Chapter 13

Forecasting
Chapter 13
© 2007 Pearson Education

How Forecasting
fits the Operations Management
Philosophy
Operations As a Competitive
Weapon
Operations Strategy
Project Management Process Strategy
Process Analysis
Process Performance and Quality
Constraint Management
Process Layout Supply Chain Strategy
Lean Systems Location
Inventory Management
Forecasting
Sales and Operations Planning
Resource Planning
Scheduling

Demand Forecasting
 Demand Forecasting:
 A prediction of future events used for planning
purposes
 Forecasting across the organization:
 Internal customers throughout the organization
depend on forecasts to formulate and execute their
plans. Examples:
 Finance: to project cash flow & capital requirements
 HR: to anticipate hiring & training needs
 Marketing: the primary source for sales forecasts
 Operations: to plan output levels, inventory levels,
schedules, and long term capabilities.
Forecasting at Unilever
 Customer demand planning (CDP), which is
critical to managing value chains, begins with
accurate forecasts.
 Unilever has a state-of-the-art CDP system that
blends historical shipment data with promotional
data and current order data.
 Statistical forecasts are adjusted with planned
promotion predictions.
 Forecasts are frequently reviewed and adjusted
with point of sale data.
 This has enabled Unilever to reduce its inventory
and improved its customer service.

Demand Patterns
 Forecasting customer demand is a difficult task.
 The demand for services and products can vary greatly.
 Changing in business conditions
 Rapid technology change
 Increasing environmental concerns
 Example:
 Demand for lawn fertilizer predictably increases in the
spring and summer months. Also demand at weekends
may depend on uncontrollable factors such as the
weather
Demand Patterns
 Time Series: The repeated observations of demand for a service
or product in their order of occurrence.
 There are five basic patterns of most time series.
a. Horizontal. The fluctuation of data around a constant mean.
b. Trend. The systematic increase or decrease in the mean of the
series over time.
c. Seasonal. A repeatable pattern of increases or decreases in
demand, depending on the time of day, week, month, or season.
d. Cyclical. The less predictable gradual increases or decreases
over longer periods of time (years or decades).
e. Random. The unforecastable variation in demand.

Demand Patterns
Horizontal Trend
Seasonal Cyclical

Designing the
Forecast System
1. Deciding what to forecast
 Level of aggregation.
 Units of measure.
2. Choosing the type of forecasting
method:
 Qualitative methods
 Judgment
 Quantitative methods
 Causal
 Time-series
1. Deciding What To Forecast
 Level of Aggregation:
 For individual service or product (for manufacturing plan)
 For total demand for groups and then divide to find
individual service or product forecast.
 Aggregation: The act of clustering several similar

services or products so that companies can obtain more
accurate forecasts.
 Few companies err by more than 5 percent when

forecasting total demand for all their services or products.
Errors in forecasts for individual items may be much
higher.
1. Deciding What To Forecast
 Units of measurement:
 Forecasts of sales revenue are not helpful because
prices fluctuate.
 Forecast the number of units of demand (SKU) then

translate into sales revenue estimates
 Stock-keeping unit (SKU): An individual item or product

that has an identifying code and is held in inventory
somewhere along the value chain.

2. Choosing the Type of
Forecasting Technique
 Judgment methods: A type of qualitative method that
translates the opinions of managers, expert opinions,
consumer surveys, and sales force estimates into
quantitative estimates.
 Causal methods: A type of quantitative method that
uses historical data on independent variables, such as
promotional campaigns, economic conditions, and
competitors’ actions, to predict demand.
 Time-series analysis: A statistical approach that relies
heavily on historical demand data to project the future
size of demand and recognizes trends and seasonal
patterns.
Judgment Methods
 Sales force estimates: The forecasts that are compiled from

estimates of future demands made periodically by members of
a company’s sales force.
 Executive opinion: A forecasting method in which the
opinions, experience, and technical knowledge of one or more
managers are summarized to arrive at a single forecast.
 Executive opinion can also be used for technological
forecasting to keep abreast of the latest advances in
technology.
 Market research: A systematic approach to determine
external consumer interest in a service or product by creating
and testing hypotheses through data-gathering surveys.
 Delphi method: A process of gaining consensus from a group
of experts while maintaining their anonymity.
Guidelines for Using Judgment
Forecasts
 Judgment forecasting is clearly needed when no

