Professional Documents
Culture Documents
CQE Study Questions Answers and Solutions
CQE Study Questions Answers and Solutions
CQE Study Questions Answers and Solutions
QReview Answers 46
INTRODUCTION
The following questions include key concepts which are representative of the CQE
examination. Each question should be worked out as completely as possible before looking
at the solution. Some questions have been taken from previously published ASQ CQE
exams.
1. Certification of a company to the ISO 9002 standard does not include which of the
following requirements?
a) Servicing
b) Design control
c) Contract review
d) Internal quality audits
4. In preparing a product quality policy for your company, you should do all of the following
except
5. What natural phenomenon created the necessity to control product and process quality?
2 QReview Study Questions
a) Gravity
b) Variation
c) Human Error
d) Management
10. Which of the following is the most important element in Statistical Quality Control?
a) document control.
b) corrective action.
c) management responsibility.
d) employee salaries.
a) A model that shows how data are distributed over a range of measurements.
b) An Analysis of Variance table.
c) A sampling plan.
d) A graph that contains data plotted on a normal curve.
14. Which of the following are two types of data used in statistical quality control?
15. The primary reason for evaluating and maintaining surveillance over a supplier's quality
program is to
a) Designer
b) Inspector
c) Machinist
d) Equipment engineer
4 QReview Study Questions
a) design requirements.
b) purchase order requirements.
c) manufacturing requirements.
d) customer use of the end product.
22. In spite of the Quality Engineer's best efforts, situations may develop in which his
decision is overruled. The most appropriate action would be to
23. If a test data does not support a Quality Engineer's expectations, the best thing to do is
a) arbitration is necessary.
b) the customer is always right.
c) good judgment should be exercised.
d) contract specifications normally apply.
28. All of the following are considerations when a total quality management (TQM) program
is implemented except
29. According to Juran, all of the following are widespread errors in perception that have led
many managers astray except
a) the work force is mainly responsible for the company's quality problems.
b) workers could do quality work but they lack the motivation to do so.
c) quality will get top priority if upper management so decrees.
d) return on investment is everything.
6 QReview Study Questions
31. An engineer has the job of providing a written plan of quality related tasks to his
manager, including a detailed timeline, for the following year. Which of the following
tools should be used?
a) Histogram
b) Flow Chart
c) Gantt Chart
d) Frequency Distribution
Basic Probability 7
BASIC PROBABILITY
32. The time it takes to answer a technical support line has a continuous uniform distribution
over an interval from 17 to 20 minutes. All of the following are true except
33. For two events, A and B, which one of the following is a true probability
statement?
a) 1/16
b) 1/4
c) 1/2
d) 1
35. What is the probability of getting a head and a tail in 2 tosses of a fair coin? And, what is
the probability of getting a head and a tail, in that order, in 2 tosses of a fair coin?
a) 1/2, 1/2
b) 1/4, 1/4
c) 1/2, 1/4
d) 1/4, 1/2
36. A coin is tossed 10 times. The first 9 tosses come up heads. What is the probability that
the 10th toss will come up heads?
a) 1/512
b) 1/256
c) 1/32
d) 1/2
37. What is the probability of obtaining exactly 2 heads in 4 tosses of a fair coin?
a) 1/4
b) 3/8
c) 1/2
d) 1/6
38. What is the probability of getting a 3 when rolling a single die? ( A die is one of a pair of
8 QReview Study Questions
dice)
a) 1/5
b) 3/5
c) 1/6
d) 1/3
39. What is the probability of getting an odd number when rolling a pair of dice? (Spots on
the two dice sum to odd number)
a) 1/4
b) 1/2
c) 1/3
d) 3/10
40. What is the probability of obtaining a sum of 7 when rolling a pair of dice?
a) 1/5
b) 3/5
c) 1/6
d) 1/3
Use the following information to answer questions 41, 42 and 43. The probability is 1/2 that
Bob will pass the CQE exam, 1/3 that Amy will pass and 3/4 that Jon will pass.
41. What is the probability that Bob, Amy and Jon will all pass the exam?
a) 1/8
b) 4/9
c) 4/11
d) 1/3
42. What is the probability that neither Bob, Amy nor Jon will pass the exam?
a) 1/9
b) 7/8
c) 1/12
d) 2/3
43. What is the probability that only one of the three will pass the exam?
a) 1/4
b) 1/3
c) 3/4
d) 3/8
44. Four people shoot at a target and the probability that each will hit the target is 1/2 (50%).
What is the probability that the target will be hit?
Basic Probability 9
a) 1/16
b) 15/16
c) 1/2
d) 1/4
45. A committee of 5 people is chosen at random from a room that contains 4 men and 6
women. What is the probability that the committee is composed of 2 men and 3 women?
a) 1/2
b) 10/21
c) 5/21
d) 1/3
46. A vendor is trying to sell you a box of 50 fuses that contains exactly 5 defective fuses.
You select 2 fuses from the box for testing. If both are good you will buy the entire box. If
one or both are defective, you will not buy the box. What is the probability that you will
buy the box?
a) .7533
b) .8082
c) .9769
d) .8531
47. What is the probability of winning the Super Lotto? (Winning = getting all 6 numbers out
of 47)
a) 1/10,737,573
b) 1/7,731,052,560
c) 1/3,457,296
d) 1/12,966,821
48. A box contains 12 connectors, 9 good and 3 defective. What is the probability of
obtaining exactly 2 good and 1 defective connector in drawing 3 parts from the box
without replacement?
a) .4219
b) .4909
c) .5022
d) .6915
10 QReview Study Questions
49. A box contains 12 connectors, 9 good ones and 3 defective ones. What is the probability
of obtaining exactly 2 good and 1 defective connector in drawing 3 parts from the box
with replacement?
a) .4219
b) .4909
c) .5022
d) .6915
50. You have been asked to sample a lot of 300 units from a vendor whose past quality has
been about 2% defective. A sample of 40 pieces is drawn from the lot and you have
been told to reject the lot if you find two or more parts defective. What is the probability
of rejecting the lot?
a) 0.953
b) 0.809
c) 0.191
d) 0.047
Use the following information to answer questions 51 and 52. A company produces
capacitors by a process that normally yields 5% defective product. A sample of 4 capacitors
is selected.
a) .9790
b) .9213
c) .8617
d) .8145
a) .1383
b) .1855
c) .0000258
d) .00000625
Use the following information to answer questions 53, 54, 55 and 56. A company makes ball
bearings that are found to be 10% defective in the long run. A sample of 10 bearings is
selected.
a) .3487
b) .3874
c) .4126
d) .1110
54. What is the probability of obtaining exactly 1 defective bearing?
Basic Probability 11
a) .3487
b) .3874
c) .4126
d) .2574
a) .0574
b) .4448
c) .7361
d) .1562
a) .3874
b) .4126
c) .2639
d) .2285
57. How many defective connectors would be expected in a sample of 200 parts if the
process averages 2% defective?
a) 1
b) 2
c) 4
d) 7
58. What is the probability of obtaining exactly 2 defective connectors in a sample of 6 parts
if the process averages 2% defective?
a) .0135
b) .0055
c) .0009
d) .0001
Use the following information to answer problems 60, 61 and 62. A company produces
integrated circuits (chips) by a process that normally yields 2000 ppm defective product for
12 QReview Study Questions
electrical test requirements (ppm = defective parts per million). A sample of 5 chips is
selected and tested.
a) .9900
b) .9603
c) .9213
d) .8563
a) .0051
b) .0009
c) .0269
d) .0100
a) .01931
b) .00510
c) .00008
d) .01000
Use the following information to answer problems 63, 64 and 65. A capability study was
made to determine the defective rate of 28AZ transistors. The study showed the rate to be
5000 ppm. Ten of the transistors were shipped to a customer.
63. What is the probability that the shipment contains no defective transistors?
a) .9511
b) .9066
c) .8512
d) .9213
64. What is the probability that the shipment contains exactly 1 defective transistor?
a) .0001
b) .0478
c) .1048
d) .1165
65. What is the probability that the shipment contains 2 or more defective transistors?
a) .0001
Basic Probability 13
b) .0478
c) .0237
d) .0011
Use the following information to answer problems 66, 67 and 68. A circuit board operation
yields 2 defects per board on the average. A sample of 1 board is selected at random.
66. What is the probability of finding exactly 2 defects on the selected board?
a) .3522
b) .2706
c) .1550
d) .0295
67. What is the probability of finding less than 2 defects on the selected board?
a) .4060
b) .6352
c) .3522
d) .3849
68. What is the probability of finding more than 2 defects on the selected board?
a) .4060
b) .2706
c) .3522
d) .3235
Use the following information to answer problems 69, 70 and 71. In manufacturing material
for automobile seats it was found that each 100-foot roll contained, on average, 2 defects
(flaws). A sample of 1 roll is selected at random from the process.
69. What is the probability that the selected roll contains 0 defects?
a) .1353
b) .2707
c) .8647
d) .7293
70. What is the probability that the selected roll contains exactly 1 defect?
a) .1353
b) .7293
14 QReview Study Questions
c) .8647
d) .2706
71. What is the probability that the selected roll contains more than 1 defect?
a) .3233
b) .5941
c) .7293
d) .8647
Use the following information to answer problems 72 and 73. A firm that makes T-shirt
decals has determined that their process yields, on average, 3 defects per day. Fifty decals
are inspected each day.
72. What is the probability of finding exactly 2 defective decals in any given day? (Assume
one defect per defective decal.)
a) .7361
b) .1494
c) .2240
d) .0746
73. What is the probability of buying a decal that contains more than 1 defect?
a) .0005
b) .0042
c) .0001
d) .0017
74. A parts dealer buys parts from a warehouse. Parts are made by either Company A or
Company B but are not identified as to which company produces them. One company
produces all parts in one shipment or lot. On the average, we know:
Company A produces 2.5% defective parts.
Company B produces 5.0% defective parts.
The warehouse states that 70% of parts will come from Company A and 30% from
Company B. If the dealer selects 4 parts at random from a lot and finds 1 defective part,
what is the probability that the lot was produced by Company A?
a) .4422
b) .5580
c) .6915
d) .3085
Sampling 15
STATISTICS
75. What is the expected value of the random variable x for the following data?
x f(x)
12 0.2
10 0.5
14 0.1
20 0.2
a) 13.6
b) 14.0
c) 12.8
d) 14.5
76. In the standard normal table, what value of z has 5% of the area in the tail beyond it?
a) 1.960
b) 1.645
c) 2.576
d) 1.282
77. Which distribution should be used to determine a confidence interval when σ is not
known and the sample size is 10?
a) z
b) t
c) F
d) χ2
78. Which of the following methods should be used to test 6 population means for statistical
significance?
