Skip to content
Tutorials Center
My learning

Topics

Lesson 10 of 46 ยท The Modelling Cycle

Chapter 2 Review: The Modelling Cycle

Review the six-step modelling cycle with a one-page cheat sheet, a recap table, a full practice lap on a decay-and-dosing example and a ten-question quiz.

Chapter 2 gave you the workflow that all later chapters reuse. Whatever the equations turn out to be (difference equations, differential equations, optimization or simulation) you will start by defining the question, writing assumptions, building with balance laws and rate rules, solving, checking and reporting.

A table of the six modelling steps with the question to ask at each step and the result you end up with
The modelling cycle on one page. Print it and keep it next to your notebook.

The chapter in a table

Lesson Key idea Worked example
Overview Six steps in a loop; a crude first lap beats a perfect plan Fermi estimate of city water use
Defining the Problem and Choosing Assumptions Sharp questions, sorted quantities, written assumptions, sensitivity "When does coffee reach 60 ยฐC?"
Building the Model Balance law plus rate rules; check units and equilibrium Bathtub; Newton's cooling law
Solving, Interpreting and Iterating Solve, translate to plain words, fit, read residuals, loop Exponential cooling, two-time-scale fix

Formulas worth remembering

  • Balance law: d(stock)/dt = inflow โˆ’ outflow.
  • Newton's cooling: dT/dt = โˆ’k (T โˆ’ T_env), solution T = T_env + (Tโ‚€ โˆ’ T_env) e^(โˆ’kt), half-life of the gap ln 2 / k.
  • Bathtub: dV/dt = a โˆ’ cV has equilibrium V* = a / c and settles on a time scale 1/c.

Try it: a full practice lap

Here is a complete lap on a new problem, drug concentration, written as a demonstration of the method only (educational model, not medical advice). The question: a 500 mg dose is taken; the amount in the body decays with a half-life of 6 hours. When does the amount fall below 100 mg?

  • Define: the target is a time, in hours.
  • Assume: one dose absorbed instantly; elimination proportional to the amount (balance law with no inflow, outflow = kA).
  • Build: dA/dt = โˆ’kA with k = ln 2 / 6.
  • Solve: A = 500 e^(โˆ’kt), so t = ln(5) / k.
Python One dose: when does it fall below the threshold?

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Now iterate. What if the dose is repeated every 8 hours? Change one assumption (single dose becomes repeated doses) and rebuild by simulation. The balance law is unchanged; only the inflow now arrives in pulses.

Python Repeated doses reach a steady pattern

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The steady peak comes from summing a geometric series: just after a dose, the body holds D(1 + r + rยฒ + โ€ฆ) = D / (1 โˆ’ r) with r = e^(โˆ’kฯ„). You can verify it against the simulation: that is the "check" step, done with two different methods.

Check your understanding

Quick check ยท 10 questions

of 10 answered You scored / 10 ยท perfect!

  1. 1. Which sequence lists the modelling steps correctly?

    The loop starts with the question and ends with a report that often leads to a new question.
  2. 2. A good modelling question contains all of these EXCEPT:

    The question must be answerable, but it cannot already contain the answer.
  3. 3. The room temperature in the coffee model is an example of:

    It could vary in real life, but the model holds it fixed on purpose.
  4. 4. dV/dt = 8 - 0.2 V. What is the equilibrium V?

    At equilibrium 8 = 0.2 V, so V = 40.
  5. 5. Why check the units of every term of a new equation?

    Every term in a sum must have the same units. A mismatch signals a mistake in the model or algebra.
  6. 6. A drug has half-life 6 h. After 12 h, what fraction of the dose remains?

    Two half-lives pass, so the amount halves twice: 1/4.
  7. 7. Residuals from a fit show a clear pattern (a run of plus signs then minus signs). What does this suggest?

    Systematic structure in residuals means there is something your model does not capture.
  8. 8. Which input matters most for improving your answer, according to a sensitivity chart?

    A long bar means the answer is sensitive to that input, so errors in it hurt most.
  9. 9. With repeated doses D every tau hours and decay factor r = exp(-k tau) per interval, the steady peak right after a dose is:

    Summing the geometric series D(1 + r + r^2 + ...) gives D/(1 - r).
  10. 10. When should you stop iterating on a model?

    The purpose sets the accuracy needed. Extra complexity needs evidence that it helps.

Summary

  • The cycle (define, assume, build, solve, check, report) is a loop, and a rough first lap is valuable.
  • Balance laws plus rate rules build most dynamic models; check units and equilibrium immediately.
  • Solve exactly when you can and numerically as a cross-check; translate results into plain statements.
  • Residual patterns and sensitivity charts decide where the next lap should go.
  • Report answer, method, limits and next step.

Next: Dimensions and Scaling: Overview