quantitative data are available to use quantitative
forecasting approaches.
 Guidelines for the use of judgment to adjust
quantitative forecasts to improve forecast quality
are as follows:
1. Adjust quantitative forecasts when they tend to be
inaccurate and the decision maker has important
contextual knowledge.
2. Make adjustments to quantitative forecasts to compensate
for specific events, such as advertising campaigns, the
actions of competitors, or international developments.
Causal Methods
Linear Regression
 Causal methods are used when historical data are
available and the relationship between the factor to
be forecasted and other external or internal factors
can be identified.
 Linear regression: A causal method in which one
variable (the dependent variable) is related to one or
more independent variables by a linear equation.
 Dependent variable: The variable that one wants to
forecast.
 Independent variables: Variables that are assumed
to affect the dependent variable and thereby “cause”
the results observed in the past.
Causal Methods
Linear Regression
Deviation,
Y Regression
Estimate of or error
Y from equation:
Dependent variable
regression Y = a + bX
equation
{ Actual
Y = dependent variable
X = independent variable
a = Y-intercept of the line
value b = slope of the line
of Y
Value of X used
to estimate Y
X
Independent variable
Linear Regression
Example 13.1
The following are sales and advertising data for the past 5
months for brass door hinges. The marketing manager says that
next month the company will spend $1,750 on advertising for the
product. Use linear regression to develop an equation and a
forecast for this product.
We use the computer to determine
Sales Advertising the best values of a, b, the correlation
Month (000 units) (000 $) coefficient (r), the coefficient of
1 264 2.5 determination (r2), and the standard
2 116 1.3 error of the estimate (syx).
3 165 1.4
a = – 8.135
4 101 1.0
5 209 2.0 b = 109.229X
r = 0.98
r2 = 0.96
syx= 15.603
Linear Regression Line for
Example 13.1
300 —
Y = a + bX
Sales (thousands of units)
250 —
Y = – 8.135 + 109.229X
200 —
150 — a = – 8.135
b = 109.229X
100 — r = 0.98
50 — r2 = 0.96
syx= 15.603
| | | |
1.0 1.5 2.0 2.5
Advertising (thousands of dollars)
Forecast for Month 6: X = $1750, Y = – 8.135 + 109.229(1.75) = 183,016

Forecasting Demand for
Example 13.1
 The production scheduler can use this

forecast of 183,016 units to determine the
quantity of brass door hinges needed for
month 6.
 If there are 62,500 units in stock, then the
requirement to be filled from production is
183,016 - 62,500 = 120,516 units.

Time Series Methods
 Naive forecast: A time-series method whereby the

forecast for the next period equals the demand for
the current period, or Forecast = Dt
 Can take into account a demand trend (increase or

decrease). Example: last week actual demand was
120, the week before was 108, so next week
demand will be = 120 +12 = 132 units
 Can take into account the seasonal pattern.

Example: demand of last July was 50,000 units, the
forecast for this coming July is 50,000 units.
Time Series Methods
 Simple moving average method: A time-series

method used to estimate the average of a demand
time series by averaging the demand for the n most
recent time periods.
 It removes the effects of random fluctuation and is

most useful when demand has no pronounced trend
or seasonal influences.

Forecasting Error
 For any forecasting method, it is important to

measure the accuracy of its forecasts.
 Forecast error is the difference found by
subtracting the forecast from actual demand
for a given period.
Et = Dt - Ft
where
Et = forecast error for period t
Dt = actual demand for period t
Ft = forecast for period t
Moving Average Method
Example 13.2
a. Compute a three-week moving average forecast for

the arrival of medical clinic patients in week 4.
The numbers of arrivals for the past 3 weeks were:
Patient
Week Arrivals
1 400
2 380
3 411
b. If the actual number of patient arrivals in week
4 is 415, what is the forecast error for week 4?
c. What is the forecast for week 5?

