Upload
patz68
View
217
Download
0
Embed Size (px)
Citation preview
8/2/2019 05 Forecasting
1/81
2008 Prentice-Hall, Inc.
Chapter 5
To accompanyQuantitative Analysis for Management, Tenth Edition,by
Render, Stair, and HannaPower Point slides created by Jeff Heyl
Forecasting
2009 Prentice-Hall, Inc.
8/2/2019 05 Forecasting
2/81
2009 Prentice-Hall, Inc. 5 2
Introduction
Managers are always trying to reduceuncertainty and make better estimates of whatwill happen in the future
This is the main purpose of forecasting Some firms use subjective methods
Seat-of-the pants methods, intuition,experience
There are also several quantitative techniques Moving averages, exponential smoothing,
trend projections, least squares regressionanalysis
8/2/2019 05 Forecasting
3/81
2009 Prentice-Hall, Inc. 5 3
Introduction
Forecasting is a method for translating pastexperience into estimates of the future
Virtually used by most management decisionssuch as:
Working capital needs Size of workforce Inventory levels Scheduling of production runs
8/2/2019 05 Forecasting
4/81
2009 Prentice-Hall, Inc. 5 4
Introduction
Judgmental Forecasting is used when situations inwhich human judgment is the only realisticforecasting method
It is especially used when: There are a few data for building a quantitative
model
Historical data are no longer representative due
to drastic environmental changes Decision to be taken is critical If the forecaster views himself as an expert in
the area
8/2/2019 05 Forecasting
5/81
2009 Prentice-Hall, Inc. 5 5
Introduction
Judgmental Forecasting Evaluated:
Trends in the data caused by seasonal patterns
may be overlooked
Human judgment may be biased based on thefollowing types:
Gamblers fallacy Conservatism
8/2/2019 05 Forecasting
6/81
2009 Prentice-Hall, Inc. 5 6
Introduction
Eight steps to forecasting :
1. Determine the use of the forecastwhatobjective are we trying to obtain?
2. Select the items or quantities that are to be
forecasted3. Determine the time horizon of the forecast
4. Select the forecasting model or models
5. Gather the data needed to make the forecast
6. Validate the forecasting model7. Make the forecast
8. Implement the results
8/2/2019 05 Forecasting
7/81 2009 Prentice-Hall, Inc. 5 7
Introduction
These steps are a systematic way of initiating,designing, and implementing a forecastingsystem
When used regularly over time, data is collected
routinely and calculations performedautomatically
There is seldom one superior forecastingsystem
Different organizations may use differenttechniques Whatever tool works best for a firm is the one
they should use
8/2/2019 05 Forecasting
8/81 2009 Prentice-Hall, Inc. 5 8
Regression Analysis
MultipleRegression
MovingAverage
ExponentialSmoothing
TrendProjections
Decomposition
DelphiMethods
Jury of ExecutiveOpinion
Sales ForceComposite
ConsumerMarket Survey
Time-Series MethodsQualitativeModels
CausalMethods
Forecasting Models
ForecastingTechniques
Figure 5.1
8/2/2019 05 Forecasting
9/81 2009 Prentice-Hall, Inc. 5 9
Time-Series Models
Time-series models attempt to predictthe future based on the past
Common time-series models are Moving average Exponential smoothing Trend projections Decomposition
Regression analysis is used in trendprojections and one type ofdecomposition model
8/2/2019 05 Forecasting
10/81 2009 Prentice-Hall, Inc. 5 10
Causal Models
Causal modelsCausal models use variables or factorsthat might influence the quantity beingforecasted
The objective is to build a model withthe best statistical relationship betweenthe variable being forecast and theindependent variables
Regression analysis is the mostcommon technique used in causalmodeling
8/2/2019 05 Forecasting
11/81 2009 Prentice-Hall, Inc. 5 11
Qualitative Models
Qualitative modelsQualitative models incorporate judgmentalor subjective factors
Useful when subjective factors are
thought to be important or when accuratequantitative data is difficult to obtain
Common qualitative techniques are Delphi method
Jury of executive opinion Sales force composite Consumer market surveys
8/2/2019 05 Forecasting
12/81 2009 Prentice-Hall, Inc. 5 12
Qualitative Models
Delphi MethodDelphi Method an iterative group process where(possibly geographically dispersed) respondentsrespondentsprovide input to decision makersdecision makers
Jury of Executive OpinionJury of Executive Opinion collects opinions of a
small group of high-level managers, possiblyusing statistical models for analysis