79. A sample size of 120 is taken from a process and is represented graphically on a
histogram. What is the appropriate number of histogram cells to use?
a) 1 - 8
b) 9 - 20
c) 21 - 35
d) 120
80. Which of the following conditions makes it possible for a process to produce a large
16 QReview Study Questions
81. For the normal probability distribution, which of the following is true about the
relationship among the median, mean and mode?
82. All of the following statistical techniques can be used to determine the effectiveness of a
supplier improvement program except
a) Pareto analysis.
b) x bar and R charts.
c) a PERT chart.
d) a flow chart.
83. A sample of n observations has a mean x and a standard deviation s > 0. If a single
observation, which equals the value of the sample mean x , is removed from the sample,
which of the following is true?
84. The factory installed brake linings for a certain kind of car have a mean lifetime of 60,000
miles with a 6,000 mile standard deviation. A sample of 100 cars has been selected for
testing. What is the standard error of x ? (Assume that the finite population correction
may be ignored.)
a) 60 miles
b) 6000 miles
c) 600 miles
d) 6100 miles
Use the following information to answer problems 85 - 90. A sample of 7 rivets was taken
from a shipment of 1000 rivets and the length was measured. The following data are
Sampling 17
obtained:
a) 3.20 inches
b) 3.36 inches
c) 4.00 inches
d) 3.65 inches
86. What is the standard deviation of the length of the rivets (estimate of population standard
deviation)?
a) 0.27 inches
b) 2.16 inches
c) 0.29 inches
d) 2.00 inches
Note: In the following 4 problems, the sample sizes are less than 30 and the t statistics
should be used to solve the problems. Analyses of this type usually involve sample sizes of
30 or greater. Handle the problems just as if the sample sizes were greater than 30 and use
the z statistics.
87. What percentage of rivets have lengths less than 2.80 inches?
a) 2.69%
b) 5.00%
c) 1.22%
d) 3.23%
88. What percentage of rivets have lengths greater than 3.65 inches?
a) 17.1%
18 QReview Study Questions
b) 14.2%
c) 15.9%
d) 7.10%
89. What percentage of rivets have lengths between 3.1 inches and 3.9 inches?
a) 89.7%
b) 78.4%
c) 52.5%
d) 99.7%
90. In the shipment of 1000 rivets, how many good parts will we find if a good part is defined
as having a minimum of 3 inches and a maximum of 4 inches?
a) 999
b) 967
c) 912
d) 878
The following information is used to answer problems 91 - 95. Data are taken from a
manufacturing process that produces optical glass. The sample size is 5 parts and the
characteristic measured is the diameter of the plates.
1 30
2 31
3 29
4 33
5 34
a) 31.4 mm
b) 29.0 mm
c) 31.0 mm
d) 34.0 mm
a) 1.00 mm
b) 2.07 mm
Sampling 19
c) 2.22 mm
d) 1.22 mm
93. The specifications for the glass plates are 30.5 ± 2 mm. What percentage of parts made
by this company will not meet specifications?
a) 32.5%
b) 5.00%
c) 35.0%
d) 37.9%
a) 17.9%
b) 7.21%
c) 15.9%
d) 24.3%
a) 78.5%
b) 24.3%
c) 15.9%
d) 22.1%
96. The Zoglen Corporation markets a product, which is a blend of 3 ingredients (A, B, C). If
the individual tolerances for the weight of the 3 ingredients are as shown, what should
the tolerance be for the net weight of the product?
97. A random sample of size n is to be taken from a large population that has a standard
deviation of 1 inch. The sample size is determined so that there will be a 95% chance
that the sample average will be within ± 0.1 inch of the true mean. Which of the following
values is nearest to the required sample size?
a) 385
20 QReview Study Questions
b) 200
c) 100
d) 40
98. All of the following conditions must be met for the process capability to be within the
specification limits except
a) Cpk ≥ 1.0
b) Cp ≥ 1.0
c) Cp = Cpk
d) a stable process.
99. A value on the abscissa of the t distribution is 1.093. What is the area to the right of this
value if the sample size is 11?
a) 0.30
b) 0.15
c) 0.05
d) 0.10
100. The spread of individual observations from a normal process capability distribution may
be expressed numerically as
a) 6 R /d2
b) 2 x A2 R
c) R /d2
d) D4 R
101. What percentage of data will normally fall within a process capability?
a) 99.00%
b) 99.73%
c) 1.00%
d) 0.27%
Use the following information to answer problems 102 - 105. A winding machine wraps wire
around a metal core to make small transformers. The design engineers have determined
that the nominal number of windings are to be 10,060 with a minimum of 10,025 and a
maximum of 10,095. A sample of 300 transformers was selected in a three month period
and the wire was unwrapped on each part to determine the number of windings. The results
were:
a) 10020 - 10100
b) 10052 - 10020
c) 10022 - 10082
d) 10020 - 10060
a) 30.0
b) 10.0
c) 0.67
d) 1.17
a) 0.90
b) 0.72
c) 3.22
d) 2.67
a) 1.00%
b) 0.35%
c) 2.13%
d) 0.49%
22 QReview Study Questions
STATISTICAL INFERENCE
107. In a sampling distribution which of the following represents the critical region?
a) α
b) β
c) 1-β
d) 1-α
108. In a hypothesis test which of the following represents the acceptance region?
a) α
b) β
c) 1-β
d) 1-α
a) a distribution of averages.
b) a distribution of variances.
c) a distribution of standard deviations.
d) a distribution of frequencies.
110. Which of the following is a number derived from sample data that describes the data in
some useful way?
a) constant
b) statistic
c) parameter
d) critical value
111. A null hypothesis assumes that a process is producing no more than the maximum
allowable rate of defective items. What does the type II error conclude about the
process?
112. A quality engineer wants to determine whether or not there is any difference between
the means of the convolute paperboard cans supplied by two vendors, A and B. A
random sample of 100 cans is selected from the output of each vendor. The sample
from vendor A yielded a mean of 13.59 with a standard deviation of 5.94. The sample
from vendor B yielded a mean of 14.43 with a standard deviation of 5.61. Which of the
following would be a suitable null hypothesis?
a) µA = µB
b) µA > µB
c) µA < µB
d) µA ≠ µB
113. A chi square test for independence has a 4 x 3 contingency table and a calculated chi
square value of 11.5. At a .05 level of significance, which of the following are true about
the decision regarding the null hypothesis?
a) 7 df, accept Ho
b) 6 df, reject Ho
c) 6 df, accept Ho
d) 11 df, accept Ho
114. A supplier states that the average diameter of optical glass plates is 37.50 mm. The
specifications are 37.50 mm ± .40 mm. A shipment of 500 plates is received and a
sample of 10 plates is selected and the diameters of each are measured. The following
data are obtained:
Does the entire population have an average diameter of 37.5 mm? Test at a level of
significance of .05.
a) t = +2.97, Reject Ho
b) t = +2.26, Accept Ho
c) t = -2.26, Accept Ho
d) t = +1.05, Accept Ho
Could this sample have possibly come from a process whose process average is .14
defects/unit? Test at a level of significance of .01.
a) z = +1.75, Accept Ho
b) z = -.075, Reject Ho
c) z = +1.01, Accept Ho
d) z = +2.47, Reject Ho
116. A study of 26 families across a certain state indicated that the average family income
24 QReview Study Questions
during 1989 was $33,400 with a Standard Deviation of $3680. Test the hypothesis that
the true average income in this state during 1989 was $32,000 against the alternative
that it was not $32,000. Test at a level of significance of .05.
a) t = +.900, Accept Ho
b) t = +1.94, Accept Ho
c) t = -2.95, Reject Ho
d) t = +3.16, Reject Ho
Could this data have reasonably come from a process which usually has an average of
11.3 lb. or greater? Test at a level of significance of .01.
a) t = +3.10, Reject Ho
b) t = -1.67, Accept Ho
c) t = -2.98, Reject Ho
d) t = +2.10, Accept Ho
118. In a hypothesis test what determines whether the test is one tailed or two tailed?
a) Critical Region
b) Alternate Hypothesis
c) Null Hypothesis
d) α
a) how well two sample data standard deviations vary from each other.
b) a test to determine if two or more samples follow the same distribution.
c) a statistical tool for analyzing data using the χ2 distribution.
d) the comparison of an observed sample distribution with a theoretical distribution.
Sampling 25
SAMPLING
a) AOQL
b) AQL
c) ATI
d) AOQ
122. What are the two unique quantities that determine a single sampling attributes plan?
123. Characteristics for which 100% inspection may be practicable include all of the
following except
a) that level of lot quality for which there is a large risk of rejecting the lot.
b) the Average Quality Limit.
c) the maximum percent defective that can be considered satisfactory as a process
average.
d) the quality level.
125. Which of the following best describes what an operating characteristic curve shows?
126. In ANSI/ASQ Z1.4, the AQL is always determined at what Pa on the OC curve?
a) 0.05
b) 0.10
c) 0.90
d) None of the above
127. Which of the following best describes what an average outgoing quality curve shows?
26 QReview Study Questions
Use the following information to answer problems 128 - 134. A sampling plan is needed
which will satisfy the following requirements:
a) 90%
b) 88%
c) 12%
d) 10%
a) 88%
b) 12%
c) 95%
d) 2.5%
a) 3%
b) 99%
c) 1%
d) 97%
a) 7%
b) 10%
c) 1%
d) 5%
132. What is the sample size that will satisfy the requirements stated above?
a) 35
Sampling 27
b) 40
c) 45
d) 50
133. What is the acceptance number that will satisfy the requirements stated above?
a) 1
b) 3
c) 5
d) 23
a) .0165
b) .2000
c) .1375
d) .9999
a) 30
b) 75
c) 140
d) 200
a) 1
b) 3
c) 6
d) 9
a) .2000
b) .0160
c) .0107
d) .0029
138. A sampling plan specifies a sample size of 50 and an acceptance number of 3. What is
the value of the AQL if α = .05?
a) .0165
b) .0107
28 QReview Study Questions
c) .1375
d) .0280
139. A sampling plan specifies a sample size of 50 and an acceptance number of 3. What is
the approximate value of the AOQL?
a) .0280
b) .0107
c) .0370
d) .0490
140. All of the following are methods for checking product except
141. All of the following are characteristics of lot by lot single sampling except
143. The following parameters are required for sampling plan construction using the
Binomial Nomograph except
a) AQL.
b) RQL.
c) α.
d) Sample size.
