You are on page 1of 81

Chapter 5

Forecasting

To accompany Quantitative Analysis for Management, Tenth Edition, by Render, Stair, and Hanna Power Point slides created by Jeff Heyl

2008 Prentice-Hall, Inc. 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

2009 Prentice-Hall, Inc.

52

Introduction
Forecasting is a method for translating past

experience into estimates of the future Virtually used by most management decisions such as: Working capital needs Size of workforce Inventory levels Scheduling of production runs

2009 Prentice-Hall, Inc.

53

Introduction
Judgmental Forecasting is used when situations in

which human judgment is the only realistic forecasting 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
2009 Prentice-Hall, Inc. 54

Introduction
Judgmental Forecasting Evaluated: Trends in the data caused by seasonal patterns

may be overlooked

Human judgment may be biased based on the

following types: Gamblers fallacy Conservatism

2009 Prentice-Hall, Inc.

55

Introduction
Eight steps to forecasting : 1. Determine the use of the forecastwhat 2. 3. 4. 5. 6. 7. 8.

objective are we trying to obtain? Select the items or quantities that are to be forecasted Determine the time horizon of the forecast Select the forecasting model or models Gather the data needed to make the forecast Validate the forecasting model Make the forecast Implement the results
2009 Prentice-Hall, Inc. 56

Introduction
These steps are a systematic way of initiating,

designing, and implementing a forecasting system When used regularly over time, data is collected routinely and calculations performed automatically There is seldom one superior forecasting system Different organizations may use different techniques Whatever tool works best for a firm is the one they should use
2009 Prentice-Hall, Inc. 57

Forecasting Models
Forecasting Techniques Qualitative Models Delphi Methods Jury of Executive Opinion Sales Force Composite Consumer Market Survey Time-Series Methods Causal Methods Regression Analysis

Moving Average Exponential Smoothing Trend Projections

Multiple Regression

Figure 5.1 Decomposition


2009 Prentice-Hall, Inc. 58

Time-Series Models
Time-series models attempt to predict

the future based on the past Common time-series models are


Moving average Exponential smoothing Trend projections Decomposition

Regression analysis is used in trend

projections and one type of decomposition model

2009 Prentice-Hall, Inc.

59

Causal Models
Causal models use variables or factors

that might influence the quantity being forecasted The objective is to build a model with the best statistical relationship between the variable being forecast and the independent variables Regression analysis is the most common technique used in causal modeling
2009 Prentice-Hall, Inc. 5 10

Qualitative Models
Qualitative models incorporate judgmental

or subjective factors Useful when subjective factors are thought to be important or when accurate quantitative data is difficult to obtain Common qualitative techniques are
Delphi method Jury of executive opinion Sales force composite Consumer market surveys

2009 Prentice-Hall, Inc.

5 11

Qualitative Models
Delphi Method an iterative group process where

(possibly geographically dispersed) respondents provide input to decision makers Jury of Executive Opinion collects opinions of a small group of high-level managers, possibly using statistical models for analysis Sales Force Composite individual salespersons estimate the sales in their region and the data is compiled at a district or national level Consumer Market Survey input is solicited from customers or potential customers regarding their purchasing plans
2009 Prentice-Hall, Inc. 5 12

Time Series Patterns


Quantitative forecasting models assume that the

time series follows some pattern called the forecast 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 increase continuously

Damped Trend - the amount of trend declines in each period and dies out to form a horizontal line
2009 Prentice-Hall, Inc. 5 13

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 are proportional to the data

2009 Prentice-Hall, Inc.

5 14

Scatter Diagrams
Scatter diagrams are helpful when forecasting time-series data because they depict the relationship between variables.
450 400 350 300 250 200 150 100 50 0 0 2 4

Radios

Annual Sales

Televisions
s pact Disc Com
6 Time (Years) 8 10 12

2009 Prentice-Hall, Inc.

5 15

Scatter Diagrams
Wacker Distributors wants to forecast sales for three

different products
250 250 250 250 250 250 250 250 250 250

YEAR 1 2 3 4 5 6 7 8 9 10
Table 5.1

TELEVISION SETS

RADIOS 300 310 320 330 340 350 360 370 380 390

COMPACT DISC PLAYERS 110 100 120 140 170 150 160 190 200 190
2009 Prentice-Hall, Inc. 5 16

Scatter Diagrams
(a)
Annual Sales of Televisions 330 250 200 150 100 50
| | | | | | | | | |

