team offers the following range for the cost of a new graphic interface software enhancement: 3 to 8 worker months and US$150 thousand to US$320 thousand.
.4 Three-Point Estimating
This technique is a more sophisticated form of the range-estimating technique, in which three separate values for the cost (or duration) of the project or the cost of individual elements of the project are provided: optimistic, pessimistic, and most likely. With three values for the estimate of the cost of the project or of the individual elements, a certain degree of clarity is added to the estimate. This technique can also serve to normalize somewhat subjective data and potentially regulate overly optimistic estimator input. In order to temper the estimates, which may be overly optimistic, an adjustment routine using historical data can be used.
Program evaluation and review technique (PERT) analysis uses a statistical probability outcome to calculate an expected proposed value based on a weighted average of the three values. It weights the most likely data by four times its value to reinforce its significance.
PERT formula:
E = (O + 4ML + P)/6
Where: E = effort, O = optimistic, ML = Most Likely, and P = pessimistic, respectively.
The premise is that all estimates are forecasts with some uncertainty. A weighted average of the expected range of durations, work, or costs is a better predictor than a single most likely estimate. Project estimators tend to be overly optimistic. Using the PERT formula, the calculation may provide a result that is statistically more accurate. An example of this technique is when the project manager offers a most likely estimate of US$100 thousand for developing a new medical testing device. To put the estimate in perspective, the project manager will highlight that, depending upon the materialization of identified project risks, the final cost could be anywhere between US$66 thousand and US$210 thousand.
Using the PERT formula, where O = 66, P = 210, and ML = 100, the project manager could propose a PERT value of US$113 thousand. The wider the range between the pessimistic and optimistic values, the greater the value is in using this technique.
4.3.2 Use of Parametric Techniques
Parametric techniques use statistical relationships between historical data and other variables (e.g., square meters in construction) to calculate an estimate for an activity cost, duration, or resource.
Activity durations or costs can be quantitatively determined by multiplying the quantity of work to be performed by the historical duration or cost of that work. For example, a laborer may take 4 hours to dig a 3-ft deep ditch that is 10 ft long. Using that same resource, it will take 8 hours to dig a 20-ft long ditch of the same depth. To apply that to cost estimates, if the labor rate for that resource is US$30 per hour, the labor cost for the 20-ft ditch is US$240.00.
The data that forms the basis for calculations, the formula that is used on those calculations, and the manner in which the model gets updated is often proprietary to the organization. There may also be factors, such as environmental or technical complexity or project risk, which may be applied to the overall estimate.
Using the parametric estimate, the estimator enters the relevant inputs, which are different from model to model. Then the model will output values of cost, schedule, and resources for the project. The output is also different from model to model.
4.3.3 Use of Bottom-Up Techniques
The most accurate and most reliable estimate for a project can be developed with the use of a bottom-up estimating technique. Prerequisite to a bottom-up estimate is a detailed WBS, activity list, and a comprehensive directory of project resources. As defined in the PMBOK® Guide—Fourth Edition, the WBS is a deliverable-oriented hierarchical decomposition of the work to be executed by the project team in order to accomplish the project objectives and create the required deliverables. It organizes and defines the total scope of the project.
The project estimate is derived from the summation of detailed estimates for all of the individual constituent components of the project. Using this technique, a detailed resource-by-resource estimate of the project can be performed. One of the many advantages of the resource-by-resource estimate is that it sets the foundation for an informed change management on occasions when the project is constrained by a shortage of resources, or conversely by a client request for a shorter project duration.
The process begins by determining what resources are required to implement a specific lowest-level element of the WBS, known as a work package. Although, in some industries, some of the resources are not explicitly mentioned as part of the resource allocation or as part of the estimate, the most logical approach would be to list all of the resources that are required regardless of the source of funds and mechanisms of payment. The list of resources can include all costs, labor, and other assets such as materials, embedded equipment, supplies, facilities, or implementation tools.
To summarize, there are many different types of resources that are required for a given project:
- People resources or labor;
- Tools, facilities, and equipment for the workers;
- Embedded materials and equipment; and
- Licenses, fees, bonds, and insurance, etc.
Using the list of resources, the estimator assigns the needed resources to each work package. For each resource, the estimator provides the optimum crew size for a particular function and the amount of time required by the optimum crew to craft that specific work package (intensity). For example, a work package could require three programmers (intensity) for five days (duration) and seven electrical engineers for one day. Then the cost of the work package is calculated by summing the product of intensity, duration, and unit price of the resource.
Rolling up the resource estimates, duration of use, and cost of those resources to the intermediate levels of the WBS, and ultimately to the top of WBS, would provide a resource-by-resource utilization estimate and the cost of those resources for the entire project.
