2.9 Further modelling functions

2.9 Further modelling functions

IB Maths: Applications and Interpretation HL: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • Half-life: m=m0(12)t/Tm=m_0\left(\frac12\right)^{t/T} where TT is the time to halve.
  • A logarithmic model f(x)=a+bln⁡xf(x)=a+b\ln x rises ever more slowly and has the yy-axis as asymptote.
  • A sinusoidal model asin⁡(b(x−c))+da\sin(b(x-c))+d has amplitude ∣a∣|a|, period 2πb\frac{2\pi}{b} and phase shift cc.
  • A logistic model L1+Ce−kx\frac{L}{1+Ce^{-kx}} is S-shaped and levels off at the carrying capacity LL.
  • A piecewise model has no jump if both rules give the same value at the join.
  • Interpret every parameter in context, with units.

Exponential models and half-life

Exponential decay has a half-life: the time for a quantity to halve, which is constant whatever the starting amount.

For half-life TT and initial amount m0m_0, m=m0(12)t/Tm=m_0\left(\frac12\right)^{t/T}.

Example: m=80(12)t/6m=80\left(\frac12\right)^{t/6} has half-life 6 hours; after 18 hours, 80(12)3=1080\left(\frac12\right)^3=10 mg. To find a time, solve 80(12)t/6=580\left(\frac12\right)^{t/6}=5 by GDC or algebra: (12)t/6=116\left(\frac12\right)^{t/6}=\frac1{16}, so t=24t=24.

5101520253020406080xy(0, 80)(6, 40)(12, 20)(18, 10)(24, 5)m = 80(½)^(t/6)
Decay with half-life 6 hours: the mass halves every 6 hours.
  • After nn half-lives(12)n\left(\frac12\right)^n of the original

A 200 g sample has a half-life of 5 days. How much is left after 15 days?

Natural logarithmic models

A logarithmic model a+bln⁡xa+b\ln x increases without bound, but ever more slowly.

A logarithmic model is f(x)=a+bln⁡xf(x)=a+b\ln x, x>0x>0. As ln⁡1=0\ln1=0, aa is the value at x=1x=1. For b>0b>0 the function rises ever more slowly and x=0x=0 is a vertical asymptote.

Find aa and bb by substituting two points. H=2H=2 at x=1x=1 and H=5H=5 at x=e2x=e^2 give a=2a=2 and 5=2+2b5=2+2b, so b=1.5b=1.5. Then H=7H=7 when x=e10/3=28.0x=e^{10/3}=28.0 (GDC).

24681012−22468xyx = 0(1, 2)(e², 5)H = 2 + 1.5 ln x
A logarithmic model: it rises ever more slowly and never meets the y-axis.

For f(x)=3+2ln⁡xf(x)=3+2\ln x, what is f(1)f(1)?

Sinusoidal models

Radians are assumed: the amplitude is ∣a∣|a|, the period 2πb\frac{2\pi}{b} and the phase shift cc moves the graph right.

A sinusoidal model is f(x)=asin⁡(b(x−c))+df(x)=a\sin(b(x-c))+d. Radians are assumed unless a degree sign is shown.

The principal axis is y=dy=d, so the maximum is d+∣a∣d+|a| and the minimum d−∣a∣d-|a|.

Example: d(t)=2.5sin⁡(π6(t−1))+6d(t)=2.5\sin\left(\frac{\pi}{6}(t-1)\right)+6 has amplitude 2.5, period 12, phase shift 1 and range 3.5≤d≤8.53.5\le d\le8.5. Solving d(t)=7d(t)=7 on the GDC in radian mode gives t=1.79t=1.79 and t=6.21t=6.21.

24681012246810xyy = 6max (4, 8.5)min (10, 3.5)(1.79, 7)(6.21, 7)d = 2.5 sin(π/6 (t − 1)) + 6
Sinusoidal model: principal axis y = 6, amplitude 2.5, period 12.

Amplitude

  • ∣a∣|a|

Period

  • 2πb\frac{2\pi}{b}

Phase shift

  • cc, right

Principal axis

  • y=dy=d

What are the amplitude and period of f(x)=3sin⁡(2x)+1f(x)=3\sin(2x)+1?

Logistic models

A logistic model describes growth with a limit: S-shaped, slow at first, fastest in the middle, then levelling off.

A logistic model is f(x)=L1+Ce−kxf(x)=\frac{L}{1+Ce^{-kx}} with L,C,k>0L,C,k>0: a population on an island, bacteria in a dish, a seedling's height.

  • y=Ly=L is the horizontal asymptote, the carrying capacity.
  • f(0)=L1+Cf(0)=\frac{L}{1+C} is the starting value.

Example: N(t)=4801+11e−0.9tN(t)=\frac{480}{1+11e^{-0.9t}} has N(0)=40N(0)=40. N=300N=300 when 1+11e−0.9t=1.61+11e^{-0.9t}=1.6, so t=3.23t=3.23.

24681012100200300400500xyy = 480(0, 40)(3.23, 300)N = 480 / (1 + 11e^(−0.9t))
Logistic growth is S-shaped and levels off at the carrying capacity 480.

In f(x)=L1+Ce−kxf(x)=\frac{L}{1+Ce^{-kx}}, what does LL represent?

Piecewise models and continuity

Make both rules give the same value at the join, and always use the piece that applies.

A piecewise model uses different rules on different intervals.

Example: f(x)=1+xf(x)=1+x for 0≤x<20\le x<2 and f(x)=ax2+xf(x)=ax^2+x for x≥2x\ge2. The first rule gives 3 at x=2x=2, so 4a+2=34a+2=3 and a=14a=\frac14.

Check which piece applies before you substitute or solve: a value found from the wrong piece must be rejected.

0.511.522.533.544.5524681012xy(2, 3)y = 1 + xy = ¼x² + x
With a=14a=\frac14 both pieces meet at (2,3)(2,3): no jump.

Worked example

Let f(x)=1+xf(x)=1+x for 0≤x<20\le x<2 and f(x)=ax2+xf(x)=ax^2+x for x≥2x\ge2. Find aa so that ff has no jump at x=2x=2.

With a=14a=\frac14, find f(3)f(3).

Choosing and interpreting models

Match the model to the context, then interpret each parameter with units.

You may also be given an unfamiliar model: treat it like the others by substituting values, using your GDC to solve, and interpreting each parameter in context.

Give answers to 3 significant figures, state units, and comment on limits such as extrapolation beyond the data.

1234562468xyexponentiallogarithmicsinusoidallogistic
Illustrative shapes of the four model types (values are not real data): exponential, logarithmic, sinusoidal and logistic.

Exponential

  • kax+cka^x+c or kerx+cke^{rx}+c
  • Constant percentage change, half-life

Logarithmic

  • a+bln⁡xa+b\ln x
  • Growth that keeps slowing

Sinusoidal

  • asin⁡(b(x−c))+da\sin(b(x-c))+d
  • Tides, temperature

Logistic

  • L1+Ce−kx\frac{L}{1+Ce^{-kx}}
  • Limited growth

Bacteria in a dish grow quickly at first, then level off at a maximum. Which model fits best?

Try an exam question

The number of bacteria in a dish is modelled by N(t)=4801+11e−0.9tN(t)=\dfrac{480}{1+11e^{-0.9t}}, where tt is in hours. Find the initial number, state the carrying capacity, and find when N=300N=300.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Write down the half-life of the isotope.

1 mark, marked against the mark scheme

Revision notes