Sales appear to be

constant over time Sales = 250 A good estimate of sales in year 11 is 250 televisions

0 1 2 3 4 5 6 7 8 9 10

Time (Years)
Figure 5.2
2009 Prentice-Hall, Inc. 5 17

Scatter Diagrams
(b)
Annual Sales of Radios 420 400 380 360 340 320 300 280
| | | | | | | | | |

Sales appear to be

0 1 2 3 4 5 6 7 8 9 10

increasing at a constant rate of 10 radios per year Sales = 290 + 10(Year) A reasonable estimate of sales in year 11 is 400 televisions

Time (Years)
Figure 5.2
2009 Prentice-Hall, Inc. 5 18

Scatter Diagrams
This trend line may
(c)
Annual Sales of CD Players 200 180 160 140 120 100
|


| | | | | | |

not be perfectly accurate because of variation from year to year Sales appear to be increasing A forecast would probably be a larger figure each year

0 1 2 3 4 5 6 7 8 9 10

Time (Years)
Figure 5.2
2009 Prentice-Hall, Inc. 5 19

Measures of Forecast Accuracy


We compare forecasted values with actual values

to see how well one model works or to compare models Forecast error = Actual value Forecast value

One measure of accuracy is the mean absolute

deviation (MAD) MAD

forecast error MAD =


n

2009 Prentice-Hall, Inc.

5 20

Measures of Forecast Accuracy


Using a nave forecasting model
YEAR 1 2 3 4 5 6 7 8 9 10 11 ACTUAL SALES OF CD PLAYERS 110 100 120 140 170 150 160 190 200 190 FORECAST SALES 110 100 120 140 170 150 160 190 200 190 ABSOLUTE VALUE OF ERRORS (DEVIATION), (ACTUAL FORECAST) |100 110| = 10 |120 110| = 20 |140 120| = 20 |170 140| = 30 |150 170| = 20 |160 150| = 10 |190 160| = 30 |200 190| = 10 |190 200| = 10 Sum of |errors| = 160 MAD = 160/9 = 17.8

Table 5.2

2009 Prentice-Hall, Inc.

5 21

Measures of Forecast Accuracy


Using a nave forecasting model
YEAR 1 2 3 4 5 6 7 8 9 10 11 ACTUAL SALES OF CD PLAYERS 110 100 120 140 170 150 160 190 200 190 FORECAST SALES 110 100 120 ABSOLUTE VALUE OF ERRORS (DEVIATION), (ACTUAL FORECAST) |100 110| = 10 |120 110| = 20 |140 120| = 20 |170 140| = 30 |150 170| = 20 |160 150| = 10 |190 160| = 30 |200 190| = 10 |190 200| = 10 Sum of |errors| = 160 MAD = 160/9 = 17.8

forecast error MAD =


n

160 140 = = 17.8 170 9


150 160 190 200 190

Table 5.2

2009 Prentice-Hall, Inc.

5 22

Measures of Forecast Accuracy


There are other popular measures of forecast

accuracy The mean squared error (error )2 MSE = n


The mean absolute percent error

error actual MAPE = 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 23

Measures of Forecast Accuracy


Year 1 2 3 4 5 6 7 8 9 10 11 Actual CD Sales 110 100 120 140 170 150 160 190 200 190 110 100 120 140 170 150 160 190 200 190 10 20 20 30 20 10 30 10 10 Forecast Sales |Actual -Forecast|

Sum of |errors| MAD

160 17.8
2009 Prentice-Hall, Inc. 5 24

Hospital Days Forecast Error Example


Ms. Smith forecasted total hospital inpatient days last year. Now that the actual data are known, she is reevaluating her forecasting model. Compute the MAD, MSE, and MAPE for her forecast.
Month JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Forecast Actual 250 320 275 260 250 275 300 325 320 350 365 380 243 315 286 256 241 298 292 333 326 378 382 396
5 25

2009 Prentice-Hall, Inc.

Hospital Days Forecast Error Example


Forecast JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC AVERAGE 250 320 275 260 250 275 300 325 320 350 365 380 Actual 243 315 286 256 241 298 292 333 326 378 382 396 MAD= 11.83 |error| 7 5 11 4 9 23 8 8 6 28 17 16 MSE= 192.83 error2 49 25 121 16 81 529 64 64 36 784 289 256 MAPE= .0381*100 = 3.81
2009 Prentice-Hall, Inc. 5 26

|error/actual| 0.03 0.02 0.04 0.02 0.04 0.08 0.03 0.02 0.02 0.07 0.04 0.04

Time-Series Forecasting Models


A time series is a sequence of evenly

spaced events (weekly, monthly, quarterly, etc.) Time-series forecasts predict the future based solely of the past values of the variable Other variables, no matter how potentially valuable, are ignored

2009 Prentice-Hall, Inc.

5 27

Decomposition of a Time-Series
A time series typically has four components 1. Trend (T) is the gradual upward or 2.

