Techniques for Estimating Work Times

Techniques for Estimating Work Times

With the work breakdown structure complete, Perry can now estimate the time required to complete each task. But remember, an estimate is just that, an estimate. The best Perry or any project manager can hope to develop are reliable estimates ones that offer confidence in being achievable.

The Benefits and Challenges of Estimating Work Times

Estimating the work times provides several benefits for the project manager. It gives an idea of the level of effort required to complete a project. This information then enables the project manager to produce a realistic plan based upon that effort. Estimating also helps the project manager anticipate the budget for the project. Allocated funds are largely based on the effort, or labor, to produce the product or deliver the service.

The estimate becomes the basis for developing a schedule. Hours are converted to flow time, which in turn is used, with the interrelationships among tasks, to calculate start and stop dates. Lastly, doing an estimate breeds commitment. If the people who will do the work also help make the estimates, they will feel more committed to their tasks and keep within the allotted time.

While it offers many benefits, estimating is not easy, for two reasons. First, it takes time and effort to develop reliable estimates. Many people take the path of least resistance and generate either an extremely pessimistic or an overly optimistic estimate. Good estimating requires extensive calculation and research to avoid skewing the calculated values. Second, estimating requires dealing with ambiguity. By its very nature, estimating has both knowns and unknowns. The unknowns can generate fear or cause people to react out of ignorance. Either way, confidence in the resulting estimate is low.

Types of Estimating Techniques

Perry and his team can use one of four techniques to estimate the time it will take to complete each task:

Scientific wildly assumed guess

Global efficiency factor

Productivity adjustment percent

Program evaluation and review, or three-point estimating, technique

Scientific Wildly Assumed Guess (SWAG)

This technique is the most popular, yet the most unreliable. The SWAG is most popular because it is quick. The estimator determines a single value for time to do the work; no long calculations are necessary. The estimator provides one figure for each task, quickly. The SWAG is also popular because it requires very little research. Often, one or two individuals can do the entire estimate. It is rarely based on in-depth analysis to derive the values.

However, the SWAG is also very unreliable for two reasons. First, it is highly subjective, based on one persons estimate of doing a task. It accounts for only a limited number of factors, relying on an hour estimate that is based on a feel for doing the work.

Second, it understates or overinflates the time. If the estimators hold themselves in high regard, then the estimate will be optimistic; if they lack confidence, then it will be pessimistic. As long as the same people do the same work, then the estimates may prove reliable. What happens, though, if someone does the work who had no input to the estimate? What happens if an obstacle arises that the new person cannot handle? Then the estimate becomes unreliable.

For these two reasons alone, Perry decides not to use the SWAG technique. Now he is thinking about using the global efficiency factor technique.

Global Efficiency Factor (GEF)

This technique is also easy to use and attempts to incorporate nonproductive time into the estimate. The estimation assumes that a person is 100 percent productive. Then the estimator accounts for nonproductive factors that are each assigned a percent relative to each other. The estimator deducts the percents from 100 percent to derive a more realistic estimate, as follows:

Task 10.4 Arrange for food and beverage Deficiency Percent to Deduct

Unsatisfactory skill level 8

Unfamiliarity with project 10

Unfamiliarity with tools 5

Lack of well-defined requirements 2

Total Deficiency 25

The GEF is not as popular but it does have its adherents. They believe that it accounts for nonproductive time and eliminates the tendency toward unwarranted optimism.

However, the GEF technique has its drawbacks. The percents to deduct are often subjective themselves, thereby skewed and subjective. The percent for each deduction will also often vary among people. Perry decides, therefore, to look at another estimating technique: the productivity adjustment percent.

Productivity Adjustment Percent (PAP)

The PAP technique attempts to do on a more global scale what the GEF does. It applies an overall productivity factor to the estimate for all tasks. For our example, we assume people are 80 percent productive:

The PAP has its adherents, for two reasons. First, it is based on historical figures. Work measurement studies are frequently used to derive the overall percent. Second, it is easy to apply this calculation. There are no percent deductions on a task-by-task basis nor any burdensome mathematical calculations.

Despite these two benefits, there are some disadvantages. The historical records are not always available to determine the productivity factor for an organization. Also, the figure is so global that it may not be relevant to a specific task. Finally, it does not account for the complexity of issues involving individual tasks. For these three reasons, Perry does not use the PAP technique. That leaves one other option: the PERT.

