Chap15 Demand Management and Forecasting
Chap15 Demand Management and Forecasting
Chapter 15
Demand Management
and
Forecasting
15-3
OBJECTIVES
• Demand Management
• Qualitative Forecasting
Methods
• Simple & Weighted Moving
Average Forecasts
• Exponential Smoothing
• Simple Linear Regression
• Web-Based Forecasting
15-4
Demand Management
Independent Demand:
Finished Goods
Dependent Demand:
A Raw Materials,
Component parts,
Sub-assemblies, etc.
B(4) C(2)
Independent Demand:
What a firm can do to manage it?
Types of Forecasts
• Qualitative (Judgmental)
• Quantitative
– Time Series Analysis
– Causal Relationships
– Simulation
15-7
Components of Demand
Seasonal
Seasonalvariation
variation
x
x x Linear
Linear
x x
x x Trend
x x Trend
Sales
x x x
x
x
xx
x xx x x
x
x
x x x x x x
x x x x x x
x x x
x xxxxx
x
x x
1 2 3 4
Year
15-9
Qualitative Methods
Delphi Method
A
A t-1 +
+ A
A t-2 +
+ AAt-3 +...+A
+...+At-n
FFtt == t-1 t-2 t-3 t-n
nn
A
A t-1 +
+ A
A t-2 +
+ AAt-3 +...+A
+...+At-n
FFtt == t-1 t-2 t-3 t-n
nn
Week Demand
1 650 Question:
Question: What
What are
are the
the 3-
3-
2 678 week
week and
and 6-week
6-week moving
moving
3 720 average
average forecasts
forecasts forfor
4 785
demand?
demand?
5 859
6 920 Assume
Assume youyou only
only have
have 33
7 850 weeks
weeks and
and 66 weeks
weeks of of
8 758 actual
actual demand
demand data
data for
for the
the
9 892 respective
10 920
respective forecasts
forecasts
11 789
12 844
15-14
Calculating the moving averages gives us:
Week Demand 3-Week 6-Week
1 650 F4=(650+678+720)/3
2 678
=682.67
3 720 F7=(650+678+720
4 785 682.67 +785+859+920)/6
5 859 727.67
=768.67
6 920 788.00
7 850 854.67 768.67
8 758 876.33 802.00
9 892 842.67 815.33
10 920 833.33 844.00
11 789 856.67 866.50
12 844 867.00 854.83
©The McGraw-Hill Companies, Inc., 2004
15-15
Plotting
Plottingthe
themoving
movingaverages
averagesandandcomparing
comparing
them
themshows
showshow
howthe
thelines
linessmooth
smooth out
outto
toreveal
reveal
the
theoverall
overallupward
upwardtrend
trendin
inthis
thisexample
example
1000
900
Demand
800
Demand
3-Week
700
6-Week
600
500 Note
Notehow
howthethe
1 2 3 4 5 6 7 8 9 10 11 12 3-Week
3-Weekisis
Week smoother
smootherthan
than
the
theDemand,
Demand,
and
and6-Week
6-Weekisis
even
evensmoother
smoother
15-16
Question:
Question: What
What is is
the
the 33 week
week
Week Demand moving
moving average
average
1 820 forecast
forecast for
for this
this
2 775 data?
data?
3 680
4 655
Assume
Assume you you only
only
5 620 have
have 33 weeks
weeks andand
6 600 55 weeks
weeks ofof actual
actual
7 575 demand
demand data
data for
for
the
the respective
respective
forecasts
forecasts
15-17
While
While the
the moving
moving average
average formula
formula implies
implies an an equal
equal
weight
weight being
being placed
placed on on each
each value
value that
that isis being
being averaged,
averaged,
the
the weighted
weighted moving
moving average
average permits
permits an
an unequal
unequal
weighting
weighting on
on prior
prior time
time periods
periods
The
The formula
formula for
for the
the moving
moving average
average is:
is:
FFt t == w
w11A
At-1 + w A t-2 ++ w
t-1 + w22 At-2 w33A
At-3 +...+w A t-n
t-3 +...+wnn At-n
nn
wwt ==weight
t weightgiven
occurrence
givento
totime
timeperiod
period“t”
“t”
ww ==11
ii
occurrence(weights
(weightsmust
mustadd
addtotoone)
one) i=1
i=1
15-19
Question:
Question:Given
Giventhetheweekly
weeklydemand
demandand
andweights,
weights,what
whatisis
the
theforecast
forecastfor
forthe
the44thperiod
th
periodor
orWeek
Week4?
4?
