Question 4.

1:
The following gives the number of pints of type B blood used at Woodlawn Hospital in the past 6 weeks:

Week of Pints Used

8/31/2014 360
9/7/2014 389
9/14/2014 410
9/21/2014 381
9/28/2014 368
10/5/2014 374

a) Forecast the demand for the week of October 12 using a 3-week moving average.
Answer:
381+368+374
= 374 pints
3

b) Use a 3-week weighted moving average, with weights of .1, .3, and .6, using .6 for the most recent week. F
Answer:
381*.1 = 38.1
368*.3 = 110.4
374*.6 = 224.4
372.9 Forecast

c) Compute the forecast for the week of October 12 using exponential smoothing with a forecast for August 3
Answer:
Week of Actual Forecast = FT = Ft-1 + a(At-1 - Ft-1)
8/31/2014 360 360
9/7/2014 389 360
9/14/2014 410 365.80
9/21/2014 381 374.64
9/28/2014 368 375.91
10/5/2014 374 374.33
10/12/2014 374.26
tal in the past 6 weeks:

g average.

using .6 for the most recent week. Forecast demand for the week of October 12.

othing with a forecast for August 31 of 360 and an a of .2.

Forecast for the Week of Oct. 12th is 374.26

Question 4.3:
Refer to Problem 4.2. Develop a forecast for years 2 through 12 using exponential smoothing with a = .4 and a foreca
on a graph with the actual data and the naive forecast. Based on a visual inspection, which forecast is better?
Answer:
Year Demand Nave Forecast Exponential
1 7.00 6.0 14.00
2 9.00 7 6.4 12.00
3 5.00 9 7.4
4 9.00 5 6.5 10.00
5 13.00 9 7.5
8.00
6 8.00 13 9.7
7 12.00 8 9.0 6.00
8 13.00 12 10.2
9 9.00 13 11.3 4.00
10 11.00 9 10.4
2.00
11 7.00 11 10.6
Forecast 7 9.2 0.00
1 2 3

While the Nave forecast goes higher on the chart, the

forecast that used exponential smoothing seems to be more
consistent.

Nave tracks the ups and downs best, but lags the data by
one period. Thus, it gives quite large errors. Exponential
smoothing is much better because it smoothens the data
and does not have as much variation.
ng with a = .4 and a forecast for year 1 of 6. Plot your new forec
forecast is better?

14.00

12.00

10.00

8.00

6.00

4.00

2.00

0.00
1 2 3 4 5 6 7 8 9 10 11

Demand Nave Forecast Exponential

Question 4.5:
The Carbondale Hospital is considering the purchase of a new ambulance. The decision will rest partly
on the anticipated mileage to be driven next year. The miles driven during the past 5 years are as follows:

Year Mileage
1 3,000
2 4,000
3 3,400
4 3,800
5 3,700

a) Forecast the mileage for next year (6th year) using a 2-year moving average.
Answer:
3,800 + 3,700
= 3,750 miles
2

b) Find the MAD based on the 2-year moving average. (Hint: You will have only 3 years of matched data.).
Answer:
2 YR Moving
Year Mileage Error Error
Average
1 3000
2 4000
3 3400 3500 -100 100
4 3800 3700 100 100
5 3700 3600 100 100
Totals 100 300
MAD 100

c) Use a weighted 2-year moving average with weights of .4 and .6 to forecast next years mileage. (The weig
recent year.). What MAD results from using this approach to forecasting? (Hint: You will have only 3 years
Answer:
Year Mileage 2 YR Moving Error Error
Average
1 3000
2 4000
3 3400 3600 -200 200
4 3800 3640 160 160
5 3700 3640 60 60
Totals 20 420
MAD 140

d) Compute the forecast for year 6 using exponential smoothing, an initial forecast for year 1 of 3,000 miles, a
Answer:
Year Milage Forecast Forecast Error Error * 0.5
1 3000 3000 0 0
 2 4000 3000 1000 500
3 3400 3500 -100 -50
4 3800 3450 350 175
5 3700 3625 75 38
Total 1325

Year 6's forecast would be 3663

ion will rest partly
years are as follows:

only 3 years of matched data.).

t next years mileage. (The weight of .6 is for the most

Hint: You will have only 3 years of matched data.)

ecast for year 1 of 3,000 miles, and a = .5.