Sales Force CompositeSales Force Composite individual salespersonsestimate the sales in their region and the data is
compiled at a district or national level Consumer Market SurveyConsumer Market Survey input is solicited from
customers or potential customers regarding theirpurchasing plans
8/2/2019 05 Forecasting
13/81 2009 Prentice-Hall, Inc. 5 13
Quantitative forecasting models assume that thetime series follows some pattern called theforecast profiles
The 4 kinds of trends are:
Constant Level-assumes no trend at all in the data Linear Trend
- forecast a straight-line trend
Exponential Trend- forecast that the amount of growth will increasecontinuously
Damped Trend- the amount of trend declines in each period and diesout to form a horizontal line
Time Series Patterns
8/2/2019 05 Forecasting
14/81 2009 Prentice-Hall, Inc. 5 14
Time-series Patterns
The 3 kinds of seasonalities:1. Nonseasonal Pattern
- assumed to have a relatively constant mean
1. Additive Seasonal Pattern- assumes that the seasonal fluctuations are
of constant size
1. Multiplicative Seasonal Pattern
- assumes that the seasonal fluctuations areproportional to the data
8/2/2019 05 Forecasting
15/81
2009 Prentice-Hall, Inc. 5 15
Scatter Diagrams
0
50100
150
200
250
300
350
400
450
0 2 4 6 8 10 12
Time (Years)
AnnualSales
Radios
Televisions
CompactDiscs
Scatter diagrams are helpful when forecasting time-series databecause they depict the relationship between variables.
8/2/2019 05 Forecasting
16/81
2009 Prentice-Hall, Inc. 5 16
Scatter Diagrams
Wacker Distributors wants to forecast sales for threedifferent products
YEAR TELEVISION SETS RADIOS COMPACT DISC PLAYERS
1 250 300 110
2 250 310 1003 250 320 120
4 250 330 140
5 250 340 170
6 250 350 150
7 250 360 160
8 250 370 190
9 250 380 200
10 250 390 190
Table 5.1
8/2/2019 05 Forecasting
17/81
2009 Prentice-Hall, Inc. 5 17
Scatter Diagrams
Figure 5.2
330
250
200
150
100
50
| | | | | | | | | |
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
AnnualSalesofT
elevisions
(a) Sales appear to be
constant over time
Sales = 250 A good estimate of
sales in year 11 is250 televisions
8/2/2019 05 Forecasting
18/81
2009 Prentice-Hall, Inc. 5 18
Scatter Diagrams
Sales appear to beincreasing at aconstant rate of 10
radios per yearSales = 290 + 10(Year)
A reasonableestimate of sales in
year 11 is 400televisions
420
400
380
360
340
320
300
280
| | | | | | | | | |
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
AnnualSalesofR
adios
(b)
Figure 5.2
8/2/2019 05 Forecasting
19/81
2009 Prentice-Hall, Inc. 5 19
Scatter Diagrams
This trend line maynot be perfectlyaccurate because ofvariation from year toyear
Sales appear to beincreasing
A forecast wouldprobably be a larger
figure each year
200
180
160
140
120
100
| | | | | | | | | |
0 1 2 3 4 5 6 7 8 9 10
Time (Years)
AnnualSalesofCD
Players
(c)
Figure 5.2
8/2/2019 05 Forecasting
20/81
8/2/2019 05 Forecasting
21/81
2009 Prentice-Hall, Inc. 5 21
Measures of Forecast Accuracy
Using a navenave forecasting modelYEAR ACTUAL
SALES OF CDPLAYERS
FORECAST SALES ABSOLUTE VALUE OFERRORS (DEVIATION), (ACTUAL
FORECAST)
1 110
2 100 110 |100 110| = 10
3 120 100 |120 110| = 20
4 140 120 |140 120| = 20
5 170 140 |170 140| = 30
6 150 170 |150 170| = 20
7 160 150 |160 150| = 10
8 190 160 |190 160| = 309 200 190 |200 190| = 10
10 190 200 |190 200| = 10
11 190
Sum of |errors| = 160
MAD = 160/9 = 17.8Table 5.2
8/2/2019 05 Forecasting
22/81
2009 Prentice-Hall, Inc. 5 22
Measures of Forecast Accuracy
Using a navenave forecasting modelYEAR ACTUAL
SALES OF CDPLAYERS
FORECAST SALES ABSOLUTE VALUE OFERRORS (DEVIATION), (ACTUAL
FORECAST)
1 110
2 100 110 |100 110| = 10
3 120 100 |120 110| = 20
4 140 120 |140 120| = 20
5 170 140 |170 140| = 30
6 150 170 |150 170| = 20
7 160 150 |160 150| = 10
8 190 160 |190 160| = 309 200 190 |200 190| = 10
10 190 200 |190 200| = 10
11 190
Sum of |errors| = 160
MAD = 160/9 = 17.8Table 5.2
8179
160errorforecast.MAD =
n
8/2/2019 05 Forecasting
23/81
2009 Prentice-Hall, Inc. 5 23
Measures of Forecast Accuracy
There are other popular measures of forecastaccuracy
The mean squared errormean squared error
n
=2error)(
MSE
The mean absolute percent errormean absolute percent error
%MAPE 100actual
error
n
= And biasbias is the average error and tells whether the
forecast tends to be too high or too low and by how
much. Thus, it can be negative or positive.
