Chapter 8 - Forecasting

Operations Management
by
R. Dan Reid & Nada R. Sanders
4th Edition © Wiley 2010

© Wiley 2010 1
Principles of Forecasting
Many types of forecasting models that
differ in complexity and amount of
data & way they generate forecasts:
1. Forecasts are rarely perfect
2. Forecasts are more accurate for
grouped data than for individual items
3. Forecast are more accurate for shorter
than longer time periods

© Wiley 2010 2
Types of Forecasting Methods
 Decide what needs to be forecast
 Level of detail, units of analysis & time horizon
required
 Evaluate and analyze appropriate data
 Identify needed data & whether it’s available
 Select and test the forecasting model
 Cost, ease of use & accuracy
 Generate the forecast
 Monitor forecast accuracy over time

© Wiley 2010 3
Types of Forecasting Methods
 Forecasting methods are classified into
two groups:

© Wiley 2010 4
 Types of Forecasting Models
 Qualitative methods – judgmental methods
 Forecasts generated subjectively by the

forecaster
 Educated guesses

 Quantitative methods – based on

mathematical modeling:
 Forecasts generated through mathematical

modeling

© Wiley 2010 5
 Qualitative Methods
Type Characteristics Strengths Weaknesses
Executive A group of managers Good for strategic or One person's opinion
opinion meet & come up with new-product can dominate the
a forecast forecasting forecast

Market Uses surveys & Good determinant of It can be difficult to

research interviews to identify customer preferences develop a good
customer preferences questionnaire

Delphi Seeks to develop a Excellent for Time consuming to

method consensus among a forecasting long-term develop
group of experts product demand,
technological
changes, and
© Wiley 2010 6
Quantitative Methods
 Time Series Models:
 Assumes information needed to generate a
forecast is contained in a time series of data
 Assumes the future will follow same patterns as
the past
 Causal Models or Associative Models
 Explores cause-and-effect relationships
 Uses leading indicators to predict the future
 Housing starts and appliance sales

© Wiley 2010 7
 Time Series Models
 Forecaster looks for data patterns as
 Data = historic pattern + random variation
 Historic pattern to be forecasted:
 Level (long-term average) – data fluctuates around a constant
mean
 Trend – data exhibits an increasing or decreasing pattern
 Seasonality – any pattern that regularly repeats itself and is of a
constant length
 Cycle – patterns created by economic fluctuations
 Random Variation cannot be predicted
© Wiley 2010 8
Time Series Patterns

© Wiley 2010 9
Time Series Models
 Naive: Ft 1  At
 The forecast is equal to the actual value observed during
the last period – good for level patterns
 Simple Mean: Ft 1   A t / n
 The average of all available data - good for level
patterns
Ft 1   A t / n
 Moving Average:
 The average value over a set time period
(e.g.: the last four weeks)
 Each new forecast drops the oldest data point & adds a
new observation
 More responsive to a trend but still lags behind actual
data

© Wiley 2010 10
Time Series Models con’t
 Weighted Moving Average: Ft 1   C t A t

 All weights must add to 100% or 1.00

e.g. Ct .5, Ct-1 .3, Ct-2 .2 (weights add to 1.0)

 Allows emphasizing one period over others; above

indicates more weight on recent data (C t=.5)

 Differs from the simple moving average that weighs all

periods equally - more responsive to trends

© Wiley 2010 11
Time Series Models con’t
 Exponential Smoothing: F  αA  1  α F
t 1 t t
Most frequently used time series method because of
ease of use and minimal amount of data needed

 Need just three pieces of data to start:

 Last period’s forecast (Ft)

 Last periods actual value (At)
 Select value of smoothing coefficient,,between 0 and 1.0
 If no last period forecast is available, average the
last few periods or use naive method
 
Higher values (e.g. .7 or .8) may place too much
weight on last period’s random variation
© Wiley 2010 12
Time Series Problem
 Determine forecast for Period Actual
periods 7 & 8 1 300
 2-period moving average 2 315
 4-period moving average
3 290
 2-period weighted moving
4 345
average with t-1 weighted 0.6
and t-2 weighted 0.4 5 320
 Exponential smoothing with 6 360
alpha=0.2 and the period 6 7 375
forecast being 375 8
© Wiley 2010 13
Time Series Problem Solution
Period Actual 2-Period 4-Period 2-Per.Wgted. Expon. Smooth.

