McGraw-Hill/I rwin Copyright 2013 by The McGraw-Hill Companies, I nc. All rights reserved.

Forecasting
Chapter 03
3-2
Learning Objectives
1. Understand the role of forecasting as a basis for
supply chain planning
2. Identify the basic components of demand:
average, trend, seasonal, and random variation
3. Show how to make a time series forecast using
moving averages, exponential smoothing, and
regression
4. Use decomposition to forecast when trend and
seasonality is present
5. Show how to measure forecast error
6. Describe the common qualitative forecasting
techniques, such as the Delphi method and
collaborative forecasting
3-3
The Role of Forecasting
Forecasting is a vital function and impacts every
significant management decision
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 roles require different forecasting
approaches
Decisions about overall directions require strategic
forecasts
Tactical forecasts are used to guide day-to-day
decisions
3-4
Types of Forecasting
Four basic types of forecasts
1. Qualitative
2. Time series analysis (primary focus of this
chapter)
3. Causal relationships
4. Simulation
Time series analysis is based on the idea that
data relating to past demand can be used to
predict future demand
3-5
Components of Demand
Average
demand for a
period of time
Trend
Seasonal
element
Cyclical
elements
Random
variation
Autocorrelation
Excel:
Components of
Demand
Instructor Slides
3-7
Trends
Identification of trend lines is a common
starting point when developing a forecast
Common trend types include linear, S-curve,
asymptotic, and exponential
3-8
Time Series Analysis
Using the past to predict the future
Used mainly for tactical decisions
Short term forecasting less than three months
Used to develop a strategy which will be implemented over the
next six to eighteen months (e.g. meeting demand)
Medium term forecasting three months to two
years
Useful for detecting general trends and identifying major turning
points
Long term forecasting greater than two years
3-9
Model Selection
Choosing an appropriate forecasting model
depends upon
1. Time horizon to be forecast
2. Data availability
3. Accuracy required
4. Size of forecasting budget
5. Availability of qualified personnel
3-10
Forecasting Method Selection
Guide
Forecasting Method Amount of Historical
Data
Data Pattern Forecast
Horizon
Simple moving average 6 to 12 months; weekly
data are often used
Stationary (i.e. no
trend or
seasonality)
Short
Weighted moving
average and simple
exponential smoothing
5 to 10 observations
needed to start
Stationary Short
Exponential smoothing
with trend
5 to 10 observations
needed to start

Stationary and
trend
Short
Linear regression 10 to 20 observations Stationary, trend,
and seasonality
Short to
Medium
3-11
Simple Moving Average
Forecast is the average of a fixed number of
past 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
3-12
Simple Moving Average
Formula

