Problems 79

Discussion 1. In the lower levels of the pyramid forecasting system, how would you prevent abdication of the
Questions responsibility for forecasting?
2. Can a grocery store capture true demand data? How might a warehouse capture demand data?
3. Some experts have argued its more important to have low bias (mean error) than to have a low
MAD. Why would they argue this way?

Problems 1. Demand for stereo headphones and MP3 players for joggers has caused Nina Industries to grow
almost 50 percent over the past year. The number of joggers continues to expand, so Nina expects
demand for headsets to also expand, because, as yet, no safety laws have been passed to prevent
joggers from wearing them. Demand for the stereo units for last year was as follows:

Demand Demand
Month (Units) Month (Units)
January 4,200 July 5,300
February 4,300 August 4,900
March 4,000 September 5,400
April 4,400 October 5,700
May 5,000 November 6,300
June 4,700 December 6,000

a. Using least squares regression analysis, what would you estimate demand to be for each
month next year? Using a spreadsheet, follow the general format in Figure 3.3. Compare your
results to those obtained by using the forecast spreadsheet function.
b. To be reasonably confident of meeting demand, Nina decides to use three standard errors
of estimate for safety. How many additional units should be held to meet this level of
confidence?
2. Historical demand for a product is

Demand
January 12
February 11
March 15
April 12
May 16
June 15

a. Using a weighted moving average with weights of 0.60, 0.30, and 0.10, find the July forecast.
b. Using a simple three-month moving average, find the July forecast.
c. Using single exponential smoothing with 0.2 and a June forecast L 13, find the July
forecast. Make whatever assumptions you wish.
d. Using simple linear regression analysis, calculate the regression equation for the preceding
demand data.
e. Using the regression equation in d, calculate the forecast for July.
80 Chapter 3 Forecasting

3. The following tabulations are actual sales of units for six months and a starting forecast in
January.
a. Calculate forecasts for the remaining five months using simple exponential smoothing with
0.2.
b. Calculate MAD for the forecasts.

Actual Forecast
January 100 80
February 94
March 106
April 80
May 68
June 94

4. Zeus Computer Chips Inc. used to have major contracts to produce the Centrino-type chips. The
market has been declining during the past three years because of the dual-core chips, which it
cannot produce, so Zeus has the unpleasant task of forecasting next year. The task is unpleasant
because the firm has not been able to find replacement chips for its product lines. Here is demand
over the past 12 quarters:

2007 2008 2009

I 4,800 I 3,500 I 3,200
II 3,500 II 2,700 II 2,100
III 4,300 III 3,500 III 2,700
IV 3,000 IV 2,400 IV 1,700

Use the decomposition technique to forecast the four quarters of 2010.

5. Sales data for two years are as follows. Data are aggregated with two months of sales in each
period.

Months Sales Months Sales

JanuaryFebruary 109 JanuaryFebruary 115
MarchApril 104 MarchApril 112
MayJune 150 MayJune 159
JulyAugust 170 JulyAugust 182
SeptemberOctober 120 SeptemberOctober 126
NovemberDecember 100 NovemberDecember 106

a. Plot the data.

b. Fit a simple linear regression model to the sales data.
c. In addition to the regression model, determine multiplicative seasonal index factors. A full
cycle is assumed to be a full year.
d. Using the results from parts b and c, prepare a forecast for the next year.
 Problems 81

6. The demand manager of Maverick Jeans is responsible for ensuring sufficient warehouse space
for the finished jeans that come from the production plants. It has occasionally been necessary to
rent public warehouse space, something that Maverick would like to avoid. In order to estimate
the space requirements the demand manager is evaluating moving-average forecasts. The
demand (in 1,000 case units) for the last fiscal year is shown below.

