Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

40 QUALITY CONTROL AND INSPECTION

Review Questions
40.1 What are the two principal aspects of product quality?
Answer. The two quality aspects are (1) product features and (2) freedom from deficiencies.
40.2 How is a process operating in statistical control distinguished from one that is not?
Answer. The process in statistical control is characterized by only random variations. A process
that is out of control exhibits additional variation that is not normal, called assignable variation.
This indicates that something is wrong with the process.
40.3 Define process capability.
Answer. Process capability is the limits of the random variations of the process when it is in
statistical control. The limits are defined as the process mean 3 standard deviations.
40.4 What are the natural tolerance limits?
Answer. The natural tolerance limits are when the tolerance on a part is set equal to the process
capability; that is 3 of the process mean.
40.5 What is the difference between control charts for variables and control charts for attributes?
Answer. In control charts for variables, measurements of the characteristic of interest are made. In
control charts for attributes, the characteristic of interest is identified as being acceptable of not
acceptable.
40.6 Identify the two types of control charts for variables.
Answer. The two charts are (1) x chart for sample means and (2) R chart for ranges.
40.7 What are the two basic types of control charts for attributes?
Answer. The two charts are (1) p chart for proportion of defects in a sample, and (2) c chart for
count of defects in a sample.
40.8 When interpreting a control chart, what does one look for to identify problems?
Answer. Problems are indicated by the following: (1) x or R lie outside their respective LCL or
UCL limits; (2) trends or cyclical patterns in the data; (3) sudden changes in average; and (4) points
consistently near the upper or lower limits.
40.9 What are the three main goals in total quality management (TQM)?
Answer. The three main goals in TQM are (1) achieving customer satisfaction, (2) encouraging
the involvement of the entire workforce, and (3) continuous improvement.
40.10 What is the difference between external customers and internal customers in TQM?
Answer. External customers are those who purchase the companys products and services.
Internal customers are inside the company, such as the companys final assembly department
which is the customer of the parts production departments.
40.11 At what company was the Six Sigma quality program first used?
Answer. Motorola Corporation.

Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-1
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

40.12 Why is the Normal statistical table used in a Six Sigma program different from the standard normal
tables found in textbooks on probability and statistics?
Answer. The Normal statistical tables used in a Six Sigma program differ from the standard normal
tables in the following two ways: (1) the Six Sigma table includes only one tail of the normal
distribution and (2) the Six Sigma table is shifted by 1.5, so that 6 in the Six Sigma table is the
same as 4.5 in the standard normal tables.
40.13 A Six Sigma program uses three measures of defects per million (DPM) to assess the performance
of a given process. Name the three measures of DPM.
Answer. The three measures of DPM are (1) defects per million opportunities, (2) defects per
million units, and (3) defective units per million units.
40.14 What is meant by robust design, as defined by Taguchi?
Answer. Robust design means that a product or process is designed so that its function and
performance are relatively insensitive to variations in design and manufacturing parameters that are
bound to occur.
40.15 Automated inspection can be integrated with the manufacturing process to accomplish certain
actions. What are these possible actions?
Answer. Possible actions discussed in text are (1) parts sortation and (2) feedback of data to
adjust the process.
40.16 Give an example of a noncontact inspection technique.
Answer. Non-contact inspection techniques include machine vision, laser measuring methods,
and electrical field techniques.
40.17 What is a coordinate measuring machine?
Answer. A CMM is an automated measuring machine consisting of a contact probe and a means
to position the probe in three dimensions relative to workpart features and surfaces; when the
probe contacts the part, the x-y-z coordinates are recorded.
40.18 Describe a scanning laser system.
Answer. The scanning laser system uses a laser beam deflected by a rotating mirror to produce a
beam of light that sweeps past an object. A photodetector on the far side of the object senses the
light beam during its sweep except for the short time when it is interrupted by the object. This time
period can be measured quickly with great accuracy. A microprocessor system measures the time
interruption that is related to the size of the object in the path of the laser beam, and converts from
time to a linear dimension.
40.19 What is a binary vision system?
Answer. In a binary vision system, the light intensity of each pixel is reduced to either of two
values (black or white, 0 or 1).
40.20 Name some of the nonoptical noncontact sensor technologies available for inspection.
Answer. The technologies include electrical fields (capacitance, inductance), radiation (X-ray), and
ultrasonic techniques (high frequency sound).
Problems
Answers to problems labeled (A) are listed in an Appendix at the back of the book.

Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-2
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

Process Capability and Tolerances

40.1 (A) (SI units) A turning operation produces parts with a mean diameter = 5.620 cm. The process is
in statistical control, and the output is normally distributed with a standard deviation = 0.005 cm.
Determine the process capability.
Solution: Process capability PC = 3 = 5.620 3(0.005) = 5.620 0.015 cm
The upper and lower limits of the process capability range are: 5.605 to 5.635 cm
40.2 A sheet-metal V-bending operation bends parts to an included angle = 46.1. The process is in
statistical control, and values of included angle are normally distributed with a standard deviation =
0.33. (a) Determine the process capability. (b) The design specification on the angle is 45 1. If
the process were adjusted so that its mean = 45.0, determine the value of the process capability
index. Also, using Table 40.1, estimate the proportion of defects that would be produced at the 45
mean.
Solution: (a) PC = 46.1 3(0.22) = 46.1 0.66
The upper and lower limits of the process capability range are: 45.44 to 46.76
(b) Given = 45 and tolerance range T = 46 - 44 = 2
PCI = 2/(6 x 0.33) = 1.01
Based on Table 40.1, this PCI indicates a fraction defect rate of less that 0.27%.
40.3 (SI units) A plastic extrusion process produces round tubular extrudate with a mean outside
diameter = 25.5 mm. The process is in statistical control, and the output is normally distributed with
standard deviation = 0.40 mm. (a) Determine the process capability. (b) The design specification on
the OD is 25.0 mm 0.75 mm. If the process were adjusted so that its mean = 45.0, determine the
value of the process capability index. Also, using Table 40.1, estimate the proportion of defects that
would be produced at the 45 mean.
Solution: (a) Process capability PC = 3 = 25.5 3(0.20) = 25.5 0.60 mm
The upper and lower limits of the process capability range are: 24.9 to 26.1 mm.
(b) Given = 25.0 mm and tolerance range T = 25.75 24.25 = 1.5
PCI = 1.5/(6 x 0.40) = 0.625
Based on Table 40.1, this PCI indicates a fraction defect rate of around 5% or more.
Control Charts
40.4 (A) (SI units) In 20 samples of size n = 6, the average value of the sample means is x = 3.781 cm
for the dimension of interest, and the mean of the ranges of the samples is R = 0.033 cm.
Determine (a) lower and upper control limits for (a) the x chart and (b) the R chart.
Solution: (a) x chart: x = 3.781 cm = CL
LCL = x - A2 R = 3.781 - 0.483(0.033) = 3.765 cm
UCL = x + A2 R = 3.781 + 0.483(0.033) = 3.797 cm
(b) R chart: R = 0.033 = CL
LCL = D3 R = 0(0.033) = 0.0
UCL = D4 R = 2.004(0.033) = 0.0661 cm
40.5 In 15 samples of size n = 9, the grand mean of the samples is x = 97 for the characteristic of
interest, and the mean of the ranges of the samples is R = 7.2. Determine (a) lower and upper
control limits for (a) the x chart and (b) the R chart.
Solution: (a) x chart: x = 97 = CL
Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-3
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