Control Charts 29
CONTROL CHARTS
a) Poisson
b) Binomial
c) Shewhart
d) Exponential
a) Poisson
b) Binomial
c) Shewhart
d) Exponential
a) Variance
b) The Law of Large Numbers
c) The Central Limit Theorem
d) The Addition Theorem
148. When used together for variables data, which of the following pair of quantities is the
most useful in preparing control charts?
a) AQL, p
b) P, n
c) x, R
d) R, σ
149. A process is in control at x = 100, R = 7.3 with n = 4. If the process level shifts to
101.5, with the same R , what is the probability that the next x point will fall outside
the old control limits?
a) .016
b) .029
c) .122
d) .360
150. A process is checked by inspection at random samples of 4 shafts after a polishing
30 QReview Study Questions
operation, x and R charts are maintained. A person making a spot check picks out 2
shafts, measures them accurately, and plots the value of each on the x chart. Both
points fall just outside the control limits. He advises the department foreman to stop the
process. This decision indicates that
151. The hardness of rivets is normally distributed with µ = 60.0 and standard deviation
σ = 1.2. What are the 3 sigma control limits for the x chart using samples of size 5?
a) 57.95, 62.15
b) 58.0, 62.0
c) 58.38, 61.62
d) 56.50, 63.5
152. A possible cause of a cycle pattern in a control chart include all of the following except
a) seasonal effects.
b) differences between suppliers.
c) different inspectors.
d) new supplier.
Use the following information to answer problems 153 - 158. The following data are obtained
from measuring the length of a metal bracket.
153. What are the values of x , UCL x and LCL x for an x chart?
a) 5, 5
b) 4, 5
c) 5, 4
d) 4, 4
a) .0012"
b) .0058"
c) .0632"
d) .0729"
a) .013"
b) .030"
c) .010"
d) 1.52"
158. What is the process capability for the individual data points?
a) 1.498" to 1.532"
b) 1.500" to 1.530"
c) 0.020" to 0.040"
d) 1.476" to 1.554"
Use the following information to answer questions (problems) 159 - 162. The following data
are the result of inspecting aircraft seat belt buckles.
a) .833
b) .024
c) .009
d) .997
a) .087, -.024
b) .024, 0
c) 1.50, .013
d) .087, 0
161. What is the appropriate type of control chart for these data?
a) u chart
b) p chart
c) c chart
d) np chart
162. An additional sample of 75 buckles contains 4 defectives. Is this sample within the
control limits?
a) Yes, p = .053
b) Yes, p = .024
c) No, p = .053
d) No, p = .024
Reliability 33
RELIABILITY
164. Which of the following best describes the failure rate in the wear out phase of the
bathtub curve?
a) Constant
b) Random
c) Improves
d) Gets Worse
a) design area.
b) manufacturing area.
c) shipping area.
d) field service area.
167. Maintainability is
168. A set of components has a MTBF of 1000 hours. What percentage will fail if the
components are tested for 500 hours?
a) 25%
b) 39%
c) 61%
d) 50%
169. What is the reliability of a system at 850 hours, if the average usage on the system was
34 QReview Study Questions
400 hours for 1650 items and the total number of failures was 145? Assume an
exponential distribution.
a) 0%
b) 36%
c) 18%
d) 83%
Use the following information to answer problems 170, 171 and 172.
A C
B D
a) .9302
b) .4000
c) .8805
d) .4403
172. What is the probability that component C will fail before 100 hours.
a) .5184
b) .2592
c) .3460
d) .1296
173. The MTBF of a complex piece of repairable radar equipment is determined to be 950
hours. The equipment has been in continuous operation for 150 hours. What is the
probability that the equipment will continue to operate without failure for at least
another 375 hours?
a) 0.5754
b) 0.6376
c) 0.6739
d) 0.8521
Regression and Correlation 35
174. All of the following about multiple regression are true except
175. All of the following statements are true about a correlation coefficient except
176. A study was conducted on the relationship between the speed of different cars and
their gasoline mileage. The correlation coefficient was found to be 0.35 from the study.
Later, it was discovered that there was a defect in the speedometers and they had all
been set 5 miles per hour too fast. The correlation coefficient was computed using the
correct scores. What is the new correlation coefficient?
a) 0.30
b) 0.35
c) 0.40
d) 0.45
$500 $2700
800 3500
1000 4600
1200 7000
1400 9500
2000 11500
a) 42.3
b) -100.7
c) -1027.4
d) 567.52
a) 6.52
b) 3.21
c) 24.4
d) 79.3
a) .52
b) .23
c) .48
d) .97
a) Y = 6.52 -1027.42x
b) Y = 79.3 + 42.3x
c) Y = -100.7 + 24.4x
d) Y = -1027.42 + 6.52x
181. What will the projected quality costs savings be when $1700 per employee is invested
in quality training?
a) $97,000.00
b) $10,056.58
c) $152,231.50
d) $212,520.63
182. If the cost savings were $8000 per employee, what are the probable training costs per
employee?
a) $1,000.00
b) $1,523.33
c) $1,384.57
d) $682.80
Cost of Quality 37
COST OF QUALITY
185. In selecting a base for measuring quality costs, which of the following should be
considered?
186. Which of the following quality cost indices is likely to have the greatest appeal to top
management as an indicator of relative cost?
187. If prevention costs are increased to pay for engineering work in quality control, and this
results in a reduction in the number of product defects, this yields a reduction in
a) appraisal costs.
b) quality costs.
c) failure costs.
d) manufacturing costs.
38 QReview Study Questions
188. A process that sorts good product from defective product falls into which of the
following quality cost categories?
a) Prevention
b) Appraisal
c) Internal failure
d) External failure
a) prevention costs.
b) appraisal costs.
c) failure costs.
d) material-procurement cost.
190. The cost of writing instructions and operating procedures for inspection and testing
should be charged to
a) prevention costs.
b) appraisal costs.
c) internal failure costs.
d) external failure costs.
192. Which of the following is least likely to be reported as a failure related cost?
193. Which of the following activities is not normally charged as a preventive cost?
a) Quality Training
b) Design and Development of Quality Measurement Equipment
c) Quality Planning
d) Laboratory Acceptance Testing
Cost of Quality 39
194. In deciding whether sampling inspection of parts would be more economical than 100%
inspection, you need to determine all of the following except
196. For a typical month, the 3D Manufacturing Company identified and reported the
following quality costs:
a) $244,000
b) $261,000
c) $205,000
d) $332,000
a) Material Repair
b) Material Scrap
c) Customer Appraisal
d) Training
40 QReview Study Questions
Use the following information to answer problems 198 - 201. A manufacturer produces an
electronic memory device. The following costs are incurred on a yearly basis.
Activity Cost
Research and Development $50,000
Manufacturing Engineering 40,000
Quality Engineering 5,000
Plant Facilities 60,000
Training 1,000
Direct Labor 80,000
Process Control (Labor) 5,000
Supervision 30,000
Materials 90,000
Inspection 13,000
Scrap Costs 21,000
Rework Costs 12,000
a) $10,000
b) $6,000
c) $4,000
d) $18,000
a) $18,000
b) $22,000
c) $12,000
d) $33,000
a) $18,000
b) $66,000
c) $33,000
d) $27,000
a) 47%
b) 8%
c) 10%
d) 14%
Metrology and Calibration 41
DESIGN OF EXPERIMENTS
a) the levels used for each factor are the only ones of interest.
b) the levels were chosen from a fixed population.
c) the equipment from which the data are collected must not be moved.
d) the factors under consideration are qualitative.
204. An experiment with two factors, in which all levels of one variable are run at each level
of the second variable, is called a
a) One-way experiment.
b) Latin square experiment.
c) factorial experiment.
d) fractional factorial experiment.
205. A two-way Analysis of Variance has r levels for the first variable and c levels for the
second variable with 2 observations per cell. The degrees of freedom for interaction is
a) 2(r x c)
b) (r - 1)(c - 1)
c) rc - 1
d) 2(r - 1)(c - 1)
206. An analysis of variance results in a calculated F value of F(10, 12) = 2.75. What is the
level of significance?
a) p < 0.01
b) p = .01
c) p = .05
d) p > .05
208. Which of the following distributions is needed to evaluate the results of analysis of
variance (ANOVA)?
a) binomial distribution
b) Chi Square distribution
c) F distribution
d) Z distribution
209. The primary advantage of the Latin Square design compared to the factorial design, is
that
210. Consider the SS and MS columns of an Analysis of Variance table for a single factor
design. The appropriate ratio for testing the null hypothesis of no treatment effect is
Use the following information to answer problems 212 - 216. The following factorial
experiment was conducted to determine the effect of study time and study material used on
test scores. The numbers in the box represent the test scores. You will need to construct the
ANOVA table to complete all of the questions.
Metrology and Calibration 43
Hours of Study
4 6
Study Material Used 1 60, 85 77, 92
2 55, 65 60, 80
a) 2
b) 4
c) 6
d) 8
213. How many levels are being examined for each factor?
a) 2
b) 4
c) 6
d) 8
a) 675.0
b) 242.0
c) 364.5
d) 1283.5
a) Study Hours
b) Study Material Used
c) Interaction
d) Residual
a) availability.
b) waste.
c) performance.
d) redundancies.
219. Which of the following terms is defined as the probability that measurement equipment
will be found to be in tolerance after a specified period of time?
a) Availability
b) Activism
c) Reproducibility
d) Reliability
220. Suppose a test procedure is performed in order to determine the value of a product
measurement. What is the order of equipment type, from lowest to highest accuracy, if
NIST traceability is to be achieved?
221. Which procedures are performed on product and test equipment in order to determine
if predefined specifications have been met?
a) Inspection
b) Identification
c) Labeling
d) Traceability
a) Reliability
b) Traceability
c) Variability
d) Dependability
223. A measurement standard with a stated uncertainty of 0.1 V, 95% confidence level, is
used to test a device that has a nominal value of 100 V, and a tolerance of ±10 V.
What is the TUR for this measurement?
a) 10
b) 100
c) 9.5
d) 95
224. A requirement of the ISO 9000 series standards is that all M&TE used for product
inspection
a) 4.1
b) 4.5
c) 4.11
d) 4.17
46 QReview Study Questions
QREVIEW ANSWERS
1. Certification of a company to the ISO 9002 standard does not include which of the
following requirements?
b) Design control
Servicing, design control, contract review and internal quality audits are all requirements
of ISO 9001. The only change in ISO 9002 is the elimination of design control. ISO
9002 is intended for organizations that do not design products.
A quality control program is more than a set of procedures, step by step lists of check
points or summaries of policies and data. It includes all activities performed to ensure
that products and services meet quality requirements.
4. In preparing a product quality policy for your company, you should do all of the following
except
Procedures are separate documents, but the quality policy may state that procedures
need to be included in the system. Functional responsibilities are also usually a
separate document but the quality policy may have a provision that they must exist.
