Introduction to Operations & Supply Chain Management, 5e (Bozarth)
Chapter 9 Forecasting
9.1 Forecast Types
1) A video game publishing company needs to predict the total sales in the European market for the next
year. This is an example of a(n):
A) firm-level demand forecast.
B) overall market demand forecast.
C) supply forecast.
D) price forecast.
9.2 Laws of Forecasting
1) Fed up with her working conditions at the call center, Lisa decides to invest in a state–of-the-art sewing
machine and produce limited quantities of her own clothing designs. After a few months of operation,
she decides to apply some of the forecasting techniques she mastered in school. Which of these
statements about her forecasts is correct?
A) Her forecasts will probably be 100% accurate.
B) Her demand forecasts for a year from now will probably be more accurate than her demand forecasts
for three months from now.
C) Her demand forecasts for each style of skirt will be more accurate than her demand forecasts for all
skirts.
D) The best way for her to determine the amount of fabric she needs is to forecast it based on her
customer orders for each type of skirt.
2) Forecasts are almost always wrong.
3) What are the laws of forecasting and what are their implications for operations and supply chain
managers?
9.3 Selecting a Forecasting Method
1) Qualitative forecasts are used when there is plenty of relevant data.
9.4 Qualitative Forecasting Methods
1) A qualitative forecasting technique well-suited for demand forecasts of a new product or service is the:
A) Delphi method.
B) build-up forecast.
C) life cycle analogy method.
D) patronage survey.
2) A qualitative forecasting technique in which individuals familiar with specific market segments
estimate the demand within these sectors that are then summed to get an overall forecast is called a:
A) market survey.
B) life cycle analogy.
C) panel consensus forecasting.
D) build-up forecast.
3) The Delphi method, panel consensus forecasting, and market surveys are all qualitative forecasting
methods, but only market surveys do NOT use experts.
4) The panel consensus forecasting approach requires that the forecasting team discuss their forecast as a
team but the ________ requires that each member of the team develop a separate forecast initially.
5) Describe three qualitative forecasting techniques and compare their strengths and weaknesses.
9.5 Time Series Forecasting Models
1) A long-term movement up or down in a time series is called:
A) seasonality.
B) trend.
C) randomness.
D) cycle.
2) A firm’s demand data from the last two quarters is displayed in the table. Use a three period moving
average to forecast demand for July.
Month
Demand
January
154
February
148
March
214
April
180
May
225
June
246
A) 206
B) 217
C) 223
D) 226
3) A firm’s demand data from the last two quarters is displayed in the table. Use a three period weighted
moving average with Wt = 0.7, Wt-1 = 0.2, and Wt-2 = 0.1 to forecast demand for July.
Month
Demand
January
154
February
148
March
214
April
180
May
225
June
246
A) 235.2
B) 195.6
C) 158.8
D) 180.4
4) Heidi favors using a two period moving average but Tim is “an exponential-smoothing man.” Tim’s
demand forecast for May was identical to Heidi’s. What value of alpha would Tim need to use in order
for his June forecast to be identical to Heidi’s if each sticks with their preferred technique? Note that Tim’s
forecast for May was identical to Heidi’s two-period moving average for May.
Month
Demand
January
154
February
148
March
214
April
180
May
225
June
246
A) .085
B) .196
C) .237
D) .348
5) A drive-in restaurant has experienced the following customer loads on the past 8 Friday nights. If their
forecast for period 7 was 59 customers, then what is their forecast for period number 8 using a smoothing
constant of 0.7?
Friday
# Customers
1
49
2
55
3
57
4
59
5
56
6
61
7
62
8
63
A) 61.10
B) 62.43
C) 59.90
D) 60.83
6) A counseling service records the number of calls to their hotline for the last year. What is the forecast
for July if the service uses a simple moving average of three periods?
Month
Demand
January
111
February
127
March
146
April
159
May
165
June
165
July
178
August
182
September
191
October
208
November
223
December
228
A) 169
B) 163
C) 157
D) 178
7) A counseling service records the number of calls to their hotline for the last year. What is the forecast
for August if the forecast for June was 164 and the service uses exponential smoothing with an alpha of
0.8?
Month
Demand
January
111
February
127
March
146
April
159
May
165
June
165
July
178
August
182
September
191
October
208
November
223
December
228
A) 164.80
B) 188.93
C) 180.67
D) 175.36
8) A counseling service records the number of calls to their hotline for the last year. What is the forecast
for August if a regression equation is used to model this data?
Month
Demand
January
111
February
127
March
146
April
159
May
165
June
165
July
178
August
182
September
191
October
208
November
223
December
228
A) 188.3
B) 179.9
C) 180.6
D) 175.7
9) A counseling service records the number of calls to their hotline for the last year. What is the forecast
for October if a weighted moving average with weights of 0.5, 0.3, and 0.2 is used to model this data?