8/2/2019 05 Forecasting
24/81
2009 Prentice-Hall, Inc. 5 24
Measures of Forecast Accuracy
Year Actual CD Sales Forecast Sales |Actual -Forecast|
1 110
2 100 110 10
3 120 100 20
4 140 120 20
5 170 140 30
6 150 170 20
7 160 150 10
8 190 160 30
9 200 190 10
10 190 200 10
11 190
Sum of |errors| 160
MAD 17.8
8/2/2019 05 Forecasting
25/81
2009 Prentice-Hall, Inc. 5 25
Hospital Days Forecast ErrorExample
Ms. Smith forecastedtotal hospital inpatientdays last year. Now thatthe actual data areknown, she isreevaluating herforecasting model.
Compute the MAD,
MSE, and MAPE for herforecast.
Month Forecast Actual
JAN 250 243
FEB 320 315
MAR 275 286
APR 260 256MAY 250 241
JUN 275 298
JUL 300 292
AUG 325 333
SEP 320 326
OCT 350 378
NOV 365 382
DEC 380 396
8/2/2019 05 Forecasting
26/81
2009 Prentice-Hall, Inc. 5 26
Hospital Days Forecast ErrorExample
Forecast Actual |error| error 2
|error/actual|
JAN 250 243 7 49 0.03
FEB 320 315 5 25 0.02
MAR 275 286 11 121 0.04
APR 260 256 4 16 0.02
MAY 250 241 9 81 0.04
JUN 275 298 23 529 0.08
JUL 300 292 8 64 0.03
AUG 325 333 8 64 0.02
SEP 320 326 6 36 0.02
OCT 350 378 28 784 0.07 NOV 365 382 17 289 0.04
DEC 380 396 16 256 0.04
AVERAGEMAD=11.83
MSE=192.83
MAPE=.0381*100 =3.81
8/2/2019 05 Forecasting
27/81
2009 Prentice-Hall, Inc. 5 27
Time-Series Forecasting Models
A time series is a sequence of evenlyspaced events (weekly, monthly, quarterly,etc.)
Time-series forecasts predict the futurebased solely of the past values of thevariable
Other variables, no matter how potentially
valuable, are ignored
8/2/2019 05 Forecasting
28/81
2009 Prentice-Hall, Inc. 5 28
Decomposition of a Time-Series
A time series typically has four components
1.1. TrendTrend(TT) is the gradual upward ordownward movement of the data over time
2.2. SeasonalitySeasonality(SS
) is a pattern of demandfluctuations above or below trend line thatrepeats at regular intervals
3.3. CyclesCycles (CC) are patterns in annual data thatoccur every several years
4.4. Random variationsRandom variations (RR) are blips in thedata caused by chance and unusualsituations
8/2/2019 05 Forecasting
29/81
2009 Prentice-Hall, Inc. 5 29
Decomposition of a Time-Series
Average Demandover 4 Years
TrendComponent
ActualDemand
Line
Time
De
mandforProduct
orService
| | | |
Year Year Year Year 1 2 3 4
Seasonal Peaks
Figure 5.3
8/2/2019 05 Forecasting
30/81
2009 Prentice-Hall, Inc. 5 30
Decomposition of a Time-Series
There are two general forms of time-seriesmodels
The multiplicative model
Demand = Tx Sx Cx R
The additive model
Demand = T+ S+ C+ R
Models may be combinations of these two forms Forecasters often assume errors are normally
distributed with a mean of zero
8/2/2019 05 Forecasting
31/81
2009 Prentice-Hall, Inc. 5 31
Moving Averages
Moving averagesMoving averages can be used when demand isrelatively steady over time
The next forecast is the average of the most
recent n data values from the time series The most recent period of data is added and the
oldest is droppedThis methods tends to smooth out short-term
irregularities in the data series
n
nperiodspreviousindemandsofSumforecastaverageMoving =
8/2/2019 05 Forecasting
32/81
2009 Prentice-Hall, Inc. 5 32
Moving Averages
Mathematically
n
YYYF
nttt
t
111
+++= ...
