What is PERT or 3 Point Estimating and how is it calculated?

In this article we will review and found out the process of estimating and defining the formulas for it's calculation.

Difficulties Estimating

It seems like nobody likes to estimate, enter 3-point estimating. It is no wonder, there are negative consequences, sometimes painful for “getting it wrong”.

From experience, getting it wrong comes with unpleasant conversations, “why did YOU get it wrong?” You told us a lie? This is only from the internal management of the company.

The customer conversations will be much stronger. The business case is eroding if not outright eliminated. The market delivery date similarly compromised, and can represent a lost opportunity.

From experience, many companies use the hours as part of the estimate of effort to determine if there was a business case for the effort. Will the company make money on this endeavor.

The estimate of hours will have some impact on how much time will be required to pay back development costs after production.

Additionally, estimates and costs will be important for the customers. The customer will need to ascertain the value obtained from the use. In equation form: π‘‰π‘Žπ‘™π‘’π‘’=𝐡𝑒𝑛𝑒𝑓𝑖𝑑−πΆπ‘œπ‘ π‘‘

The role of the Work Breakdown Structure

From experience, one of the problems with estimating is understanding of the size and range of the work takes some effort. There is a saying, the devil’s in the details. Estimating without this understanding is risky. That is not to suggest that estimating knowing the details is entirely safe, just more likely to be successful than without understanding what we are doing and how.

The breakdown of the scope of the project is accomplished via the Work Breakdown Structure (WBS). We should start with what is a WBS. We provide a definition from The Project Management Institute:

A WBS, as defined in the PMBOK® Guide—Third Edition is “a deliverable-oriented hierarchical decomposition of the work to be executed by the project team to accomplish the project objectives and create the required deliverables. It organizes and defines the total scope of the project.

Task Variation

Projects have a defined objective, ideally, and a collection of deliverables, with tasks identified to produce those deliverables. These identified tasks are the contents from which we will build the project schedule, an example of which is a Gantt Chart. 

Point source estimates of duration are not very helpful, and represent considerable risk to the schedule. An analog of task variation, is mechanical measurements and the variation in the range of pieces that will comprise an assembly. No doubt you have experiences that sometimes a thing takes a certain amount of time, but when you do that same thing a few days or weeks later, the time to accomplish is much different. 

When we get this wrong, the parts will not fit or consume area beyond that which we allocated. In schedule parlance, this would mean the project conclusion date is later than desired. This is the reason for PERT of 3-point estimating. 

3-Point Estimating

Task variation is connected to probability of the actual completion date. For example, if I say I can get something completed in 1 day, that has some associated probability. If I make my estimate 1 month, the probability of successfully completing the task within that time frame increases as we extend the range of dates to accomplish.

The rationale for using 3 point estimates is to address this variation in task completion.

The PERT equation is below.

𝑃𝐸𝑅𝑇=4∗ML+O+P6

Where:

ML = Most Likely = 80 hours

O = Optimistic = 64 hours

P = Pessimistic = 120 hours

𝑃𝐸𝑅𝑇=(4∗80)+64+1206 𝑃𝐸𝑅𝑇=5046 𝑃𝐸𝑅𝑇=84 hours

For this example, we see a range of durations for a specific task. We weigh the Most Likely estimation more heavily by multiplying this duration estimate by 4.

The other 2 estimates are stand alone.

The 6 in the denominator accounts for 6 unique durations.

So, we end up with an average of the 84 from the 3-point estimates. Essentially, we are making an average out of the range of estimates possible for the task we are exploring. We attempt to find a duration that fits within this range, without settling in on the most likely or optimistic. For sure there are more techniques for decision analysis and risk management.

Benefits

Developing a project schedule on single point source estimates is not the best solution as it does not account for the task variation possible. PERT accounts for some of this variation by assigning a value to that is based upon a range of possibilities rather than a single number.

Additionally, the difference between optimistic and pessimistic (task variance), can tell us something about the confidence of the person making the estimate. For example, a tight clustered estimate, implies, that there is a degree of confidence in the estimates.

The larger this difference, the less confidence in the estimates, which can be interpreted as a greater risk. π‘‡π‘Žπ‘ π‘˜π‘‰π‘Žπ‘Ÿπ‘–π‘Žπ‘›π‘π‘’=π‘ƒπ‘’π‘ π‘ π‘–π‘šπ‘–π‘ π‘‘π‘–π‘−π‘‚π‘π‘‘π‘–π‘šπ‘–π‘ π‘‘π‘–π‘

Digging into the tasks and the effort to accomplish, helps creating and managing the project schedule. We can account for Timmy being out sick, or some system malfunction when it comes to accomplishing the specific task.

Probably more importantly, is observing the range of estimates tells the project manager (provided they are paying attention) the degree of risk associated with the estimate.