Estimation is the process of finding an estimate, or approximation, which is a value that is usable for some purpose even if input data may be incomplete, uncertain, or unstable. Generally, estimation involves using the estimated value to approximate the resources needed for completion of the project.
These approximate value are computed through various formal or informal processes. Any estimate that turns out to be incorrect will either return an over estimated or underestimated value. Overestimate value occurs if the estimate exceeded the actual result, and an underestimate if the estimate fell short of the actual result. An estimated value is a quantitative assessment of an attempt likely cost or outcome. Ideally people use these estimates to forecast a project’s cost, size, resources, effort, or duration.
The basis of sound project planning are accurate estimates. Various processes have been developed to aid engineers in making accurate estimates, like
- Analogy based estimation/Top down estimation
- Expert Judgment
- Compartmentalization
- Three point estimate
- Reverse Analysis
- Cost estimate
- Delphi method
- Documenting estimation results
- Educated assumptions
- Examining historical data
- Identifying dependencies
- Parametric estimating
- Risk assessment
- Structured planning
These are the techniques used for estimation of costs involved, activity durations and resources needed.
Estimation Techniques
- Top-Down Estimation – The top-down estimation technique is often used to estimates large, high-level, chunks of work, usually at the feature or function level. This approach is often used in the early stages of a project before getting the detailed information about the project work. Top-down estimation technique provides a “ball-park” estimate and is primarily used only to provide an early viewpoint into the amount of potential project work. As we further divide the project into smaller, and more workable pieces, varied estimation techniques are used where the estimates often vary drastically from the initial top-down estimate.
- Bottom-Up Estimating – The bottom-up estimation technique is often used to estimate detailed pieces of work, and those estimates are then aggregated into a total estimate for the entire project.
- Analogous Estimating – Analogous estimating is a technique that uses expert judgment and historical information of similar activities as the basis for estimating future schedule activity.
- Parametric Estimating – Parametric estimating is a technique that uses independent variables to help estimate expected project work. These independent variables may be feature definitions, design or specification documents, or any other variables that describes the scope or type of project work.
- What-If Analysis – It is a technique that is used to evaluate the effects of changing selected factors within the project schedule such as resources, scope and quality to determine what effects those changes may have on the outcome of the project.
- Three-Point Estimating – It is a technique that uses a mathematical formula to determine a weighted average of three types of estimates and uses the formula to calculate a weighted average estimate. This method is also referred to as the Program Evaluation and Review Technique (PERT).
(Estimate=[Optimistic+[4*Most Likely]+Pessimistic]/6)
- Optimistic – It is an estimate that is based on the best-case scenario for work completion, such as low risk, small chance of risk occurring and low impact.
- Most likely – It is an estimate that, given the resources available to do the work, is the most realistically expected.
- Pessimistic – It is an estimate that is based on the worst-case scenario for work completion such as high risk, greater chance of risk occurring and high impact.
Expert Judgment
Experts have specialized skills and knowledge along with years of experience in the specific domain. Experts use their long drawn experience and skills to estimate resource requirement for the project or task under consideration. Usually they are involved since project initiation and are part of project planning and implementation, so as to better apply skills and experience of experts.
Their capability helps project managers in analyzing the historical information, standard definition, evaluation of various options and zeroing on the apt option. Utilizing expert judgments usually involves brainstorming.
Brainstorming – Brainstorming is a group creativity technique by which efforts are made to find a conclusion for a specific problem by gathering a list of ideas spontaneously contributed by its members. The term was popularized by Alex Faickney Osborn in the 1953 book Applied Imagination. Osborn claimed that two principles contribute to “ideative efficacy,” these being : defer judgment and reach for quantity.
Following these two principles were his four general rules of brainstorming, established with intention to :
reduce social inhibitions among group members.
stimulate idea generation.
increase overall creativity of the group.
Go for quantity: This rule is a means of enhancing divergent production, aiming to facilitate problem solving through the maxim quantity breeds quality. The assumption is that the greater the number of ideas generate the bigger the chance of producing a radical and effective solution.
