1.1: What is modelling?

1.1: What is modelling?#

We model all the time without noticing. Choosing a route to avoid traffic, judging whether a project will finish on time, or guessing how a change in one part of an organisation will ripple through the rest all involve building a simplified picture of a system and reasoning from it. These everyday models stay inside our heads, vague and uncheckable.

A model, in the scientific sense, is a simplified representation of a system, built on purpose to answer a specific question [1]. What makes scientific modelling different from everyday intuition is that it makes the representation explicit: the assumptions are written down precisely, in code or mathematics, so their consequences can be examined and so that others can check and build on them [2][3].

The short activity below draws that line for yourself: each statement describes either the informal picture we reason with in our heads or the explicit representation a scientific model makes.

Learning Activity: Everyday intuition or scientific model?
tell what makes a scientific model different from everyday intuition.
Six statements about how a system is represented, each belonging either to everyday intuition or to a scientific model.
For each row, click Everyday intuition or Scientific model; when every row is labelled, press Check answers.

Both are simplified pictures of a system; the difference that matters is that a scientific model is made explicit enough to be examined, checked, and built upon, whereas everyday intuition stays hidden and unverifiable.

This part of the chapter builds up the foundations in three steps:

1.1.1 · From chaos to clarity

The different purposes a model can serve, and the complex systems that make simulation necessary.

1.1.1: From chaos to clarity: why model?
1.1.2 · Building blocks

The elements every model shares, and the main types of model by how they represent a system.

1.1.2: The building blocks of a model
1.1.3 · When to model

The situations in which modelling is the right tool, and when it is not.

1.1.3: When do we use models?

Throughout the chapter, modelling and simulation are treated as one connected activity: we build a model to run it, and we run it to learn something we could not work out by reasoning alone.

When is a model the right tool?#

Before walking those three steps, it helps to settle a prior question: whether a model is even needed. Not every question calls for one. Some can be answered by observing, counting, or looking something up, and reaching for a model there only adds effort and uncertainty. A model earns its place when the answer lies somewhere you cannot look directly, in the future, in a system not yet built, or in a what-if that cannot be tried for real. Telling these two kinds of question apart is a habit worth forming early, because it decides when the rest of this book is the right tool at all. In the warm-up below, decide for each task whether you could measure the answer directly or would need to build a model.

Learning Activity: Model or not?
tell what counts as a model from what does not.
Eight everyday questions, one at a time, each answerable either by direct measurement or only by building a model.
For each question, press Just measure if you could observe or look up the answer now, or Build a model if it lies in the future, the unbuilt, or a what-if, then press Next.
Question 1 of 8Correct 0   Time 0.0s

Key takeaways#

Key takeaways

  • A scientific model is a purposeful simplification whose assumptions are explicit enough to inspect and challenge.

  • Modelling is most useful when the answer cannot be observed directly, such as a future condition, an unbuilt system, or a what-if scenario.

  • A model should not replace direct observation when the answer can be measured reliably.

That dividing line is worth carrying forward as we turn to the modelling itself. Begin with 1.1.1: From chaos to clarity: why model?, which asks what a model is actually for and why complex systems make simulation necessary.

References#