File: ch12, Chapter 12: Forecasting
True/False
1. Forecasts based on mathematical formulas are referred to as qualitative forecasts.
Ans: False
Difficulty: Easy
Learning Objective: LO 1
2. Because of globalization of markets, managers are finding it increasingly more
difficult to create accurate demand forecasts.
Ans: True
Difficulty: Moderate
Learning Objective: LO 1
3. Forecasting customer demand is often a key to providing good quality service.
Ans: True
Difficulty: Easy
Learning Objective: LO 1
4. One way to deal with the bullwhip effect is to develop and share the forecasts with
other supply chain members.
Ans: True
Difficulty: Easy
Learning Objective: LO 1
5. Qualitative forecasts use mathematical techniques and statistical formulas.
Ans: False
Difficulty: Moderate
Learning Objective: LO 1
6. In today’s competitive environment, effective supply chain management requires
absolute demand forecasts.
Ans: False
Difficulty: Moderate
Learning Objective: LO 1
7. Sharing demand forecasts with supply chain members has resulted in an increased
bullwhip effect.
Ans: False
Difficulty: Moderate
Learning Objective: LO 1
8. Because of advances in technology, many service industries no longer require
accurate forecasts to provide high quality service.
Ans: False
Difficulty: Moderate
Learning Objective: LO 1
9. The type of forecasting method used depends entirely on whether the supply chain is
continuous replenishment or not.
Ans: False
Difficulty: Moderate
Learning Objective: LO 1
10. Continuous replenishment systems rely heavily on extremely accurate long-term
forecasts.
Ans: False
Difficulty: Easy
Learning Objective: LO 1
11. The type of forecasting method selected depends on time frame, demand behavior
and causes of behavior.
Ans: True
Difficulty: Hard
Learning Objective: LO 2
12. A gradual, long-term up or down movement of demand over time is referred to as a
trend.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
13. A seasonal pattern is an oscillating movement in demand that occurs periodically over
the short-run and is repetitive.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
14. Short-midrange forecasts tend to use quantitative models that forecast demand based
on historical demand.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
15. Long-range quantitative forecasts are used to determine future demand for new
products, markets, and customers.
Ans: False
Difficulty: Moderate
Learning Objective: LO 2
16. The trend toward continuous replenishment in supply chain design has shifted the
need for accurate forecasts from long-term to short-term.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
17. The type of forecasting method selected depends on time frame, demand behavior,
and causes of behavior.
Ans: True
Difficulty: Hard
Learning Objective: LO 2
18. Because of heightened competition resulting from globalization most companies find
little strategic value in long-range forecasts.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
19. Movements in demand that do not follow a given pattern are referred to as random
variations.
Ans: True
Difficulty: Easy
Learning Objective: LO 2
20. Many companies are shifting from long-term to short-term forecasts for strategic
planning.
Ans: False
Difficulty: Moderate
Learning Objective: LO 2
21. The demand behavior for skis is considered cyclical.
Ans: False
Difficulty: Moderate
Learning Objective: LO 2
22. The long-term strategic planning process is dependent upon qualitative forecasting
methods.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
23. The Delphi method generates forecasts based on informed judgments and opinions
from knowledgeable individuals.
Ans: True
Difficulty: Moderate
Learning Objective: LO 2
24. The most common type of forecasting method for long-term strategic planning is
based on quantitative modeling
Ans: False
Difficulty: Moderate
Learning Objective: LO 2
25. Time series methods use historical data to predict future demand.
Ans: True
Difficulty: Moderate
Learning Objective: LO 3
26. One reason time series methods are popular for forecasting is that they are relatively
easy to use and understand.
Ans: True
Difficulty: Moderate
Learning Objective: LO 3
27. Exponential smoothing is an averaging method for forecasting that reacts more
strongly to recent changes in demand.
Ans: True
Difficulty: Moderate
Feedback: Times Series Methods
28. Time series methods assume that demand patterns in the past are a good predictor of
demand in the future.
Ans: True
Difficulty: Moderate
Learning Objective: LO 3
29. The moving average method is used for creating forecasts when there is no variation
in demand.