Example 13.2
Solution
a. Forecast for week

Week Arrivals Average 4 is the average of
1 400 the arrivals for
2 380
3 411 weeks 1,2 and 3
4 415 397 411 + 380 + 400
5 ? 402 F4 =
3
b. c.
Forecast error for week 4 is 18. Forecast for week 5
It is the difference between the
is the average of
actual arrivals (415) for week 4
and the average of 397 that was the arrivals for
used as a forecast for week 4. weeks 2,3 and 4 =
(415 – 397 = 18) 402
Example 13.2
The moving average method may involve the use of as many

periods of past demand as desired. The stability of the
demand series generally determines how many periods to
450 — include. Large value of n with stable demand & small value of
n with fluctuated demand.
430 —
410 —
Patient arrivals
390 —
370 —
Actual patient
arrivals
| | | | | |
0 5 10 15 20 25 30
Week
Comparison of
3- and 6-Week MA Forecasts
3-week moving 6-week moving

average forecast average forecast
Patient Arrivals
Actual patient arrivals
Week
Application 13.1
 We will use the following customer-arrival

data in this moving average application:

Application 13.1a Moving Average Method
D 4  D3  D 2 790 810 740

F5    780
3 3
780 customer arrivals

D 5  D4  D3 805 790 810

F6    801.667
3 3

© 2007 Pearson
Weighted Moving Averages
 Weighted moving average method: A time-

series method in which each historical
demand in the average can have its own
weight; the sum of the weights equals 1.0
Ft+1 = W1Dt + W2Dt-1 + …+ WnDt-n+1
 The advantage is that it allows you to emphasize

recent demand over earlier demand

Application 13.1b Weighted Moving Average
F5  W1D4  W 2 D3  W 3 D2  0.50790  0.30810  0.20740  786


F6  W1D 5  W 2 D 4  W 3 D 3  0.50805  0.30790  0.20810  801.5

© 2007 Pearson 
Exponential Smoothing
 Exponential smoothing method: A sophisticated

weighted moving average method that calculates the
average of a time series by giving recent demands more
weight than earlier demands.
Ft+1 =  (Demand this period) + (1 –  )(Forecast calculated last period)
=  Dt + (1– )Ft
Or an equivalent equation: Ft+1 = Ft +   (Dt – Ft )

Where alpha (is a smoothing parameter with a value between 0 and 1.0
Exponential smoothing is the most frequently used formal forecasting
method because of its simplicity and the small amount of data needed
to support it.
Example 13.3
Week Arrivals
Reconsider the medical clinic patient 1 400
arrival data. It is now the end of week 3. 2 380
a. Using  = 0.10, calculate the 3 411
exponential smoothing forecast for 4 415
week 4. 5 ?
Ft+1 =  Dt + (1-)Ft Ft= (400+380)/2
F4 = 0.10(411) + 0.90(390) = 392.1 = 390
b. What is the forecast error for week 4 if the

actual demand turned out to be 415?
E4 = 415 - 392 = 23
c. What is the forecast for week 5?
F5 = 0.10(415) + 0.90(392.1) = 394.4
Application 13.1c Exponential Smoothing
Ft1  Ft  Dt  Ft   783 0.20790 783  784.4


Ft1  Ft  D t  Ft   784.4  0.20805 784.4  788.52

© 2007 Pearson 
 Large  is used with large fluctuations in demand & to

emphasize recent demand.
 Example: if 
Ft+1 =  (Demand this period) + (1 –  )(Forecast calculated last period)

=  Dt + (1– )Ft
 Dt + (0.8)Ft
=  Dt + (0.8)(  Dt-1 + (0.8)Ft-1) =  Dt + 0.16 Dt-1 + 0.64Ft-1
=  Dt + 0.16 Dt-1 +  Dt-2 + 0.1024 Dt-3 + …….
Recent Demand has more weight

Trend-Adjusted
 A trend in a time series is a systematic increase or
decrease in the average of the series over time.
 Where a significant trend is present, exponential smoothing
approaches must be modified; otherwise, the forecasts tend to
be below or above the actual demand.
 Trend-adjusted exponential smoothing method: The

method for incorporating a trend in an exponentially
smoothed forecast.
 With this approach, the estimates for both the average and the
trend are smoothed, requiring two smoothing constants. For
each period, we calculate the average and the trend.