1 300        

2 315        

3 290        

4 345        

5 320        

6 360        

7 375 340.0 328.8 344.0 372.0

8   367.5 350.0 369.0 372.6

© Wiley 2010 14
 Forecasting trend problem: a company uses exponential smoothing with
trend to forecast usage of its lawn care products. At the end of July the
company wishes to forecast sales for August. July demand was 62. The
trend through June has been 15 additional gallons of product sold per
month. Average sales have been 57 gallons per month. The company uses
alpha+0.2 and beta +0.10. Forecast for August.

 Smooth the level of the series:

S July  αA t  (1  α)(S t 1  Tt 1 )  0.262  0.857  15  70

 Smooth the trend:

TJuly  β(St  St 1 )  (1  β)Tt 1  0.170  57   0.915  14.8

 Forecast including trend:

FITAugust  S t  Tt  70  14.8  84.8 gallons

© Wiley 2010 15
 Linear Trend Line
A time series technique that computes a
forecast with trend by drawing a straight line
through a set of data using this formula:
Y = a + bx where
Y = forecast for period X
X = the number of time periods from X = 0
A = value of y at X = 0 (Y intercept)
B = slope of the line

© Wiley 2010 16
Forecasting Trend
 Basic forecasting models for trends compensate for the lagging
that would otherwise occur
 One model, trend-adjusted exponential smoothing uses a
three step process
 Step 1 - Smoothing the level of the series

S t  αA t  (1  α)(S t 1  Tt 1 )
 Step 2 – Smoothing the trend
Tt  β(S t  S t 1 )  (1  β)Tt 1
 Forecast including the trend
FITt 1  S t  Tt
© Wiley 2010 17
Forecasting Seasonality
 Calculate the average demand per season
 E.g.: average quarterly demand
 Calculate a seasonal index for each season of
each year:
 Divide the actual demand of each season by the
average demand per season for that year
 Average the indexes by season
 E.g.: take the average of all Spring indexes, then
of all Summer indexes, ...

© Wiley 2010 18
Seasonality con’t
 Forecast demand for the next year & divide
by the number of seasons
 Use regular forecasting method & divide by four
for average quarterly demand
 Multiply next year’s average seasonal demand
by each average seasonal index
 Result is a forecast of demand for each season of
next year

© Wiley 2010 19
 Seasonality problem: a university must develop forecasts for the
next year’s quarterly enrollments. It has collected quarterly
enrollments for the past two years. It has also forecast total
enrollment for next year to be 90,000 students. What is the
forecast for each quarter of next year?

Quarter Year Seasonal Year Seasonal Avg. Year3

1 Index 2 Index Index
Fall 24000 1.2 26000 1.238 1.22 27450

Winter 23000 22000

Spring 19000 19000

Summer 14000 17000

Total 80000 84000 90000

Average 20000 21000 22500

© Wiley 2010 20
Causal Models
 Often, leading indicators can help to predict
changes in future demand e.g. housing starts
 Causal models establish a cause-and-effect
relationship between independent and dependent
variables
 A common tool of causal modeling is linear
regression: Y  a  bx
 Additional related variables may require multiple
regression modeling

© Wiley 2010 21
Linear Regression
 Identify dependent (y) and
independent (x) variables
b
 XY  X  Y   Solve for the slope of the
 X 2  X  X 
line
b
 XY  n XY
2
 X  nX
2

 Solve for the y intercept

a  Y  bX
 Develop your equation for
the trend line
Y=a + bX

© Wiley 2010 22
 Linear Regression Problem: A maker of golf shirts has been
tracking the relationship between sales and advertising dollars. Use
linear regression to find out what sales might be if the company
invested $53,000 in advertising next year.