3-13
Simple Moving Average -
Example
Week Demand 3 Week 9 Week

Week Demand 3 Week 9 Week
1 800


16 1700 2200 1811
2 1400


17 1800 2000 1800
3 1000


18 2200 1833 1811
4 1500 1067


19 1900 1900 1911
5 1500 1300


20 2400 1967 1933
6 1300 1333


21 2400 2167 2011
7 1800 1433


22 2600 2233 2111
8 1700 1533


23 2000 2467 2144
9 1300 1600


24 2500 2333 2111
10 1700 1600 1367

25 2600 2367 2167
11 1700 1567 1467

26 2200 2367 2267
12 1500 1567 1500

27 2200 2433 2311
13 2300 1633 1556

28 2500 2333 2311
14 2300 1833 1644

29 2400 2300 2378
15 2000 2033 1733

30 2100 2367 2378

4
5
800 1400 1000
1067
3
1400 1000 1500
1300
3
F
F




10
800 1400 1000 ... 1300
1367
9
F


3-14
Weighted Moving Average
The simple moving average formula implies
equal weighting for all periods
A weighted moving average allows unequal
weighting of prior time periods
The sum of the weights must be equal to one
Often more recent periods are given higher
weights than periods farther in the past
3-15
Selecting Weights
Experience and/or trial-and-error are the
simplest approaches
The recent past is often the best indicator of
the future, so weights are generally higher for
more recent data
If the data are seasonal, weights should reflect
this appropriately
3-16
Exponential Smoothing
A weighted average method which includes all past
data in the forecasting calculation
More recent results are weighted more heavily
The most used of all forecasting techniques
An integral part of computerized forecasting
Well accepted for six 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
Instructor Slides
3-18
Exponential Smoothing Model
3-19
Exponential Smoothing
Example
Week Demand Forecas
t
1 820 820
2 775 820
3 680 811
4 655 785
5 750 759
6 802 757
7 798 766
8 689 772
9 775 756
10 760
3-20
Linear Regression Analysis
Regression is used to identify the functional
relationship between two or more 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
Y=a + bt
3-21
Example 3.2 Least Squares
Method
Quarter Sales Quarter Sales
1 600 7 2,600
2 1,550 8 2,900
3 1,500 9 3,800
4 1,500 10 4,500
5 2,400 11 4,000
6 3,100 12 4,900
The least squares method
determines the parameters a and b
such that the sum of the squared
errors is minimized least squares
3-22
Example 3.2 - Calculations
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
Sum 78 33,350 268,200 650 112,502,500
The forecast is extended to periods 13-16
3-23
Time Series Decomposition
Chronologically ordered data is referred to as
a time series
A time series may contain one or many
elements
Trend, seasonal, cyclical, autocorrelation, and
random
Identifying these elements and separating the
time series data into these components is
known as decomposition

3-24
Seasonal Variation
Seasonal variation may be either additive or
multiplicative (shown here with a changing
trend)
3-25
Determining Seasonal Factors
The seasonal factor (or index) is the ratio of
the amount sold during each season divided
by the average for all seasons
Season Past Sales Average
Sales for
Each
Season
Seasonal
Factor
Spring 200
Summer 350
Fall 300
Winter 150
Total 1000
3-26
Decomposition Using Least
Squares Regression
1. Decompose the time series into its
components
a. Find seasonal component
b. Deseasonalize the demand
c. Find trend component
2. Forecast future values of each component
a. Project trend component into the future
b. Multiply trend component by seasonal
component
3-27
Decomposition -Steps 1 & 2
(1)
Perio
d (t)
(2)
Quarte
r
(3)
Actual
Demand
(y)
(4)
Average of Same Quarters of
Each Year
(5)
Seasonal
Factor
1 I 600 0.82 1
736
2 II 1,550 1.10 4
5,650
3 III 1,500 0.97 9
13,896
4 IV 1,500 1.12 16
21,517
5 I 2,400 0.82 2,942.6 25 73,565
6 II 3,100 1.10 2,824.7 36 101,689
7 III 2,600 0.97 2,676.2 49 131,134
8 IV 2,900 1.12 2,599.9 64 166,394
9 I 3,800 0.82 4,659.2 81 377,395
10 II 4,500 1.10 4,100.4 100 410,040
11 III 4,000 0.97 4,117.3 121 498,193
12 IV 4,900 1.12 4,392.9 144 632,578
Column 3
Column
5
(Column
1)
2
Column 1 x
Column 6
3-28
Decompostion Step 5
Create the final forecast by adjusting the
regression line by the seasonal factor
Perio
d
Quarte
r
Y from
Regression
Seasonal
Factor
Forecast (F x
Seasonal Factor
13 I 5,003.5 0.82 4,102.87
14 II 5,345.7 1.10 5,880.27
15 III 5,687.9 0.97 5,517.26
16 IV 6,030.1 1.12 6,753.71
3-29
Forecast Errors
Forecast error is 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
3-30
Forecast Error Measurements
Ideally, MAD will be zero
(no forecasting error)
Larger values of MAD
indicate a less accurate
model
MAPE scales the forecast error
to the magnitude of demand

Instructor Slides
Mont
h
Foreca
st
Actua
l
Deviatio
n
Abs.
Dev.
1 1,000 950 -50 50
2 1,000 1,070 +70 70
3 1,000 1,100 +100 100
4 1,000 960 -40 40
5 1,000 1,090 +90 90
6 1,000 1,050 +50 50
6220 400
6220/6=1036.7 Average demand
MAD=400/6=66.7
MAPE=66.7/1036.7=6.43%
3-32
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
Market research
Panel consensus
Historical analogy
Delphi method