MG 403 Operations Management — Exam 2
with correct answers and rationale
Section 1: Forecasting (Questions 1–50)
1. Which of the following is not a type of qualitative forecasting?
A. Jury of executive opinion
B. Delphi method
C. Sales force composite
D. Average moving
Rationale: Average moving (or moving average) is a quantitative time-series forecasting method that
uses historical numerical data. Qualitative methods rely on subjective judgment, experience, and
opinion rather than numerical data -1-6.
2. A naive forecast for September sales of a product would be equal to the sales in August.
A. True
B. False
Rationale: A naive forecast simply uses the most recent period’s actual value as the forecast for the
next period. Therefore, the naive forecast for September equals August sales -1.
3. Given that the current period's actual and forecast sales figures are $103M and $99M
respectively, if we used an alpha of 0.4, what is the approximate simple exponential smoothing
forecast for the next period?
A. $99.4M
B. **$100.6M**
C. $101.4M
D. $102.4M
Rationale: Simple exponential smoothing formula: F(t+1) = F(t) + α[A(t) − F(t)]. Calculation: 99 +
0.4(103 − 99) = 99 + 1.6 = $100.6M -1-6.
4. Which of the following values of alpha would cause exponential smoothing to respond the most
slowly to forecast errors?
A. 0.90
B. 0.50
C. 0.30
D. 0.10
Rationale: A smaller alpha places more weight on the previous forecast and less on the most recent
actual value, making the forecast respond more slowly to changes. Alpha = 0.10 produces the
slowest response among the choices -1.
5. Changing α in simple exponential smoothing from 0.2 to 0.9 would make:
A. Forecasts more dependent on older actual values
B. Forecasts more dependent on most recent actual values
C. Forecasts less responsive to demand changes
D. Forecasts equivalent to a moving average
Rationale: A higher alpha gives greater weight to the most recent actual demand, making the
forecast more responsive to recent changes -1.
,6. With regards to the four components of time-series discussed in class, the fundamental
difference between "cyclical" and "seasonal" is the:
A. Magnitude of the repeating patterns
B. Direction of the repeating patterns
C. Frequency of the repeating patterns
D. Predictability of the repeating patterns
Rationale: Seasonal patterns repeat within one year (e.g., quarterly, monthly), while cyclical
patterns repeat over longer, often multi-year periods. The key distinction is the frequency (or
duration) of the repeating patterns -1.
7. __________ expresses the error as a percent of the actual values, undistorted by a single large
value.
A. MAD
B. MSE
C. MAPE
D. Tracking signal
Rationale: Mean Absolute Percent Error (MAPE) expresses forecast error as a percentage of actual
values. Because it is a percentage, it is not disproportionately affected by a single large actual value,
unlike MAD or MSE -1.
8. Which of the following uses three types of participants: decision makers, staff personnel, and
respondents?
A. Sales force composite
B. Jury of executive opinion
C. The Delphi method
D. Consumer market survey
Rationale: The Delphi method is a structured qualitative forecasting technique that involves decision
makers, staff personnel, and respondents (experts) who provide anonymous input through iterative
rounds -1.
9. In a normal distribution, what percentage of values falls between μ − 2σ and μ + 2σ?
A. 68%
B. 95%
C. 99.7%
D. 50%
Rationale: In a normal distribution: approximately 68% of values fall within ±1σ, 95% within ±2σ, and
99.7% within ±3σ -2.
10. A forecasting method that uses a weighted average of past values with weights that decrease
exponentially is called:
A. Moving average
B. Exponential smoothing
C. Linear regression
D. Naive forecasting
Rationale: Exponential smoothing assigns exponentially decreasing weights to past observations,
with the most recent values receiving the highest weights.
11. Which forecasting method is most appropriate when little historical data is available?
A. Time-series analysis
B. Qualitative forecasting
,C. Causal modeling
D. Exponential smoothing
Rationale: Qualitative forecasting methods rely on expert judgment and opinion rather than
historical data, making them suitable when data is scarce (e.g., new product forecasting).
12. The tracking signal is used to:
A. Measure the average forecast error
B. Detect when a forecast is consistently biased
C. Calculate the seasonal index
D. Determine the optimal alpha value
Rationale: The tracking signal is the ratio of cumulative forecast error (RSFE) to MAD. It monitors
whether forecasts are consistently over- or under-predicting, indicating bias.
13. A forecast that consistently overestimates demand has a:
A. Positive tracking signal
B. Negative tracking signal
C. Zero tracking signal
D. Random tracking signal
Rationale: If actual demand is consistently below forecast, the cumulative error is negative, resulting
in a negative tracking signal.
14. Which of the following is a causal forecasting method?
A. Exponential smoothing
B. Moving average
C. Linear regression
D. Delphi method
Rationale: Linear regression is a causal method because it assumes the forecast variable is related to
other independent variables (e.g., price, advertising).