3. 4.

downward movement of the data over time Seasonality (S) is a pattern of demand fluctuations above or below trend line that repeats at regular intervals Cycles (C) are patterns in annual data that occur every several years Random variations (R) are blips in the data caused by chance and unusual situations

2009 Prentice-Hall, Inc.

5 28

Decomposition of a Time-Series
Demand for Product or Service Trend Component Seasonal Peaks Actual Demand Line Average Demand over 4 Years

Year 1 Figure 5.3

Year 2

Time

Year 3

Year 4

2009 Prentice-Hall, Inc.

5 29

Decomposition of a Time-Series
There are two general forms of time-series

models The multiplicative model

Demand = T x S x C x 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

2009 Prentice-Hall, Inc.

5 30

Moving Averages
Moving averages can be used when demand is

relatively 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 dropped
This methods tends to smooth out short-term

irregularities in the data series

Moving average forecast =

Sum of demands in previous n periods n


2009 Prentice-Hall, Inc. 5 31

Moving Averages
Mathematically

Ft +1 = where

Yt + Yt 1 + ... + Yt n+1 n

Ft +1 = forecast for time period t + 1 Y = actualt value in time period t n = number of periods to average

2009 Prentice-Hall, Inc.

5 32

Wallace Garden Supply Example


Wallace Garden Supply wants to forecast

demand for its Storage Shed They have collected data for the past year They are using a three-month moving average to forecast demand (n = 3)

2009 Prentice-Hall, Inc.

5 33

Wallace Garden Supply Example


MONTH January February March April May June July August September October Table 5.3 November ACTUAL SHED SALES 10 12 13 16 19 23 26 30 28 18 16
2009 Prentice-Hall, Inc. 5 34

THREE-MONTH MOVING AVERAGE

(10 + 12 + 13)/3 = 11.67 (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

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 +1

( Weight in period i )( Actual value in period ) = ( Weights )


w1Yt + w2Yt 1 + ... + w nYt n+1 Ft +1 = w1 + w2 + ... + w n

Mathematically

here wi = weight for the ith observation


2009 Prentice-Hall, Inc. 5 35

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 36

Wallace Garden Supply Example


Wallace Garden Supply decides to try a weighted

moving average model to forecast demand for its Storage Shed They decide on the following weighting scheme
WEIGHTS APPLIED 3 2 PERIOD Last month

Two months ago

1 Three months ago 3 x Sales last month + 2 x Sales two months ago + 1 X Sales three months ago Sum of the weights
2009 Prentice-Hall, Inc. 5 37

Wallace Garden Supply Example


MONTH January February March April May June July August September Table 5.4 October ACTUAL SHED SALES 10 12 13 16 19 23 26 30 28 18
2009 Prentice-Hall, Inc. 5 38

THREE-MONTH WEIGHTED MOVING AVERAGE

[(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

Wallace Garden Supply Example

Program 5.1A
2009 Prentice-Hall, Inc. 5 39

Wallace Garden Supply Example

Program 5.1B
2009 Prentice-Hall, Inc. 5 40

Exponential Smoothing
Exponential smoothing is easy to use and

requires little record keeping of data It is a type of moving average

recast = Last periods forecast + (Last periods actual demand Last periods forecast) Where is a weight (or smoothing constant) with constant a value between 0 and 1 inclusive A larger gives more importance to recent data while a smaller value gives more importance to past data 2009 Prentice-Hall, Inc.

