FORECASTING

rat
nine Manion wenn
Forecasting
intoyourmind
p
• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, ?

 Forecasting
6828

• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.

7
Forecasting

• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.

7
b) 2.5, 4.5, 6.5, 8.5, 10.5, ?
Tcu'sown21
Forecasting
6828

• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.

7
b) 2.5, 4.5, 6.5, 8.5, 10.5, 12.
5
Forecasting

• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.

7
b) 2.5, 4.5, 6.5, 8.5, 10.5, 12.
5
c) 5.0, 7.5, 6.0, 4.5, 7.0, 9.5, 8.0, 6.5, ?
we
Forecasting

• Predict the next number in the pattern:

a) 3.7, 3.7, 3.7, 3.7, 3.7, 3. intourial

7
b) 2.5, 4.5, 6.5, 8.5, 10.5, 12. Danto 2

Fe
c) 5.0, 7.5, 6.0, 4.5, 7.0, 9.5, 8.0, 6.5, 9.
0
aJwalainulinown
No11
 Outline
d WOW
• What is forecasting? Homonymy
should know what happen and forecasting first

• Types of forecasts
Mann
• Qualitative (technological) methods:
down'doejnnl.esoimidv inward
• Forecasts generated subjectively by the forecaster
Jima
• Quantitative (statistical) methods:
www.nnmdnidiusiweinriooeh
• Forecasts generated through mathematical modeling
• Good forecasts
 What is Forecasting? Ho
n Process of predicting a future
event based on historical data CLUMANDVELMAAME
actionminnow news rate
n Educated Guessing
FWIWburngas
n Underlying basis of
all business decisions
ü Production main Wiman
ü Inventory reunionon udmondon
ü Personnel linuwin grand
ü Facilities nowDenon
 organization manegement
O N
Importance of Forecasting in OM
www.ATPN
e
ANNA OSAN are
Departments throughout the organization depend on
inward oineswargritionio
forecasts to formulate and execute their plans.
verbal ituound
Finance needs forecasts to project cash flows and capital
nourished
requirements.
undoniwenn no Douronine in
Human resources need forecasts to anticipate hiring needs.

Production needs forecasts to plan production levels,O

workforce,o material requirements,o inventories, etc.
Importance of Forecasting in OM
Potvin
e
aimoh
Manufacturers also forecast worker absenteeism,
lumen'IN O
machine availability, material costs,O transportation
niulinini
and production lead times, etc.
O

Besides demand,O service providers are also interested

manusinioniuddautern
in forecasts of population, of other demographic
Winwood C
variables, of weather, etc.
o o
 Types of Forecasting Models
•
2Mainmmann
Qualitative (technological) methods:
Type
duoyriunuinuurian DAMN
• Forecasts generated subjectively by the forecaster

Imo
• Quantitative (statistical) methods:mepredatadfana
• Forecasts generated through mathematical modeling
Qualitative Forecasting Methods Type
grow
Qualitative
gnaw
Forecasting

Models
Executive Market Delphi
Judgement/opi Research/ Method
nion Survey

6
anuranchasm Medina
professernial Smoothing
opinion expert
experaint

moreexperiant
Qualitative Methods Type
YMMV
Inanelom nuisuniscopious copious

gnomon
no

autism

win wusovan

n
londoninvonnosin
vitamin
mnnwimndn.us
 Type
Quantitative Forecasting Methods
Nimr
Quantitative
Winn
Forecasting

Opara1007 Judulin
nanoeudo

Time Series Regression

Models Models
pi
handsurinal nnwv.sn io
inputmannnnoooo

undisurnundlawnwiunton coverninioninorite

2. Moving 3. Exponential
1. Naive
mom
Average Smoothing
a) simple a) level
b) weighted b) trend mornin damsawww
c) seasonality
Forecasting During the Life Cycle W
Introductio
n
Growt
h mm
Maturit
y
1
Declin
e

panin Jima
gnaw Venn
Qualitative Quantitative
-1 Executive judgment want
modelsrinonunial models
1- Time series
2- Market research analysis
3-Survey of sales force
2- Regression analysis
minoruis nivdometrin
Sales

Time
 Type
Quantitative Forecasting Methods
Nima
Nima
Quantitative
Forecasting

Time Series Regression

Models Models

2. Moving 3. Exponential
1. Naive
Average Smoothing
a) simple
b) weighted
 USMA
Time Series Models
Time
mind
• Try to predict the future based on past data
n'Winonaironies
JIM
• Assume that factors influencing the past will continue to
indwa
influence the future
 limned
viaurenumnd In'D
USMA
1. Naive Approach Time 1 NA
roubon
n Demand in next period is the same as demand
intocroyuYainward
in most recent period
ü May sales = 48 → June forecast = 48
game
prior
pgwoop
seasonialaffect
iggy
n Usually not good
M flower
min'osine
brown
Tourismand
Hotel

laid
 USMA
2a. Simple Moving Average Time 2 MOV
Nina
• Assumes an average is a good estimator of future
behavior
Hugon88w
worthNNMJ HON
• Used if little or no trend jp nrwunoinfo8e
WenÉb
• Used for smoothing