Month 1 2 3 4 5 6 7 8 9 10 11 12
Demand 20 18 21 25 24 27 22 30 23 20 29 22

a. Use a three-month moving average to estimate the month-in-advance forecast of demand for
months 412 and generate a forecast for the first month of next year. Calculate the average
forecast error and mean absolute error.
b. Use a three-month weighted moving average with weights of 0.6, 0.3, 0.1 (most recent to
last recent, respectively) to calculate month-in-advance forecasts for months 412 and
forecast for the first month of next year. Calculate the average forecast error and mean ab-
solute error.
c. Compare the average forecast error and MAD for the forecasting methods in parts a and b.
Based on these error calculations, which of the two forecast methods would you recommend?
Why?
7. Maverick Jeans demand manager decided to evaluate exponential smoothing. To maintain com-
parability, she used the data from problem 6.
a. Use a starting forecast of 20 for month 4 to develop forecasts for months 512 and the first
month of next year. Use smoothing constants () of 0.2, 0.5, and 0.8. Calculate the average
forecast error and mean absolute error.
b. What observations do you have? How does exponential smoothing compare to the results for
moving averages from problem 6?
8. Talbot Publishing Companys production planning manager has provided the following historical
sales data for its leading textbook on forecasting:

Year 4 5 6 7
Sales (in 1,000 units) 21 18 20 17

The firm is considering using a basic exponential smoothing model with 0.2 to forecast this
items sales.
a. Use the sales average of 20,000 units through year 3 as the forecast for period 4. Prepare fore-
casts for years 5 through 7 as of the end of year 4.
b. Calculate the average error and MAD value for the three forecasts using the actual sales data
provided. Estimate the standard deviation of the forecast errors using the calculated MAD.
c. Redo the forecasts and MAD calculations, updating the forecasts for years 6 and 7 at the end
of years 5 and 6, respectively. What do you observe?
9. Use a spreadsheet program to compare a three-period moving average forecasting model with a
basic exponential smoothing model (ESF). Ten periods of past actual demand data are available
(27, 26, 32, 41, 28, 35, 43, 47, 28, and 38). Use the first five periods data to seed the model (as
described next) and the last 5 periods to test the model.
a. Use the average of the first five periods to seed the ESF model (i.e., the past forecast) and
a smoothing model with 0.3. Compare this ESF model with a 3-period moving
82 Chapter 3 Forecasting

average model using the last 5 periods of data. Use MAD as the criterion for comparing the
techniques.
b. Does your comparison change if you use the past three periods data to seed the ESF model?
What is the best alpha (smoothing) value to use for this model?
10. The following two demand sets are to be used to test two different basic exponential smoothing
models. The first model uses 0.1, and the second uses 0.5. In both cases, the model
should be initialized with a beginning forecast value of 50; that is, the ESF forecast for period 1
made at the end of period 0 is 50 units. In each of the four cases (two models on two demand
sets), compute the average forecast error and MAD. What do the results mean?

Demand Set I Demand Set II

Period Demand Period Demand
1 51 1 77
2 46 2 83
3 49 3 90
4 55 4 22
5 52 5 10
6 47 6 80
7 51 7 16
8 48 8 19
9 56 9 27
10 51 10 79
11 45 11 73
12 52 12 88
13 49 13 15
14 48 14 21
15 43 15 85
16 46 16 22
17 55 17 88
18 53 18 75
19 54 19 14
20 49 20 16

11. In an effort to improve customer service and reduce the cost of inventory, the Thanskavel Com-
pany invested in a number of lean manufacturing initiatives. One result of these investments was
a reduction of the manufacturing lead time for the companys specialty product, Eggsbar, from six
weeks to four weeks. A study of Eggsbar disclosed that the demand averages 200 units per week,
with a standard deviation of 22 units. The demand from one week to the next week was found to
be independent.
a. What was the mean and standard deviation of the demand during manufacturing lead time for
this product before and after the initiatives?
b. Thanskavel has a policy of holding safety stock of finished goods inventory in the amount of
2.5 times the standard deviation of demand during manufacturing lead time. In the past, this
amounted to $10,000 of safety stock inventory for this product. What reduction in safety stock
for this product resulted from the reduction of the manufacturing lead time?
12. Five individual products in a product family of the Cumberland Company have identical sales
patterns. Each averages 100 units per month, with a standard deviation of 10 units. Assuming
normal distributions and independent demands: