In his article “The Systems Engineering Mindset, Problem Solving and Critical Thinking,” James Lackey makes the case that the principles of Systems Engineering can and should be practiced by everyone. In this article I’d like to expand on his point by taking a Systems Engineering tool and applying it to everyday situations: Models.
While you would be hard pressed to find a spherical cow in nature, every first year engineer and physicist is familiar with the idea. The idea is that even though a cow isn’t spherical, modelling a cow as spherical is useful because the math for spheres is much easier than math for cow shaped objects.
How does such modeling apply to everyday life? At its root, you can think of it as an estimation. A good example is rounding prices. It is much easier to calculate the cost of ten $10 dollar items than ten $8.97 items.
Similarly, a building contractor can model a house as a rectangular box to determine the rough cost of building materials, even though most houses aren’t exactly rectangular boxes.
At the heart of modeling are assumptions. We assume that the cow is spherical, that $8.97 is close enough to $10, and that the house is mostly rectangular. Choosing the right assumptions is key to the process. And it’s important to be clear about your assumptions when you publish your results, either to a scientific journal or the client who is paying for the building material.
Of course, refining the model is important as well, and there are many tools available for that. A building contractor might use a Computer Aided Design (CAD) software package to better calculate the building material. The satellite systems engineer will use finite element analysis software to determine heat transfer between two electronic devices. And you can always use a calculator to determine the exact cost of the 10 $8.97 items. The spherical cow, alas, will always be spherical.
As George E. P. Box said, “… all models are wrong but some are useful.” Modeling is a valuable tool for everyone, but the key is to make the right assumptions and refine the model as needed.