where
= forecast for time period t+ 1
= actual value in time period t
n = number of periods to average
tY
1tF
8/2/2019 05 Forecasting
33/81
2009 Prentice-Hall, Inc. 5 33
Wallace Garden Supply Example
Wallace Garden Supply wants to forecastdemand for its Storage Shed
They have collected data for the past year
They are using a three-month movingaverage to forecast demand (n = 3)
8/2/2019 05 Forecasting
34/81
2009 Prentice-Hall, Inc. 5 34
Wallace Garden Supply Example
Table 5.3
MONTH ACTUAL SHED SALES THREE-MONTH MOVING AVERAGE
January 10
February 12
March 13
April 16
May 19
June 23
July 26
August 30
September 28
October 18
November 16
(12 + 13 + 16)/3 = 13.67
(13 + 16 + 19)/3 = 16.00
(16 + 19 + 23)/3 = 19.33
(19 + 23 + 26)/3 = 22.67
(23 + 26 + 30)/3 = 26.33
(26 + 30 + 28)/3 = 28.00
(30 + 28 + 18)/3 = 25.33
(28 + 18 + 16)/3 = 20.67
(18 + 16 + 14)/3 = 16.00
(10 + 12 + 13)/3 = 11.67
8/2/2019 05 Forecasting
35/81
2009 Prentice-Hall, Inc. 5 35
Weighted Moving Averages
Weighted moving averagesWeighted moving averages use weights to put moreemphasis on recent periods
Often used when a trend or other pattern is emerging
= )( ))(( Weights periodinvalueActualperiodinWeight1 iFt Mathematically
n
ntntt
twww
YwYwYwF +
+= ++...
...
21
11211
ere
wi= weight for the ith observation
8/2/2019 05 Forecasting
36/81
2009 Prentice-Hall, Inc. 5 36
Weighted Moving Averages
Both simple and weighted averages areeffective in smoothing out fluctuations in thedemand pattern in order to provide stableestimates
ProblemsIncreasing the size ofn smoothes outfluctuations better, but makes the method lesssensitive to real changes in the data
Moving averages can not pick up trends verywell they will always stay within past levels andnot predict a change to a higher or lower level
8/2/2019 05 Forecasting
37/81
2009 Prentice-Hall, Inc. 5 37
Wallace Garden Supply Example
Wallace Garden Supply decides to try a weightedmoving average model to forecast demand for itsStorage Shed
They decide on the following weighting scheme
WEIGHTS APPLIED PERIOD
3 Last month
2 Two months ago
1 Three months ago
6
3 x Sales last month + 2 x Sales two months ago + 1 X Sales three months ago
Sum of the weights
8/2/2019 05 Forecasting
38/81
2009 Prentice-Hall, Inc. 5 38
Wallace Garden Supply Example
Table 5.4
MONTH ACTUAL SHED SALES THREE-MONTH WEIGHTEDMOVING AVERAGE
January 10
February 12
March 13
April 16
May 19
June 23
July 26
August 30
September 28
October 18
[(3 X 13) + (2 X 12) + (10)]/6 = 12.17[(3 X 16) + (2 X 13) + (12)]/6 = 14.33
[(3 X 19) + (2 X 16) + (13)]/6 = 17.00
[(3 X 23) + (2 X 19) + (16)]/6 = 20.50
[(3 X 26) + (2 X 23) + (19)]/6 = 23.83
[(3 X 30) + (2 X 26) + (23)]/6 = 27.50
[(3 X 28) + (2 X 30) + (26)]/6 = 28.33
[(3 X 18) + (2 X 28) + (30)]/6 = 23.33
[(3 X 16) + (2 X 18) + (28)]/6 = 18.67
[(3 X 14) + (2 X 16) + (18)]/6 = 15.33
8/2/2019 05 Forecasting
39/81
2009 Prentice-Hall, Inc. 5 39
Wallace Garden Supply Example
Program 5.1A
8/2/2019 05 Forecasting
40/81
2009 Prentice-Hall, Inc. 5 40
Wallace Garden Supply Example
Program 5.1B
8/2/2019 05 Forecasting
41/81
2009 Prentice-Hall, Inc. 5 41
Exponential Smoothing
Exponential smoothingExponential smoothingis easy to use andrequires little record keeping of data
It is a type of moving average
ecast = Last periods forecast+ (Last periods actual demand Last periods forecast)
Where is a weight (orsmoothing constantsmoothing constant) witha value between 0 and 1 inclusive
A largergives more importance to recent datawhile a smaller value gives more importance to
past data
8/2/2019 05 Forecasting
42/81
8/2/2019 05 Forecasting
43/81
2009 Prentice-Hall, Inc. 5 43
Exponential Smoothing Example
In January, Februarys demand for a certaincar model was predicted to be 142
Actual February demand was 153 autos Using a smoothing constant of = 0.20, what
is the forecast for March?