15. The primary disadvantage of the moving average method is that:
A. It is too complex to compute
B. It requires storing all historical data
C. It cannot be used for short-term forecasting
D. It always lags behind trends
Rationale: A moving average requires retaining n periods of historical data for each forecast. It also
lags trends, but the storage requirement is a key operational disadvantage.
16. If a time series has a consistent upward trend, which forecasting method would be most
appropriate?
A. Simple moving average
B. Linear trend projection
C. Naive forecast
D. Simple exponential smoothing
Rationale: Linear trend projection explicitly models the trend component, while simple exponential
smoothing and moving averages lag behind trends.
17. The seasonal index for a quarter with sales 20% above the annual average is:
A. 0.80
B. 1.20
C. 20.0
, D. 0.20
Rationale: Seasonal index = (Quarterly average / Overall average) × 100 or expressed as a decimal. If
a quarter is 20% above average, the index is 1.20.
18. Which of the following is true about forecast accuracy?
A. Forecasts are always exact
B. Forecast accuracy decreases as the forecast horizon increases
C. Qualitative forecasts are always more accurate than quantitative forecasts
D. Forecasts for individual products are more accurate than aggregate forecasts
Rationale: Uncertainty increases with the time horizon, so long-range forecasts are less accurate
than short-range forecasts. Aggregate forecasts (e.g., product families) are generally more accurate
than individual item forecasts.
19. A company uses a 3-period moving average. If sales for the last three periods were 100, 110,
and 120, the forecast for the next period is:
A. 100
B. 105
C. 110
D. 120
Rationale: (100 + 110 + 120) / 3 = = 110.
20. The value of alpha that gives equal weight to all past observations is:
A. 0.0
B. Not possible with exponential smoothing
C. 1.0
D. 0.5
Rationale: Exponential smoothing always gives more weight to recent observations. There is no
alpha value that gives equal weight to all past observations (a moving average with n periods does
this).
21. Which forecasting approach is most suitable for predicting demand for a brand-new product
with no historical data?
A. Exponential smoothing
B. Market survey
C. Moving average
D. Trend projection
Rationale: Market surveys gather consumer intentions and opinions, which is useful when no
historical data exists.
22. The RSFE (Running Sum of Forecast Errors) is used to calculate the:
A. MAD
B. MSE
C. Tracking signal
D. MAPE
Rationale: Tracking signal = RSFE / MAD. RSFE alone indicates bias; dividing by MAD normalizes it.
23. A forecast error of +5 means:
A. Actual demand was 5 units higher than forecast
B. Forecast was 5 units higher than actual demand
C. The forecast was perfect
with correct answers and rationale
Section 1: Forecasting (Questions 1–50)
1. Which of the following is not a type of qualitative forecasting?
A. Jury of executive opinion
B. Delphi method
C. Sales force composite
D. Average moving
Rationale: Average moving (or moving average) is a quantitative time-series forecasting method that
uses historical numerical data. Qualitative methods rely on subjective judgment, experience, and
opinion rather than numerical data -1-6.
2. A naive forecast for September sales of a product would be equal to the sales in August.
A. True
B. False
Rationale: A naive forecast simply uses the most recent period’s actual value as the forecast for the
next period. Therefore, the naive forecast for September equals August sales -1.
3. Given that the current period's actual and forecast sales figures are $103M and $99M
respectively, if we used an alpha of 0.4, what is the approximate simple exponential smoothing
forecast for the next period?
A. $99.4M
B. **$100.6M**
C. $101.4M
D. $102.4M
Rationale: Simple exponential smoothing formula: F(t+1) = F(t) + α[A(t) − F(t)]. Calculation: 99 +
0.4(103 − 99) = 99 + 1.6 = $100.6M -1-6.
4. Which of the following values of alpha would cause exponential smoothing to respond the most
slowly to forecast errors?
A. 0.90
B. 0.50
C. 0.30
D. 0.10
Rationale: A smaller alpha places more weight on the previous forecast and less on the most recent
actual value, making the forecast respond more slowly to changes. Alpha = 0.10 produces the
slowest response among the choices -1.
5. Changing α in simple exponential smoothing from 0.2 to 0.9 would make:
A. Forecasts more dependent on older actual values
B. Forecasts more dependent on most recent actual values
C. Forecasts less responsive to demand changes
D. Forecasts equivalent to a moving average
Rationale: A higher alpha gives greater weight to the most recent actual demand, making the
forecast more responsive to recent changes -1.
,6. With regards to the four components of time-series discussed in class, the fundamental
difference between "cyclical" and "seasonal" is the:
A. Magnitude of the repeating patterns
B. Direction of the repeating patterns
C. Frequency of the repeating patterns
D. Predictability of the repeating patterns
Rationale: Seasonal patterns repeat within one year (e.g., quarterly, monthly), while cyclical
patterns repeat over longer, often multi-year periods. The key distinction is the frequency (or
duration) of the repeating patterns -1.
7. __________ expresses the error as a percent of the actual values, undistorted by a single large
value.