Ans: False
Difficulty: Moderate
Learning Objective: LO 3
30. Because of ease of use and simplicity, exponential smoothing is preferred over
smoothing average.
Ans: False
Difficulty: Moderate
Learning Objective: LO 3
31. The average, absolute difference between the forecast and demand is a popular
measure of forecast error.
Ans: True
Difficulty: Moderate
Learning Objective: LO 4
32. The larger the mean absolute deviation (MAD) the more accurate the forecast.
Ans: False
Difficulty: Moderate
Learning Objective: LO 4
33. Forecast bias is measured by the per-period average of the sum of forecast errors.
Ans: True
Difficulty: Moderate
Learning Objective: LO 4
34. Because of the development of advanced forecasting models managers no longer
track forecast error.
Ans: False
Difficulty: Moderate
Learning Objective: LO 4
35. Regression is used for forecasting when there is a relationship between the dependent
variable, demand, and one or more independent (explanatory) variables.
Ans: True
Difficulty: Moderate
Learning Objective: LO 6
36. Correlation in linear regression is a measure of the strength of the relationship
between the dependent variable, demand, and an independent (explanatory) variable.
Ans: True
Difficulty: Moderate
Learning Objective: LO 6
37. A linear regression model that relates demand to time is known as a linear trend line.
Ans: True
Difficulty: Moderate
Learning Objective: LO 6
38. Linear regression relates two variables using a linear model.
Ans: True
Difficulty: Moderate
Learning Objective: LO 6
39. A correlation coefficient is a measure of the strength of the linear relationship
between an independent and a dependent variable.
Ans: True
Difficulty: Moderate
Learning Objective: LO 6
40. Multiple regression analysis can be used to relate demand to two or more dependent
variables.
Ans: False
Difficulty: Moderate
Learning Objective: LO 6
Multiple Choice
41. Forecast methods based on judgment, opinion, past experiences, or best guesses
are known as ___________ methods.
a. quantitative
b. qualitative
c. time series
d. regression
Difficulty: Easy
Learning Objective: LO 1
42. A forecast
a. predicts what will occur in the future.
b. results from an uncertain process.
c. support strategic planning.
d. All of these answer choices are correct.
Difficulty: Easy
Learning Objective: LO 1
43. Forecasts of product demand determine how much
a. inventory is needed.
b. product to make.
c. material to purchase from suppliers.
d. All of these answer choices are correct.
Difficulty: Easy
Learning Objective: LO 1
44. The ______________ effect is caused in part by distortion in product demand
information caused by inaccurate forecasts.
a. bullwhip
b. regression
c. error
d. None of these answer choices is correct.
Difficulty: Easy
Learning Objective: LO 1
45. Continuous replenishment relies heavily on ____________term forecast.
a. short-
b. medium-
c. long-
d. All of these answer choices are correct.
Difficulty: Easy
Learning Objective: LO 1
46. In ___________________ replenishment, the supplier and customer care
continuously update data.
a. demand
b. ongoing
c. continuous
d. forecasted
Difficulty: Easy
Learning Objective: LO 1
47. ________________ demand is a key to providing good-quality service.
a. Predicted
b. Forecasted
c. Anticipated
d. Unknown
Difficulty: Easy
Learning Objective: LO 1
48. A long-range forecast would normally not be used to
a. design the supply chain.
b. implement strategic programs.
c. determine production schedules.
d. plan new products for changing markets.
Difficulty: Moderate
Learning Objective: LO 2
49. A forecast where the current period’s demand is used as the next period’s forecast is
known as a
a. moving average forecast.
b. naïve forecast.
c. weighted moving average forecast.
d. Delphi method.
Difficulty: Moderate
Learning Objective: LO 2
50. Which of the following is not a type of predictable demand behavior?
a. trend
b. random variation
c. cycle
d. seasonal pattern
Difficulty: Moderate
Learning Objective: LO 2
51. A ___________ is an up-and-down movement in demand that repeats itself over a
period of more than a year.
a. trend
b. seasonal pattern
c. random variation
d. cycle
Difficulty: Easy
Learning Objective: LO 2
52. Selecting the type of forecasting method to use depends on
a. the time frame of the forecast.
b. the behavior of demand and demand patterns.
c. the causes of demand behavior.
d. All of these answer choices are correct.