New forecast (for March demand) = 142 + 0.2(153 142)
= 144.2 or 144 autos
If actual demand in March was 136 autos, theApril forecast would be
New forecast (for April demand) = 144.2 + 0.2(136 144.2)
= 142.6 or 143 autos
8/2/2019 05 Forecasting
44/81
2009 Prentice-Hall, Inc. 5 44
Selecting the Smoothing Constant
Selecting the appropriate value foris key toobtaining a good forecast
The objective is always to generate an
accurate forecast The general approach is to develop trial
forecasts with different values ofand selectthe that results in the lowest MAD
8/2/2019 05 Forecasting
45/81
2009 Prentice-Hall, Inc. 5 45
Port of Baltimore Example
QUARTER ACTUALTONNAGE
UNLOADED
FORECASTUSING =0.10
FORECASTUSING =0.50
1 180 175 175
2 168 175.5 = 175.00 + 0.10(180 175) 177.5
3 159 174.75 = 175.50 + 0.10(168 175.50) 172.75
4 175 173.18 = 174.75 + 0.10(159 174.75) 165.88
5 190 173.36 = 173.18 + 0.10(175 173.18) 170.44
6 205 175.02 = 173.36 + 0.10(190 173.36) 180.22
7 180 178.02 = 175.02 + 0.10(205 175.02) 192.61
8 182 178.22 = 178.02 + 0.10(180 178.02) 186.30
9 ? 178.60 = 178.22 + 0.10(182 178.22) 184.15
Table 5.5
Exponential smoothing forecast for two values of
8/2/2019 05 Forecasting
46/81
2009 Prentice-Hall, Inc. 5 46
Selecting the Best Value ofQUARTER ACTUAL
TONNAGEUNLOADED
FORECASTWITH = 0.10
ABSOLUTEDEVIATIONSFOR = 0.10
FORECASTWITH = 0.50
ABSOLUTEDEVIATIONSFOR = 0.50
1 180 175 5.. 175 5.
2 168 175.57.5..
177.59.5..
3 159 174.7515.75
172.7513.75
4 175 173.181.82
165.889.12
5 190 173.3616.64
170.4419.56
6 205 175.0229.98
180.2224.78
7 180 178.021.98
192.6112.61
8 182 178.223.78
186.304.3..
Sum of absolute deviations 82.45 98.63
MAD = |deviations| = 10.31 MAD = 12.33
Table 5.6Best choiceBest choice
8/2/2019 05 Forecasting
47/81
2009 Prentice-Hall, Inc. 5 47
Port of Baltimore Example
Program 5.2A
8/2/2019 05 Forecasting
48/81
2009 Prentice-Hall, Inc. 5 48
Port of Baltimore Example
Program 5.2B
8/2/2019 05 Forecasting
49/81
2009 Prentice-Hall, Inc. 5 49
PM Computer: Moving AverageExample
PM Computer assembles customized personalcomputers from generic parts
The owners purchase generic computer parts involume at a discount from a variety of sourceswhenever they see a good deal.
It is important that they develop a good forecast ofdemand for their computers so they can purchasecomponent parts efficiently.