5. What natural phenomenon created the necessity to control product and process quality?
b) Variation
It is because of variation that the science of quality control exists. All measurements
exhibit variation and this is why it is necessary to control product and process quality.
Quality Management is the means of implementing and carrying out quality policy.
Quality Control monitors and improves the conformance of products, processes and
services. Quality Assurance takes action to assure that products or services will satisfy
the specified requirements.
Design quality defines the level of characteristics that go into the product. Conformance
quality measures how well the design intent is met.
The number of marriages, births and deaths in the United States next year can be
predicted with some degree of accuracy, but exactly who will get married, who will be
born or who will die cannot be predicted. This concept can be applied to a manufacturing
process. A statistical study can determine that products from a certain process are on
average two percent defective, but in any sample, the specific parts that will be defective
cannot be predicted.
c) the study of the characteristics of a product or process, with the help of numbers,
to make them behave the way we want them to behave.
The word statistical means having to do with numbers, or more specifically, with drawing
conclusions from numbers. The word quality means much more than the goodness or
defectiveness of the product. It refers to the qualities or characteristics of the product or
process being studied. The word control means to keep something within boundaries or
to regulate it so that its outcome may be predicted with some degree of accuracy.
Basic Quality Concepts 49
10. Which of the following is the most important element in Statistical Quality Control?
In statistical process control, the feedback loop is between the process control function
and the device that regulates the process or the person responsible for adjustments.
Continuous feedback and the appropriate corrective action are what make statistical
quality and process control work to achieve the desired results.
Accurate measurements fall within an acceptable range but are distributed randomly
within that range. Precise measurements have minimum variation, which means that
they are clustered close together. Accurate and precise measurements would then
imply that the data is clustered closely around the central value.
d) employee salaries.
Document control, corrective action and management responsibility are all essential
parts of a quality system according to the ISO 9001/Q9001 quality systems standards.
Companies usually develop a specific policy for employee salaries.
a) A model that shows how data are distributed over a range of measurements.
Because of variation between measurements of individual parts, data when plotted will
form a distribution. A distribution model describes how the data are dispersed. A plot of
the distribution will show a center value and the range of measurements.
14. Which of the following are two types of data used in statistical quality control?
Data classified as good/bad, pass/fail, go/no-go, etc., are called attribute or discrete
data. When actual measurements are taken and recorded, the data are called variables
or continuous data.
15. The primary reason for evaluating and maintaining surveillance over a supplier's quality
50 QReview Study Questions
program is to
Purchased parts have a significant impact on the quality of a final product. It is important
that supplier's have an effective quality program in place. An effective quality program
and specific product quality are closely related.
b) Quality Manual
A quality manual will usually address the elements of the ISO standard. Specific quality
tools and instructions are not included as part of the ISO requirements.
b) Inspector
A designer, machinist and equipment engineer all play a role in the manufacture of a
product and therefore have the ability to generate product-quality characteristics. An
inspector only examines a product and cannot alter it in any way.
In since incoming inspection is concerned with the quality of purchased products, the
requirements should be listed on the purchase order.
c) equal.
The International Organization for Standardization is also known as ISO. ISO is not an
acronym for this organization but it is a Latin word which translates to equal.
Products need to be tested in the environment where they are used. Various other types
of tests can be performed for prove in, but every condition varies to some degree and
the best way to prove a product is under actual working conditions.
21. The advantage of a written procedure is
Basic Quality Concepts 51
A written procedure allows everyone concerned to know exactly what is required. It may
always be referred to when coordinating activities. Unwritten procedures usually cause
confusion because of communication failure.
22. In spite of the Quality Engineer's best efforts, situations may develop in which his
decision is overruled. The most appropriate action would be to
c) document findings, report them to superiors and move on the next assignment.
Human conflict can be minimized but never eliminated. There may be times when
engineering decisions are overruled or employee suggestions not adopted. In these
cases, the findings should be documented for possible review at a later date, then the
engineer or employee can move on to the next assignment. In any conflict, whether it is
between management and employees, between management and unions, or between
employees, good judgement must be used.
23. If a test data does not support a Quality Engineer's expectations, the best thing to do is
Objective data that has been collected should never be altered or the analysis of the
data should never be modified to fit expectations. The expectations were just probably
not realizable for the application under test.
Contract specifications take precedence over all other practices. Any deviation from the
specifications must be negotiated with the customer.
All quality programs must have clear policies, objectives and documented procedures.
Without these, the effort will be aimless.
It is good policy for auditors to not have a vested interest in the activity being audited.
For example, the fox guarding the chicken coop will usually result in fewer chickens.
27. Analysis of data on all product returns is important because
Analysis of data on all product returns is important because it may provide feedback as
to how a customer's specific application of the product may affect the product quality. It
will also help to determine failure rates based on the longevity of the product. Immediate
feedback and analysis may be obtained for the specific products returned as opposed to
collecting data over some specified time and generalizing product problems.
28. All of the following are considerations when a total quality management (TQM) program
is implemented except
No one person is responsible for product quality in a TQM program. TQM involves the
use of statistical techniques, continuous improvement and total involvement by all.
29. According to Juran, all of the following are widespread errors in perception that have led
many managers astray except
Juran has listed many widespread errors in perception in his books (Juran on
Leadership for Quality, 1989), but does not include "return on investment is everything."
This statement may not necessarily lead managers astray.
Feedback lets an employee know whether the concepts learned from training have been
grasped. This helps the employee know where their strengths and weaknesses are. It
also can help the employee gain confidence in their abilities.
31. An engineer has the job of providing a written plan of quality related tasks to his
manager, including a detailed timeline, for the following year. Which of the following
tools should be used?
c) Gantt Chart
Gantt charts are typically horizontal bar charts, where each bar represents a task that
spans from the beginning time to the ending time. Histograms and frequency
distributions are analytical tools and a flow chart is typically used for relationship
modeling.
Basic Probability 53
BASIC PROBABILITY
32. The time it takes to answer a technical support line has a continuous uniform distribution
over an interval from 17 to 20 minutes. All of the following are true except
33. For two events, A and B, which one of the following is a true probability
statement?
For any two events, P(A or B) = P(A) + P(B) - P(A & B). For independent events
P(A & B) = P(A) x P(B) and for mutually exclusive events P(A & B) = 0 as
illustrated below.
Mutually Exclusive
Independent
A
A
A&B
B B
d) 1
35. What is the probability of getting a head and a tail in 2 tosses of a fair coin? And, What is
the probability of getting a head and a tail, in that order, in 2 tosses of a fair coin?
c) 1/2, 1/4
P(Head & Tail in any order) = P(Head) & P(Tail) or P(Tail) & P(Head)
= P(Head) x P(Tail) + P(Tail) x P(Head)
= 1/2 x 1/2 + 1/2 x 1/2 = 1/4 + 1/4
= 1/2
d) 1/2
All tosses are independent events. The probability of getting a head on the 10th toss is
the same as the probability of getting a head on any toss, which is 50%.
37. What is the probability of obtaining exactly 2 heads in 4 tosses of a fair coin?
b) 3/8
The total number of events is 16 and the number of favorable events is 6, therefore
P(2 Heads in 4 tosses) = number of favorable events / total number of events
= 6/16 = 3/8
38. What is the probability of getting a 3 when rolling a single die? ( A die is one of a pair of
dice)
c) 1/6
39. What is the probability of getting an odd number when rolling a pair of dice? (Spots on
the two dice sum to odd number)
b) 1/2
The total number of events is 36 and the number of favorable events is 18, therefore
P(odd number) = number of favorable events / total number of events
= 18/36 = ½
40. What is the probability of obtaining a sum of 7 when rolling a pair of dice?
c) 1/6
The total number of events is 36 and the number of favorable events is 6, therefore
P(rolling a 7) = number of favorable events / total number of events
= 6/36 = 1/6
Use the following information to answer questions 41, 42 and 43. The probability is 1/2 that
Bob will pass the CQE exam, 1/3 that Amy will pass and 3/4 that Jon will pass.
41. What is the probability that Bob, Amy and Jon will all pass the exam?
a) 1/8
P(all will pass) = P(Bob will pass) x P(Amy will pass) x P(Jon will pass)
= 1/2 x 1/3 x 3/4
= 1/8
56 QReview Study Questions
42. What is the probability that neither Bob, Amy nor Jon will pass the exam?
c) 1/12
P(none will pass) = [1 - P(Bob will pass)] x [1 - P(Amy will pass)] x [1 - P(Jon will pass)]
= (1 - 1/2) x (1 - 1/3) x (1 - 3/4)
= 1/2 x 2/3 x 1/4
= 1/12
43. What is the probability that only one of the three will pass the exam?
d) 3/8
P(only one will pass) = [P(Bob will pass) & P(Amy will fail) & P(Jon will fail)] or
[P(Bob will fail) & P(Amy will pass) & P(Jon will fail)] or
[P(Bob will fail) & P(Amy will fail) & P(Jon will pass)]
= [1/2 x (1 - 1/3) x (1 - 3/4)] +
[(1 - 1/2) x 1/3 x (1 - 3/4)] +
[(1 - 1/2) x (1 - 1/3) x 3/4]
= 2/24 + 1/24 + 6/24
= 3/8
44. Four people shoot at a target and the probability that each will hit the target is 1/2 (50%).
What is the probability that the target will be hit?
b) 15/16
The target can be hit 0 times, 1 time, 2 times, 3 times or 4 times, therefore
P(0) + P(1) + P(2) + P(3) + P(4) = 1
P(0) = 1/2 x 1/2 x 1/2 x 1/2 = 1/16
P(hitting target) = P(1) + P(2) + P(3) + P(4)
= 1 - P(0) = 1 - 1/16
= 15/16
45. A committee of 5 people is chosen at random from a room that contains 4 men and 6
women. What is the probability that the committee is composed of 2 men and 3 women?
b) 10/21
46. A vendor is trying to sell you a box of 50 fuses that contains exactly 5 defective fuses.
You select 2 fuses from the box for testing. If both are good you will buy the entire box. If
one or both are defective, you will not buy the box. What is the probability that you will
Basic Probability 57
47. What is the probability of winning the Super Lotto? (Winning = getting all 6 numbers out
of 47)
a) 1/10,737,573
Use the hypergeometric probability distribution:
6 41 47 (1)(1)
P(winning) = = = .00000009313
6 0 6 10,737,573
48. A box contains 12 connectors, 9 good ones and 3 defective ones. What is the probability
of obtaining exactly 2 good and 1 defective connector in drawing 3 parts from the box
without replacement?
b) .4909
Use the hypergeometric probability distribution:
9 3
P(2 good and 1 defective) =
2 1
=
(36)(3) = 108 = 27 =.4909
12 220 220 55
3
49. A box contains 12 connectors, 9 good ones and 3 defective ones. What is the probability
of obtaining exactly 2 good and 1 defective connector in drawing 3 parts from the box
with replacement?
a) .4219
2 1
3 9 3
P(2 good and 1 defective) = = 3(.5625)(.25) =.4219
2 12 12
50. You have been asked to sample a lot of 300 units from a vendor whose past quality has
been about 2% defective. A sample of 40 pieces is drawn from the lot and you have
been told to reject the lot if you find two or more parts defective. What is the probability
of rejecting the lot?