A. MAD
B. MSE
C. MAPE
D. Tracking signal
Rationale: Mean Absolute Percent Error (MAPE) expresses forecast error as a percentage of actual
values. Because it is a percentage, it is not disproportionately affected by a single large actual value,
unlike MAD or MSE -1.
8. Which of the following uses three types of participants: decision makers, staff personnel, and
respondents?
A. Sales force composite
B. Jury of executive opinion
C. The Delphi method
D. Consumer market survey
Rationale: The Delphi method is a structured qualitative forecasting technique that involves decision
makers, staff personnel, and respondents (experts) who provide anonymous input through iterative
rounds -1.
9. In a normal distribution, what percentage of values falls between μ − 2σ and μ + 2σ?
A. 68%
B. 95%
C. 99.7%
D. 50%
Rationale: In a normal distribution: approximately 68% of values fall within ±1σ, 95% within ±2σ, and
99.7% within ±3σ -2.
10. A forecasting method that uses a weighted average of past values with weights that decrease
exponentially is called:
A. Moving average
B. Exponential smoothing
C. Linear regression
D. Naive forecasting
Rationale: Exponential smoothing assigns exponentially decreasing weights to past observations,
with the most recent values receiving the highest weights.
11. Which forecasting method is most appropriate when little historical data is available?
A. Time-series analysis
B. Qualitative forecasting
,C. Causal modeling
D. Exponential smoothing
Rationale: Qualitative forecasting methods rely on expert judgment and opinion rather than
historical data, making them suitable when data is scarce (e.g., new product forecasting).
12. The tracking signal is used to:
A. Measure the average forecast error
B. Detect when a forecast is consistently biased
C. Calculate the seasonal index
D. Determine the optimal alpha value
Rationale: The tracking signal is the ratio of cumulative forecast error (RSFE) to MAD. It monitors
whether forecasts are consistently over- or under-predicting, indicating bias.
13. A forecast that consistently overestimates demand has a:
A. Positive tracking signal
B. Negative tracking signal
C. Zero tracking signal
D. Random tracking signal
Rationale: If actual demand is consistently below forecast, the cumulative error is negative, resulting
in a negative tracking signal.
14. Which of the following is a causal forecasting method?
A. Exponential smoothing
B. Moving average
C. Linear regression
D. Delphi method
Rationale: Linear regression is a causal method because it assumes the forecast variable is related to
other independent variables (e.g., price, advertising).
15. The primary disadvantage of the moving average method is that:
A. It is too complex to compute
B. It requires storing all historical data
C. It cannot be used for short-term forecasting
D. It always lags behind trends
Rationale: A moving average requires retaining n periods of historical data for each forecast. It also
lags trends, but the storage requirement is a key operational disadvantage.
16. If a time series has a consistent upward trend, which forecasting method would be most
appropriate?
A. Simple moving average
B. Linear trend projection
C. Naive forecast
D. Simple exponential smoothing
Rationale: Linear trend projection explicitly models the trend component, while simple exponential
smoothing and moving averages lag behind trends.
17. The seasonal index for a quarter with sales 20% above the annual average is:
A. 0.80
B. 1.20
C. 20.0
, D. 0.20
Rationale: Seasonal index = (Quarterly average / Overall average) × 100 or expressed as a decimal. If
a quarter is 20% above average, the index is 1.20.
18. Which of the following is true about forecast accuracy?
A. Forecasts are always exact
B. Forecast accuracy decreases as the forecast horizon increases
C. Qualitative forecasts are always more accurate than quantitative forecasts
D. Forecasts for individual products are more accurate than aggregate forecasts
Rationale: Uncertainty increases with the time horizon, so long-range forecasts are less accurate
than short-range forecasts. Aggregate forecasts (e.g., product families) are generally more accurate
than individual item forecasts.
19. A company uses a 3-period moving average. If sales for the last three periods were 100, 110,
and 120, the forecast for the next period is:
A. 100
B. 105
C. 110
D. 120
Rationale: (100 + 110 + 120) / 3 = = 110.
20. The value of alpha that gives equal weight to all past observations is:
A. 0.0
B. Not possible with exponential smoothing
C. 1.0
D. 0.5
Rationale: Exponential smoothing always gives more weight to recent observations. There is no
alpha value that gives equal weight to all past observations (a moving average with n periods does
this).
21. Which forecasting approach is most suitable for predicting demand for a brand-new product
with no historical data?
A. Exponential smoothing
B. Market survey
C. Moving average
D. Trend projection
Rationale: Market surveys gather consumer intentions and opinions, which is useful when no
historical data exists.
22. The RSFE (Running Sum of Forecast Errors) is used to calculate the:
A. MAD
B. MSE
C. Tracking signal
D. MAPE
Rationale: Tracking signal = RSFE / MAD. RSFE alone indicates bias; dividing by MAD normalizes it.
23. A forecast error of +5 means:
A. Actual demand was 5 units higher than forecast
B. Forecast was 5 units higher than actual demand
C. The forecast was perfect
