Professional Documents
Culture Documents
Forecasting: To Accompany
Forecasting: To Accompany
Forecasting
To accompany
Quantitative Analysis for Management, Tenth Edition,
by Render, Stair, and Hanna © 2008 Prentice-Hall, Inc.
Power Point slides created by Jeff Heyl © 2009 Prentice-Hall, Inc.
Introduction
Managers are always trying to reduce
uncertainty and make better estimates of what
will 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 regression
analysis
Techniques
Qualitative Time-Series Causal
Models Methods Methods
Figure 5.1
Consumer
Decomposition
Market Survey
450
400
Radios
350
Annual Sales
300
250
Televisions
200
pac t Discs
150
100
Com
50
0
0 2 4 6 8 10 12
Time (Years)
(a)
Sales appear to be
Annual Sales of Televisions
330 –
constant over
250 – time
Sales = 250
200 – A good estimate of
150 – sales in year 11 is
100 –
250 televisions
50 –
| | | |
| | | |
| |
0 1 2 3 4 5 6
7 8 9 10
Time (Years)
© 2009 Prentice-Hall, Inc. 5 – 12
Scatter Diagrams
(b)
420 – Sales appear to be
400 – increasing at a
Annual Sales of Radios
380 – constant rate of 10
360 – radios per year
340 –
Sales = 290 +
320 –
10(Year)
300 – A reasonable
280 – estimate of sales in
| |
|
|
|
|
|
|
|
year 11 is 400
|
televisions
0 1 2 3 4 5 6
7 8 9 10
Time (Years)
200 –
accurate because
180 – of variation from
160 – year to year
Sales appear to be
140 –
increasing
120 –
A forecast would
100 – probably be a
| | | | |
larger figure each
|
|
| | |
year
0 1 2 3 4 5 6
7 8 9 10
Time (Years)
MAD
forecast error
n
MAD
5
forecast
170 error 140
160
17.8
|170 – 140| = 30
6 150 170 |150 – 170| = 20
7 160 n 150 |160 – 150| = 10
8 190 160 |190 – 160| = 30
9 200 190 |200 – 190| = 10
10 190
9 200 |190 – 200| = 10
11 — 190 —
Sum of |errors| = 160
MAD = 160/9 = 17.8
MSE (error)2
n
The mean absolute percent error
error
MAPE actual 100%
n
And bias 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 .
© 2009 Prentice-Hall, Inc. 5 – 18
Measures of Forecast Accuracy
Trend
Component
Seasonal Peaks
Actual
Demand
Line
Average Demand
over 4 Years
| | | |
Year Year Year Year
1 2 3 4
Time
Figure 5.3
Demand = T + S + C + R
Mathematically
Ft Yt Yt 1 ... Yt n1
n
1
here
= forecFats1t for time period t + 1
= actualYvtalue in time period t
n = number of periods
to average
Table 5.3
© 2009 Prentice-Hall, Inc. 5 – 29
Weighted Moving Averages
Weighted moving averages use weights to put
more emphasis on recent periods
Often used when a trend or other pattern is
emerging
Ft
(Weight in period i)(Actual value in
1
period) (Weights)
Mathematically
Ft w1Yt w2Yt 1 ... wnYt n1
w1 w2 ... w n
1
re
wi = weight for the ith observation
© 2009 Prentice-Hall, Inc. 5 – 30
Weighted Moving Averages
Both simple and weighted averages are
effective in smoothing out fluctuations in
the demand pattern in order to provide
stable estimates
Problems
Increasing the size of n smoothes out
fluctuations better, but makes the method
less sensitive to real changes in the data
Moving averages can not pick up trends
very well – they will always stay within past
levels and not predict a change to a higher or
lower level
© 2009 Prentice-Hall, Inc. 5 – 31
Wallace Garden Supply Example
Program 5.1A
© 2009 Prentice-Hall, Inc. 5 – 34
Wallace Garden Supply Example
Program 5.1B
© 2009 Prentice-Hall, Inc. 5 – 35
Exponential Smoothing
Exponential smoothing is easy to use and
requires little record keeping of data
It is a type of moving average
Mathematically
(Yt Ft )
Ft 1 Ft
re