8/2/2019 05 Forecasting
50/81
2009 Prentice-Hall, Inc. 5 50
PM Computers: Data
Period Month Actual Demand1 Jan 37
2 Feb 40
3 Mar 41
4 Apr 37
5 May 456 June 50
7 July 43
8 Aug 47
9 Sept 56
Compute a 2-month moving average Compute a 3-month weighted average using weights of
4,2,1 for the past three months of data Compute an exponential smoothing forecast using = 0.7,
previous forecast of 40
Using MAD, what forecast is most accurate?
8/2/2019 05 Forecasting
51/81
2009 Prentice-Hall, Inc. 5 51
PM Computers: Moving AverageSolution
2 monthMA Abs. Dev 3monthWMA Abs. Dev Exp.Sm. Abs. Dev
37.00
37.00 3.00
38.50 2.50 39.10 1.90
40.50 3.50 40.14 3.14 40.43 3.43
39.00 6.00 38.57 6.43 38.03 6.97
41.00 9.00 42.14 7.86 42.91 7.09
47.50 4.50 46.71 3.71 47.87 4.87
46.50 0.50 45.29 1.71 44.46 2.54
45.00 11.00 46.29 9.71 46.24 9.76
51.50 51.57 53.07
5.29 5.43 4.95MADExponential smoothing resulted in the lowest MAD.
8/2/2019 05 Forecasting
52/81
2009 Prentice-Hall, Inc. 5 52
Exponential Smoothing withTrend Adjustment
Like all averaging techniques, exponentialsmoothing does not respond to trends
A more complex model can be used thatadjusts for trends
The basic approach is to develop anexponential smoothing forecast then adjust itfor the trend
New forecast (Ft)+ Trend correction (T
t)
S
8/2/2019 05 Forecasting
53/81
2009 Prentice-Hall, Inc. 5 53
Exponential Smoothing withTrend Adjustment
The equation for the trend correction uses anew smoothing constant
Ttis computed by
)()1( 11 tttt FFTT += ++
where
Tt+1 = smoothed trend for period t+ 1
Tt= smoothed trend for preceding period
= trend smooth constant that we select
Ft+1 = simple exponential smoothed forecast
for period t+ 1
Ft= forecast for pervious period
8/2/2019 05 Forecasting
54/81
2009 Prentice-Hall, Inc. 5 54
Selecting a Smoothing Constant
As with exponential smoothing, a high value ofmakes the forecast more responsive to changes intrend
A low value ofgives less weight to the recent trend
and tends to smooth out the trend Values are generally selected using a trial-and-error
approach based on the value of the MAD for differentvalues of
Simple exponential smoothing is often referred to asfirst-order smoothingfirst-order smoothing
Trend-adjusted smoothing is called second-ordersecond-order,double smoothingdouble smoothing, orHolts methodHolts method
8/2/2019 05 Forecasting
55/81
2009 Prentice-Hall, Inc. 5 55
Trend Projection
Trend projection fits a trend line to a seriesof historical data points
The line is projected into the future for
medium- to long-range forecasts Several trend equations can be developed
based on exponential or quadratic models
The simplest is a linear model developed
using regression analysis
8/2/2019 05 Forecasting
56/81
8/2/2019 05 Forecasting
57/81
2009 Prentice-Hall, Inc. 5 57
Trend Projection
The mathematical form is
XbbY 10 +
where
= predicted value
b0 = intercept
b1 = slope of the line
X = time period (i.e., X= 1, 2, 3, , n)
Y
8/2/2019 05 Forecasting
58/81
2009 Prentice-Hall, Inc. 5 58
Trend Projection
ValueofDependentVariable
Time