5 41

Exponential Smoothing
Mathematically

Ft +1 = Ft + (Yt Ft )
Ft+1 = new forecast (for time period t + 1) Ft = pervious forecast (for time period t) = smoothing constant (0 1) Yt = pervious periods actual demand

ere

The idea is simple the new estimate is the old

estimate plus some fraction of the error in the last period


2009 Prentice-Hall, Inc. 5 42

Exponential Smoothing Example


In January, Februarys demand for a certain

car 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, the

April forecast would be

New forecast (for April demand)

= 144.2 + 0.2(136 144.2) = 142.6 or 143 autos


2009 Prentice-Hall, Inc. 5 43

Selecting the Smoothing Constant


Selecting the appropriate value for

obtaining a good forecast The objective is always to generate an accurate forecast The general approach is to develop trial forecasts with different values of and select the that results in the lowest MAD

is key to

2009 Prentice-Hall, Inc.

5 44

Port of Baltimore Example


Exponential smoothing forecast for two values of
QUARTER ACTUAL TONNAGE UNLOADED 180 168 159 175 190 205 180 182 ? 175 175.5 = 175.00 + 0.10(180 175) 174.75 = 175.50 + 0.10(168 175.50) 173.18 = 174.75 + 0.10(159 174.75) 173.36 = 173.18 + 0.10(175 173.18) 175.02 = 173.36 + 0.10(190 173.36) 178.02 = 175.02 + 0.10(205 175.02) 178.22 = 178.02 + 0.10(180 178.02) 178.60 = 178.22 + 0.10(182 178.22) FORECAST USING =0.10 FORECAST USING =0.50 175 177.5 172.75 165.88 170.44 180.22 192.61 186.30 184.15

1 2 3 4 5 6 7 8 9 Table 5.5

2009 Prentice-Hall, Inc.

5 45

Selecting the Best Value of


QUARTER ACTUAL TONNAGE UNLOADED 180 168 159 175 190 205 180 182 FORECAST WITH = 0.10 175 175.5 174.75 173.18 173.36 175.02 178.02 ABSOLUTE DEVIATIONS FOR = 0.10 5.. FORECAST WITH = 0.50 175 177.5 172.75 165.88 170.44 180.22 192.61 186.30 ABSOLUTE DEVIATIONS FOR = 0.50 5.

1 2 3 4 5 6 7 8 Table 5.6

7.5.. 15.75 1.82 16.64 29.98 1.98 3.78 82.45 10.31

9.5.. 13.75 9.12 19.56 24.78 12.61 4.3.. 98.63


MAD = 2009 Prentice-Hall, Inc. 12.33 5 46

Sum of absolute deviations

Best choice
|deviations| =

178.22

MAD =

Port of Baltimore Example

Program 5.2A
2009 Prentice-Hall, Inc. 5 47

Port of Baltimore Example

Program 5.2B
2009 Prentice-Hall, Inc. 5 48

PM Computer: Moving Average Example


PM Computer assembles customized personal

computers from generic parts

The owners purchase generic computer parts in volume at a discount from a variety of sources whenever they see a good deal.
It is important that they develop a good forecast of

demand for their computers so they can purchase component parts efficiently.

2009 Prentice-Hall, Inc.

5 49

PM Computers: Data
Period Month Actual Demand 1 Jan 37 2 Feb 40 3 Mar 41 4 Apr 37 5 May 45 6 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? 2009 Prentice-Hall, Inc. 5 50

PM Computers: Moving Average Solution


2mn o th M A A s Dv b. e 3mn W A o th M A s Dv b. e EpS . x. m 3. 0 70 3. 0 70 3 .5 8 0 4 .5 0 0 3 .0 9 0 4 .0 1 0 4 .5 7 0 4 .5 6 0 4 .0 5 0 5 .5 1 0 2 0 .5 3 0 .5 6 0 .0 9 0 .0 4 0 .5 0 0 .5 1 .0 1 0 5 9 .2 4 .1 0 4 3 .5 8 7 4 .1 2 4 4 .7 6 1 4 .2 5 9 4 .2 6 9 5 .5 1 7 5 3 .4 3 4 .1 6 3 .4 7 6 .8 3 1 .7 1 1 .7 9 1 .7 3. 0 91 4. 3 04 3. 3 80 4. 1 29 4. 7 78 4. 6 44 4. 4 62 5. 7 30 49 .5 30 .0 19 .0 34 .3 69 .7 70 .9 48 .7 25 .4 97 .6 A s Dv b. e

Exponential smoothing resulted in the lowest MAD.