LCL = x - A2 R = 97 - 0.337(7.2) = 94.57

UCL = x + A2 R = 97 + 0.337(7.2) = 99.43
(b) R chart: R = 8.5 = CL
LCL = D3 R = 0.184(7.2) = 1.325
UCL = D4 R = 1.816(7.2) = 13.075
40.6 (SI units) Ten samples of size n = 7 have been collected from a process in statistical control, and the
dimension of interest has been measured for each part. The calculated values of x for each sample
are (mm) 9.22, 9.15, 9.20, 9.28, 9.19, 9.12, 9.20, 9.24, 9.17, and 9.23. The values of R are (mm)
0.24, 0.17, 0.30, 0.26, 0.26, 0.19, 0.21, 0.32, 0.21, and 0.23, respectively. (a) Determine the values
of the center, LCL, and UCL for the x and R charts. (b) Construct the control charts and plot the
sample data on the charts.
Solution: x = x /m = x /10
= (9.22 + 9.15 + 9.20 + 9.28 + 9.19 + 9.12 + 9.20 + 9.24 + 9.17 + 9.23)/10 = 9.20
R = R/10 = (0.24 + 0.17 + 0.30 + 0.26 + 0.27 + 0.19 + 0.21 + 0.32 + 0.21 + 0.23)/10 = 0.24
(a) x chart: x = 9.20 mm = CL
LCL = x - A2 R = 9.20 - 0.419(0.24) = 9.099 mm
UCL = x + A2 R = 9.20 + 0.419(0.24) = 9.301 mm
R chart: R = 0.024 = CL
LCL = D3 R = 0.076(0.24) = 0.0182 mm
UCL = D4 R = 1.924(0.24) = 0.4618 mm
(b) Student exercise.
40.7 (USCS units) Seven samples of 5 parts each have been collected from an extrusion process that is in
statistical control, and the diameter of the extrudate has been measured for each part. The calculated
values of x for each sample are 1.002, 0.999, 0.995, 1.004, 0.996, 0.998, and 1.006 (in). The
values of R are 0.010, 0.011, 0.014, 0.020, 0.008, 0.013, and 0.017 (in), respectively. (a) Determine
the values of the center, LCL, and UCL for x and R charts. (b) Construct the control charts and
plot the sample data on the charts.
Solution: x = x /7 = (1.002 + 0.999 + 0.995 + 1.004 + 0.996 + 0.998 + 1.006)/7 = 1.000
R = R/7 = (0.010 + 0.011 + 0.014 + 0.020 + 0.008 + 0.013 + 0.017)/7 = 0.0133
(a) x chart: x = 1.000 in = CL
LCL = x - A2 R = 1.000 - 0.577(0.0133) = 0.9923 in
UCL = x + A2 R = 1.000 + 0.577(0.0133) = 1.0077 in
R chart: R = 0.0133 = CL
LCL = D3 R = 0(0.0133) = 0
UCL = D4 R = 2.114(0.0133) = 0.0281 in
(b) Student exercise.
40.8 In an x chart used for an industrial process, UCL = 53.5 and LCL = 39.9 for a critical process
variable of interest. The following ten sample means were collected at random times from the
process: 45.6, 47.2, 49.3, 46.8, 48.8, 51.0, 46.7, 50.1, 49.5, and 48.9. (a) Plot the data together with
the CL, LCL, and UCL on a piece of graph paper. (b) Interpret the data. Is anything amiss?
Solution: The central line CL = (UCL + LCL)/2 = (53.5 + 39.9)/2 = 46.7. All of the data points lie
inside the upper and lower control limits. However, only the first data point (45.6) is below the CL.
The rest are above CL. That is like flipping 10 coins and getting 9 out of 10 heads or 9 out of 10
Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-4
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

tails. In addition, if the first five data values are averaged (47.54) and compared to the average of
the last five points (49.24), an upward moving trend is suggested. To sum up, the data are not
consistent with a process that is in statistical control. An investigation should be launched to figure
out what is causing the process to behave like this.
40.9 (A) A p chart is based on six samples of 40 parts each. The average number of defects per sample is
1.6. Determine the center, LCL and UCL for the p chart.
Solution: p = 1.6/40 = 0.04 = CL
LCL = p - 3 p (1 p ) / n = 0.04 - 3 0.04(0.96) / 40 = 0.04 - 3(0.031) = -0.053 0
UCL = p +3 p (1 p ) / n = 0.04 + 3 0.04(0.96) / 40 = 0.04 + 3(0.031) = 0.133
40.10 Eight samples of equal size are taken to prepare a p chart. The total number of parts in these eight
samples = 480 and the total number of defects counted was 72. Determine the center, LCL and UCL
for the p chart.
Solution: d = 72/8 = 9.0 and n = 480/8 = 60
p = d /n = 9.0/60 = 0.15 = CL

