Mathematical modelling

Mathematical modelling

Spec OT3.1–5 · AQA A-Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • A mathematical model translates a real situation into mathematics, using modelling assumptions about what is ignored or held constant.
  • Put in inputs to get outputs; run a model backwards to find the input for a target output.
  • Interpret constants and coefficients in context, such as £40 fixed charge and £25 per day.
  • Outputs outside the valid domain, such as a negative height, show the model is being misused.
  • Refine a model by changing assumptions, parameters or form after comparing with data.

What a mathematical model is

Identify the quantities, make simplifying assumptions, choose a relationship, and write it as an equation or inequality.

A mathematical model translates a real situation into mathematics so it can be explored.

  1. Identify the quantities involved and the question to answer.
  2. Make modelling assumptions: statements about what is ignored or held constant (for example "air resistance is negligible").
  3. Choose variables and a type of relationship (linear, quadratic, exponential).
  4. Write the model as an equation or inequality.

Example: a ball thrown upwards from 1.5 m at 12 m s−1^{-1} is modelled by h=1.5+12t−5t2h=1.5+12t-5t^2, assuming the ball is a particle and gravity is constant.

0.511.522.53246810xy(0, 1.5)(1.2, 8.7)(2.52, 0)h = 1.5 + 12t − 5t²
Height h (m) of the ball against time t (s).
  1. 1

    Identify

    the quantities and the question

  2. 2

    Assume

    what is ignored or held constant

  3. 3

    Choose

    variables and a type of relationship

  4. 4

    Write

    an equation or inequality

Building a model

Which of these is a modelling assumption for the thrown ball?

Using a model

Put in inputs to get outputs, or run the model backwards to find the input that gives a target output.

Once a model exists you put in inputs and obtain outputs.

For h=1.5+12t−5t2h=1.5+12t-5t^2: t=0t=0 gives the starting height 1.51.5 m; completing the square, h=−5(t−1.2)2+8.7h=-5(t-1.2)^2+8.7, so the maximum height is 8.78.7 m after 1.21.2 s; solving h=0h=0 gives the landing time t=2.52t=2.52 s.

For a van hire model C=40+25dC=40+25d, d=4d=4 gives C=£140C=\text{£}140. The model can also be run backwards: £240 allows d=8d=8 days. Always give outputs with units and appropriate accuracy.

05010015020025012345678Days dCost (£)
Cost C=40+25dC=40+25d for 1 to 8 days of hire.

Worked example

Using C=40+25dC=40+25d, how many days of hire can you afford with £240?

Using C=40+25dC=40+25d, what is the cost of 6 days?

Interpreting outputs and parameters

Each number in a model has a meaning in context, and outputs must be possible for the real situation.

In C=40+25dC=40+25d the constant £40 is a fixed charge and the coefficient £25 is the cost per day.

In P=200×1.5tP=200\times1.5^t, 200200 is the initial value and 1.51.5 is the hourly growth factor (a 50% increase each hour).

In N=−2t2+60t+100N=-2t^2+60t+100 for website visitors, N(0)=100N(0)=100 is the number on launch day and the maximum N=550N=550 occurs on day 15.

Check that outputs are possible: a negative number of visitors shows the model is being used outside its valid domain.

5101520253035−200−100100200300400500600xy(0, 100)(15, 550)N = −2t² + 60t + 100
Visitors N against day t: maximum on day 15, negative from day 32.

C=40+25dC=40+25d

  • £40: fixed charge
  • £25: cost per day

P=200×1.5tP=200\times1.5^t

  • 200: initial population
  • 1.5: growth factor, a 50% rise each hour

In P=200×1.5tP=200\times1.5^t, what does 1.5 represent?

Refining and evaluating a model

Compare outputs with real data; if they disagree, change the assumptions, parameters or type of function.

Compare outputs with real data. If they disagree, refine the model.

The exponential model P=200×1.5tP=200\times1.5^t predicts about 665 000665\,000 bacteria after 20 hours with no limit; a better model has an upper limit because food and space run out. A van hire model C=40+25dC=40+25d that ignores discounts can be refined: if days after the 7th cost £15, then C=110+15dC=110+15d for d>7d>7.

For website visitors the quadratic gives N(40)=−700N(40)=-700, which is impossible, so it is restricted to 0≤t≤300\le t\le30.

To evaluate a model, ask whether the assumptions are reasonable, whether outputs match data, and for which inputs it stays valid.

246810121450100150200250300350400xy(7, 215)original: C = 40 + 25drefined model
Van hire: C = 40 + 25d (dashed) is refined to C = 110 + 15d after day 7, so the cost rises more slowly from then on.

Original model

  • P=200×1.5tP=200\times1.5^t has no limit
  • C=40+25dC=40+25d ignores discounts

Refined model

  • An upper limit as food and space run out
  • C=110+15dC=110+15d for d>7d>7

The visitor model gives N(40)=−700N(40)=-700. What does this show?

Try an exam question

The height hh metres of a ball tt seconds after it is thrown upwards is modelled by h=2+10t−5t2h=2+10t-5t^2. State one modelling assumption, find the maximum height predicted, and explain why the model cannot be used for large tt.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Which modelling assumption is made in this model?

1 mark, marked against the mark scheme

Revision notes