2009 Prentice-Hall, Inc. 5 51

MAD

Exponential Smoothing with Trend Adjustment


Like all averaging techniques, exponential

smoothing does not respond to trends A more complex model can be used that adjusts for trends The basic approach is to develop an exponential smoothing forecast then adjust it for the trend New forecast (Ft) + Trend correction (Tt)

2009 Prentice-Hall, Inc.

5 52

Exponential Smoothing with Trend Adjustment


The equation for the trend correction uses a

new smoothing constant Tt is computed by

Tt +1 = (1 )Tt + ( Ft +1 Ft )
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
2009 Prentice-Hall, Inc. 5 53

Selecting a Smoothing Constant


As with exponential smoothing, a high value of

makes the forecast more responsive to changes in trend A low value of gives 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 different values of Simple exponential smoothing is often referred to as first-order smoothing Trend-adjusted smoothing is called second-order, second-order double smoothing, or Holts method smoothing
2009 Prentice-Hall, Inc. 5 54

Trend Projection
Trend projection fits a trend line to a series

of 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

2009 Prentice-Hall, Inc.

5 55

Trend Projection
Trend projections are used to forecast time-series

data that exhibit a linear trend.

A trend line is simply a linear regression equation in

which the independent variable (X) is the time period Least squares may be used to determine a trend projection for future forecasts.
Least squares determines the trend line forecast by

minimizing the mean squared error between the trend line forecasts and the actual observed values.

The independent variable is the time period and the

dependent variable is the actual observed value in the time series.


2009 Prentice-Hall, Inc.

5 56

Trend Projection
The mathematical form is

Y = b0 + b1 X
where b0 b1 X = predicted value Y = intercept = slope of the line = time period (i.e., X = 1, 2, 3, , n)

2009 Prentice-Hall, Inc.

5 57

Trend Projection
Dist7

Value of Dependent Variable

*
Dist1

Dist5 Dist3

Dist6

Dist2

Dist4

Time

Figure 5.4
2009 Prentice-Hall, Inc. 5 58

Midwestern Manufacturing Company Example


Midwestern Manufacturing Company has experienced

the following demand for its electrical generators over the period of 2001 2007
YEAR 2001 2002 2003 2004 2005 2006 2007 ELECTRICAL GENERATORS SOLD 74 79 80 90 105 142 122
Table 5.7

2009 Prentice-Hall, Inc.

5 59

Midwestern Manufacturing Company Example

Notice code instead of actual years

Program 5.3A
2009 Prentice-Hall, Inc. 5 60

Midwestern Manufacturing Company Example

r2 says model predicts about 80% of the variability in demand Significance level for Ftest indicates a definite relationship

Program 5.3B
2009 Prentice-Hall, Inc. 5 61

Midwestern Manufacturing Company Example


The forecast equation is

Y = 56.71 + 10.54 X
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 for X = 9

(sales in 2009) = 56.71 + 10.54(9) = 151.57, or 152 generators


2009 Prentice-Hall, Inc. 5 62

Midwestern Manufacturing Company Example


160 150 140 Generator Demand 130 120 110 100 90 80 70 60 50
| | | | | | | | |


Actual Demand Line Trend Line Y = 56.71 + 10.54 X

Figure 5.5

2001 2002 2003 2004 2005 2006 2007 2008 2009 Year
2009 Prentice-Hall, Inc. 5 63

Midwestern Manufacturing Company Example

Program 5.4A
2009 Prentice-Hall, Inc. 5 64

Midwestern Manufacturing Company Example

Program 5.4B
2009 Prentice-Hall, Inc. 5 65

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

2009 Prentice-Hall, Inc.

5 66

Seasonal Variations
Eichler Supplies sells telephone

answering machines Data has been collected for the past two years sales of one particular model They want to create a forecast that includes seasonality

2009 Prentice-Hall, Inc.

5 67

Seasonal Variations
MONTH SALES DEMAND YEAR 1 YEAR 2 AVERAGE TWOYEAR DEMAND MONTHLY DEMAND AVERAGE SEASONAL INDEX 0.957 0.851 0.904 1.064 1.309 1.223 1.117

January February March April May June July

80 85 80 110 115 120 100

100 75 90 90 131 110 110 90 1,128 = 94 12 months 95

90 80 85 100 123 115 105 100 Seasonal index = 90

94 94 94 94 94 94 94

August 110 Average monthly demand = September 85

Table 5.8

94 1.064 Average two-year demand Average monthly demand 94 0.957


2009 Prentice-Hall, Inc. 5 68

Seasonal Variations
The calculations for the seasonal indices are
Jan. Feb. Mar. Apr. May June
1,200 0.957 = 96 12 1,200 0.851 = 85 12 1,200 0.904 = 90 12 1,200 1.064 = 106 12 1,200 1.309 = 131 12 1,200 1.223 = 122 12

July Aug. Sept. Oct. Nov. Dec.