Program Evaluation and Review Technique (PERT)

The PERT, also known as the three-point estimating technique, uses three estimates of time to complete a task. The three estimates are the most likely, most pessimistic, and most optimistic. The most likely time is the effort (usually in hours) to complete a task under normal or reasonable conditions. The most pessimistic time is the effort to complete a task under the worst conceivable circumstances. The most optimistic is the effort to complete a task under the best or ideal circumstances. The three variables are then used to calculate an expected time to complete a task, as shown below:

Task 1.3.2.1 Determine type of entertainment/music

This estimating technique accounts for the level of effort to complete a task after accounting for all the parameters to do the work. The estimator assumes that a person is 100 percent productive during the time to complete the task. Realistically, of course, no one is 100 percent productive. Some time is inevitably spent being nonproductive, so the hour estimates are adjusted to account for this nonproductive time (e.g., telephone calls, meetings, break times). This time has no direct relationship to the work being done; the results have no impact on progress on the actual work. Below is an example of how to calculate the revised expected time:

The three-point estimating technique has its problems. For one, it is time consuming; performing the calculations can take a long time, even when a computer is used, especially for a large project. Also, it is laborious; performing the calculations requires considerable thinking and searching for reliable information. Lastly, not too many people understand the reasons for taking this approach; its underpinnings are probability and bell curve analysis, which can be intimidating and too academic for some people.

The technique, however, does offer four benefits. It forces people to think seriously about the time to complete a task; the three variables require looking at as many parameters as possible to calculate a realistic estimate. The estimate is more reliable than other estimating techniques; it accounts for many parameters to compensate for being too optimistic or pessimistic. It improves communication; discussion over the parameters that relate to each variable forces people to communicate to come to a conclusion. It identifies issues and assumptions early. When calculating each variable, people must identify issues and assumptions. To ignore issues and assumptions adds to the cost of addressing them later in the project cycle.

Of the four estimating techniques, Perry elects the three-point estimating technique. He believes that it will provide more-reliable estimates and also offer many other benefits. So, using the three-point estimating technique requires Perry to keep in mind the following points.

He must get input estimates from the people who will do the actual work. He appreciates their knowledge of what is required to complete the work. He also knows that getting their input will encourage commitment to their assigned tasks. It is one thing to follow someones dictate of the hours to do the work; it is another when he does the estimate himself.

He must look at the historical records of other projects. Rather than apply the estimating technique for all tasks, he may uncover some reliable estimates that can be reused for similar tasks. He is cautious enough to realize that circumstances are not always exactly the same from one project to another. Consequently, the reusable estimate may require revision. Also, he reminds himself that it is still a good idea to get input from people who will do the work.

He must identify and document the assumptions and parameters used to derive the estimates. Doing this is important for two reasons. First, he and others will better understand the rationale behind the estimates. Second, he can also determine what has changed since making the estimate and make revisions accordingly.

He must maintain consistency in the estimating process. He avoids using the three-point estimate technique for some tasks and not for others. Otherwise, he will mix apples with oranges, with some estimates being realistic, optimistic, or pessimistic. A lack of discipline in this regard can result in an unrealistic schedule.

He must make the estimates public. After getting feedback, he will publish the estimates. He does not hide the estimates, as though they were his personal golf scores or bowling averages. By publishing the estimates, Perry knows that people will feel a subtle pressure to use them.

He must understand that the estimates are approximations, not accuracies. Circumstances change that make estimates irrelevant. That requires reestimating during the project to increase the reliability and validity of the overall estimate. Ideally, an estimate has the highest confidence level, which becomes more possible as a project progresses through its cycle.

Factors to Consider in Drawing Up Estimates

Estimators consider many factors when performing their calculations. These include:

Availability of nonlabor support

Clarity and definitiveness of scope

Complexity of the work

Degree of available information to estimate

Degree of uncertainty or risk in achieving the outcome

Experience, knowledge, and expertise of the team members

Financial constraints on the project

History of similar work performed

Legal constraints on the project

Location of team members working on the task

Number of people assigned to the task

Number of potential interruptions

Priority of the task

Productivity of team members

Project size

Standardization of processes related to the task

Structure versus unstructured nature of the work to be performed

Whether the completion date of the task is dictated

When estimating, Perry will consider Parkinsons law. He knows that too much available time to perform a task is almost as troublesome as too little. Parkinsons law, of course, says that work expands to fill the time available. In other words, if given ten hours to complete a task when it ordinarily takes five hours, people will take ten hours. So Perry treats Parkinsons law seriously.

Perry is now ready for the next big planning action: developing a schedule.

Questions for Getting Started

Did you identify all the work package items in the work breakdown structure?

Did you select the most appropriate estimating technique? What are the reasons for your choice?

With at least the core team members, did you get time estimates for the tasks they are responsible for completing?

If you meet resistance, how do you plan to overcome it?

If necessary, how will you reevaluate estimates? Review historical records? Apply the Delphi approach?