Difficulty: Easy
Learning Objective: LO 2
53. A qualitative procedure used to develop a consensus forecast is known as
a. exponential smoothing.
b. regression methods.
c. the Delphi technique.
d. naïve forecasting.
Difficulty: Moderate
Learning Objective: LO 2
54. The sum of the weights in a weighted moving average forecast must
a. equal the number of periods being averaged.
b. equal 1.00.
c. be less than 1.00.
d. be greater than 1.00.
Difficulty: Easy
Learning Objective: LO 3
55. An exponential smoothing forecasting technique requires all of the following
except
a. the forecast for the current period.
b. the actual demand for the current period.
c. a smoothing constant.
d. large amounts of historical demand data.
Difficulty: Moderate
Learning Objective: LO 3
56. Given the demand and forecast values below, the naïve forecast for September is
Period Demand Forecast
April 100 97
May 105 103
June 97 98
July 102 105
August 99 102
September
a. 100.6.
b. 99.0.
c. 102.0.
d. cannot be determined.
Difficulty: Moderate
Solution: Naïve forecast=99.0
Learning Objective: LO 3
57. The smoothing constant, α, in the exponential smoothing forecast
a. must always be a value greater than 1.0.
b. must always be a value less than 0.10.
c. must be a value between 0.0 and 1.0.
d. should be equal to the time frame for the forecast.
Difficulty: Moderate
Learning Objective: LO 3
58. The closer the smoothing constant, α, is to 1.0 the
a. greater the reaction to the most recent demand.
b. greater the dampening, or smoothing, effect.
c. more accurate the forecast.
d. less accurate the forecast.
Difficulty: Moderate
Learning Objective: LO 3
59. The exponential smoothing model produces a naïve forecast when the smoothing
constant, α, is equal to
a. 0.00.
b. 1.00.
c. 0.50.
d. 2.00
Difficulty: Moderate
Learning Objective: LO 3
60. The _______ method uses demand in the first period to forecast demand in the next
period.
a) naïve
b) moving average
c) exponential smoothing
d) linear trend
Difficulty: Moderate
Learning Objective: LO 3
61. The _________________ forecast method consists of an exponential smoothing
forecast with a trend adjustment factor added to it.
a) exponentially smoothed
b) adjusted exponentially smoothing
c) time series
d) moving average
Difficulty: Moderate
Learning Objective: LO 3
62. Given the following demand data for the past five months, the three-period moving
average forecast for June is
Period Demand
January 120
February 90
March 100
April 75
May 110
a. 103.33.
b. 99.00.
c. 95.00.
d. 92.50
Difficulty: Moderate
Solution: Moving Average, MA3=95.00
Learning Objective: LO 3
63. Given the following demand data for the past five months, the four-period moving
average forecast for June is
Period Demand
January 120
February 90
March 100
April 75
May 110
a. 96.25.
b. 99.00.
c. 110.00.
d. 93.75.
Difficulty: Moderate
Solution: Moving Average, MA4=93.75
Learning Objective: LO 3
64. A company wants to produce a weighted moving average forecast for April with the
weights 0.40, 0.35, and 0.25 assigned to March, February, and January respectively.
If the company had demands of 5,000 in January, 4,750 in February, and 5,200 in
March, then April’s forecast is
a. 4983.33.
b. 4992.50.
c. 4962.50.
d. 5000.00.
Difficulty: Moderate
Solution: Weighted Moving Average, WMA3=4992.50
Learning Objective: LO 3
65. The weighted moving average forecast for the fifth period, with weights of 0.15
for period 1, 0.20 for period 2, 0.25 for period 3, and 0.40 for period 4, using the
demand data shown below is
Period Demand
1 3,500
2 3,800
3 3,500
4 4,000
a. 3,760.
b. 3,700.
c. 3,650.
d. 3,325.
Difficulty: Moderate
Solution: Weighted Moving Average, WMA3=3,760
Learning Objective: LO 3
66. For the demand values and the January forecast shown in the table below the
exponential smoothing forecast for March using α = 0.30 is
Period Demand Forecast
January 500 480
February 476
March 503
April
a. 489.
b. 486.
c. 483.
d. 480.