Withhold criticism: In brainstorming, criticism of ideas generated should be put ‘on hold’. Instead, participants should focus on extending or adding to ideas, reserving criticism for a later ‘critical stage’ of the process. By suspending judgment, participants will feel free to generate unusual ideas.
Welcome wild ideas: To get a good long list of suggestions, wild ideas are encouraged. They can be generated by looking from new perspectives and suspending assumptions. These new ways of thinking might give you better solutions.
Combine and improve ideas: As suggested by the slogan “1+1=3”. It is believed to stimulate the building of ideas by a process of association.
Variations of brainstorming
Nominal group technique – Participants are asked to write their ideas anonymously. Then the facilitator collects the ideas and the group votes on each idea. The vote can be as simple as a show of hands in favor of a given idea. This process is called distillation. After distillation, the top ranked ideas may be sent back to the group or to subgroups for further brainstorming. For example, one group may work on the color required in a product. Another group may work on the size, and so forth. Each group will come back to the whole group for ranking the listed ideas. Sometimes ideas that were previously dropped may be brought forward again once the group has re-evaluated the ideas. It is important that the facilitator be trained in this process before attempting to facilitate this technique. The group should be primed and encouraged to embrace the process. Like all team efforts it may take a few practice sessions to train the team in the method before tackling the important ideas.
Group passing technique – Each person in a circular group writes down one idea, and then passes the piece of paper to the next person, who adds some thoughts. This continues until everybody gets his or her original piece of paper back. By this time, it is likely that the group will have extensively elaborated on each idea. The group may also create an “idea book” and post a distribution list or routing slip to the front of the book. On the first page is a description of the problem. The first person to receive the book lists his or her ideas and then routes the book to the next person on the distribution list. The second person can log new ideas or add to the ideas of the previous person. This continues until the distribution list is exhausted. A follow-up “read out” meeting is then held to discuss the ideas logged in the book. This technique takes longer, but it allows individuals time to think deeply about the problem.
Team idea mapping method – This method of brainstorming works by the method of association. It may improve collaboration and increase the quantity of ideas, and is designed so that all attendees participate and no ideas are rejected.
The process begins with a well-defined topic. Each participant brainstorms individually, then all the ideas are merged onto one large idea map. During this consolidation phase, participants may discover a common understanding of the issues as they share the meanings behind their ideas. During this sharing, new ideas may arise by the association, and they are added to the map as well. Once all the ideas are captured, the group can prioritize and/or take action.
Directed brainstorming – Directed brainstorming is a variation of electronic brainstorming (described below). It can be done manually or with computers. Directed brainstorming works when the solution space (that is, the set of criteria for evaluating a good idea) is known prior to the session. If known, those criteria can be used to constrain the Ideation process intentionally. In directed brainstorming, each participant is given one sheet of paper (or electronic form) and told the brainstorming question. They are asked to produce one response and stop, then all of the papers (or forms) are randomly swapped among the participants. The participants are asked to look at the idea they received and to create a new idea that improves on that idea based on the initial criteria. The forms are then swapped again and respondents are asked to improve upon the ideas, and the process is repeated for three or more rounds.
Individual brainstorming – “Individual brainstorming” is the use of brainstorming in solitary situations. It typically includes such techniques as free writing, free speaking, word association, and drawing a mind map, which is a visual note taking technique in which people diagram their thoughts. Individual brainstorming is a useful method in creative writing and has been shown to be superior to traditional group brainstorming.
Question brainstorming – This process involves brainstorming the questions, rather than trying to come up with immediate answers and short term solutions. Theoretically, this technique should not inhibit participation as there is no need to provide solutions. The answers to the questions form the framework for constructing future action plans. Once the list of questions is set, it may be necessary to prioritize them to reach to the best solution in an orderly way.
Analogy Estimation
Analogous estimating applies historical data for similar tasks or project, and an estimation is based on same by experts. It is a top-down approach to estimation and is usually less accurate in estimation.