New Forecast
3000
3500
3450
3625
3663
Question 4.25:
The following gives the number of accidents that occurred on Florida State Highway 101 during the past 4 months:

MONTH NUMBER OF ACCIDENTS

January 30
February 40
April 60
March 90

Forecast the number of accidents that will occur in May, using least-squares regression to derive a trend equation.

Answer:
Month (X) Number of Accidents (Y)
1 Jan 30
2 Feb 40
3 Mar 60
4 Apr 90

X Y X2 XY
1 30 1 30
2 40 4 80
3 60 9 180
4 90 16 360
Sum 30 650
Average 2.5 55

Trend Equation Y = A+BX

b 20
a 5

Trend Equation Y = 5 + 20x

The regression line is Y = 5+20x. The forecast for May (X = 5) is:

Y= 105
ring the past 4 months:

rive a trend equation.

Slope(b) = (NXY - (X)(Y)) / (NX2 - (X)2)

Question 4.27
George Kyparisis owns a company that manufactures sailboats. Actual demand for Georges sailboats during each of

YEAR
SEASON 1 2 3 4
Winter 1400 1200 1000 900
Spring 1500 1400 1600 1500
Summer 1000 2100 2000 1900
Fall 600 750 650 500

George has forecasted that annual demand for his sailboats in year 5 will equal 5,600 sailboat
Based on this data and the multiplicative seasonal model, what will the demand level be for
Georges sailboats in the spring of year 5?
Answer:

YEAR

SEASON 1 2 3 4
Winter 1,400 1,200 1,000 900
Spring 1,500 1,400 1,600 1,500
Summer 1,000 2,100 2,000 1,900
Fall 600 750 650 500
Total Average Annual Demand =
r Georges sailboats during each of the past four seasons was as follows:

AVG.
AVG. Yearly Seasonal
Monthly Year 5
Demand Index
Demand

1,125 1,250 0.9 1,260

1,500 1,250 1.2 1,680
1,750 1,250 1.4 1,960
625 1,250 0.5 700 Demand level for the spring of year 5 will be 1,68
5,000
of year 5 will be 1,68
Question 6.12:

Sr No Type of complaint Freq Cumulative Freq Cum Percent

1 Parking /Drives 38 38 38.8%
2 Pool 22 60 61.2%
3 Tenant Issues 18 78 79.6%
4 Grounds 16 94 95.9%
5 Electrical/Plumbing 4 98 100.0%

Steps:
1) First add up each type of compliant over 12 weeks period. This gives us Freq of Each complaint.
2) We now see that we have 5 type of complaints.
3) We sort the table using Excel Home Sort & Filter Largest to smallest. We expand selection to include Ty
compliant column.
4) Now we have table sorted by largest no of compliants to smallest.
5) Insert a column Cumulative freq
6) Add freq from prev column one by one to get cumulative freq. Last cell gives 98 which is total of all complia
7) Insert another column Cumulative percent
8) Now divide each number of cumulative freq by the last number (which is total 98).
9) We get the cum percent nos which indicates that the compliant is what % of total complaint.
10) Now we plot them in excel using Graph function.

From Pareto chart we see that 80% of the problems are due to Parking/Drives, Pool and Tenant issues.
So Mary should first address Parking/Drive issues followed by Pool and then Tenant issues. This will
address 80% of the complaints. Post this Mary should again record the complaints in next 12 weeks &
then repeat the Pareto chart to identify the new issues which should be addressed on priority.
 N Pareto Chart
o
40 100.0% C
o u
f 35 90.0% m
u
C 80.0% l
30
o 70.0% a
m t
p 25 60.0% i
l v
a 20 50.0% e
complaint. i
15 40.0% %
n
t 30.0%
d selection to include Type of s 10
20.0%
5 10.0%
0 0.0%
h is total of all compliants s ol s
i ve ue ds ng
Po s un bi
Dr Is o m
g/ nt Gr lu
in na l /P
a rk Te ca
plaint. P tri
ec
El

