Estimating Parametrically for Project Management

In order to properly plan projects, every project, regardless of industry, adopts a methodology to estimate time and cost. The technique is frequently employed in the engineering, construction, and life sciences sectors.

It can be challenging to determine with any degree of certainty how long a project will take and what kind of financial commitment will be necessary to make it successful when working with a project that has numerous moving parts and variables, such as construction. Due to its reliability, parametric estimation is frequently employed.

What Is Parametric Estimating?

Parametric estimating is a method to estimate cost, duration and effort on a project. It uses a set of algorithms, statistics or models to describe the project and is one of four primary methods that project managers use when estimating a project.

While usually a more accurate method, parametric estimating requires a lot of effort upfront. The method is well-established and can be found in the Project Management Institute’s Project Management Body of Knowledge (PMBOK).

It works by correlating a parameter with a cost or time value. That correlation is scaled to the size of the project; for example, if it takes one minute to hammer a shingle on a roof, it’ll take an hour to hammer in 60 shingles to the roof.

A parametric estimation requires historical data which can be many things such as old projects that are similar can be used. Market data is usually available to the public while some agencies provide statistics for benchmarking.

As common as it is to use parametric estimating, there are many ways to use it. Some use complex statistical models and perform regression analysis. It’s also possible to develop an algorithm and use it, but this is usually only done for large projects where even small miscalculations can have a big impact.

If you’re managing a smaller project, though, parametric estimating can also prove helpful. For example, you can apply the rule of three; if something didn’t occur within a situation or event, you can scale that result with 95 percent confidence that the ratio will be the same for a larger situation or event.

Once you have an accurate estimate, it informs your project plan. Project management software makes it easier to use parametric estimating to create project schedules.  

Parametric Estimating Formula

Before we get to the formula, it’s important to know that there are two types of parametric estimating results: deterministic estimates and probabilistic estimates.

Deterministic estimates are a single number for the project resources. It’s based on parametric scaling, though it can be manually adjusted to make up for differences between the current and past projects.

Probabilistic estimates are a range that comes from the probability of different costs and durations. It’s usually based on a probability density curve with three benchmarks: most likely estimate, optimistic estimate and pessimistic estimate.

While larger projects will likely require more complicated statistical models or algorithms, smaller projects follow a more straightforward formula.

a_old
E_parametric = ——————————- x p_curr
p_old

In this case, E_parametric = parametric estimate, a_old is the historic amount of cost or time and p_old is the value of that parameter in your current project.

Parametric Estimating Example

To better illustrate the parametric formula, we’ll use a parametric estimating example. To keep it simple, let’s say you’re building a dog house. After researching, you’ve determined that for a dog house covering a total area of five feet, materials and labor cost $100.

The current area for the dog house you want to build is 10 feet. First divide 100 by five, which is the cost of the historic model and its total area. That gives you 20. Now you multiply 20 by the size of the current space for the dog house, which is 10. That gives you 200.

The new dog house will cost ~$200 to build which is the parametric estimate. Although we kept the example simple, projects are rarely so straightforward. It will take more research to determine the various historical data you’ll need to have an accurate estimate.

Parametric Estimating vs. Analogous Estimating

As we’ve defined above, parametric estimating is a technique that uses algorithms to calculate cost or duration with the help of historical data and project parameters. There are many different types of estimating methods and many of them incorporate historical data.

For example, analogous estimating is also used to forecast the duration or cost of an activity or project by the use of historical data from like-minded projects of the past. It also uses parameters from previous, similar projects to estimate a future project.

Even though both are based on historical data, analogous estimating is considered a top-down approach and therefore less accurate than parametric estimating. While parametric is more accurate when comparing projects or activities that are similar and can be scaled, it requires more effort upfront.

Advantages of Parametric Estimating

Parametric estimating has clear benefits, but let’s take a moment to explore those advantages more closely.

Accuracy

Accuracy is likely the most important benefit when talking about an estimating technique. This is done by referencing years of historical data to improve the accuracy of parametric estimating. The more data you have, the better your estimate.

Flexibility

The parametric estimating technique is also flexible. Once a model has been established, you can reuse it for similar projects. As noted, the more you use it, the more accurate your estimation is. It’s truly a gift that keeps on giving from one project to the next.

Usefulness

Naturally, parametric estimating is useful, but that usefulness extends beyond accurate and flexible estimations of duration and cost. Parametric estimating is useful for stakeholders, team members and customers who are all more likely to be on board with your estimates knowing that it rests on past evidence. By seeing how past projects evolved, these parties can expect similar results from the current ones.

Disadvantages of Parametric Estimating

Parametric estimating isn’t without its critics. Let’s take a look at a few of the disadvantages of using parametric estimating.

Relies on Data Quality

Quality data is essential to project success. If you’re unable to gather good data, then the parametric estimating technique can’t work its magic because it’s a data-driven estimation. Without any historical data on which to build your parametric estimating, it’s not going to be useful.

Ignores Certain Types of Tasks

Parametric estimating isn’t good for every project. For example, if you’re working on a project that delivers intangible outputs such as coding, parametric estimating isn’t going to work. Each line of code requires different amounts of thought and creativity.

Time-Consuming

We’ve mentioned this above, but it’s important to note that using parametric estimating will take a lot of work and effort, even money, during the planning stage. It might not be worthwhile if you’re not managing a large, complicated project.

‍