1,200 1.117 = 112 12 1,200 1.064 = 106 12 1,200 0.957 = 96 12 1,200 0.851 = 85 12 1,200 0.851 = 85 12 1,200 0.851 = 85 12
2009 Prentice-Hall, Inc. 5 69

Regression with Trend and Seasonal Components


Multiple regression can be used to forecast both

trend 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

Y = a + b1 X 1 + b2 X 2 + b3 X 3 + b4 X 4
where X1 X2 X3 X4 = time period = 1 if quarter 2, 0 otherwise = 1 if quarter 3, 0 otherwise = 1 if quarter 4, 0 otherwise
2009 Prentice-Hall, Inc. 5 70

Regression with Trend and Seasonal Components

Program 5.6A
2009 Prentice-Hall, Inc. 5 71

Regression with Trend and Seasonal Components

Program 5.6B (partial)


2009 Prentice-Hall, Inc. 5 72

Regression with Trend and Seasonal Components


The resulting regression equation is

Y = 104 .1 + 2.3 X 1 + 15.7 X 2 + 38.7 X 3 + 30.1X 4


Using the model to forecast sales for the first two

quarters of next year

Y = 104 .1 + 2.3(13 ) + 15.7(0) + 38.7(0) + 30.1(0) = 134 Y = 104 .1 + 2.3(14 ) + 15.7(1) + 38.7(0) + 30.1(0) = 152
These are different from the results obtained using the

multiplicative decomposition method Use MAD and MSE to determine the best model

2009 Prentice-Hall, Inc.

5 73

Regression with Trend and Seasonal Components


American Airlines original spare parts inventory

system used only time-series methods to forecast the demand for spare parts
This method was slow to responds to even moderate

changes in aircraft utilization let alone major fleet expansions

They developed a PC-based system named RAPS

which uses linear regression to establish a relationship between monthly part removals and various functions of monthly flying hours
The computation now takes only one hour instead of

the days the old system needed Using RAPS provided a one time savings of $7 million and a recurring annual savings of nearly $1 million
2009 Prentice-Hall, Inc. 5 74

Monitoring and Controlling Forecasts


Tracking signals can be used to monitor the

performance of a forecast Tacking signals are computed using the following equation
RSFE Tracking signal = MAD where

forecast error MAD =


n
2009 Prentice-Hall, Inc. 5 75

Monitoring and Controlling Forecasts


Signal Tripped + 0 MADs Upper Control Limit Tracking Signal Acceptable Range

Lower Control Limit

Time
Figure 5.7
2009 Prentice-Hall, Inc. 5 76

Monitoring and Controlling Forecasts


Positive tracking signals indicate demand is greater

than forecast Negative tracking signals indicate demand is less than forecast Some variation is expected, but a good forecast will have about as much positive error as negative error Problems are indicated when the signal trips either the upper or lower predetermined limits This indicates there has been an unacceptable amount of variation Limits should be reasonable and may vary from item to item
2009 Prentice-Hall, Inc. 5 77

Regression with Trend and Seasonal 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 lower volume 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 errors

are expected to fall within 2 MADs, 98% with 3 MADs or 99.9% within 4 MADs whenever the errors are approximately normally distributed
2009 Prentice-Hall, Inc.

5 78

Kimballs Bakery Example


Tracking signal for quarterly sales of croissants
TIME PERIOD FORECAST DEMAND ACTUAL DEMAND ERROR RSFE |FORECAST | | ERROR | CUMULATIVE ERROR MAD TRACKING SIGNAL

1 2 3 4 5 6

100 100 100 110 110 110

90 95 115 100 125 140

10 5 +15 10 +15 +30

10 15 0 10 +5 +35

10 5 15 10 15 30

10 15 30 40 55 85

10.0 7.5 10.0 10.0 11.0 14.2

1 2 0 1 +0.5 +2.5

85 = = 14.2 n 6 RSFE 35 Tracking signal = = = 2.5MAD s MAD 14.2


2009 Prentice-Hall, Inc. 5 79

forecast error MAD =

Forecasting at Disney
The Disney chairman receives a daily

report from his main theme parks that contains only two numbers the forecast of yesterdays 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

2009 Prentice-Hall, Inc.

5 80

Using The Computer to Forecast


Spreadsheets can be used by small and

medium-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
2009 Prentice-Hall, Inc. 5 81