58 QReview Study Questions
c) 0.191
Use the following information to answer questions 51 and 52. A company produces
capacitors by a process that normally yields 5% defective product. A sample of 4 capacitors
is selected.
d) .8145
4
d) .00000625
4
Basic Probability 59
Use the following information to answer questions 53, 54, 55 and 56. A company makes
ball bearings that are found to be 10% defective in the long run. A sample of 10 bearings is
selected.
a) .3487
0
b) .3874
1
a) .0574
3
57. How many defective connectors would be expected in a sample of 200 parts if the
process averages 2% defective?
c) 4
The average number of defective parts is equal to the sample size multiplied by the
average percent defective. The average number of defective parts = 200 x .02 = 4.
58. What is the probability of obtaining exactly 2 defective connectors in a sample of 6 parts
60 QReview Study Questions
b) .0055
Use the Binomial probability distribution and the average number of defective parts from
the previous problem:
6
P(2) = ( 02) (.98) = 15(.0004)(.9224) =.0055
2 4
2
The number rolled in a game of dice, the number of defects in a random sample, and the
number of games played in the world series are all probabilistic events because the
exact outcome is not predictable 100% of the time. The acceleration of an apple when it
drops from a tree is a deterministic event because it will always accelerate at the same
rate (9.81 m/s2).
Use the following information to answer problems 60, 61 and 62. A company produces
integrated circuits (chips) by a process that normally yields 2000 ppm defective product for
electrical test requirements (ppm = defective parts per million). A sample of 5 chips is
selected and tested.
a) .9900
P(all 5 are good) = P(0 defective) = (50 )(.002)0 (.998)5 = 1 (1)(.99) = .99
d) .0100
c) .00008
Use the following information to answer problems 63, 64 and 65. A capability study was
made to determine the defective rate of 28AZ transistors. The study showed the rate to be
5000 ppm. Ten of the transistors were shipped to a customer.
63. What is the probability that the shipment contains no defective transistors?
a) .9511
64. What is the probability that the shipment contains exactly 1 defective transistor?
b) .0478
1
65. What is the probability that the shipment contains 2 or more defective transistors?
d) .0011
Use the following information to answer problems 66, 67 and 68. A circuit board operation
yields 2 defects per board on the average. A sample of 1 board is selected at random.
66. What is the probability of finding exactly 2 defects on the selected board?
b) .2706
67. What is the probability of finding less than 2 defects on the selected board?
a) .4060
e −2 2 0 e −2 21 (.1353)(1) (.1353)(2)
P(x < 2) = + = + =.1353 +.2706 =.4059
0! 1! 1 1
68. What is the probability of finding more than 2 defects on the selected board?
d) .3235
e −2 2 0 e − 2 2 1 e − 2 2 2
P(x > 2) = 1 − + +
0! 1! 2!
= 1 - (.1353)(1)/1 - (.1353)(2)/1 - (.1353)(4)/2
= 1 - .1353 - .2706 - .2706
= .3235
Basic Probability 63
Use the following information to answer problems 69, 70 and 71. In manufacturing material
for automobile seats it was found that each 100-foot roll contained 2 defects (flaws) on the
average. A sample of 1 roll is selected at random from the process.
69. What is the probability that the selected roll contains 0 defects?
a) .1353
70. What is the probability that the selected roll contains exactly 1 defect?
d) .2706
71. What is the probability that the selected roll contains more than 1 defect?
b) .5941
72. What is the probability of finding exactly 2 defective decals in any given day? (Assume
one defect per defective decal.)
c) .2240
73. What is the probability of buying a decal that contains more than 1 defect?
64 QReview Study Questions
d) .0017
74. A parts dealer buys parts from a warehouse. Parts are made by either Company A or
Company B but are not identified as to which company produces them. One company
produces all parts in one shipment or lot. On the average, we know:
Company A produces 2.5% defective parts.
Company B produces 5.0% defective parts.
The warehouse states that 70% of parts will come from Company A and 30% from
Company B. If the dealer selects 4 parts at random from a lot and finds 1 defective part,
what is the probability that the lot was produced by Company A?
b) .5580
The question to be answered is “What is the probability that the lot came from company
A given that the shipment contains one defective part or P(A/D)?”
STATISTICS
75. What is the expected value of the random variable x for the following data?
x f(x)
12 0.2
10 0.5
14 0.1
20 0.2
a) 12.8
76. In the standard normal table, what value of z has 5% of the area in the tail beyond it?
b) 1.645
The value that has 5% of the area to the right must have 95% of the area to the left.
Most normal tables provide the total area to the left of the z value, but in some cases the
table may show the area from the left of the z value to the midpoint of the normal curve
for positive z values. In the initial case, the z value that corresponds to an area of .9500
is the desired value. In the latter case, the z value that corresponds to an area of .4500
is the desired value.
77. Which distribution should be used to determine a confidence interval when σ is not
known and the sample size is 10?
b) t
When the population variance is estimated by S2, and the sample size is small, then S2
will fluctuate considerably from sample to sample and the distribution of the random
variable will deviate from the normal distribution. The t distribution is derived from the
random variable containing the sample standard deviation and is used for small sample
sizes. The t distribution should be used for sample sizes of 30 or less.
78. Which of the following methods should be used to test 6 population means for statistical
significance?
b) Analysis of Variance
The Analysis of Variance or ANOVA is a widely used technique for dealing with more
than two populations. Analysis of Variance analyzes the total variation by dividing it into
meaningful components.
79. A sample size of 120 is taken from a process and is represented graphically on a
66 QReview Study Questions
b) 9 - 20
Grouping data into cells condenses large amounts of data that can be hard to manage.
Typically, between 5 and 20 cells should be used choosing the number approximately
equal to the square root of the sample size. Also, the cells should be of uniform width
and the lower limit of the first cell should be just below the smallest data value.
80. Which of the following conditions makes it possible for a process to produce a large
number of defective units while it is in statistical control?
If the process capability is wider than the tolerance, a percentage of parts will be outside
of the specification limits. Any parts outside of the specification limits are considered to
be defective.
81. For the normal probability distribution, which of the following is true about the
relationship among the median, mean and mode?
82. All of the following statistical techniques can be used to determine the effectiveness of a
supplier improvement program except
d) a flow chart.
A flow chart is used to illustrate the order in which a sequence of events will occur and
therefore cannot supply any meaningful data for evaluating a supplier improvement
program. Pareto analysis and x bar and R charts are effective tools for evaluating a
program based on numerical data collected from the entity being evaluated. The
program evaluation and review technique (PERT) is also an effective tool for evaluating
a particular program, particularly in evaluating scheduling requirements.
83. A sample of n observations has a mean x and a standard deviation s > 0. If a single
observation, which equals the value of the sample mean x , is removed from the sample,
which of the following is true?
Removing an observation that is equal to the average does not change the average.
The standard deviation increases because the removed point helped to keep the
variation of the data from the mean to a minimum.
84. The factory installed brake linings for a certain kind of car has a mean lifetime of 60,000
miles with a 6,000 mile standard deviation. A sample of 100 cars has been selected for
Statistics 67
testing. What is the standard error of x ? (Assume that the finite population correction
may be ignored.)
c) 600 miles
Use the following information to answer problems 85 - 90. A sample of 7 rivets was taken
from a shipment of 1000 rivets and the length was measured. The following data are
obtained:
1 3.1
2 3.1
3 3.2
4 3.7
5 3.6
6 3.7
7 3.1
b) 3.36 inches
x=
∑x i
=
23.5
= 3.36"
n 7
86. What is the standard deviation of the length of the rivets (estimate of population standard
deviation)?
c) 0.29 inches
∑ (x − x )
2
i 1.76
s= = = 0.29"
n −1 6
Note: In the following 4 problems, the sample sizes are less than 30 and the t statistics
should be used to solve the problems. Analyses of this type usually involve sample sizes of
30 or greater. Handle the problems just as if the sample sizes were greater than 30 and use
68 QReview Study Questions
the z statistics.
87. What percentage of rivets have lengths less than 2.80 inches?
a) 2.69%
The percentage of rivets that are less than 2.80" is represented by area B. The area of B
can be found by obtaining the area that corresponds to the following z value from the
standard normal table.
x i − x 2.80 − 3.36 −.56
z= = = = −193
.
s .29 .29
A
B
x
2.80 3.36
z
-1.93 0
Area B = .0269, therefore 2.69% of the rivets have lengths less than 2.80".
88. What percentage of rivets have lengths greater than 3.65 inches?
c) 15.9%
The percentage of rivets that are greater than 3.65" is represented by area D. The area
of D can be found by subtracting the area to the left of D from the total area under the
curve. The area to the left D can be found by obtaining the area that corresponds to the
following z value from the standard normal table.
x i − x 3.65 − 3.36 .29
z= = = = 10
.
s .29 .29
C
D
x
3.36 3.65
z
0 1.0
The area to the left of D = .8413 and the area of D = 1 - .8413 = .1587, therefore 15.9%
of the rivets have lengths greater than 3.65".
89. What percentage of rivets have lengths between 3.1 inches and 3.9 inches?
b) 78.4%
Statistics 69
Area A plus area C represent the percentage of rivets that are between 3.1" and 3.9".
The area of A plus the area of C can be found by obtaining the areas that correspond to
the following two z values from the standard normal table and subtracting the first from
the second.
x i − x 3.1 − 3.36 x i − x 3.9 − 3.36
z1 = = = −.90 z2 = = = 186
.
s .29 s .29
A C
B D
x
3.1 3.36 3.9
z
-.90 0 +1.86
Area B = .1841, area BAC = .9685 and area BAC - B = .9685 - .1841 = .7844, therefore
78.4% of the rivets have lengths between 3.1" and 3.9".
90. In the shipment of 1000 rivets, how many good parts will we find if a good part is defined
as having a minimum of 3 inches and a maximum of 4 inches?
d) 878
Area A plus area C represent the percentage of rivets that are between 3.0" and 4.0".