It is applied when deficient data is present but technical and program specifications are clear. Any changes or modification done on historic data so as to be applicable on the present one are called as scaling parameter. Such adjustment factors are due to the variation in size of the project, achievements, allocated budget, etc. Any variation in quantifiable parameters can be done easily but qualitative values like preference by end-user, requires expert inputs, as well.
Project managers should be aware of the factors influencing the estimation and how it links earlier project with the present project. Study and analysis of such influencing factors should be done with extreme care as any deviation would result in erroneous estimate. Hence, skilled project managers should undertake this estimation.
Analogy estimation is also used for validating estimations, made by other methods to conclude whether over or under estimation is done.
Parametric or Mathematical Estimation
Mathematical algorithms or parametric equations are used to estimate the costs of a product or project. The results of the models are typically necessary to obtain approval to proceed, and are factored into business plans, budgets, and other financial planning and tracking mechanisms.
Models typically function through the input of parameters that describe the attributes of the product or project in question, and possibly physical resource requirements. The model then provides as output various resources requirements in cost and time. Some models concentrate only on estimating project costs (often a single monetary value).
If the project consists of activities that are common to many other projects, average costs are available per unit. For example, if you ask a construction company how much it would cost to build a standard office building, the estimator will ask for the size of the building in square feet and the city in which the building will be built. From these two factors—size and location—the company’s estimator can predict the cost of the building. Factors like size and location are parameters—measurable factors that can be used in an equation to calculate a result. The estimator knows the average cost per square foot of a typical office building and adjustments for local labor costs.
Other models are used to calculate costs of equipment, which provides more accuracy than models on a project level. Usage of these models is dependent on required accuracy and the phase of the project and its estimate.
Parametric models are contrasted with the semi-parametric, semi-nonparametric, and non-parametric models, all of which consist of an infinite set of “parameters” for description. The distinction between these four classes is as
in a “parametric” model all the parameters are in finite-dimensional parameter spaces;
a model is “non-parametric” if all the parameters are in infinite-dimensional parameter spaces;
a “semi-parametric” model contains finite-dimensional parameters of interest and infinite-dimensional nuisance parameters;
a “semi-nonparametric” model has both finite-dimensional and infinite-dimensional unknown parameters of interest.
Bottom-Up Estimating
The most accurate and time-consuming estimating method is to identify the cost of each item in each activity of the schedule, including labor and materials. If you view the project schedule as a hierarchy where the general descriptions of tasks are at the top and the lower levels become more detailed, finding the price of each item at the lowest level and then summing them to determine the cost of higher levels is called bottom-up estimating.
This type of estimate is typically more accurate than an analogous or parametric estimate. The estimate can be rolled up—subtotaled—to display less detail. This process is made easier using computer software. On projects with low complexity, the cost estimates can be done on spreadsheet software. On larger projects, software that manages schedules can also manage costs and display them by activity and category.
Activity-Based Estimates
An activity can have costs from multiple vendors in addition to internal costs for labor and materials. Detailed estimates from all sources can be reorganized so those costs associated with a particular activity can be grouped by adding the activity code to the detailed estimate.
The detailed cost estimates can be sorted and then subtotaled by activity to determine the cost for each activity.
Work Breakdown Structure – WBS
It decomposes a complex task or project into manageable tasks. As discussed earlier, it involves steps to break down the given task or project.
Best Practices
The following best practices are recommended for project estimating
Expectations – Set expectations regarding estimates based on the information available to make those estimates. At the beginning of a project when information is less available, estimates are often less accurate. Later, as more information becomes available, estimates become more accurate.
Multiple Estimates – For the best possible overall estimate, use multiple estimating techniques and different sources for each estimate.
History – Maintaining and analyzing the historical accuracy of related estimates may help make future estimates more accurate.
Review – Do a sanity check on the estimates. Have others review them as well.
Document – Document all assumptions and the estimating techniques used to derive the project estimates.