Difficulty: Hard
Solution: Exponential Smoothing, F3=489
Learning Objective: LO 3
67. For the demand values and the January forecast shown in the table below the
exponential smoothing forecast for March using α = 0.40 is
Period Demand Forecast
January 1,250 1,200
February 1,225
March
a. 1,200.
b. 1,220.
c. 1,222.
d. 1,225.
Difficulty: Hard
Solution: Exponential Smoothing, F3=1,222.
Learning Objective: LO 3
68. If the forecast for July was 3,300 and the actual demand for July was 3,250, then
the exponential smoothing forecast for August using α = 0.20 is
a. 3,300.
b. 3,290.
c. 3,275.
d. 3,250.
Difficulty: Moderate
Solution: Exponential Smoothing, F2=3,290.
Learning Objective: LO 3
69. Given the demand and forecast values shown in the following table,
Period Demand Forecast
June 495 484
July 515 506
August 519 528
September 496 506
October 557 550
calculate the three-period moving average forecast for November.
a. 516.
b. 528.
c. 524.
d. 515.
Difficulty: Moderate
Solution: Moving Average, MA3=524
Learning Objective: LO 3
70. Given the demand and forecast values shown in the table below,
Period Demand Forecast
June 495 484
July 515 506
August 519 528
September 496 506
October 557 550
the exponential smoothing forecast for November using α = 0.35 is
a. 552.45.
b. 553.50.
c. 554.55.
d. 557.50.
Difficulty: Moderate
Solution: Exponential Smoothing, F6=552.45
Learning Objective: LO 3
71. Given the demand and forecast values shown in the table below,
Period Demand Forecast
June 495 484
July 515 506
August 519 528
September 496 506
October 557 550
the forecast error for September is
a. 10.00.
b. -10.00.
c. 1.00.
d. 39.00.
Difficulty: Moderate
Solution: E=-10.00
Learning Objective: LO 4
72. If forecast errors are normally distributed then
a. 1 MAD = 1σ
b. 1 MAD ≈ 0.8 σ
c. 0.8 MAD ≈ 1σ
d. 1 MAD ≈ 1.96 σ
Difficulty: Moderate
Learning Objective: LO 4
73. A forecasting model has produced the following forecasts,
Period Demand Forecast Error
January 120 110
February 110 115
March 115 120
April 125 115
May 130 125
the forecast error for February is
a. 10.
b. -10.
c. -15.
d. -5
Difficulty: Moderate
Solution: E=-5
Learning Objective: LO 4
74. Given the demand and forecast values shown in the table below,
Period Demand Forecast
June 495 484
July 515 506
August 519 528
September 496 506
October 557 550
the mean absolute deviation (MAD) through the end of October is
a. 9.20
b. -9.20
c. 1.00
d. 7.00
Difficulty: Hard
Solution: MAD=9.20
Learning Objective: LO 4
75. The per-period average of cumulative error is called
a. cumulative forecast variation.
b. absolute error.
c. average error.
d. noise.
Difficulty: Easy
Learning Objective: LO 4
76. A forecasting model has produced the following forecasts:
Period Demand Forecast Error
January 120 110
February 110 115
March 115 120
April 125 115
May 130 125
The mean absolute deviation (MAD) for the end of May is
a. 7.0.
b. 7.5.
c. 10.0.
d. 3.0
Difficulty: Hard
Solution: MAD=7.0
Learning Objective: LO 4
77. A forecasting model has produced the following forecasts:
Period Demand Forecast Error
January 120 110
February 110 115
March 115 120
April 125 115
May 130 125
The mean absolute percent deviation (MAPD) for the end of May is
a. 0.0250.
b. 0.0583.
c. 0.5830.
d. 0.6670.
Difficulty: Hard
Solution: MAPD=0.0583
Learning Objective: LO 4
78. A forecasting model has produced the following forecasts:
Period Demand Forecast Error
January 120 110
February 110 115
March 115 120
April 125 115
May 130 125
At the end of May the average error would be
a. 7.
b. 5.
c. 3.
d. 1.
Difficulty: Moderate
Solution: Ebar=3
Learning Objective: LO 4
79. A forecasting model has produced the following forecasts:
Period Demand Forecast Error
January 120 110
February 110 115
March 115 120
April 125 115
May 130 125
At the end of May the tracking signal would be
a. 0.000.
b. 0.667.
c. 1.333.
d. 2.143.
Difficulty: Hard
Solution: Tracking Signal=2.143
Learning Objective: LO 4
80. The mean absolute percentage deviation (MAPD) measures the absolute error as a
percentage of
a. all errors.
b. per-period demand.
c. total demand.
d. the average error.
Difficulty: Moderate
Learning Objective: LO 4
81. A large positive cumulative error indicates that the forecast is probably
a. higher than the actual demand.
b. lower than the actual demand.
c. unbiased.
d. biased.
Difficulty: Moderate
Learning Objective: LO 4
82. Which of the following statements concerning average error is true?
a. A positive value indicates high bias, and a negative value indicates low bias.
b. A positive value indicates zero bias, and a negative value indicates low bias.
c. A negative value indicates zero bias, and a negative value indicates high bias.
d. A positive value indicates low bias, and a negative value indicates high bias.
Difficulty: Moderate
Learning Objective: LO 4
83. Which of the following is a reason why a forecast can go out of control?
a. a change in trend
b. an irregular variation such as unseasonable weather
c. a promotional campaign
d. All of these answer choices are correct.
Difficulty: Moderate
Learning Objective: LO 4
84. Which of the following can be used to monitor a forecast to see if it is biased high
or low?
a. a tracking signal
b. the mean absolute deviation (MAD)
c. the mean absolute percentage deviation (MAPD)
d. a linear trend line model
Difficulty: Moderate
Learning Objective: LO 4
85. A tracking signal is computed by
a. multiplying the cumulative error by MAD.
b. multiplying the absolute error by MAD.
c. dividing MAD by the cumulative absolute error.
d. dividing the cumulative error by MAD.
Difficulty: Moderate
Learning Objective: LO 4
86. Regression forecasting methods relate _________to other factors that cause demand
behavior.
a) supply
b) demand
c) time
d) money
Difficulty: Moderate
Learning Objective: LO 6
87. Correlation is a measure of the strength of the
a. nonlinear relationship between two dependent variables.
b. nonlinear relationship between a dependent and independent variable.
c. linear relationship between two dependent variables.
d. linear relationship between a dependent and independent variable.
Difficulty: Moderate
Learning Objective: LO 6
88. A mathematical technique for forecasting that relates the dependent variable to an
independent variable is
a. correlation analysis.
b. exponential smoothing.
c. linear regression.
d. weighted moving average.
Difficulty: Easy
Learning Objective: LO 6
(Use the following information for the next five problems.)
The owner of Koffi, the sole coffee house located in a resort area, wants to develop a
forecast based on the relationship between tourism and coffee drinks sold. He has
generated the following data over the past 12 months:
Tourists Coffee Drinks
Month (thousands) (per day)
January 22 132
February 25 175
March 34 210
April 30 150
May 15 60
June 10 50
July 8 45
August 6 40
September 10 35
October 15 75
November 18 110
December 20 140
The data from using Data Analysis on Excel is as follows:
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.954141355
R Square 0.910385725
Adjusted R Square 0.901424297
Standard Error 18.57063782
Observations 12
ANOVA
df SS MS F
Regression 1 35034.98077 35034.98 101.5894
Residual 10 3448.685892 344.8686
Total 11 38483.66667
Coefficients Standard Error t Stat P-value
Intercept -11.5743276 12.46354188 -0.92865 0.37494
Tourists (thousands) 6.389163997 0.633898777 10.07915 1.48E-06
89. What is the approximate intercept, a?
a. -11.6
b. 11.6
c. 6.4
d. -6.4
Difficulty: Moderate
Learning Objective: LO 6
90. What is the approximate slope, b?
a. -11.6
b. 11.6
c. 6.4
d. -6.4
Difficulty: Moderate
Learning Objective: LO 6
91. What is the forecasted number of coffee drinks sold if the number of tourists is 25
(thousand)?
a. 128
b. 138
c. 148
d. 158
Difficulty: Moderate
Learning Objective: LO 6
92. What is the correlation?
a. 0.95
b. 0.91
c. 0.90
d. 19
Difficulty: Moderate
Learning Objective: LO 6
93. What is the coefficient of determination?
a. 0.95
b. 0.91
c. 0.90
d. 18
Difficulty: Moderate
Learning Objective: LO 6
94. Data mining uses and analyzes data that is stored in
a. databases.
b. data warehouses.
c. data marts.
d. All of these answer choices are correct.
Difficulty: Moderate
Learning Objective: LO 6
95. _______________ can be subdivided into ________________ that store subsets of
data.
a. Databases, data warehouses
b. Data warehouses, data marts
c. Databases, data marts
d. Data warehouses, databases
Difficulty: Moderate
Learning Objective: LO 6
96. Association rule learning is a data mining technique that
a. discovers trends, predicts future events, and assesses possible course of action.
b. searches for relationships between variables.
c. identifies groups of data that fall naturally together.
d. None of these answer choices is correct.
Difficulty: Moderate
Learning Objective: LO 6
97. Cluster analysis is a tool that
a. discovers trends, predicts future events, and assesses possible course of action.
b. searches for exact relationships between variables.
c. identifies groups of data that fall naturally together.
d. None of these answer choices is correct.
Difficulty: Moderate
Learning Objective: LO 6
98. Data mining
a. discovers trends, predicts future events, and assesses possible course of action.
b. searches for exact relationships between variables.
c. identifies groups of data that fall naturally together.
d. None of these answer choices is correct.
Difficulty: Moderate
Learning Objective: LO 6
Short Answer
99. Discuss the importance of accurate forecasts in supply chain management.
Ans: A company’s supply chain encompasses all of the facilities, functions, and activities
involved in producing a product or service from suppliers to customers. Supply chain
functions such as purchasing, inventory, production, scheduling, and transportation are all
affected in the short run by product demand and in the long run by new products and
processes. Forecasts of product demand determine how much inventory is needed, how
much product to make, and how much material to purchase from suppliers to meet
forecasted customer needs. This in turn determines the kind of transportation that will be
needed and where plants, warehouses and distribution centers will be located. Without
accurate forecasts, large costs of costly inventory must be kept at each stage of the supply
chain to compensate for the uncertainties of customer demand. If there are insufficient
inventories, customer service suffers because of late deliveries and stock outs. This can
be quite damaging to customer service and the need to have on-time delivery to compete
in today’s competitive environment.
Difficulty: Hard
Learning Objective: LO 1
100. Compare and contrast short-mid-range forecasts and long-range forecasts.
Ans: Forecasts are either short-to mid-range, or long-range. Short-range (to mid-range)
forecasts are typically for daily, weekly, or monthly sales demand for up to
approximately two years into the future, depending on the company and the type of
industry. They are primarily used to determine production and delivery schedules and to
establish inventory levels. A long-range forecast is usually for a period longer than two
years into the future. A long-range forecast is normally used for strategic planning—to
establish long-term goals, plan new products for changing markets, enter new markets,
develop new facilities, develop technology, design the supply chain and implement
strategic programs, such as TQM. It is important to remember that these classifications
are generalizations and the line between short-and long-range forecasts is not always so
distinct.
Difficulty: Moderate
Learning Objective: LO 2
101. Explain the difference between qualitative and quantitative forecasting methods.
Ans: Qualitative forecasting methods are based on judgment, opinion, past experience, or
best guesses. They are most useful for predicting the future when there is no past
historical data to model or when the need exists to predict for the very long-term.
Quantitative forecasting methods are based on mathematical formulas. Examples of
quantitative forecasting methods are time series and regression analysis.
Difficulty: Moderate
Learning Objective: LO 2
102. Explain how and why time series and regression forecasting methods differ.
Ans: Time series methods are statistical techniques that use historical data to predict
future demand. Time series techniques assume that what has occurred in the past will
continue to occur in the future. Regression (or causal) forecasting methods attempt to
develop a mathematical relationship (in the form of a regression model) between demand
and factors that cause demand to behave the way it does.
Difficulty: Moderate
Learning Objective: LO 3