The area of A plus the area of C can be found by obtaining the areas that correspond to
the following two z values from the standard normal table and subtracting the first from
the second.
x i − x 3.0 − 3.36 x i − x 4.0 − 3.36
z1 = = = −124
. z2 = = = +2.20
s .29 s .29
A C
B D
x
3.0 3.36 4.0
z
-1.24 0 +2.20
Area B = .1076, area BAC = .9861 and area BAC - B = .9861 - .1076 = .8785, therefore
87.85% of the rivets in the shipment have lengths between 3.0" and 4.0" which equates
to 878 good parts. Remember, do not round up because you cannot have a partially
good part.
The following information is used to answer problems 91 - 95. Data are taken from a
manufacturing process that produces optical glass. The sample size is 5 parts and the
characteristic measured is the diameter of the plates.
70 QReview Study Questions
a) 31.4 mm
x=
∑x i
=
157
.
= 314
. mm
n 5
b) 2.07 mm
∑ (x − x)
2
i 8.29
s= = = 2.07mm
n−1 4
93. The specifications for the glass plates are 30.5 ± 2 mm. What percentage of parts made
by this company will not meet specifications?
d) 37.9%
Area B plus area D represent the percentage of parts that will not meet specification. To
find the area of B plus D, it is first necessary to obtain the areas that correspond to the
following two z values from the standard normal table.
x i − x 28.5 − 314
. x i − x 32.5 − 314
.
z1 = = = −14
. z2 = = = +.53
s 2.07 s 2.07
A C
B D
x
28.5 31.4 32.5
z
-1.4 0 +.53
Area B = .0808, area BAC = .7019, area D = 1 - BAC = 1 - .7019 = .2981 and area
B + D = .0808 + .2981 = .3789, therefore 37.9% will not meet specifications.
94. What percentage of parts will be less than 29.5 mm?
a) 17.9%
Statistics 71
The percentage of rivets that are less than 29.5 mm is represented by area B. The area
of B can be found by obtaining the area that corresponds to the following z value from
the standard normal table.
x i − x 29.5 − 314
.
z= = = −.92
s 2.07
A
B
x
29.5 31.4
z
-.92 0
Area B = .1788, therefore 17.9% of the parts will be less than 29.5 mm.
d) 22.1%
The percentage of parts that are greater than 33 mm is represented by area D. The
area of D can be found by subtracting the area to the left of D from the total area under
the curve. The area to the left D can be found by obtaining the area that corresponds to
the following z value from the standard normal table.
x i − x 33.0 − 316
.
z= = = +.77
s 2.07
C
D
x
31.4 33.0
z
0 .77
The area to the left of D = .7793 and the area of D = 1 - .7793 = .2207, therefore 22.1%
of the parts will be greater than 33 mm.
96. The Zoglen Corporation markets a product, which is a blend of 3 ingredients (A, B, C). If
the individual tolerances for the weight of the 3 ingredients are as shown, what should
the tolerance be for the net weight of the product?
72 QReview Study Questions
97. A random sample of size n is to be taken from a large population that has a standard
deviation of 1 inch. The sample size is determined so that there will be a 95% chance
that the sample average will be within ± 0.1 inch of the true mean. Which of the following
values is nearest to the required sample size?
a) 385
The sample size is determined by using the formula for variables data. The z value
corresponding to 95% of the area under the normal curve is 1.96.
2 2
Zs (196
. )(1)
n= = = 385
E 0.1
98. All of the following conditions must be met for the process capability to be within the
specification limits except
c) Cp = Cpk
The value of Cp is based on a ratio of the specification limits to the measured process
variability (6σ). A Cp less than 1.0 (large variance) will yield a distribution that falls
outside of the specification limits. A value of Cpk less than 1.0 will represent a shift of the
mean value of the data from the target value, which will cause some of the data to
possibly fall outside of a specification limit. A stable process implies that all the data will
fall within the specification limits. The process capability index and the process
performance index may equal each other and fall outside of the specification limits if they
are both less than 1.0.
99. A value on the abscissa of the t distribution is 1.093. What is the area to the right of this
value if the sample size is 11?
b) 0.15
100. The spread of individual observations from a normal process capability distribution may
be expressed numerically as
a) 6 R /d2
If only control chart information is available, the standard deviation may be calculated
by R /d2, where d2 is a factor corresponding to the sample size. The process
capability is defined as 6σ, therefore the expression may be defined as 6 R /d2.
101. What percentage of data will normally fall within a process capability?
b) 99.73%
Process capability is defined as 6σ. The area under the normal curve from -3σ to +3σ,
which is 6σ, is 99.73% of the total area.
Use the following information to answer problems 102 - 105. A winding machine wraps wire
around a metal core to make small transformers. The design engineers have determined
that the nominal number of windings is to be 10,060 with a minimum of 10,025 and a
maximum of 10,095. A sample of 300 transformers was selected in a three month period
and the wire was unwrapped on each part to determine the number of windings. The results
were:
c) 10022 - 10082
d) 1.17
USL − LSL 10095 − 10025 70
Cp = = = = 117
.
6σ 6(10) 60
74 QReview Study Questions
a) .90
x − nearest spec lim it 10052 − 10025
C pk = = =.90
3σ 3(10)
105. What is the expected percent defective?
b) 0.35%
The area outside of the specification limits is where defective product can be found.
x i − x 10025 − 10052
z1 = = = −2.7
s 10
x − x 10095 − 10052
z2 = i = = +4.3
s 10
10022 10082
Area ≈ 0
area = .0035
x
10025 10052 10095
z
-2.7 0 +4.3
Statistical Inference 75
STATISTICAL INFERENCE
107. In a sampling distribution which of the following represents the critical region?
a) α
108. In a hypothesis test which of the following represents the acceptance region?
d) 1 - α
The acceptance region is denoted as 1 - α or one minus the critical region. This can
also be termed as the area under the distribution curve where one would accept a true
null hypothesis.
b) a distribution of variances.
The chi square distribution is a special case of the gamma distribution. The variance
of the chi square distribution is known as two times the number of degrees of freedom,
therefore the variance can be substituted back into the density function to give a
distribution of variances.
110. Which of the following is a number derived from sample data that describes the data in
some useful way?
b) statistic
Any function of the random variables constituting a random sample is called a statistic.
111. A null hypothesis assumes that a process is producing no more than the maximum
allowable rate of defective items. What does the type II error conclude about the
process?
76 QReview Study Questions
The type II error or β is the acceptance of a false null hypothesis. If the null hypothesis
is false in this case then the process will actually be producing too many defectives
and the type II error will conclude that the process is acceptable.
112. A quality engineer wants to determine whether or not there is any difference between
the means of the convolute paperboard cans supplied by two vendors, A and B. A
random sample of 100 cans is selected from the output of each vendor. The sample
from vendor A yielded a mean of 13.59 with a standard deviation of 5.94. The sample
from vendor B yielded a mean of 14.43 with a standard deviation of 5.61. Which of the
following would be a suitable null hypothesis?
a) µA = µB
The null hypothesis is the statement we are trying to prove or disprove and the
alternate hypothesis is the statement we will accept if the null hypothesis is not
accepted. Typically, if the null hypothesis is not accepted we will accept a range of
values as provided in the wrong answers to this question.
113. A chi square test for independence has a 4 x 3 contingency table and a calculated chi
square value of 11.5. At a .05 level of significance, which of the following are true about
the decision regarding the null hypothesis?
c) 6 df, accept Ho
The degrees of freedom for a contingency table are (r - 1)(c - 1), therefore
df = 3 x 2 = 6. From the chi square table (α = .05), the critical value is 12.59. The
calculated value of 11.5 falls within the acceptance region, therefore accept Ho.
114. A supplier states that the average diameter of optical glass plates is 37.50 mm. The
specifications are 37.50 mm ± .40 mm. A shipment of 500 plates is received and a
sample of 10 plates is selected and the diameters of each are measured. The following
data are obtained:
Does the entire population have an average diameter of 37.5 mm? Test at a level of
significance of .05.
a) t = +2.97, Reject Ho
The null and alternate hypotheses are H0 : x = 37.5 and H1: x = 37.5 . The t statistic
must be computed to determine if it falls in the critical region of the t distribution. The
mean and standard deviation need to be computed to determine the t statistic.
Statistical Inference 77
x − µ 37.652 − 37.5
t= = = +2.967
s .162
n 10
t
37.5 x -2.262 0 2.262
The t statistic falls in the critical region, therefore reject Ho in favor of H1.
Could this sample have possibly come from a process whose process average is .14
defects/unit? Test at a level of significance of .01.
The null and alternate hypotheses are H0: p = .14 and H1: p > .14. The z statistic must
be computed to determine if it falls within the critical region. The critical z value, from
the normal curve table, is +2.325. The computation of the z statistic is as follows:
np (40 )(.14 )
σ= = =.059
n 40
p − p .20 −.14
z= = = +101
.
σ .059
.01
x
P=.14
z
0 1.01 2.325
The z statistic falls in the acceptance region, therefore do not reject Ho.
116. A study of 26 families across a certain state indicated that the average family income
during 1989 was $33,400 with a Standard Deviation of $3680. Test the hypothesis that
the true average income in this state during 1989 was $32,000 against the alternative
that it was not $32,000. Test at a level of significance of .05.
b) t = +1.94, Accept Ho
78 QReview Study Questions
The null and alternate hypotheses are H0: µ = 32,000 and H1: µ ≠ 32,000. The t
statistic must be computed to determine if it falls within the critical region. The critical t
values, from the student t table, are ±2.060. The computation of the t statistic is as
follows:
x − µ 33,400 − 32,000
t= = = +194
.
σ 3,680
n 26
1.94
.025 .025
x t
32,000 -2.060 0 +2.060
The t statistic falls in the acceptance region, therefore accept Ho.
Could this data have reasonably come from a process which usually has an average of
11.3 lb. or greater? Test at a level of significance of .01.
c) t = -2.98, Reject Ho
The null and alternate hypotheses are H0: µ = 11.3 and H1: µ < 11.3. The t statistic
must be computed to determine if it falls within the critical region. The critical t value,
from the student t table, is -2.539. The computation of the t statistic is as follows:
x − µ 10.5 − 113
.
t= = = −2.98
σ 12
.
n 20
.01
-2.98
x t
11.3 -2.539 0
The t statistic falls in the critical region, therefore reject Ho in favor of H1.
118. In a hypothesis test what determines whether the test is one tailed or two tailed?
b) Alternate Hypothesis
If the alternate hypothesis is written so that the population being tested is less than
(e.g. µ < .90) or greater than (e.g. µ > .90) some value then the test will be one tailed.
Statistical Inference 79
If it is stated so that the population is not equal to some value (e.g. µA ≠ µB) then the
test will be two tailed.
The pooled variance is used when two or more samples are considered to be
independent. Independent is defined as when the individual data in one sample
cannot be related to or associated with the other in any meaningful manner.
If some sample data are taken, one might want to know what distribution the data
follows. Tests for goodness of fit can be performed to determine if the data follows any
particular distribution.
80 QReview Study Questions
Sampling 81
SAMPLING
c) ATI
Dodge and Romig define ATI as Average Total Inspection. The sampling plans in the
Doge Romig tables attempt to minimize the ATI.
122. What are the two unique quantities that determine a single sampling attributes plan?
An OC curve that describes the sampling plan may be developed using the sample
size, acceptance number and various values of incoming quality.
123. Characteristics for which 100% inspection may be practicable include all of the
following except
Typically, testing these types of properties can take a long time and are usually
unnecessary. They also usually require some type of destructive test. If destructive
testing is performed on 100% of the product then no product will ever be available for
use.
AQL is the acceptable quality level. If the incoming quality level is at or below the
AQL, the lots have a high probability of being accepted.
125. Which of the following best describes what an operating characteristic curve shows?
0.7
0.6
0.5
Approximately two
0.4 The Consumer’s times the AOQL
0.3 Risk (β)
Rejectable
0.2 Acceptable Quality
Quality Level (RQL)
Level (AQL)
0.1
0
0% 1% 2% 3% 4% 5% 6% 7%
p - Quality of lots submitted to inspection (AIQ)
82 QReview Study Questions
126. In ANSI/ASQ Z1.4, the AQL is always determined at what Pa on the OC curve?
The probability of acceptance is not set at a specific value. AQL's may have a different
Pa depending on what sample code is chosen.
127. Which of the following best describes what an average outgoing quality curve shows?
0.03
AOQL = .028
0.025
Average Outgoing Quality (AOQ)
0.02
0.015
0.01
0.005
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12
Use the following information to answer problems 128 - 134. A sampling plan is needed
which will satisfy the following requirements:
d) 10%
Alpha is known as the producer's risk or when acceptable quality product is rejected
and is 1 - .90.
b) 12%
c) 1%
a) 7%
The lot tolerance percent defective is the same as RQL expressed as percent
defective and is given.
132. What is the sample size that will satisfy the requirements stated above?
d) 50
133. What is the acceptance number that will satisfy the requirements stated above?
a) 1
a) .0165
The Binomial Nomograph may not be drawn to scale. AOQL ≈ .033 / 2 = .0165
The point where the lines meet is where the sample size and acceptance number can
be found.
.01
.12
.033
.07
.50
.90
Given: AQL = .015 (1.5% defective), RQL = .050 (5.0% defective), α = .05, β = .10
d) 200
c) 6
b) .0160
The Binomial Nomograph may not be drawn to scale. AOQL ≈ .032 / 2 = .0160
The point where the lines meet is where the sample size and acceptance number can
be found.
.015 .05
.032
.05
.50
.95
138. A sampling plan specifies a sample size of 50 and an acceptance number of 3. What is
the value of the AQL if α = .05?
d) .0280
139. A sampling plan specifies a sample size of 50 and an acceptance number of 3. What is
the approximate value of the AOQL?
Sampling 85
c) .0370
The Binomial Nomograph may not be drawn to scale. AOQL ≈ .0614 / 2 = .037
The Nomograph must be constructed and the lines must drawn through the sample
size and acceptance number so that the AQL and AOQL can be obtained.
n = 50
.028 c=3
.0614
.50
.95
140. All of the following are methods for checking product except
d) safeguard checking.
Checking product is a safegaurd but it is not a particular method for checking product.
Constant percentage sampling, spot checking and no checking at all are among the
many methods for checking product.
141. All of the following are characteristics of lot by lot single sampling except
c) rejected lots are immediately scrapped.
In lot by lot single sampling, rejected lots are usually screened 100% for the cause of
rejection. In some cases, the lot may be scrapped.
Continuous sampling is used where product flow is continuous and it is not feasible to
be formed into lots.
143. The following parameters are required for sampling plan construction using the
Binomial Nomograph except
d) Sample size.
The sample size and acceptance number are the two values that describe a sampling
plan and can be obtained by using the Binomial Nomograph. Input values include
AQL, RQL, α and β.
86 QReview Study Questions
CONTROL CHARTS
When assignable causes of variation are detected, changes in the inputs can be made
to bring the process back in control. Control charts also provide feedback to operators
and engineers to help reduce process variability.
b) Binomial
a) Poisson
The c and u values are the number of defects in a sample. The Poisson distribution is
used to calculate probabilities associated with "defects." The Binomial cannot be used
to calculate the probability of a certain number of defects.
The central limit theorem states that the distribution of averages always tend to follow
a normal distribution even though the distribution of individual data values may be
skewed.
148. When used together for variables data, which of the following pair of quantities is the
most useful in preparing control charts?
c) x, R
These two statistics are derived from variable data, where x is the average of a
sample of measurements and R is the range or difference between the highest and
lowest value in the sample.
149. A process is in control at x = 100, R = 7.3 with n = 4. If the process level shifts to
Control Charts 87
101.5, with the same R , what is the probability that the next x point will fall outside the
old control limits?
a) .016
106.829
UCL = 105.329
101.5
x = 100
96.171
LCL = 94.671
The plotted points must be the average of the sample or x . Individual data points are
not to be plotted on a x chart.
151. The hardness of rivets is normally distributed with µ = 60.0 and standard deviation
σ = 1.2. What are the 3 sigma control limits for the x chart using samples of size 5?
88 QReview Study Questions
c) 58.38, 61.62
152. A possible cause of a cycle pattern in a control chart include all of the following except
d) new supplier.
A new supplier may cause a sudden shift in the pattern. The shift will appear exactly
when the new supplier's products are used.
Use the following information to answer problems 153 - 158. The following data are obtained
from measuring the length of a metal bracket.
153. What are the values of x , UCL x and LCL x for an x chart?
. + 150
151 . + 152
. + 153
.
x= = 1515
.
4
UCL x = x + A 2 R = 1515
. + 0.58(0.03) = 1532
.
LCL x = x − A 2 R = 1515
. − 0.58(0.03) = 1498
.
c) .030", .0634", 0
Four samples were taken. Each sample consisted of 5 brackets, therefore the sample
size is 5.
b) .0058"
A 2 R (0.58)(0.03)
S tan dard Error = S.E. = = =.0058
3 3
a) .013"
R .030
S tan dard Deviation = s = = =.013
d 2 2.326
158. What is the process capability for the individual data points?
d) 1.476" to 1.554"
Use the following information to answer problems 159 - 162. The following data are the
result of inspecting aircraft seat belt buckles.
b) .024
number of defective buckles 10
p= = =.024
total sample 420
d) .087, 0
p (1 − p ) (.024 )(.9760 )
UCL = p + 3 =.024 + 3 =.087
n 52.5
p(1 − p ) (.024)(.9760)
LCL = p − 3 =.024 − 3 =0
n 52.5
b) p chart
The points plotted are values of fraction defective or percent defective. These types of
data require the use of p charts.
a) Yes, p = .053
The new sample point (p = 4 / 75 = .053) value does not exceed the upper control limit,
therefore the sample point is in control.
Regression and Correlation 91
RELIABILITY
164. Which of the following best describes the failure rate in the wear out phase of the
bathtub curve?
d) Gets Worse
The bathtub curve has three phases: infant mortality, useful life and wear out.
Typically, if a product fails in its infant mortality phase it will never reach the customer.
As the product goes through its useful life phase it will begin and continue to wear out,
increasing its failure rate through time.
a) design area.
A product will not become more reliable after it has been designed. Certain
characteristics must be designed into products to ensure the smallest probability of
failure. Various statistical techniques can be used to help determine designs that will
meet the required expectations.
b) the process of estimating the probability that a product will perform its intended
function for a stated time.
167. Maintainability is
One of the biggest problems with system failures is the downtime while a system is
being fixed. An important factor in designing a system is to minimize the time to
restore the system to functional operation.
168. A set of components has a MTBF of 1000 hours. What percentage will fail if the
92 QReview Study Questions
b) 39%
1 1
λ= = =.001
MTBF 1000
Re liability = e − λt = e −(.001)( 500 ) =.6065
probability of failure = 1 − Re liability = 1−.6065 =.3900
169. What is the reliability of a system at 850 hours, if the average usage on the system was
400 hours for 1650 items and the total number of failures was 145? Assume an
exponential distribution.
d) 83%
145
failure rate = λ = =.0002196
(400 )(1650 )
reliability = e − λt = e − (.0002196 )( 850 ) =.8297 ≈ 83%
Use the following information to answer problems 170, 171 and 172.
A C
B D
c) .8805
1 1
MTBFA = = = 1000
λ A .001
1 1
MTBFB = = = 500
λ B .002
1 1
MTBFC = = = 333.33
λ C .003
1 1
MTBFD = = = 400
λ D .0025
172. What is the probability that component C will fail before 100 hours.
b) .2592
The probability that component C will work for at least 100 hours is
R C = e −λ C t = e.003(100 ) =.7408
therefore, the probability that component C will fail before 100 hours is
1 - RC = 1 - .7408 = .2592 or 25.9%
173. The MTBF of a complex piece of repairable radar equipment is determined to be 950
hours. The equipment has been in continuous operation for 150 hours. What is the
probability that the equipment will continue to operate without failure for at least
another 375 hours?
a) 0.5754
1 1
λ= = =.0010526
MTBF 950
t = 150 + 375 = 525 hours
R = e − λt = e −(.0010526 )( 525 ) =.5754
94 QReview Study Questions
174. All of the following about multiple regression are true except
Multiple regression implies that more than one independent variable is being used to
observe only one dependent variable.
175. All of the following statements are true about a correlation coefficient except
A correlation coefficient of ±1 indicates perfect correlation, but it does not imply that the
independent variables are related, as to cause and effect, to the dependant variable.
176. A study was conducted on the relationship between the speed of different cars and
their gasoline mileage. The correlation coefficient was found to be 0.35 from the study.