Month
Demand
January
111
February
127
March
146
April
159
May
165
June
165
July
178
August
182
September
191
October
208
November
223
December
228
A) 177.4
B) 185.7
C) 197.7
D) 190.3
10) A counseling service records the number of calls to their hotline for the last year. What is the forecast
for March if an adjusted exponential smoothing model is used with α=0.8 and β=0.7? The unadjusted
forecast for January is 123.5.
Month
Demand
January
111
February
127
March
146
April
159
May
165
June
165
July
178
August
182
September
191
October
208
November
223
December
228
A) 135.44
B) 138.53
C) 129.76
D) 144.34
11) A company keeps track of unit sales and notes a strong trend during the past eight periods. They use
an adjusted exponential smoothing model with an alpha equal to 0.7 and a beta equal to 0.6. Using the
demand data and previous forecasts shown in the table, develop a forecast for period 4.
Week
# Units
Unadjusted
Forecast
Trend
Adjusted
Forecast
1
68
65.0
0.0
65.0
2
72
67.1
1.3
68.4
3
80
4
93
5
108
6
121
7
125
8
139
A) 82.2
B) 84.9
C) 87.5
D) 91.6
12) Consider the data that generate this plot covering time for months numbered 1 to 50. What
characteristic is most prominent in the pattern?
A) trend
B) seasonality
C) randomness
D) none
13) Fluctuations in demand due to seasonality are greater than those due to randomness.
14) A seasonal pattern in time series data is evident when the level of the variable of interest moves
erratically up or down from one period to the next.
15) The greater the randomness in the model, the greater the number of periods that should be used in a
moving average forecast.
16) Exponential smoothing with an alpha of one will yield identical results to a last period forecast.
17) When there is a significant upward or downward trend in the data, the two of the best forecasting
models to use are adjusted exponential smoothing and linear regression.
18) The independent variable is the quantity the forecaster is interested in estimating with a linear
regression model.
19) A seasonal index less than 1.0 means that the model is overforecasted.
20) The slope of the regression equation is positive if the r-squared value is greater than 0.0.
21) ________ is unpredictable movement from one time period to the next.
22) Two smoothing models that yield identical forecasts are exponential smoothing with an alpha equal
to ________ and a moving average with n equal to ________.
23) The greater the randomness in the data, the ________ the value of the alpha should be in an
exponential smoothing forecast.
24) Two time series techniques that are appropriate when the data display a strong upward or downward
trend are ________ and ________.
25) Dividing actual demand by the model’s forecast yields an index that can be used to adjust for
________ in the data.
26) A(n) ________ forecasting model bases all forecasts on past actual values all the way back to the first
period.
27) “A simple moving average was good enough for my dad, and it’s good enough for me,” Ethan
declared as he prepared his forecast. The assistant dean knew that the size of incoming MBA classes had
been increasing dramatically over the previous few semesters and that the forecasts Ethan would prepare
using his father’s method would ________ the actual size of the incoming class.
28) What patterns are common in time series data? Describe each one and draw one plot that displays all
of these patterns. Label the patterns on your plot.
19
29) Examine these two graphs and based on the demand pattern and axis scaling, recommend a
forecasting technique (and the required parameters) that would work best for each one. Justify your
recommendations.
Answer: Demand Pattern A shows a strong linear trend from 103 up to 207 but Demand Pattern B
fluctuates around 169 and (with a standard deviation of 4) is purely random. Adjusted exponential
smoothing or linear regression are the only suitable methods for Demand Pattern A. A linear equation of
the form Y = 95.98 + 5.82X yields an R squared of 0.99. The randomness in Demand Pattern B is pure noise
and therefore not predictable. Moving averages with a large n or exponential smoothing with a low alpha
will smooth this randomness but don’t truly add much in the way of predictive ability.
Diff: 3
Reference: 9.5 Time Series Forecasting Models
Keywords: time series, trend, adjusted exponential smoothing, regression, moving average, randomness, exponential
smoothing
AACSB: Application of Knowledge
LO: 9.2: Apply a variety of time series forecasting models, including moving average, exponential smoothing, and
linear regression models.
30) Develop forecasts for periods 7 through 10 for the demand data in the table using a three period
moving average, a weighted moving average using 0.6, 0.3, and 0.1, and exponential smoothing with
alpha = 0.7. Use a 6th period forecast of 135 as the starting point for the exponential smoothing technique.
Period
Actual
MA n = 3
WMA
Exp. Smoothing
1
64
2
84
3
91
4
97
5
115
6
135
7
137
8
144
9
153
10
171
1
2
3
4
5
115
6
135
7
137
8
144
9
153