In bottom-up estimating, at both the activity and work package level, three-point estimating and/or PERT techniques can be effectively deployed in place of single-point estimates. Once the estimates are aggregated up to the project level, a more accurate and clear depiction of the total work, time, and cost required for the project can be determined. This allows for more detailed tracking and overall project control.
In summary, each of these techniques can be used alone or in combination and at any level of the WBS and activity.
4.4 Outputs
Two outputs at the end of the Create Estimates stage would be:
4.4.1 Completed Estimates
Estimates for activity duration, activity resources, and costs are the key outputs of the Create Estimates stage. The proposed values are a prediction of the project outcome and not definitive values. Therefore, it is considered to be a living document that is managed and controlled throughout the project life cycle.
4.4.2 Basis of Estimates
The amount and type of supporting details for the estimate vary by application area. Whatever the level of detail, the supporting documentation provides a clear and complete understanding of how the estimate was derived. This can include assumptions made, factors or units used, or comparative information. Identification of risks should also be included.
4.5 Considerations
When creating the estimate, there are a few considerations that should be made by the project manager, estimators, team, and stakeholders:
4.5.1 Optimism in Estimates
Estimates need to be tempered with reality when creating project estimates. For example, there is often rework, which can impact project commitments if not originally accounted for in the estimate. Evidence of this optimism in planning and estimating is proven by the number of projects that overrun their cost and time estimates against the number of projects that underrun their cost and time estimates. For project management purposes, success is often considered to be achieved when the project meets its budgetary and schedule estimates, while delivering the full scope. Project stakeholders, such as customers and sponsors, need to also think realistically.
4.5.2 Progressive Elaboration of Estimates
Estimates should be refined during the course of the project to reflect additional details as they become available. The accuracy of a project estimate will increase as the project progresses through the project life cycle. Therefore, project estimating becomes an iterative process. The estimates serve as the basis for the project budget; therefore, the budget should have the ability to adjust and fine tune, such that the current budget is realistic and in sync with the revised estimates.
4.5.3 Cost-Schedule Interrelationship
Development of the baseline estimate of resources and costs is predicated on certain assumptions about the pace of the project. The cost of the project is often impacted by the duration of the project; therefore, the project manager should always be sensitive to the fact that virtually any change in project schedule and the amount of work completed in the project could trigger a change in costs for the project. For example, there could be an increase in costs for the project if a shortage of resources forces a longer duration for the project, or if the urgency of the deliverable forces a shorter duration for the project. Earned value is one method for monitoring this relationship.
4.5.4 Normalization of Historical Data
If the historical data that is available to the project managers cover projects that were completed for several years and in many different locations, then the data needs to be normalized for time and for location before it is used in the estimating model. The adjustment needs to be performed in the light of the differences between the proposed project and those projects that provided the basis for the model. The indices to be normalized are the time at which the project was completed and the location where the project was executed.
The adjustment of the input data for the models can be conducted by a simple ratio adjustment. The time-adjusted cost of a project is determined through a multiplication by the inflation or escalation rate. Then, the time-adjusted cost is adjusted again by another multiplier in order to account for differences in project costs in the two locations. For example, labor rates, productivity, and work practices may vary by location. The time adjustment multiplier is almost always larger than one, although the location adjustment multiplier may be either larger or smaller than one.
4.5.5 Accounting for Time
If the project being estimated spans several years, the estimator or estimating models should consider factors such as inflation in their models. This can include wages, resource costs, and materials. The project estimating team could consider inflators; however, deflators are also an element for consideration because “falling prices” could make a bid uncompetitive.
4.5.6 Expert Judgment
Project estimators should consult with one or more experts in order to validate the calculated estimate against the experience and understanding of the experts. If a large number of experts are consulted in this manner, then statistical analysis and refinement techniques can be employed to enhance the results.
Obtaining the opinion of an expert is similar to using a parametric technique for validating the estimate. In this case, the parameters of the parametric technique are intuition and experience.
A variation of expert judgment is known as the Delphi method, in which more than one expert is consulted, and the collective judgment of these experts undergoes several refinement iterations by the same experts. These experts work anonymously with one another in order to eliminate bias.
Expert judgment is used to check the validity and reasonableness of an estimate. However, this method is most useful and most effective when applied during the early stages of project estimating.
4.6 Summary
The project estimating process develops a prediction of how many resources the project will use, how much the project will cost, and how long it will take to complete.
Various methods of estimating, using analogous (top-down), parametric, and bottom-up techniques were described along with their impacts on project management. The estimates can vary significantly depending on the phase of project life cycle in which the estimate was made and the tools and techniques that were used in the process.
The estimation techniques used may differ based on the amount of time allocated for the estimating process, the amount of information available at the time of estimation, the industry, the application area, the complexity, and identified risks. Whatever techniques are used to derive the estimates, it is important that the estimating should always be performed by the people who will be doing the work or the people most familiar with the work to be accomplished.