Trend-Adjusted Exponential
Smoothing Formula
Ft+1 = At +Tt
where At = Dt + (1 – )(At-1 + Tt-1)

Tt = (At – At-1) + (1 – )Tt-1
At = exponentially smoothed average of the series in period t

Tt = exponentially smoothed average of the trend in period t
 = smoothing parameter for the average
 = smoothing parameter for the trend
Dt = demand for period t
Ft+1 = forecast for period t + 1

Trend-Adjusted
Example 13.4 Medanalysis ran an average of 28
blood tests per week during the past four weeks. The trend
over that period was 3 additional patients per week. This
week’s demand was for 27 blood tests. We use  = 0.20 and
 = 0.20 to calculate the forecast for next week.
A0 = 28 patients and Tt = 3 patients Week Demand Forecast
At = Dt + (1 – )(At-1 + Tt-1)
1 D1
A1= 0.20(27) + 0.80(28 + 3) = 30.2
2 D2 Average = 28
Tt = (At – At-1) + (1 – )Tt-1 3 D3
T1 = 0.20(30.2 – 28) + 0.80(3) = 2.8
4 D4
5 D5= 27
Ft+1 = At + Tt
F2 = 30.2 + 2.8 = 33 blood tests 6 ?
Example 13.4 Medanalysis
Trend-Adjusted Exponential Smoothing
80 —
Trend-adjusted
70 — forecast
60 —
Patient arrivals
50 —
40 — Actual blood
test requests
30 —
| | | | | | | | | | | | | | |
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Week
Forecast for Medanalysis Using the
Trend-Adjusted Exponential Smoothing Model

© 2007 Pearson
Application 13.2
The forecaster for product A estimated (in

March) the sales average to be 300,000
cases sold per month and the trend to be
+8,000 per month.
The actual sales for April were 330,000 cases.
What is the forecast for May,
assuming  = 0.20 and  = 0.10?

Application 13.2 Solution
thousand
thousand
To make forecasts for periods beyond the next period, multiply the trend
estimate by the number of additional periods, and add the result to the
current
© 2007 average
Pearson Education
© 2007 Pearson
Seasonal Patterns
 Seasonal patterns are regularly repeated upward or

downward movements in demand measured in
periods of less than one year.
 An easy way to account for seasonal effects is to use one
of the techniques already described but to limit the data in
the time series to those time periods in the same season.
 If the weighted moving average method is used,

high weights are placed on prior periods belonging
to the same season.
 Multiplicative seasonal method is a method whereby
seasonal factors are multiplied by an estimate of average
demand to arrive at a seasonal forecast.
 Additive seasonal method is a method whereby
seasonal forecasts are generated by adding a constant to
the estimate of the average demand per season.
Multiplicative Seasonal
Method
 Step 0: Estimate the average total demand of the next
period (year)
 Step 1: For each year, calculate the average demand for
each season by dividing annual demand by the number of
seasons per year.
 Step 2: For each year, divide the actual demand for each
season by the average demand per season, resulting in a
seasonal index for each season of the year, indicating the
level of demand relative to the average demand.
 Step 3: Calculate the average seasonal index for each
season using the results from Step 2. Add the seasonal
indices for each season and divide by the number of years
of data.
 Step 4: Calculate each season’s forecast for next year.
Using the Multiplicative
Seasonal Method
Example 13.5: Stanley Steemer, a carpet cleaning company
needs a quarterly forecast of the number of customers expected next
year. The business is seasonal, with a peak in the third quarter and a
trough in the first quarter.
Forecast customer demand for each quarter of year 5, based on an
estimate of total year 5 demand of 2,600 customers.
Quarter Year 1 Year 2 Year 3 Year 4

1 45 70 100 100
2 335 370 585 725
3 520 590 830 1160
4 100 170 285 215
Total 1000 1200 1800 2200
Step 0: Demand has been increasing by an average of 400 customers

each year. The forecast demand is found by extending that trend, and
projecting
© 2007 an annual demand in year 5 of 2,200 + 400 = 2,600 customers.
Pearson Education
Example 13.5 (contd.)
 Step 1: Average seasonal demand
Average seasonal demand

Year 1 1000/4 = 250
Year 2 1200/4 = 300
Year 3 1800/4 = 450
Year 4 2200/4 = 550

 Step 2: seasonal Index .
 Example for Q1, Year 1: 45/250 = 0.18

1 0.18 0.233 0.222 0.1818
2 1.34 1.23 1.3 1.318
3 2.08 1.96 1.84 2.109
4 0.4 .566 .63 .390

 Step 3: The average seasonal Index .

