Probability Theory and
Applied Statistics
Exam exercise-4: Quality Control
by
Md. Al-Helal
Part I Control Charts
Question 1
Use the data to:
Explain and illustrate
the construction of an S-
chart, where the control
limits are requested to
give average run length
equal to 1000 (set
`nsigmas` in the `qcc()`
command to the proper
value (default is
`nsigmas=3`).
library(qcc):It is R command for loading quality control charts package.
Control Charts
A certain type of integrated circuit is connected to its frame by five wires. Thirty samples of five
units each were taken, and the pull strength (Pull strength is a measure of the force required to pull an object
apart or separate its components. ) (in grams) was measured
Quality Control
Question 2
explain and illustrate the construction of a CUSUM-chart.
Explanation of CUSUM chart
• The cumulative sum chart is constructed when there is a small change
in the process mean µ.
• The cumulative sum of residuals must vary around 0 for a process.
• CUSUM chart is built to monitor operation from [Link] includes decision
interval which has a default value of 5 and standard error shift with a
default value 1.
• It has two operations namely upward and downward.
• Upward:U(i)=max{0,U(i-1)+(zi-k)}
• Downward:D(i)=max{0,D(i-1)+(-k-zi)}, k is a positive number.
• If zi<-k, i.e. zi is greater than k below 0; then D is [Link] this
happens a number of times in a row, then D grows big.
• If zi>k,i.e. zi is greater than k above 0; then U is [Link] this
happens a number of times in a row, then U grows big.
Part II Sampling Plan
Question 1
You are confronted with the demand to construct a sampling plan, that requires that
Producer: The acceptance probability should be (at least) 90%, when the failure rate is 2%.
Consumer: The acceptance probability should be (at most) 1%, when the failure rate is 5%.You are
requested to
Determination of the sample size n and the number of defective units c.
If the lot size is constant,then n and c are determined by specifying PRP (p1=.02,q2=.09) and CRP (p2=.05,q2=.01)
from the Rstudio.
Question 2
Determine a binomial sampling plan that fulfills these requirements.
For the sample size,n=504 and the acceptance number,c=14;we get the rejection number 15 from the Rstudio.
This means that,if the number of observed defectives x is less than or equal to 14,the lot will be accepted. On the other
hand, if x is greater than 14,the lot will be rejected.
Question-2:To determine the binomial sampling plan,the underneath
terms are defined:
• x~b(n,p)
• N: The number of units produced in a lot
• n: The sample size drawn from a lot
• x: The number of defective units in the sample \random variable
count
• c: The number of defective units in the lot
• P: The fraction of defective items in the lot
• If x<=c: Accept the lot
• If x>c: Reject the lot
Question 3
Display and explain the OC-curve of the plan. This includes determining the acceptance probabilities for
failure rates 2% and 5%.
OC-curve
For the failure rates 2% and 5%, we get the acceptance probabilities of the lot is 0.914434993 and 0.009787141 respectively.
This means that,if 100 lots from a process that manufactures 2% defective product are submitted to this sampling plan,we will
expect to accept 91 of the lots and reject 9 of [Link] the other hand, when manufacturers 5% defective product 5%
defective product are submitted to this sampling plan, we will expect to accept approx. 1 of the lot and reject 99 of them.
Size of n
Change of n