OCR MEI S2 2014 June — Question 2 17 marks

Exam BoardOCR MEI
ModuleS2 (Statistics 2)
Year2014
SessionJune
Marks17
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicPoisson distribution
TypePoisson with binomial combination
DifficultyStandard +0.3 This is a straightforward application of Poisson distribution with standard techniques: basic probability calculations, scaling parameters, combining independent Poisson variables, and applying normal approximation to binomial. All parts follow textbook methods with no novel problem-solving required, though the multi-part structure and distribution combination in (iv) elevate it slightly above average difficulty.
Spec2.04d Normal approximation to binomial5.02i Poisson distribution: random events model5.02j Poisson formula: P(X=x) = e^(-lambda)*lambda^x/x!5.02k Calculate Poisson probabilities5.02l Poisson conditions: for modelling5.02m Poisson: mean = variance = lambda5.02n Sum of Poisson variables: is Poisson

2 Manufacturing defects occur in a particular type of aluminium sheeting randomly, independently and at a constant average rate of 1.7 defects per square metre.
  1. Explain the meaning of the term 'independently' and name the distribution that models this situation.
  2. Find the probability that there are exactly 2 defects in a sheet of area 1 square metre.
  3. Find the probability that there are exactly 12 defects in a sheet of area 7 square metres. In another type of aluminium sheet, defects occur randomly, independently and at a constant average rate of 0.8 defects per square metre.
  4. A large box is made from 2 square metres of the first type of sheet and 2 square metres of the second type of sheet, chosen independently. Show that the probability that there are at least 8 defects altogether in the box is 0.1334 . A random sample of 100 of these boxes is selected.
  5. State the exact distribution of the number of boxes which have at least 8 defects.
  6. Use a suitable approximating distribution to find the probability that there are at least 20 boxes in the sample which have at least 8 defects.

Question 2:
Part (i)
AnswerMarks Guidance
AnswerMarks Guidance
'Independently' means that the occurrence of one defect does not affect the probability of another defect occurringE1 Allow e.g. "event" for "defect"
Poisson distributionE1 Allow Po(...)
Part (ii)
AnswerMarks Guidance
AnswerMarks Guidance
\(P(2 \text{ defects}) = \frac{e^{-1.7}1.7^2}{2!}\) OR from tables \(P(2 \text{ defects}) = 0.7572 - 0.4932\)M1 For calculation of \(P(X=2)\)
\(= 0.2640\)A1 CAO; allow 0.264; Do not allow 0.2639
Part (iii)
AnswerMarks Guidance
AnswerMarks Guidance
New \(\lambda = 7 \times 1.7 = 11.9\)B1 For mean (SOI)
\(P(12 \text{ defects}) = \frac{e^{-11.9}11.9^{12}}{12!}\)M1 For calculation with their \(\lambda \neq 1.7\)
\(= 0.1143\) caoA1 Allow 0.114 www; Note B0M1A0 for 0.114 from use of \(\lambda = 11.7\)
Part (iv)
AnswerMarks Guidance
AnswerMarks Guidance
New \(\lambda = (2 \times 1.7) + (2 \times 0.8) = 5.0\)B1* For new mean \(= 5\) seen
\(P(\text{at least 8 defects}) = 1 - P(7 \text{ or fewer defects})\)B1* For \(1 - P(X \leq 7)\) or \(1 - P(X < 8)\) seen
\(= 1 - 0.8666\)B1dep* For \(1 - 0.8666\) seen
\((= 0.1334)\) NB Answer given
Part (v)
AnswerMarks Guidance
AnswerMarks Guidance
Binomial\((100, 0.1334)\) or \(B(100, 0.1334)\)B1* For binomial
B1dep*For parameters
Part (vi)
AnswerMarks Guidance
AnswerMarks Guidance
Mean \(= 13.34\)B1 For mean (soi)
Variance \(= 100 \times 0.1334 \times 0.8666 = 11.56\)B1 For variance (soi); For \(B(13.34, 11.56)\) seen, award B0B0 unless used correctly as part of a Normal calculation
Using Normal approx. to the binomial, \(X \sim N(13.34, 11.56)\)
\(P(X \geq 20) = P\left(Z \geq \frac{19.5 - 13.34}{\sqrt{11.56}}\right)\)B1 For 19.5 seen
M1For probability using correct tail and a sensible calculation with their mean and variance; e.g. using standard deviation \(= 11.56\) or finding \(P(Z < 1.812)\) gets M0A0
\(= P(Z > 1.812) = 1 - \Phi(1.812) = 1 - 0.9650\)
\(= 0.035\)A1 CAO (Do not FT wrong or omitted CC)
# Question 2:

## Part (i)
| Answer | Marks | Guidance |
|--------|-------|----------|
| 'Independently' means that the occurrence of one defect does not affect the **probability** of another defect occurring | E1 | Allow e.g. "event" for "defect" |
| Poisson distribution | E1 | Allow Po(...) |

## Part (ii)
| Answer | Marks | Guidance |
|--------|-------|----------|
| $P(2 \text{ defects}) = \frac{e^{-1.7}1.7^2}{2!}$ **OR** from tables $P(2 \text{ defects}) = 0.7572 - 0.4932$ | M1 | For calculation of $P(X=2)$ |
| $= 0.2640$ | A1 | CAO; allow 0.264; Do not allow 0.2639 |

## Part (iii)
| Answer | Marks | Guidance |
|--------|-------|----------|
| New $\lambda = 7 \times 1.7 = 11.9$ | B1 | For mean (SOI) |
| $P(12 \text{ defects}) = \frac{e^{-11.9}11.9^{12}}{12!}$ | M1 | For calculation with their $\lambda \neq 1.7$ |
| $= 0.1143$ cao | A1 | Allow 0.114 www; Note B0M1A0 for 0.114 from use of $\lambda = 11.7$ |

## Part (iv)
| Answer | Marks | Guidance |
|--------|-------|----------|
| New $\lambda = (2 \times 1.7) + (2 \times 0.8) = 5.0$ | B1* | For new mean $= 5$ seen |
| $P(\text{at least 8 defects}) = 1 - P(7 \text{ or fewer defects})$ | B1* | For $1 - P(X \leq 7)$ or $1 - P(X < 8)$ seen |
| $= 1 - 0.8666$ | B1dep* | For $1 - 0.8666$ seen |
| $(= 0.1334)$ **NB Answer given** | | |

## Part (v)
| Answer | Marks | Guidance |
|--------|-------|----------|
| Binomial$(100, 0.1334)$ or $B(100, 0.1334)$ | B1* | For binomial |
| | B1dep* | For parameters |

## Part (vi)
| Answer | Marks | Guidance |
|--------|-------|----------|
| Mean $= 13.34$ | B1 | For mean (soi) |
| Variance $= 100 \times 0.1334 \times 0.8666 = 11.56$ | B1 | For variance (soi); For $B(13.34, 11.56)$ seen, award B0B0 unless used correctly as part of a Normal calculation |
| Using Normal approx. to the binomial, $X \sim N(13.34, 11.56)$ | | |
| $P(X \geq 20) = P\left(Z \geq \frac{19.5 - 13.34}{\sqrt{11.56}}\right)$ | B1 | For 19.5 seen |
| | M1 | For probability using correct tail and a sensible calculation with their mean and variance; e.g. using standard deviation $= 11.56$ or finding $P(Z < 1.812)$ gets M0A0 |
| $= P(Z > 1.812) = 1 - \Phi(1.812) = 1 - 0.9650$ | | |
| $= 0.035$ | A1 | CAO (Do not FT wrong or omitted CC) |

---
2 Manufacturing defects occur in a particular type of aluminium sheeting randomly, independently and at a constant average rate of 1.7 defects per square metre.\\
(i) Explain the meaning of the term 'independently' and name the distribution that models this situation.\\
(ii) Find the probability that there are exactly 2 defects in a sheet of area 1 square metre.\\
(iii) Find the probability that there are exactly 12 defects in a sheet of area 7 square metres.

In another type of aluminium sheet, defects occur randomly, independently and at a constant average rate of 0.8 defects per square metre.\\
(iv) A large box is made from 2 square metres of the first type of sheet and 2 square metres of the second type of sheet, chosen independently. Show that the probability that there are at least 8 defects altogether in the box is 0.1334 .

A random sample of 100 of these boxes is selected.\\
(v) State the exact distribution of the number of boxes which have at least 8 defects.\\
(vi) Use a suitable approximating distribution to find the probability that there are at least 20 boxes in the sample which have at least 8 defects.

\hfill \mbox{\textit{OCR MEI S2 2014 Q2 [17]}}