Sales $ Adv.$ b
 XY  n XY
XY X^2 Y^2 2
(Y) (X)  X  nX 2

1 130 32 4160 2304 16,900

28202  447.25147.25
b  1.15
2 151 52 7852 2704 22,801 9253  447.25
2

3 150 50 7500 2500 22,500 a  Y  b X  147.25  1.1547.25

a  92.9
4 158 55 8690 3025 24964 Y  a  bX  92.9  1.15X
5 153.85 53 Y  92.9  1.1553  153.85

Tot 589 189 28202 9253 87165

Avg 147.25 47.25
© Wiley 2010 23
 Correlation Coefficient
How Good is the Fit?
 Correlation coefficient (r) measures the direction and strength of the linear
relationship between two variables. The closer the r value is to 1.0 the better
the regression line fits the data points.
n XY    X  Y 
r
 X    X   Y   Y 
2 2
2 2
n * n
428,202  189589
r 2
 .982
4(9253) - (189) * 487,165  589
2

r 2  .982  .964
2

 Coefficient of determination ( r 2) measures the amount of variation in the

dependent variable about its mean that is explained by the regression line.
2
Values of ( r ) close to 1.0 are desirable.

© Wiley 2010 24
Multiple Regression
 An extension of linear regression but:
 Multiple regression develops a relationship
between a dependent variable and multiple
independent variables. The general formula
is:

© Wiley 2010 25
Measuring Forecast Error
 Forecasts are never perfect
 Need to know how much we should
rely on our chosen forecasting method
 Measuring forecast error:
E t  A t  Ft
 Note that over-forecasts = negative
errors and under-forecasts = positive
errors

© Wiley 2010 26
 Measuring Forecasting Accuracy
 Mean Absolute Deviation (MAD) MAD   actual  forecast
 measures the total error in a n
forecast without regard to sign
Cumulative Forecast Error (CFE) CFE  actual  forecast 

 Measures any bias in the forecast

 actual - forecast 2

 Mean Square Error (MSE) MSE 

n
 Penalizes larger errors
CFE
 Tracking Signal TS 
MAD
 Measures if your model is working

© Wiley 2010 27
 Accuracy & Tracking Signal Problem: A company is comparing the
accuracy of two forecasting methods. Forecasts using both methods are
shown below along with the actual values for January through May. The
company also uses a tracking signal with ±4 limits to decide when a
forecast should be reviewed. Which forecasting method is best?

Method A Method B
Month Actual F’cast Error Cum. Tracking F’cast Error Cum. Tracking
sales Signal Error Signal
Error

Jan. 30 28 2 2 2 27 2 2 1

Feb. 26 25 1 3 3 25 1 3 1.5
March 32 32 0 3 3 29 3 6 3
April 29 30 -1 2 2 27 2 8 4
May 31 30 1 3 3 29 2 10 5

MAD 1 2
MSE 1.4 4.4
© Wiley 2010 28
Forecasting Software
 Spreadsheets
 Microsoft Excel, Quattro Pro, Lotus 1-2-3
 Limited statistical analysis of forecast data
 Statistical packages
 SPSS, SAS, NCSS, Minitab
 Forecasting plus statistical and graphics
 Specialty forecasting packages
 Forecast Master, Forecast Pro, Autobox, SCA

© Wiley 2010 29
 Forecasting Across the
Organization
 Forecasting is critical to management of all
organizational functional areas
 Marketing relies on forecasting to predict demand
and future sales
 Finance forecasts stock prices, financial performance,
capital investment needs..
 Information systems provides ability to share
databases and information
 Human resources forecasts future hiring
requirements

© Wiley 2010 30