*
*
*
*
*
*
*Dist2
Dist4
Dist6
Dist1
Dist3
Dist5
Dist7
Figure 5.4
Mid t M f t i
8/2/2019 05 Forecasting
59/81
2009 Prentice-Hall, Inc. 5 59
Midwestern ManufacturingCompany Example
Midwestern Manufacturing Company has experiencedthe following demand for its electrical generators overthe period of 2001 2007
YEAR ELECTRICAL GENERATORS SOLD
2001 74
2002 79
2003 80
2004 90
2005 1052006 142
2007 122
Table 5.7
Mid t M f t i
8/2/2019 05 Forecasting
60/81
2009 Prentice-Hall, Inc. 5 60
Midwestern ManufacturingCompany Example
Program 5.3A
Notice codeinstead of
actual years
8/2/2019 05 Forecasting
61/81
Mid t M f t i
8/2/2019 05 Forecasting
62/81
2009 Prentice-Hall, Inc. 5 62
Midwestern ManufacturingCompany Example
The forecast equation is
XY 54107156 .. +
To project demand for 2008, we use the coding
system to define X= 8
(sales in 2008) = 56.71 + 10.54(8)
= 141.03, or 141 generators
Likewise forX= 9
(sales in 2009) = 56.71 + 10.54(9)
= 151.57, or 152 generators
Mid t M f t i
8/2/2019 05 Forecasting
63/81
2009 Prentice-Hall, Inc. 5 63
Midwestern ManufacturingCompany Example
GeneratorDem
and
Year
160
150
140
130
120 110
100
90
80
70 60
50
| | | | | | | | |
2001 2002 2003 2004 2005 2006 2007 2008 2009
Actual Demand Line
Trend LineXY 54107156 .. +
Figure 5.5
Mid t M f t i
8/2/2019 05 Forecasting
64/81
2009 Prentice-Hall, Inc. 5 64
Midwestern ManufacturingCompany Example
Program 5.4A
Mid t M f t i
8/2/2019 05 Forecasting
65/81
2009 Prentice-Hall, Inc. 5 65
Midwestern ManufacturingCompany Example
Program 5.4B
8/2/2019 05 Forecasting
66/81
2009 Prentice-Hall, Inc. 5 66
Seasonal Variations
Recurring variations over time mayindicate the need for seasonaladjustments in the trend line
A seasonal index indicates how aparticular season compares with anaverage season
When no trend is present, the seasonalindex can be found by dividing theaverage value for a particular season bythe average of all the data
8/2/2019 05 Forecasting
67/81
2009 Prentice-Hall, Inc. 5 67
Seasonal Variations
Eichler Supplies sells telephoneanswering machines
Data has been collected for the past two
years sales of one particular model They want to create a forecast that
includes seasonality
8/2/2019 05 Forecasting
68/81
2009 Prentice-Hall, Inc. 5 68
Seasonal Variations
MONTH SALES DEMAND AVERAGE TWO-YEAR DEMAND
MONTHLYDEMAND
AVERAGESEASONAL
INDEXYEAR 1 YEAR 2
January 80 10090
94 0.957
February 85 7580
94 0.851
March 80 9085
94 0.904
April 110 90100
94 1.064
May 115 131123
94 1.309
June 120 110115
94 1.223
July 100 110105
94 1.117
August 110 90100
94 1.064
September 85 95
90
94 0.957Seasonal index =
Average two-year demand
Average monthly demandAverage monthly demand = = 94
1,128
12 months
Table 5.8
8/2/2019 05 Forecasting
69/81
2009 Prentice-Hall, Inc. 5 69
Seasonal Variations
The calculations for the seasonal indices are
Jan. July96957012
2001=.
,1121171
12
2001 =.,
Feb. Aug.85851012
2001
=.,
106064112
2001=.
,
Mar. Sept.90904012
2001=.
,969570
12
2001=.
,
Apr. Oct.106064112
2001=.
,858510
12
2001=.
,
May Nov.131309112
2001 =., 85851012
2001=.