Complaints
00.0% C
u
0.0% m
u
0.0% l
0.0% a
t
0.0% i
v
0.0% e
0.0% %
0.0%
0.0%
0.0%
.0%
Question S6.11:

a) Sample (n) Sample Mean (A2) Range (in.) UCLx

1 10.002 0.011 10.11552
2 10.002 0.014 10.11552
3 9.991 0.007 10.11552
4 10.006 0.022 10.11557
5 9.997 0.013 10.11547
6 9.999 0.012 10.11549
7 10.001 0.008 10.11551
8 10.005 0.013 10.11556
9 9.995 0.004 10.11544
10 10.001 0.011 10.11551
11 10.001 0.014 10.11551
12 10.006 0.009 10.11557
10.0005 0.0115

b) 0.025
Sample Mean
0.02
10.01
10.005 0.015
10 Sample Mean
9.995 0.01
9.99
9.985 0.005
9.98
0 2 4 6 8 10 12 14 0
0 2 4 6 8
c) Yes

d) The process is in control because all values fall within the upper and lower limits
 LCLx UCLR LCLR
9.88548 0.116329 0.113683
9.88548 0.116329 0.113683
9.88548 0.116329 0.113683
9.88543 0.116329 0.113682
9.88553 0.116328 0.113684
9.88551 0.116328 0.113683
9.88549 0.116328 0.113683
9.88544 0.116329 0.113683
9.88556 0.116328 0.113684
9.88549 0.116328 0.113683
9.88549 0.116328 0.113683
9.88543 0.116329 0.113682

Column D

0 2 4 6 8 10 12
Question S6.20:

a)
Quality Control p chart

Number of samples 10
Sample size 200 Enter
Enter the
the sample
sample size
size then
then enter
enter the
the number
number of
of defects
defects in
in
each
each sample.
sample.

Data Results
# Defects % Defects Total Sample Size 2000
Sample 1 5 0.025 Total Defects 50
Sample 2 7 0.035 Percentage defects 0.025
Sample 3 4 0.02 Std dev of p-bar 0.0110397011
Sample 4 4 0.02 z value 3
Sample 5 6 0.03
Sample 6 3 0.015 Upper Control Limit 0.0581191032
Sample 7 5 0.025 Center Line 0.025
Sample 8 6 0.03 Lower Control Limit 0
Sample 9 2 0.01
Sample 10 8 0.04

0.07

0.06

Graph information P LCL CL UCL 0.05

Sample 1 0.025 0 0.025 0.0581191 0.04
Mean

Sample 2 0.035 0 0.025 0.0581191

0.03
Sample 3 0.02 0 0.025 0.0581191
Sample 4 0.02 0 0.025 0.0581191 0.02
Sample 5 0.03 0 0.025 0.0581191 0.01
Sample 6 0.015 0 0.025 0.0581191
Sample 7 0.025 0 0.025 0.0581191 0
1 2 3
Sample 8 0.03 0 0.025 0.0581191
Sample 9 0.01 0 0.025 0.0581191
Sample 10 0.04 0 0.025 0.0581191
 b) The process is in control
c) For 100 sample size the control limits will change

p-chart

P
LCL
CL
UCL

2 3 4 5 6 7 8 9 10
Sample
Question S6.23:

The school board is trying to evaluate a new math program introduced to second-graders in five elementary schools a
standardized math tests in each elementary school yielded the following data: School No. Of Test Errors A 52 B 27 C
control limits to contain 99.73% of the random variation in test scores. What does the chart tell you? Has the new ma

c School No. of test Errors

52 A 52
27 B 27
35 C 35
44 D 44
55 E 55

c-bar = 42.6

See the sheet c- chart. The control limits are LCL = 23.02 and UCL = 62.181

Quality Control Process Charts:

Number of samples 5

Defects per sample

Upper Control Limit, UCL 62.181
Center 42.6
Lower Control Limit, LCL 23.02

Control Chart for the Number of Defects

70

60

50

40

30

20

10

0
1 2 3 4 5
ond-graders in five elementary schools across the county this year. A sample of the student scores on
: School No. Of Test Errors A 52 B 27 C 35 D 44 E 55 Construct a c-chart for test errors, and set the
does the chart tell you? Has the new math program been effective?