LCL = p - 3 p(1 p) / n = 0.15 - 3 0.15(0.85) / 60 = 0.15 - 3(0.0461) = 0.0117

UCL = p + 3 p(1 p) / n = 0.15 + 3 0.15(0.85) / 60 = 0.15 + 3(0.0461) = 0.2883
40.11 The yield of good chips during a critical step in silicon processing of integrated circuits averages
96%. The number of chips per wafer is 250. Determine the center, LCL, and UCL for the p chart
that might be used for this process.
Solution: Use p = 1 - 0.96 = 0.04 = CL
LCL = p - 3 p(1 p) / n = 0.04 - 3 0.04(0.96) / 250 = 0.04 - 3(0.0124) = 0.00282
UCL = p + 3 p(1 p) / n = 0.04 + 3 0.04(0.96) / 250 = 0.04 + 3(0.0124) = 0.07718
40.12 The upper and lower control limits for a p chart are: LCL = 0.15 and UCL = 0.27. Determine the
sample size n that is used with this control chart.
Solution: p = 0.5(LCL + UCL) = 0.5(.15 + .27) = 0.21
UCL - LCL = 0.27 - 0.15 = 0.12 = 6 p(1 p) / n = 6 0.12(0.88) / n
(0.12)2 = 62 (0.12 x 0.88/n)
0.0144 = 36(0.12)(0.88)/n = 3.8016/n
n = 3.8016/0.0144 = 264
40.13 The upper and lower control limits for a p chart are: LCL = 0 and UCL = 0.30. Determine the
minimum possible sample size n that is compatible with this control chart.
Solution: CL = p = 0.5(UCL + LCL) = 0.5(.30 + 0) = 0.15
LCL = 0 = p - 3 p(1 p) / n
Therefore, p = 3 p(1 p) / n
0.15 = 3 0.15( 0.85 ) / n
(0.15)2 0.0225 = 0.01 = 32 (0.15)(0.85)/n = 1.1475/n
n = 1.1475/0.0225 = 51
40.14 Twelve cars were inspected after final assembly. The number of defects found ranged between 87
and 139 defect per car with an average of 110. Determine the center and upper and lower control
limits for the c chart that might be used in this situation.
Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-5
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

Solution: CL = 110
LCL = c - 3 c = 110 - 3 110 = 78.5 78
UCL = c + 3 c = 110 + 3 110 = 141.5 142
Quality Programs
40.15 (A) A foundry that casts turbine blades inspects for six features considered critical-to-quality.
During the previous month, 948 castings were produced. During inspection, 37 defects among the
six features were found, and 21 castings had one or more defects. Determine DPMO, DPM, and
DUPM in a Six Sigma program for these data and convert each to its corresponding sigma level.

Solution:Summarizing the data, Nu = 948, No = 6, Nd = 37, and Ndu = 21. Thus,

37
DPMO = 1,000,000 = 6505
948 ( 6 )
The corresponding sigma level is about 4.0 from Table 40.3.
37
DPM = 1,000,000 = 39,030
948
The corresponding sigma level is about 3.2.
21
DUPM = 1,000 ,000 = 22,152
948
The corresponding sigma level is about 3.5.
40.16 In the previous problem, if the foundry desired to improve its quality performance to the 5.0
sigma level in all three measures of DPM, how many defects and defective units would they
produce in an annual production quantity of 10,000 castings? Assume the same six features are
used to assess quality.

Solution:Summarizing the data, Nu = 10,000, No = 6, and Nd and Ndu are unknown. To achieve a
5.0 sigma level, they would produce 233 dpm, according to Table 40.3. Thus, for defects per
million opportunities,
Nd
DPMO = 1,000,000 = 233
10 ,000 ( 6 )
Rearranging, Nd = 233(10,000)(6)/1,000,000 = 13.98, rounded up to 14 defects in 10,000 total
castings.
Similarly, if the 5.0 sigma criterion were used for defects per million units,
Nd
DPM = 1,000 ,000 = 233
10 ,000
Rearranging, Nd = 233(10,000)/1,000,000 = 2.33 defects in 10,000 total castings.
Finally, if the 5.0 sigma criterion were used for defective units per million units,
N du
DUPM = 1, 000,000 = 233
10,000
Ndu = 233(10,000)/1,000,000 = 2.33 defective units in 10,000 total castings.
40.17 The inspection department in an automobile final assembly plant inspects cars coming off the
production line against 55 quality features considered important to customer satisfaction. The
department counts the number of defects found per 100 cars, which is the same type of metric
used by a national consumer advocate agency. During a one-month period, a total of 16,582 cars
rolled off the assembly line. These cars included a total of 6045 defects of the 55 features, which
translates to 36.5 defects per 100 cars. In addition, a total of 1955 cars had one or more of the

Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-6
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

defects during this month. Determine DPMO, DPM, and DUPM in a Six Sigma program for
these data and convert each to its corresponding sigma level.