,
June Dec.122223112
2001 =.,
85851012
2001 =.,
Regression with Trend and
8/2/2019 05 Forecasting
70/81
2009 Prentice-Hall, Inc. 5 70
Regression with Trend andSeasonal Components
Multiple regressionMultiple regression can be used to forecast bothtrend and seasonal components in a time series
One independent variable is time Dummy independent variables are used to represent
the seasons
The model is an additive decomposition model
where
X1 = time periodX2 = 1 if quarter 2, 0 otherwise
X3 = 1 if quarter 3, 0 otherwise
X4 = 1 if quarter 4, 0 otherwise
44332211 XbXbXbXbaY +
Regression with Trend and
8/2/2019 05 Forecasting
71/81
2009 Prentice-Hall, Inc. 5 71
Regression with Trend andSeasonal Components
Program 5.6A
Regression with Trend and
8/2/2019 05 Forecasting
72/81
2009 Prentice-Hall, Inc. 5 72
Regression with Trend andSeasonal Components
Program 5.6B (partial)
Regression with Trend and
8/2/2019 05 Forecasting
73/81
2009 Prentice-Hall, Inc. 5 73
Regression with Trend andSeasonal Components
The resulting regression equation is
4321 130738715321104 XXXXY ..... +
Using the model to forecast sales for the first two
quarters of next year
These are different from the results obtained using themultiplicative decomposition method
Use MAD and MSE to determine the best model
13401300738071513321104 =)(.)(.)(.)(..Y15201300738171514321104 =)(.)(.)(.)(..Y
Regression with Trend and
8/2/2019 05 Forecasting
74/81
2009 Prentice-Hall, Inc. 5 74
Regression with Trend andSeasonal Components
American Airlines original spare parts inventorysystem used only time-series methods to forecastthe demand for spare parts
This method was slow to responds to even moderatechanges in aircraft utilization let alone major fleet
expansions They developed a PC-based system named RAPS
which uses linear regression to establish arelationship between monthly part removals andvarious functions of monthly flying hours
The computation now takes only one hour instead ofthe days the old system needed
Using RAPS provided a one time savings of $7 millionand a recurring annual savings of nearly $1 million
8/2/2019 05 Forecasting
75/81
2009 Prentice-Hall, Inc. 5 75
Monitoring and Controlling Forecasts
Tracking signalsTracking signalscan be used to monitor theperformance of a forecast
Tacking signals are computed using thefollowing equation
MAD
RSFE=signalTracking
n
= errorforecastMADwhere
8/2/2019 05 Forecasting
76/81
2009 Prentice-Hall, Inc. 5 76
Monitoring and Controlling Forecasts
AcceptableRange
Signal Tripped
Upper Control Limit
Lower Control Limit
0 MADs
+
Time
Figure 5.7
Tracking Signal
8/2/2019 05 Forecasting
77/81
2009 Prentice-Hall, Inc. 5 77
Monitoring and Controlling Forecasts
Positive tracking signals indicate demand is greaterthan forecast
Negative tracking signals indicate demand is less thanforecast
Some variation is expected, but a good forecast willhave about as much positive error as negative error
Problems are indicated when the signal trips either theupper or lower predetermined limits
This indicates there has been an unacceptable amountof variation
Limits should be reasonable and may vary from item toitem
Regression with Trend and
8/2/2019 05 Forecasting
78/81
2009 Prentice-Hall, Inc. 5 78
Regression with Trend andSeasonal Components
How do you decide on the upper and lower limits? Too small a value will trip the signal too often and
too large will cause a bad forecast
Plossl & Wight use maximums of 4 MADs for
high volume stock items and 8 MADs for lowervolume items One MAD is equivalent to approximately 0.8
standard deviation so that 4 MADs =3.2 s.d.
For a forecast to be in control, 89% of the errorsare expected to fall within 2 MADs, 98% with 3MADs or 99.9% within 4 MADs whenever theerrors are approximately normally distributed
Ki b ll B k E l
8/2/2019 05 Forecasting
79/81
2009 Prentice-Hall, Inc. 5 79
Kimballs Bakery Example
Tracking signal for quarterly sales of croissantsTIME
PERIODFORECASTDEMAND
ACTUALDEMAND
ERROR RSFE |FORECAST || ERROR |
CUMULATIVEERROR
MAD TRACKINGSIGNAL
1 100 90 10 10 10 10 10.0 1
2 100 95 5 15 5 15 7.5 2
3 100 115 +15 0 15 30 10.0 0
4 110 100 10 10 10 40 10.0 1
5 110 125 +15 +5 15 55 11.0 +0.5
6 110 140 +30 +35 30 85 14.2 +2.5
2146
85errorforecast .MAD =n
sMAD..MAD
RSFE52
214
35signalTracking =
8/2/2019 05 Forecasting
80/81
2009 Prentice-Hall, Inc. 5 80
Forecasting at Disney
The Disney chairman receives a dailyThe Disney chairman receives a dailyreport from his main theme parks thatreport from his main theme parks thatcontains only two numbers thecontains only two numbers the forecastforecastof yesterdays attendance at the parks andof yesterdays attendance at the parks andthethe actualactualattendanceattendance
An error close to zero (using MAPE as themeasure) is expected
The annual forecast of total volume
conducted in 1999 for the year 2000resulted in a MAPE of 0
8/2/2019 05 Forecasting
81/81
Using The Computer to Forecast
Spreadsheets can be used by small andmedium-sized forecasting problems
More advanced programs (SAS, SPSS,Minitab) handle time-series and causal
models May automatically select best model
parameters Dedicated forecasting packages may be
fully automatic May be integrated with inventory planning
and control