= new forecast (for time period t + 1)
Ft+1
Ft = pervious forecast (for time period t)
= smoothing constant (0 ≤ ≤ 1)
Yt = pervious period’s actual demand
Table 5.5
© 2009 Prentice-Hall, Inc. 5 – 40
Selecting the Best Value of
ACTUAL FORECAST ABSOLUTE ABSOLUTE
TONNAGE WITH = DEVIATIONS DEVIATIONS
QUARTE 0.10 FORECAST FOR = 0.10 FOR =
R WITH = 0.50 0.50
175 175
UNLOADE 5 . 5
D
175.5 177.5
7.5 . 9.5
1
180 174.75 172.75
15.75 13.75
2 173.18 165.88
1.82 9.12
168
173.36 170.44
16.64 19.56
3
159 175.02 180.22
29.98 24.78
4 178.02 192.61
1.98 12.61
175
Table 5.6
8 186.30
3.78 4.3
5
182
Sum 190
Best cc h h oi
of absolute deviations
17 8. 2 2
82.45 98.63
ce Σ|deviations|
6 MAD = = ©M20A09DPr=entic1e-
5 – 41
Port of Baltimore Example
Program 5.2A
© 2009 Prentice-Hall, Inc. 5 – 42
Port of Baltimore Example
Program 5.2B
3 7 .0 0
3 7 .0 0 3 .0 0
3 8 .5 0 2 .5 0 3 9 .1 0 1 .9 0
4 0 .5 0 3 .5 0 4 0 .1 4 3 .1 4 4 0 .4 3 3 .4 3
3 9 .0 0 6 .0 0 3 8 .5 7 6 .4 3 3 8 .0 3 6 .9 7
4 1 .0 0 9 .0 0 4 2 .1 4 7 .8 6 4 2 .9 1 7 .0 9
4 7 .5 0 4 .5 0 4 6 .7 1 3 .7 1 4 7 .8 7 4 .8 7
4 6 .5 0 0 .5 0 4 5 .2 9 1 .7 1 4 4 .4 6 2 .5 4
4 5 .0 0 1 1 .0 0 4 6 .2 9 9 .7 1 4 6 .2 4 9 .7 6
5 1 .5 0 5 1 .5 7 5 3 .0 7
MAD 5 .2 9 5 .4 3 4 .9 5
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
© 2009 Prentice-Hall, Inc. 5 – 48
Selecting a Smoothing Constant
Yˆ b0 b1 X
where
= pYrˆedicted value
b0 = intercept
b1 = slope of the line
X = time period (i.e., X = 1, 2,
3, …, n)
© 2009 Prentice-Hall, Inc. 5 – 52
Trend Projection
Value of Dependent Variable
Dist7 *
Dist5 * Dist6
* Dist3 *
Dist4
Dist1 * Dist2
*
*
Table 5.7
Notice code
instead of
actual years
Program 5.3A
© 2009 Prentice-Hall, Inc. 5 – 55
Midwestern Manufacturing
Company Example
Program 5.3B
© 2009 Prentice-Hall, Inc. 5 – 56
Midwestern Manufacturing
Company Example
The forecast equation is
Likewise for X = 9
160 –
150 –
140 –
Trend Line
130 –
Generator Demand
Yˆ 56.71 10.54
120 –
X
110 –
100 –
90 –
80 –
70 – Actual Demand Line
60 –
50 –
| | | | | | | | |
Program 5.4A
Program 5.4B
© 2009 Prentice-Hall, Inc. 5 – 60
Seasonal Variations
■ Recurring variations over time may
indicate the need for seasonal
adjustments in the trend line
■ A seasonal index indicates how a
particular season compares with an
average season
■ When no trend is present, the seasonal
index can be found by dividing the
average value for a particular season by
the average of all the data
Yˆ a b1 X 1 b2 X 2 b3 X 3 b4 X 4
where
X1 = time period
X2 = 1 if quarter 2, 0 otherwise
X3 = 1 if quarter 3, 0 otherwise
X4 = 1 if quarter 4, 0 otherwise
© 2009 Prentice-Hall, Inc. 5 – 65
Regression with Trend and
Seasonal Components
Program 5.6A
RSFE
Tracking signal MAD
where
MAD
forecast error
n
Signal Tripped
Upper Control Limit Tracking Signal
+
Acceptable
0 MADs Range
–
Lower Control Limit
Time
Figure 5.7
2 100 95 –5 –15 5 15 0
7.5 –1
3 100 115 +15 0 15 30 +0.5
10.0
+2.5
4 110 100 –10 –10 10 40
10.0
5 110 125 +15 +5 15 55
11.0
6 110 140 +30 +35 30 85
14.2 RSFE 35
Tracking signal
2.5MADs
forecastMAD
error 85
© 2009 Prentice-Hall, Inc. 5 – 74
Forecasting at Disney
■ The Disney chairman receives a daily
report from his main theme parks that
contains only two numbers – the forecast
of yesterday’s attendance at the parks and
the actual attendance
■ An error close to zero (using MAPE as the
measure) is expected
■ The annual forecast of total volume
conducted in 1999 for the year 2000
resulted in a MAPE of 0