Note
Notethat
thatthe
theweights
weightsplace
placemore
moreemphasis
emphasison
onthe
the
most
mostrecent
recentdata,
data,that
thatisistime
timeperiod
period“t-1”
“t-1”
15-20
F4 = 0.5(720)+0.3(678)+0.2(650)=693.4
15-21
Question:
Question:Given
Giventhe theweekly
weeklydemand
demandinformation
informationand
and
weights,
weights,what
whatisisthe
theweighted
weightedmoving
movingaverage
averageforecast
forecast
of
ofthe
the55thperiod
th
periodor orweek?
week?
F5 = (0.1)(755)+(0.2)(680)+(0.7)(655)= 672
15-23
FFtt == FFt-1
t-1
+
+ (A
(A t-1
t-1
-
- F
F )
t-1)
t-1
Where :
Ft Forcast value for the coming t time period
Ft - 1 Forecast value in 1 past time period
At - 1 Actual occurance in the past t time period
Alpha smoothing constant
• Premise: The most recent observations might
have the highest predictive value
• Therefore, we should give more weight to the
more recent time periods when forecasting
15-24
Answer:
Answer:The
Therespective
respectivealphas
alphascolumns
columnsdenote
denotethe
theforecast
forecastvalues.
values. Note
Note
that
thatyou
youcan
canonly
onlyforecast
forecastone
onetime
timeperiod
periodinto
intothe
thefuture.
future.
Week Demand 0.1 0.6
1 820 820.00 820.00
2 775 820.00 820.00
3 680 815.50 793.00
4 655 801.95 725.20
5 750 787.26 683.08
6 802 783.53 723.23
7 798 785.38 770.49
8 689 786.64 787.00
9 775 776.88 728.20
10 776.69 756.28
15-26
Note
Notehow
howthat
thatthe
thesmaller
smalleralpha
alpharesults
resultsin
inaa smoother
smootherline
line
in
inthis
thisexample
example
900
800 Demand
Demand
700 0.1
600 0.6
500
1 2 3 4 5 6 7 8 9 10
Week
15-27
F1=820+(0.5)(820-820)=820 F3=820+(0.5)(775-820)=797.75
nn
1 MAD 0.8 standard deviation
AA --FF
t=1
tt tt
1 standard deviation 1.25 MAD
MAD
MAD == t=1
nn
Question:
Question: What
What isis the
the MAD
MAD value
value given
given
the
the forecast
forecast values
values inin the
the table
table below?
below?
40
nn
Note
Notethat
thatby
byitself,
itself,the
theMAD
AA --FF
tt tt
40 only
onlylets
letsus
usknow
knowthe
MAD
themean
mean
MAD
MAD==
t=1
t=1 == 40 ==10 error
errorininaaset
setof
offorecasts
nn 44 10 forecasts
15-32
RSFE
RSFE Running
Running sum
sum of
of forecast
forecast errors
errors
TS
TS == ==
MAD
MAD Mean
Mean absolute
absolute deviation
deviation
15-33
The
Thesimple
simplelinear
linearregression
regression Y
model
modelseeks
seeksto
tofit
fitaaline
line
through
throughvarious
variousdata
dataover
over
time
a
time
0 1 2 3 4 5 x (Time)
Yt = a + bx Is
Isthe
thelinear
linearregression
regressionmodel
model
aa == yy-- bx
bx
xy
xy -- n(y)(x)
n(y)(x)
bb == 22 22
xx -- n(x
n(x))
15-35
Question:
Question:Given
Giventhe
thedata
databelow,
below,what
whatisisthe
thesimple
simplelinear
linear
regression
regressionmodel
modelthat
thatcan
canbe
beused
usedto
topredict
predictsales
salesin
infuture
future
weeks?
weeks?
Week Sales
1 150
2 157
3 162
4 166
5 177
15-36
Answer:
Answer: First,
First, using
using the
thelinear
linear regression
regressionformulas,
formulas, we
we
can
can compute
compute“a” “a”and
and“b”
“b”
Week Week*Week Sales Week*Sales
1 1 150 150
2 4 157 314
3 9 162 486
4 16 166 664
5 25 177 885
3 55 162.4 2499
Average Sum Average Sum
bb==
xy
xy--n(n(y)(x)
y)(x) 2499
== 2499--5(162.4)(3)
5(162.4)(3) 63
63==6.3
6.3
x - n(x )
x22
- n(x 22
) 55 5( 9
55 5(9 )) 10
10
aa== yy--bx
bx==162.4
162.4--(6.3)(3)
(6.3)(3)==143.5
143.5
15-37
155 Forecast
150
145
140
135
1 2 3 4 5
Period
15-38
Web-Based Forecasting:
Steps in CPFR
Question Bowl
Question Bowl
Question Bowl
Question Bowl
Question Bowl
Question Bowl
Question Bowl
Answer: c. 0 to 1
15-47
Question Bowl
Question Bowl
Answer: e. 0
15-49
Question Bowl
Answer: b. 75 (Y=25+5(10)=75)
15-50
Question Bowl
End of Chapter 15
1-51