Because learning changes everything.

Chapter 3

FORECASTING

© 2023 McGraw Hill, LLC. All rights reserved. Authorized only for instructor use in the classroom. No reproduction or further distribution permitted without the prior written consent of McGraw Hill, LLC.
 Learning Objectives
1. Understand how forecasting is essential to supply chain
planning.
2. Evaluate demand using quantitative forecasting models.
3. Apply qualitative techniques to forecast demand.
4. Apply collaborative techniques to forecast demand.

© McGraw Hill, LLC 2

 Role of Forecasting in Starbucks
Starbucks, the largest coffee chain in the world with over 30000 stores in 80 countries.
• Variety of products beyond coffee, coffee beans, salads, sandwiches, mugs, etc.
• Product offerings varies by season and some are store location-specific.
• Many are perishable, some runs the risk of becoming obsolete.
• Starbucks branded coffee and ice cream are sold in grocery stores.
• Need for forecasting regional and global demand as well as store-specific or demand.

Far-flung Supply chain at Starbucks.

• The company brings raw beans from Latin America, Africa and Asia in giant shipping containers to
USA and European 6 storage locations near roasting facilities .
• Roasted coffee is packaged and shipped to distribution centers (DCs) 4 in USA, 2 in Europe and 2 in
Asia).
• DCs also store many more items that are sold through Starbucks besides coffee.

Various types of forecasting (global, regional, store-specific) will help make decisions as to the quantity to
be bought and shipped and processed through various steps in the supply chain.
Analytical exercise on Forecasting SC demand –Starbucks corp. gives a numerical illustration of this
interesting forecasting challenge.

© McGraw Hill, LLC 3

 Forecasting in Operations and Supply Chain
Management
Forecasting is vital to every business organization and impacts every significant
management decision.
• Forecasting is the basis of corporate planning and control .
• Finance and accounting use forecasts as the basis for budgeting and cost
control.
• Marketing relies on forecasts to make key decisions such as new product
planning and personnel compensation.
• Production uses forecasts to select suppliers, determine capacity
requirements, and to drive decisions about purchasing, staffing, and inventory.

Different activities require different forecasting approaches.

• Decisions about overall directions require strategic forecasts.
• Tactical forecasts are used to guide day-to-day decisions and the goal is to
estimate demand in short term.

© McGraw Hill, LLC 4

 Decoupling Points
• Decoupling points occur when inventory is positioned
in the supply chain to allow processes or entities to
operate independently.
• For example, inventory at a retail store separates
(buffers) the manufacturer from the actions of
individual consumers.
• Forecasts of demand at these decoupling points allow
inventory to be set to the proper level.

© McGraw Hill, LLC 5

 Types of Forecasting
Basic types of forecasts:
• Qualitative.
• Quantitative.
• Time series analysis (primary focus of this chapter).
• Causal relationships.
• Simulation.
Time series analysis is based on the idea that data
relating to past demand can be used to predict future
demand.

© McGraw Hill, LLC 6

 Components of Demand
Average demand for the period

Trend

Seasonal element

Cyclical elements

Random variation

Autocorrelation

© McGraw Hill, LLC 7

 Time series forecasting overview
Trend line is the usual starting point of the time series
forecasting.
Trend line is adjusted for
• Seasonal effects.
• Cyclical elements.
• Any other expected elements that may influence the final
forecast.

Estimation of autocorrelation allows us the improve

forecasting accuracy.
Estimation of random effects allows us to give a range for the
forecast.
© McGraw Hill, LLC 8
 Components of Demand – Growth &
Seasonal
Exhibit 3.1

Access the text alternative for slide images.

© McGraw Hill, LLC 9
 Trends
• Identification of trend lines is a common starting point
when developing a forecast.
• Common trend types include linear, S-curve, asymptotic,
and exponential.

Exhibit 3.2

Access the text alternative for slide images.

© McGraw Hill, LLC 10
 Time Series Analysis
• Using the past to predict the future.

Short term – forecasting less than three months

• Used mainly for tactical decisions (for example, replenishing inventory).
Medium term – forecasting three months to two years
• Used to develop a strategy which will be implemented over the next six to
eighteen months (for example, meeting demand).
Long term – forecasting greater than two years
• Useful for detecting general trends and identifying major turning points.

© McGraw Hill, LLC 11

 Model Selection
• Choosing an appropriate forecasting model depends upon:
1. Time horizon to forecast.
2. Data availability.
3. Accuracy required.
4. Size of forecasting budget.
5. Availability of qualified personnel.

• Other factors may also be considered.

1. Degree of flexibility (can the firm react quickly if the forecast
is inaccurate?).
2. Consequence of a bad forecast (important or costly decisions
require a good forecast).
© McGraw Hill, LLC 12
 Forecasting Method Selection Guide
Amount of
Forecasting Method Historical Data Data Pattern Forecast Horizon
Simple moving 6 to 12 months; Stationary (i.e. no Short
average weekly data are trend or seasonality)
often used
Weighted moving 5 to 10 observations Stationary Short
average and simple needed to start
exponential
smoothing
Exponential 5 to 10 observations Stationary and trend Short
smoothing with trend needed to start
Linear regression 10 to 20 Stationary, trend, Short to Medium
observations and seasonality
Trend and seasonal 2 to 3 observations Stationary, trend and Short to Medium
models per season seasonality

© McGraw Hill, LLC 13

 Simple Moving Average
Forecast is based on average demand over the most recent periods
Useful when demand is not growing or declining rapidly, and no seasonality is
present.
Removes some of the random fluctuation from the data.
Selecting the period length is important.
• Longer periods provide more smoothing.
• Shorter periods react to trends more quickly.

• - forecast for the coming period t.

• – number of periods to be averaged.
• - actual occurrence in the past period.
• , , and - actual occurrences two periods ago, three periods ago, and so on
up to n periods ago.

© McGraw Hill, LLC 14

 Simple Moving Average – Example (p. 52 to 54)
Note that no forecast is possible until “n” periods have passed
Exhibit 3.4
Week Demand 3 -Week 9 -Week Week Demand 3-Week 9-Week
1 800 16 1,700 2,200 1,811
2 1,400 17 1,800 2,000 1,800
3 1,000 18 2,200 1,833 1,811
4 1,500 1,067 19 1,900 1,900 1,911
5 1,500 1,300 20 2,400 1,967 1,933
6 1,300 1,333 21 2,400 2,167 2,011
7 1,800 1,433 22 2,600 2,233 2,111
8 1,700 1,533 23 2,000 2,467 2,144
9 1,300 1,600 24 2,500 2,333 2,111
10 1,700 1,600 1,367 25 2,600 2,367 2,167
11 1,700 1,567 1,467 26 2,200 2,367 2,267
12 1,500 1,567 1,500 27 2,200 2,433 2,311
13 2,300 1,633 1,556 28 2,500 2,333 2,311
14 2,300 1,833 1,644 29 2,400 2,300 2,378
15 2,000 2,033 1,733 30 2,100 2,367 2,378

Access the text alternative for slide images.

© McGraw Hill, LLC 15
 Weighted Moving Average
The simple moving
average formula implies all
periods are equally
important. • W1 – weight to be given to
A weighted moving the actual occurrence for the
average allows unequal period t – 1.
weighting of prior time • W2 – weight to be given to
periods. the actual occurrence for the
• The sum of the weights period t − 2.
must be equal to one. • W1 – weight to be given to
• More recent data the actual occurrence for the
(periods) are given more period t − n.
significance (higher • n – total number of prior
weights) than older data. periods in the forecast.
© McGraw Hill, LLC 16
 Choosing Weights
• Experience and trial and error are the simplest
approaches.
• The most recent past is the most important indicator of
what to expect in the future, so weights are generally
higher for more recent data.
• If the data are seasonal, weights should be established
appropriately.
• The weighted moving average has an advantage over the
simple moving average.

© McGraw Hill, LLC 17

 Exponential Smoothing
• The importance of data diminishes as the past becomes more
distant.
• The most logical and easiest method to use.
• An integral part of all computerized forecasting programs.
• Well accepted for six major reasons:
1. Exponential models are surprisingly accurate.
2. Formulating an exponential model is relatively easy.
3. The user can understand how the model works.
4. Little computation is required to use the model.
5. Computer storage requirements are small.
6. Tests for accuracy are easy to compute.

© McGraw Hill, LLC 18

 Exponential Smoothing Model

Ft Ft  1   ( At  1  Ft  1 )

• Ft -forecast value for time period t.

• F(t  1) -forecast value for the previous time period t − 1.

• A(t  1) -actual occurrence for time period t − 1.

•  -alpha, the smoothing constant.

© McGraw Hill, LLC 19

 Exponential Smoothing Example

Week Demand Forecast F2 F1   ( A1  F1 ) 820  0.2(820  820)

1 820 820
2 775 820 F4 F3   ( A3  F3 ) 811  0.2(680  811)
3 680 811
4 655 785 F6 F5   ( A5  F5 ) 759  0.2(750  759)
5 750 759
6 802 757 F8 F7   ( A7  F7 ) 766  0.2(798  766)
7 798 766
8 689 772 F10 F9   ( A9  F9 ) 756  0.2(775  756)
9 775 756
10 760

© McGraw Hill, LLC 20

 Exponential Smoothing with Trend
The presence of a trend in the data causes the exponential
smoothing forecast to always lag behind the actual
occurrence.
This can be corrected by adding a trend adjustment.
• The trend smoothing constant is delta ( ).

Both alpha and delta reduce the impact of the error that
occurs between the actual and the forecast.

© McGraw Hill, LLC 21

 Exponential Smoothing with Trend – Example 3.1

Calculate the new forecast, assuming the following:

• The previous forecast including trend (FITt − 1) is 110 and
the previous estimate of the trend (Tt − 1) is 10.
•  0.2 and  0.3.
• Actual demand for period t − 1 was 115.

© McGraw Hill, LLC 22

 Choosing Alpha and Delta

Relatively small values for 𝛼 and 𝛿 are common.

• Usually in the range 0.1 to 0.3.

𝛼 depends upon how much random variation is present.

𝛿 depends upon how steady the trend is.

of 𝛼 and 𝛿 to minimize overall forecast error.

Measurements of forecast error can be used to select values

© McGraw Hill, LLC 23

 Linear Regression Analysis
• Regression - the functional
relationship between two or more Yt a  bt
correlated variables, usually from
observed data.
• One variable (the dependent
variable) is predicted for given
values of the other variable (the
independent variable).
• Linear regression is a special case
which assumes the relationship
between the variables can be
explained with a straight line.

© McGraw Hill, LLC 24

 Linear Regression Forecasting
• Past data and future projections are assumed to fall
around straight line.
• Used both for time series forecasting and for causal
relationship forecasting.
• Least squares method used for linear regression
forecasting.

© McGraw Hill, LLC 25

 Least Squares Method – Example 3.2
• The least squares method Exhibit 3.6
determines the parameters a
and b such that the sum of the
squared errors is minimized –
the “least squares.”

Quarter Sales Quarter Sales

1 600 7 2,600
2 1,550 8 2,900 The best line to use is one
that minimizes the sum of
3 1,500 9 3,800 the squared errors.
4 1,500 10 4,500
5 2,400 11 4,000
6 3,100 12 4,900

Access the text alternative for slide images.

© McGraw Hill, LLC 26

 Calculations - Example 3.2
The forecast is then extended to periods 13 to 16.
Exhibit 3.7
t y t×y t squared y squared Y
1 600 600 1 360,000 801.3
2 1,550 3,100 4 2,402,500 1,160.9
3 1,500 4,500 9 2,250,000 1,520.5
4 1,500 6,000 16 2,250,000 1,880.1
5 2,400 12,000 25 5,760,000 2,239.7
6 3,100 18,600 36 9,610,000 2,599.4
7 2,600 18,200 49 6,760,000 2,959.0
8 2,900 23,200 64 8,410,000 3,318.6
9 3,800 34,200 81 14.440,000 3,678.2
10 4,500 45,000 100 20,250,000 4,037.8
11 4,000 44,000 121 16,000,000 4,397.4
12 4,900 58,800 144 24,010,000 4,757.1

© McGraw Hill, LLC 27

 Excel Regression Tool
• Microsoft Excel includes
Exhibit 3.8
data analysis tools,
which can perform least
squares regression on a
data set.

© McGraw Hill, LLC 28

 Decomposition of a Time Series
Time series - chronologically ordered data.
A time series may contain one or many elements.
• Trend, seasonal, cyclical, autocorrelation, and random.

Decomposition – the process of identifying and

separating time series data into fundamental
components:
• Trend.
• Seasonality.

© McGraw Hill, LLC 29

 Seasonal Variation
• Seasonal variation may be either additive or multiplicative
(superimposed on changing trend).

Exhibit 3.9

Trend + Seasonal factor Trend × Seasonal index

Access the text alternative for slide images.

© McGraw Hill, LLC 30
 Determining Seasonal Index – Simple Proportion
Example 3.3

The seasonal index is the ratio of the amount sold during

each season divided by the average for all seasons.

Example 3.3

Average Sales Next

for Each Expected Average Seasonal Year’s
Past Season Demand for Sales Index Seasonal
Season Sales (1,000 / 4) Seasonal Index Next Year (1,100 / 4) Forecast
Spring 200 1,000 / 4 = 250 200 / 250 = 0.8 275 × 0.8 = 220

Summer 350 1,000 / 4 = 250 350 / 250 = 1.4 275 × 1.4 = 385
Fall 300 1,000 / 4 = 250 300 / 250 = 1.2 275 × 1.2 = 330
Winter 150 1,000 / 4 = 250 150 / 250 = 0.6 275 × 0/6 = 165
Total 1,000 1,100

© McGraw Hill, LLC 31

 Computing Trend and Seasonal Indexes from a
Linear Regression – Example 3.4 1

Demand for the past two years: Exhibit 3.10

Quarter Amount Quarter Amount

1 300 7 520
2 200 8 420
3 220 9 400
4 530 10 700

Forecast including trend:

FIT = 176.1 + 52.3t

Access the text alternative for slide images.

© McGraw Hill, LLC 32
 Computing Trend and Seasonal Indexes
from a Linear Regression – Example 3.4 2

Exhibit 3.10

Access the text alternative for slide images.

© McGraw Hill, LLC 33
 Computing Trend and Seasonal Indexes
from a Linear Regression – Example 3.4 3

Example 3.4

• Forecast , including trend and seasonal index (FITS):

FITSt = FIT × Seasonal.

• Forecast for next year:

I – FITS9 = [176.1 + 52.3(9)]1.25 = 808
II – FITS10 = [176.1 + 52.3(10)]0.79 = 552
III – FITS11 = [176.1 + 52.3(11)]0.70 = 526
IV – FITS12 = [176.1 + 52.3(12)]1.28 = 1,029

© McGraw Hill, LLC 34

 Forecast Errors
Forecast error - the difference between the forecast value and
what actually occurred.
All forecasts contain some level of error.
Sources of error.
• Bias – when a consistent mistake is made.
• Random – errors that are not explained by the model being
used.

Measures of error.
• Mean absolute deviation (MAD).
• Mean absolute percent error (MAPE).
• Tracking signal.

© McGraw Hill, LLC 35

 Measurement of Error
• Ideally, MAD will be zero
• MAPE scales the forecast error
(no forecasting error).
to the magnitude of demand.
• Larger values of MAD
indicate a less accurate MAD
MAPE 
model. Average Demand

• Tracking signal indicates whether

forecast errors are accumulating
over time (either positive or negative
errors).

Running sum of forecast errors

TS 
Average mean absolute deviation

© McGraw Hill, LLC 36

 Computing Forecast Error – Example (p. 69)

Exhibit 3.12

Month Forecast Actual Deviation RSFE Abs. Dev. Sum of Abs. Dev. MAD TS
1 1,000 950 −50 −50 50 50 50 −1
2 1,000 1,070 +70 +20 70 120 60 0.33
3 1,000 1,100 +100 +120 100 220 73.3 1.64
4 1,000 960 −40 +80 40 260 65 1.2
5 1,000 1,090 +90 +170 90 350 70 2.4
6 1,000 1,050 +50 +220 50 400 66.7 3.3
Overall

© McGraw Hill, LLC 37

 Causal Relationship Forecasting
Causal relationship forecasting uses independent variables
other than time to predict future demand.
• This independent variable must be a leading indicator.

Many apparently causal relationships are actually just

correlated events – care must be taken when selecting
causal variables.

© McGraw Hill, LLC 38

 Multiple Regression Analysis
Appropriate when a number of factors influence
a variable of interest.
In this case, the forecast analyst may utilize multiple
regression.
• Analogous to linear regression analysis, but with multiple
independent variables.
• Multiple regression is supported by statistical software
packages.

© McGraw Hill, LLC 39

 Qualitative Forecasting Techniques
Generally used to take advantage of expert knowledge.
Useful when judgment is required, when products are new, or
if the firm has little experience in a new market.
Examples of techniques:
• Market research.
• Panel consensus.
• Historical analogy.
• Delphi method.

© McGraw Hill, LLC 40

 Collaborative Planning, Forecasting, and
Replenishment (CPFR)
An Internet tool used to coordinate the efforts of a supply
chain.
• Demand forecasting.
• Production and purchasing.
• Inventory replenishment.

Integrates all members of a supply chain – manufacturers,

distributors, and retailers.
Depends upon the exchange of internal information to
provide a more reliable view of demand.

© McGraw Hill, LLC 41

 N-Tier Supply Chain
Exhibit 3.14

Access the text alternative for slide images.

© McGraw Hill, LLC 42
 CPFR Steps
Step 1
Creation of a front-end partnership agreement.
Step 2
Joint business planning.
Step 3
Development of demand forecasts.
Step 4
Sharing forecasts.
Step 5
Inventory replenishment.

© McGraw Hill, LLC 43

 End of Main Content

Because learning changes everything. ®

www.mheducation.com

© 2023 McGraw Hill, LLC. All rights reserved. Authorized only for instructor use in the classroom. No reproduction or further distribution permitted without the prior written consent of McGraw Hill, LLC.
 Accessibility Content: Text Alternatives for Images

© McGraw Hill, LLC 45

 Components of Demand – Growth &
Seasonal – Text Alternative
Return to parent-slide containing images.

A graph is plotted for number of units demanded versus

years. The years range from 0 to 4. The graph shows a
horizontal line labeled average and a rising line labeled
trend. A rising oscillating curve, rises along the trend
line. Dots are plotted along the rising curve. The dots
along the peaks of the curve are indicated as seasonal.

Return to parent-slide containing images

.
© McGraw Hill, LLC 46
 Trends – Text Alternative
Return to parent-slide containing images.

The first image shows, the graphs are plotted for sales in
thousands versus years. The first graph for linear trend
shows a straight rising line. The second graph shows a S
shaped curve which has shallow rise, then a steep rise and
then flattens at the end.
The second image shows, the graphs are plotted for sales in
thousands versus years. The first graph for asymptotic trend
shown in inward opening rising curve with a steep slope at
the beginning which flattens as time increases. The second
graph shows an exponential curve.

Return to parent-slide containing images

.
© McGraw Hill, LLC 47
 Simple Moving Average – Example (p. 52 to
54) – Text Alternative
Return to parent-slide containing images.

The graph shows three lines for actual, 3-week and 9-week.
The actual line begins at 1, 800 and has an oscillating
upward trend it rises and ends at 31, 2000. The 3-week line
begins at 4, 1000 and also has a rising oscillating trend with
less up and downs. The line ends at 30, 2200. The 9-week
line begins at 10, 1200 and has a rising trend. The line ends
at 30, 2600.

Return to parent-slide containing images

.
© McGraw Hill, LLC 48
 Least Squares Method – Example 3.2 – Text
Alternative
Return to parent-slide containing images.

A graph is plotted for sales ranging from 0 to 5000 dollars

versus quarters ranging from 0 to 12. The graph shows a line
rising from 0, 500 to 12, 4800. The line has points ranging
from Y 1 to Y 12 scattered along it.

Return to parent-slide containing images

.
© McGraw Hill, LLC 49
 Seasonal Variation – Text Alternative
Return to parent-slide containing images.

The graph is plotted for amount versus months. The first

graph shows a rising oscillating curve which is bound
within two parallel rising lines. The amplitude of each
segment is same.
The second graph shows a rising oscillating curve which
is bound within two diverging rising lines. The amplitude
of each segment increases as time increases.

Return to parent-slide containing images

.
© McGraw Hill, LLC 50
 Computing Trend and Seasonal Indexes from
a Linear Regression – Example 3.4 – Text 1

Alternative
Return to parent-slide containing images.

A graph is plotted for amount ranging from 100 to 700

versus quarters. The graph shows a rising line from 0,
180 to 600, quarter 4 of last year. 8 points are scattered
around the line, 4 above and 4 below.

Return to parent-slide containing images

.
© McGraw Hill, LLC 51
 Computing Trend and Seasonal Indexes from
a Linear Regression – Example 3.4 – Text 2

Alternative
Return to parent-slide containing images.

The table has 5 columns for quarter, actual amount, from trend
equation F I T sub I = 176.1 + 52.3 t, ratio of actual over trend and
seasonal index (average of same quarters in both years). The data
in the table is as follows:
Quarter 2 years ago: 1, 2, 3 and 4. Actual amount for each quarter:
300, 200, 220, 530. From trend equation F I T sub I = 176.1 + 52.3
t: 228.3, 280.6, 335.9, 385.1. Ratio of actual over trend: 1.31, 0.71,
0.66, 1.38.
Last year: 1, 2, 3 and 4. Actual amount for each quarter: 520, 420,
400, 700. From trend equation F I T sub I = 176.1 + 52.3 t: 437.4,
489.6, 541.9, 594.2. Ratio of actual over trend: 1.19, 0.86, 0.74,
1.18. Seasonal index (average of same quarters in both years is
as follows: 1, 1.25; 2, 0.79; 3, 0.70; 4, 1.28.
Return to parent-slide containing images
.
© McGraw Hill, LLC 52
 N-Tier Supply Chain – Text Alternative
Return to parent-slide containing images.

In the flow diagram the solid arrows indicate the material

flows and the dashed arrows indicate information flows.
The material flows from Tier n vendors till tier 1 vendors
and then to fabrication and final assembly. From
fabrication and final assembly it flows to the distribution
center and finally to the retailer. From the retailer
forecast information is relayed to the distribution center.
Distribution center relays replenishment information to
fabrication and final assembly. Fabrication and final
assembly relays production planning and purchase
information to the vendors.
Return to parent-slide containing images
.
© McGraw Hill, LLC 53