 Example:
The average seasonal Index for Q1, All Years
=(0.18 + 0.233+ 0.222+ 0.1818 )/4 = 0.2043
 Step 4:
 Forecast = (The average seasonal Index) *(Average
seasonal demand for year 5).
Example: forecast for Q1, year 5 = 0.2042 X 650 = 132.795

 Step 2: seasonal Index .
 Example for Q1, Year 1: 45/250 = 0.18

The average
seasonal Inde

1 0.18 0.233 0.222 0.1818 .2042
2 1.34 1.23 1.3 1.318 1.297
3 2.08 1.96 1.84 2.109 1.997
4 0.4 .566 .63 .390 .4965

Example 13.5 OM Explorer Solution

© 2007 Pearson
Application 13.3 Multiplicative Seasonal Method
1320/4 quarters = 330

330 *.52 = 171.6 330 *1.24= 409.2
330 *1.48= 488.24

330 *.76= 250.8

© 2007 Pearson
Measures of
Forecast Error
 Cumulative sum of forecast errors (CFE): A
measurement of the total forecast error that
assesses the bias in a forecast. CFE = Et
 Mean squared error (MSE): A measurement of the
dispersion of forecast errors.
MSE =
 Et2
n
 Mean absolute deviation (MAD): A measurement
of the dispersion of forecast errors.
MAD = |E
n
t|
 Standard deviation (): A measurement

of the dispersion of forecast errors.  =  (Et – E )2
n–1
Measures of
Forecast Error
Mean absolute percent error (MAPE): A
measurement that relates the forecast error to the
level of demand and is useful for putting forecast
performance in the proper perspective.
 [ |Et | / Dt ](100)
MAPE =
n

Calculating Forecast Error
Example 13.6
The following table shows the actual sales of
upholstered chairs for a furniture manufacturer and
the forecasts made for each of the last eight months.
Calculate CFE, MSE, MAD, and MAPE for this product.
Absolute
Error Absolute Percent
Month, Demand, Forecast, Error, Squared, Error, Error,
t Dt Ft Et E t2 |Et| (|Et|/Dt)(100)
1 200 225 -25 625 25 12.5%
2 240 220 20 400 20 8.3
3 300 285 15 225 15 5.0
4 270 290 –20 400 20 7.4
5 230 250 –20 400 20 8.7
6 260 240 20 400 20 7.7
7 210 250 –40 1600 40 19.0
8 275 240 35 1225 35 12.7
Total –15 5275 195 81.3%
Example 13.6 Forecast Error Measures
Cumulative forecast error (bias): CFE = – 15
– 15
Average forecast error (mean bias): E= = – 1.875
8
5275
Mean squared error: MSE = = 659.4
8
Standard deviation:  = 27.4
195
Mean absolute deviation: MAD = = 24.4
8
81.3%
Mean absolute percent error: MAPE = = 10.2%
8

© 2007 Pearson
OM Solver Output for Medical Clinic Patient Arrivals

Results Sheet
Moving Average
Forecast for 7/17/06

Results Sheet
Weighted Moving Average

Results Sheet

Results Sheet
Trend-Adjusted

Forecast for 8/7/06

Comparing Forecasting Errors
Method 1 Method 2 Method 3 Method 4

Criteria for Selecting
Time-Series Methods
 Forecast error measures provide important information for
choosing the best forecasting method for a service or product.
 They also guide managers in selecting the best values for the
parameters needed for the method:
 n for the moving average method, the weights for the weighted
moving average method, and  for exponential smoothing.
 The criteria to use in making forecast method and parameter
choices include
1. minimizing bias
2. minimizing MAPE, MAD, or MSE
3. meeting managerial expectations of changes in the
components of demand
4. minimizing the forecast error last period
Using Multiple Techniques
 Research during the last two decades suggests that

combining forecasts from multiple sources often
produces more accurate forecasts.
 Combination forecasts: Forecasts that are produced

by averaging independent forecasts based on different
methods or different data or both.

Forecasting as a Process
The forecast process itself, typically done on a
monthly basis, consists of structured steps. They
often are facilitated by someone who might be called
a demand manager, forecast analyst, or
demand/supply planner.

Some Principles for the Forecasting
Process
• Better processes yield better forecasts.
• Demand forecasting is being done in virtually every company.
The challenge is to do it better than the competition.
• Better forecasts result in better customer service and lower
costs, as well as better relationships with suppliers and
customers.
• The forecast can and must make sense based on the big
picture, economic outlook, market share, and so on.
• The best way to improve forecast accuracy is to focus on
reducing forecast error.
• Bias is the worst kind of forecast error; strive for zero bias.
• Whenever possible, forecast at higher, aggregate levels.
Forecast in detail only where necessary.
• Far more can be gained by people collaborating and
communicating well than by using the most advanced
© 2007forecasting
Pearson Educationtechnique or model.
© 2007 Pearson
Demand Forecast Applications
Time Horizon
Application Short Term Medium Term Long Term
(0–3 months) (3 months– 2 years) (more than 2 years)
Forecast • Individual • Total sales • Total sales
Quality products or • Groups or families
services of products or
services
Decision • Inventory • Staff planning • Facility location
management • Production • Capacity planning
Area
• Final assembly planning • Process
scheduling • Master production management
• Workforce scheduling
scheduling • Purchasing
• Master production • Distribution
scheduling
Forecasting • Time series • Causal • Causal
Technique • Causal • Judgment • Judgment
• Judgment

© 2007 Pearson

Chapter 13

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Chapter 13

Încărcat de

Drepturi de autor:

Formate disponibile

Forecasting

© 2007 Pearson Education

© 2007 Pearson Education

© 2007 Pearson Education

© 2007 Pearson Education

© 2007 Pearson Education

 Aggregation: The act of clustering several similar

 Few companies err by more than 5 percent when

 Forecast the number of units of demand (SKU) then

 Stock-keeping unit (SKU): An individual item or product

© 2007 Pearson Education

 Sales force estimates: The forecasts that are compiled from

 Judgment forecasting is clearly needed when no

Forecast for Month 6: X = $1750, Y = – 8.135 + 109.229(1.75) = 183,016

 The production scheduler can use this

© 2007 Pearson Education

 Naive forecast: A time-series method whereby the

 Can take into account a demand trend (increase or

 Can take into account the seasonal pattern.

 Simple moving average method: A time-series

 It removes the effects of random fluctuation and is

© 2007 Pearson Education

 For any forecasting method, it is important to

a. Compute a three-week moving average forecast for

c. What is the forecast for week 5?

a. Forecast for week

The moving average method may involve the use of as many

3-week moving 6-week moving

Actual patient arrivals

 We will use the following customer-arrival

© 2007 Pearson Education

D 4  D3  D 2 790 810 740

D 5  D4  D3 805 790 810

 Weighted moving average method: A time-

Ft+1 = W1Dt + W2Dt-1 + …+ WnDt-n+1

 The advantage is that it allows you to emphasize

© 2007 Pearson Education

F5  W1D4  W 2 D3  W 3 D2  0.50790  0.30810  0.20740  786

786 customer arrivals

F6  W1D 5  W 2 D 4  W 3 D 3  0.50805  0.30790  0.20810  801.5

802 customer arrivals

 Exponential smoothing method: A sophisticated

Or an equivalent equation: Ft+1 = Ft +   (Dt – Ft )

b. What is the forecast error for week 4 if the

Ft1  Ft  Dt  Ft   783 0.20790 783  784.4

Ft1  Ft  D t  Ft   784.4  0.20805 784.4  788.52

 Large  is used with large fluctuations in demand & to

Ft+1 =  (Demand this period) + (1 –  )(Forecast calculated last period)

Recent Demand has more weight

 Trend-adjusted exponential smoothing method: The

© 2007 Pearson Education

where At = Dt + (1 – )(At-1 + Tt-1)

At = exponentially smoothed average of the series in period t

© 2007 Pearson Education

© 2007 Pearson Education

The forecaster for product A estimated (in

© 2007 Pearson Education

 Seasonal patterns are regularly repeated upward or

 If the weighted moving average method is used,

Quarter Year 1 Year 2 Year 3 Year 4

Step 0: Demand has been increasing by an average of 400 customers

 Step 1: Average seasonal demand

Average seasonal demand

© 2007 Pearson Education