moochingivovemNo
Ft+1 = Forecast for the upcoming period, t+1
n = Number of periods to be averaged
minor
A t = Actual occurrence in period t
 USMA
2a. Simple Moving AverageTime 2 MOV
You’re manager in Amazon’s electronics department.
You want to forecast ipod sales for months 4-6 using a
3-period moving average.
Sale
Mont s
h1 (000
4 4
2 6) 6
3 5 s
4 ?
5 ?
6 ?
 USMA
2a. Simple Moving Average
Time 2 MOV
You’re manager in Amazon’s electronics department.
You want to forecast ipod sales for months 4-6 using a
3-period moving average.
Sale Moving
Mont s (n=3
Average
h1 (000
4 )N
2 6) A
N
3 5 A
N
4 ? (4+6+5)/3
A
5 ? =5
6 ?
 USMA
2a. Simple Moving Average
Time 2 MOV
What if ipod sales were actually 3 in month
4 monoid main CAV
Sale Moving
Mont s (n=3
Average
h1 (000
4 )N
4t6 s
2 6) 4 A
N
3 5 A
N
4 3 A5 s
5 ?
6 ?
 USMA
2a. Simple Moving Average Time 2 MOV

Forecast for Month 5?

Sale Moving
Mont s (n=3
Average
h1 (000
4 )N
2 6) A
N
3 5 A
N
4 3 A5 s
5 ? (6+5+3)/3=4.6 p
6 ? 67
 USMA
2a. Simple Moving Average
Time 2 MOV

Actual Demand for Month 5 = 7

Sale Moving
Mont s (n=3
Average
h1 (000
4 )N
2 6) A
N
3 5 A
N
4 3 A5
5 7
6 ? 4.667
 USMA
2a. Simple Moving Average
Time 2 MOV

Forecast for Month 6?

Actual sale for calculate
Sale Moving
Mont s (n=3
Average
h1 (000
4 )N
2 6) A
N
3 5 A
N
4 3 A5
5 7
6 ? (5+3+7)/3
4.667
 USMA
2b. Weighted Moving Average
Time 2 MOV
• Gives more emphasis to recent data

• Weights
• decrease for older data
• sum to 1.0
pillowanddown Simple moving
average models
weight all previous
periods equally
 USMA
3. Exponential
Times EXPO
• Assumes the most recent observations have the
highest predictive value
• gives more weight to recent time periods

Ft+1 = Ft + a(At -
Ft) e
t Need initial
Ft+1 = Forecast value for time t+1 forecast Ft
to start.
At = Actual value at time t
= Smoothing constant
 3. Exponential – Example USMA
Times EXPO
Company A, a personal computer producer
purchases generic parts and assembles them to
final product. Even though most of the orders
require customization, they have many common
components. Thus, managers of Company A need
a good forecast of demand so that they can
purchase computer parts accordingly to minimize
inventory cost while meeting acceptable service
level. Demand data for its computers for the past 9
months is given in the following table.
 A prove a nowatwow compareerr

3. Exponential – Example I prior datadove

itwould

i Ai Ft+1 = Ft + a(At -
Ft)
Given the monthly demand
data what are the exponential
smoothing forecasts for
periods 1-9 using a =0.10?

Assume F1=D1
 USMA
3. Exponential – Example
Times EXPO
Ft+1 = Ft + a(At -
i
Ft) a
Ai Fi

F2 = F1+ a(A1–F1) =820+.1(820–820)

=820
 NS MN
3. Exponential – Example
Times EXPO
Ft+1 = Ft + a(At -
Ft)
i Ai
a
Fi

=
Fy
Fa
F3 = F2+ a(A2–F2) =820+.1(775–820)

=815.5

Fro though
3. Exponential – Example Say

Ft+1 = Ft + a(At - p
Ft)
i Ai
a
Fi

This process
continues
through month 10
 USMA
3. Exponential – Example
Times EXPO
Ft+1 = Ft + a(At -
Ft)
i Ai
a
Fi
a
= =
if the
a constant
equals 0.6
 Usma
Quantitative Forecasting Methods
Quantitative Regress
Forecasting

Time Series Regression

Models Models

2. Moving 3. Exponential
1. Naive
Average Smoothing
a) simple
b) weighted
 Linear regression in forecasting
Usma
Regress
Linear regression is based on
1. Fitting a straight line to data
2. Explaining the change in one variable through changes in
other variables.
to
mfs't y
y
dependent variable = a + b (independent variable)

By using linear regression, we are trying to explore which

independent variables affect the dependent variable
ummmm
Nanterre mo x wingman
maintain
Example: do people drink more when it’sUsma
cold?
Regress
Alcohol Sales

Which line best

fits the data?

elvity point is best

line pass
linear

Average Monthly
f Temperature
mm
 regress
Example: do people drink more when it’s cold?

Alcohol Sales

Which line best

fits the data?

The best line is the one

that minimizes the error
Temp
go dow
Average Monthly
Temperature
Formulas Usma
pm stope
Regress
y=a+bx
binconstanty any
Caprio

where,
Mean
 slope madwomen
U My
Regression – Example
y=a+b
not
Regress
X
 Conclusion: General Guiding Principles for
Forecasting

nindsomordnignmnin
1. Forecasts are more accurate for larger groups of items.
convoor
2. Forecasts are more accurate for shorter periods of time.
nominalerror
3. Every forecast should include an estimate of error.
Wilohrioshio b
1. Before applying any forecasting method, the total system
should be understood.
2. Before applying any forecasting method, the method should be
tested and evaluated. ariosidonahal union W'DerroratorD
6. Be aware of people; they can prove you wrong very easily in
forecasting
Taatanolds