Solution:Although the inspection department uses number of defects per 100 cars, a Six Sigma
program uses defects per million as its metric. Summarizing the data, Nu = 16,582, No = 55, Nd =
6045, and Ndu = 1955. Thus,
6045
DPMO = 1,000,000 = 6628
16582 ( 55 )
The corresponding sigma level is about 4.0 from Table 40.3.
6045
DPM = 1,000 ,000 = 364,552
16582
The corresponding sigma level is about 1.8.
1955
DUPM = 1,000 ,000 = 117,899
16582
The corresponding sigma level is about 2.7.
40.18 (A) (USCS units) A company produces a certain part whose most important dimension is 37.50
0.025 in. If the tolerance is exceeded, the customer returns the part to the manufacturer at a cost
of $200 in rework and replacement expenses. (a) Determine the constant k in the Taguchi loss
function, Equation (40.13). (b) The company can add a finish grinding operation that will allow the
tolerance to be reduced to 0.010 in. Using the loss function from part (a), what is the value of the
loss associated with this new tolerance?
Solution:(a) In Equation (40.13), the value of (x N) is the tolerance 0.025 in. The loss is the
expected cost of rework and replacement, which is $200. Using this cost in the loss function, the
value of k can be determined as follows:
200 = k(0.025)2 = 0.000625k
k = 200/0.000625 = $320,000
Accordingly, the Taguchi loss function is L(x) = 320,000(x N)

(b) The value of the loss for a tolerance of 0.010 is L(x) = 320,000(0.010)2 = $32.00
40.19 (USCS units) The additional operation in the preceding problem will add $3.00 to the current cost
of the part, which is $15.00. If the rate of returns from the customer at the tolerance of 0.025 in
is 3.1%, and it is expected to drop to zero returns using the new tolerance, should the company
add the finish grinding operation to the manufacturing sequence for the part? Answer this
question using the basic cost and return rate data without consideration of the Taguchi loss
function.
Solution: The basic cost and return rate data are that 3.1% of the parts are returned, and it costs
$200 for each one that is returned. Comparing the cost per part with and without the grinding
operation, we have
Without the grinding operation, cost/pc = $15.00 + 0.031($200) = 15.00 + 6.20 = $21.20
With the added operation, cost/pc = 15.00 + 3.00 = $18.00
The calculation favors the addition of the grinding operation.
Laser Measurement Technologies
40.20 (USCS units) A laser triangulation system has the laser mounted at a 35 angle from the vertical.
The distance between the worktable and the photodetector is 24.0000 in. Determine (a) the
Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-7
Solutions for Fundamentals of Modern Manufacturing, 5e (published by Wiley) MPGroover 2012

distance between the laser and the photodetector when no part is present and (b) the height of a
part when the distance between the laser and photodetector is 12.0250 in.
Solution: (a) L with no part when D = 0: D=H L cot A;
0 = H L cot A; L = H/cot A = H tan A
L = 24.0000 tan(35) = 16.8050 in
(b) D = H L cot A = 24.0000 12.0250 cot(35)
D = 24.0000 12.0250(1.42815) = 6.8265 in
40.21 (SI units) A laser triangulation system is used to determine the height of a steel block. The system
has a photosensitive detector that is located 750.000 mm above the working surface and the laser
is mounted at a 30.00 angle from the vertical. With no part on the worktable, the position of the
laser reflection on the photosensor is recorded. After a part is placed on the worktable, the laser
reflection shifts 70.000 mm toward the laser. Determine the height of the object.
Solution: Need to find the distance of the reflection from the laser.
L with no part when D=0: D = H L cot A; 0 = H L cot A; L = H/cot A = H tan A
L = 750 tan(30) = 433.0127 mm
L with part = 433.0127 70 = 363.0127 mm
D = H L cot A = 750 363.0127 cot(30) = 750 363.0127(1.732) = 121.262 mm

Excerpts from this work may be reproduced by instructors for distribution on a not-for-profit basis for testing or instructional purposes only to
students enrolled in courses for which the textbook has been adopted. Any other reproduction or translation of this work beyond that permitted
by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful.
40-8