Results UCL LCL

Total Defects 213 62.180 23.02
Defect Rate 42.6 62.180 23.02
Standard Deviation 6.53 62.180 23.02
z value 3 62.180 23.02
UCL 62.18 62.180 23.02
LCL 42.6

5
Question S6.27:

Meena Chavan Corp.s computer chip production process yields DRAM chips with an average life of 1,800
hours and s = 100 hours. The tolerance upper and lower specification limits are 2,400 hours and 1,600 hours,
respectively. Is this process capable of producing DRAM chips to specification?

Answer:

Average life 1,800 Upper Specification 2

100 Lower Specification 0.67
UCL 2,400
LCL 1,600

Because the lower limit of the Cpk is 0.67, the process is not capable of producing DRAM chips to specification
age life of 1,800
and 1,600 hours,

chips to specification
 Question S6.35:

b)
Sample Late Proportion Airline Airline
(week) Flights Late p UCL LCL Industry UCL Industry LCL
1 2 0.02 0.04 0.08 0.00 0.10 0.04
2 4 0.04 0.04 0.08 0.00 0.10 0.04
3 10 0.1 0.04 0.08 0.00 0.10 0.04
4 4 0.04 0.04 0.08 0.00 0.10 0.04
5 1 0.01 0.04 0.08 0.00 0.10 0.04
6 1 0.01 0.04 0.08 0.00 0.10 0.04
7 13 0.13 0.04 0.08 0.00 0.10 0.04
8 9 0.09 0.04 0.08 0.00 0.10 0.04
9 11 0.11 0.04 0.08 0.00 0.10 0.04
10 0 0 0.04 0.08 0.00 0.10 0.04
11 3 0.03 0.04 0.08 0.00 0.10 0.04
12 4 0.04 0.04 0.08 0.00 0.10 0.04
13 2 0.02 0.04 0.08 0.00 0.10 0.04
14 2 0.02 0.04 0.08 0.00 0.10 0.04
15 8 0.08 0.04 0.08 0.00 0.10 0.04
16 2 0.02 0.04 0.08 0.00 0.10 0.04
17 3 0.03 0.04 0.08 0.00 0.10 0.04
18 7 0.07 0.04 0.08 0.00 0.10 0.04
19 3 0.03 0.04 0.08 0.00 0.10 0.04
20 2 0.02 0.04 0.08 0.00 0.10 0.04
21 3 0.03 0.04 0.08 0.00 0.10 0.04
22 7 0.07 0.04 0.08 0.00 0.10 0.04
23 4 0.04 0.04 0.08 0.00 0.10 0.04
24 3 0.03 0.04 0.08 0.00 0.10 0.04
25 2 0.02 0.04 0.08 0.00 0.10 0.04
26 2 0.02 0.04 0.08 0.00 0.10 0.04
27 0 0 0.04 0.08 0.00 0.10 0.04
28 1 0.01 0.04 0.08 0.00 0.10 0.04
29 3 0.03 0.04 0.08 0.00 0.10 0.04
30 4 0.04 0.04 0.08 0.00 0.10 0.04
a) Number of Samples = 30
Sample size = 100
For 95% confidence level, z value = 1.96

p= 0.04
UCL = 0.08
LCL = 0.00

c)
Chart Title
0.14

0.12

0.10
P-bar
Proportion Defective

UCL
0.08 LCL
Airline UCL
Airline LCL
0.06
Percentage of Late Flight

0.04

0.02

0.00
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930

d)
The quality of service is within the 95% confidence limits most of the time, however in
weeks 3,7,8,9 the number of delays exceeded the upper contol limit.
The airlne exceeded the upper limit of delay for airline industry only in week 7 and week
9. There might be some special circumstances liek weathe problems, etc during that time
leading to delay. In other weeks the delay rates where within 95% confidence limit.
e UCL
e LCL
entage of Late Flight