Later, it was discovered that there was a defect in the speedometers and they had all
been set 5 miles per hour too fast. The correlation coefficient was computed using the
correct data. What is the new correlation coefficient?
b) 0.35
$500 $2700
800 3500
1000 4600
1200 7000
1400 9500
2000 11500
c) -1027.4
x x2 Y xY
500 250,000 2,700 1,350,000
800 640,000 3,500 2,800,000
1,000 1,000,000 4,600 4,600,000
1,200 1,440,000 7,000 8,400,000
1,400 1,960,000 9,500 13,160,000
2,000 4,000,000 11,500 23,000,000
6,900 9,290,000 38,800 53,450,000
int ercept = a =
(∑ x)(∑ xY) − ∑ Y(∑ x ) 2
(∑ x) − n(∑ x )
2
2
a) 6.52
(∑ x)(∑ Y)
∑ xY −
n
slope = b =
(∑ x)
2
∑ x −
2
n
(6,900)( 38,800)
53,450,000 − 8,830,000
6 =
b= = 6.52
6900 2 1335
, ,000
9,290,000 −
6
d) .97
96 QReview Study Questions
X (x − x ) Y ( Y − Y) (x − x ) ( Y − Y)
500 -650 2,700 -3,766.67 2,448,335.50
800 -350 3,500 -2,966.67 1,038,334.50
1,000 -150 4,600 -1,866.67 280,000.50
1,200 50 7,000 533.33 26,666.50
1,400 250 9,500 3033.33 758,332.50
2,000 850 11,500 5033.33 4,278,330.50
6,900 38,800 8,829,999.50
1
R= n
∑ [( x − x)( Y − Y)]
σ xσ Y
σ x = 475.22, σ Y = 3,196.18
1 (8,830,000)
6 1471666
, , .7
R= = =.97
(475.22)(3,196.18) 1518, ,887.8
d) Y = -1027.42 + 6.52x
181. What will the projected quality costs savings be when $1700 per employee is invested
in quality training?
b) $10,056.58
182. If the cost savings were $8000 per employee, what are the probable training costs per
employee?
c) $1,384.57
COST OF QUALITY
A quality cost program can be used to formally evaluate the costs associated quality.
Having this financial control enables management to control these costs and identify
opportunities to reduce costs.
d) examining each cost element in relation to other elements and the total.
Quality costs are broken down into different areas: prevention costs, appraisals costs,
internal failure costs and external failure costs. These areas are compared to one
another and to the total so that the relationship can be seen and a balance between
the costs can be obtained.
185. In selecting a base for measuring quality costs, which of the following should be
considered?
186. Which of the following quality cost indices is likely to have the greatest appeal to top
management as an indicator of relative cost?
Comparing quality costs directly to sales is the only true relative measure. Comparing
quality costs to revenue received from sales provides a good measure to determine if
the appropriate amount is being spent quality.
187. If prevention costs are increased to pay for engineering work in quality control, and this
results in a reduction in the number of product defects, this yields a reduction in
c) failure costs.
All costs associated with parts failing to meet quality standards or that cause
manufacturing loss are considered failure costs.
98 QReview Study Questions
188. A process that sorts good product from defective product falls into which of the
following quality cost categories?
c) Internal failure
Any costs associated with defective products are considered failure costs. Any failure
costs discovered before the customer receives the product are considered internal
failure costs.
a) prevention costs.
Prevention costs are costs associated with designing, implementing and maintaining a
quality system. These are costs incurred to reduce other quality costs.
190. The cost of writing instructions and operating procedures for inspection and testing
should be charged to
a) prevention costs.
Writing instructions and operating procedures are prevention costs because they may
significantly reduce problems and the costs associated with resolving the problems, if
properly adhered to.
The fact that the analysis was done on non-conforming hardware and not on the
hardware before it was known to be non-conforming constitutes this as a failure cost.
192. Which of the following is least likely to be reported as a failure related cost?
Typically, this cost is overlooked because there is no direct method to obtain the cost
associated with the late delivery. Often times, other events can take place while
waiting for the part, therefore resulting in a smaller failure cost.
193. Which of the following activities is not normally charged as a preventive cost?
194. In deciding whether sampling inspection of parts would be more economical than 100%
inspection, you need to determine all of the following except
The cost of improving a production process may be of interest after inspection has
been completed and it has been determined that the process needs to be modified to
increase the quality of product.
196. For a typical month, the 3D Manufacturing Company identified and reported the
following quality costs:
d) $332,000
Failure Costs
In-plant scrap and rework 88,000
Retest and troubleshooting 39,000
Field warranty cost 205,000
Total 332,000
a) Material Repair
This cost is primarily the value of time spent on the return and repair of goods that
have been in the customer's hands.
Use the following information to answer problems 198 - 201. A manufacturer produces an
electronic memory device. The following costs are incurred on a yearly basis.
100 QReview Study Questions
Activity Cost
Research and Development $50,000
Manufacturing Engineering 40,000
Quality Engineering 5,000
Plant Facilities 60,000
Training 1,000
Direct Labor 80,000
Process Control (Labor) 5,000
Supervision 30,000
Materials 90,000
Inspection 13,000
Scrap Costs 21,000
Rework Costs 12,000
a) $18,000
Appraisal Costs
Process Control 5,000
Inspection 13,000
Total 18,000
c) $33,000
Failure Costs
Scrap 21,000
Rework 12,000
Total 33,000
DESIGN OF EXPERIMENTS
The model errors are assumed to be normally and independently distributed random
variables with mean zero and variance σ2. The variance σ2 is also assumed constant
for all levels of the factor, which is necessary to test means.
a) the levels used for each factor are the only ones of interest.
The conclusions of the hypothesis test will only apply to the factor levels considered.
The conclusions cannot be extended to similar factor levels that were not considered.
204. An experiment with two factors, in which all levels of one variable are run at each level
of the second variable, is called a
c) factorial experiment.
Factorial experiments are used when several factors are of interest. More specifically,
in each complete trial or replicate of the experiment combinations of the levels of the
factors are investigated.
205. A two-way Analysis of Variance has r levels for the first variable and c levels for the
second variable with 2 observations per cell. The degrees of freedom for interaction is
b) (r - 1)(c - 1)
206. An analysis of variance results in a calculated F value of F(10, 12) = 2.75. What is the
level of significance?
c) p = .05
The F.05 table will yield 2.75 for 10 degrees of freedom in the numerator and 12
degrees of freedom for the denominator.
102 QReview Study Questions
The Xy notation is commonly used in experimental designs. The base (X) indicates the
number of levels. The exponent (y) indicates the number of factors.
208. Which of the following distributions is needed to evaluate the results of analysis of
variance (ANOVA)?
c) F distribution
The F distribution is a ratio of two independent chi square random variables. The chi
square distribution is a distribution of sample variances.
209. The primary advantage of the Latin Square design compared to the factorial design, is
that
A Latin square is an experimental design in which each level of each is combined only
once with each level of two other factors or variables. Also, no interactions exist
between the variables.
210. Consider the SS and MS columns of an Analysis of Variance table for a single factor
design. The appropriate ratio for testing the null hypothesis of no treatment effect is
The MS residual estimates the variance of the experimental error. The MS treatments
will estimate σ2 only if all factor level means are equal, but the value of MS treatments
will be greater than σ2 if the factor level means are different. Dividing one by the other
will provide an F ratio that can be tested against a critical F value to determine if the
factor level means are different.
Use the following information to answer problems 212 - 216. The following factorial
experiment was conducted to determine the effect of study time and study material used on
test scores. The numbers in the box represent the test scores. You will need to construct the
ANOVA table to complete all of the questions.
Hours of Study
4 6
Study Material Used 1 60, 85 77, 92
2 55, 65 60, 80
a) 2
The two factors are hours of study and study material used.
213. How many levels are being examined for each factor?
a) 2
The number of levels is the number of settings for being analyzed for each factor.
a) 675.0
A complete ANOVA table will be constructed to answer this problem and the following
problems.
(∑ x)
2
574 2
CM = = = 41184.5
nT 8
SS total = ∑ x − CM = 42468 − 41184.5 = 1283.5
2
∑ (∑ rows)
2
314 2 260 2
SSmaterial = − CM = + − 41184.5 = 364.5
n rows 4 4
∑ (∑ cells)
2
Source SS df MS F Fcritical
Hours 242.0 1 242.0 1.43 7.71
Material 364.5 1 364.5 2.16 7.71
Interaction 2.0 1 2.0 0.01 7.71
Residual 675.0 4 168.75
SS = Sum of Squares
df = Degrees of Freedom
MS = Mean Square = SS / df = Variance
F = F Ratio = MSsource / MSresidual
The computed F value of the study material used is the largest of any of the factors,
therefore it has the biggest impact on the outcome.
None of the computed F values are larger than the critical value of F, therefore none of
the factors are significant.
Metrology and Calibration 105
The concept is to manufacture the highest quality product in the most efficient manner
possible. For example, a program which requires daily testing of highly accurate
electronic test equipment would probably not be necessary, would be very expensive
to implement and therefore would substantially increase production costs.
c) performance.
219. Which of the following terms is defined as the probability that measurement equipment
will be found to be in tolerance after a specified period of time?
d) Reliability
If a certain class of equipment is tested every six months, and 95% of the units pass
the full verification test, then this class is said to be 95% reliable on a six month
interval.
220. Suppose a test procedure is performed in order to determine the value of a product
measurement. What is the order of equipment type, from lowest to highest accuracy, if
NIST traceability is to be achieved?
Working standards can be calibrated with more accurate transfer standards. Transfer
standards are then calibrated with primary standards of the highest known accuracy, i.
e. NIST standards. The reason for the use of transfer standards is to lower overall test
costs. It could be very expensive to have a complete inventory of working standards
periodically calibrated against a primary standard.
221. Which procedures are performed on product and test equipment in order to determine
if predefined specifications have been met?
106 QReview Study Questions
a) Inspection
A "calibration" can consist of several components. The first step is the "inspection" or
verification portion of the procedure. An adjustment and retest may or may not be
necessary based upon the results of the initial inspection, which normally requires
compliance to predefined specifications.
b) Traceability
Traceability is the chain that ensures measurements have been made in relation to
recognized standards. This effort, which is an ISO standard requirement, is made in
order to promote consistency of measurement and therefore increase overall product
quality.
223. A measurement standard with a stated uncertainty of 0.1 V, 95% confidence level, is
used to test a device that has a nominal value of 100 V, and a tolerance of ±10 V.
What is the TUR for this measurement?
b) 100
The TUR of 100 is the uncertainty (0.1 V) divided by the allowable tolerance (10 V).
TUR = 0.1 V /10 V = 100
The 95% confidence level reflects an industry consensus to express the uncertainty
portion (0.1 V) of the calculation at a 2σ confidence level. The nominal value of 100 V
has no bearing on the calculation.
224. A requirement of the ISO 9000 series standards is that all M&TE used for product
inspection
c) 4.11
This ISO element requires identifying inspection, measuring and test equipment. It also
requires establishing, documenting and maintaining calibration procedures and
records.