Forecasting

Session 3
Chapter 3
(W. Stevenson)
FORECAST
 A statement about the future value of a variable of
interest such as demand (Prediction about the
future)
 Forecasting is used to make informed decisions.
 Long-range( Planning for several years or more)
 Short-range(Planning and scheduling day to day
activities)
 Forecasts affect decisions and activities throughout
an organization
 2 uses of forecast-
 1. Plan the system and
 2. Plan the use of the system.
 Features of Forecasts
 Forecasting techniques generally assume that the
same underlying casual system that existed in the
past will continue to exist in the future.
 Forecasts are rarely perfect because of
randomness. Allowances should be made for
forecast errors.
 Forecast for groups of items tend to be more
accurate than forecasts for individual items.
Forecasting errors among items in a group usually
have a cancelling effect.
 Forecastaccuracy decreases as the time period
covered by the forecast-the time horizon increases.
Elements of a good forecast

Timely Accurate Reliable

Simple to
Meaningful
In writing understand
Units
and use

Cost-
effective
Steps in the Forecasting Process

“The forecast”

Step 6 “Monitor the forecast”

Step 5 “Make the forecast”
Step 4 “Obtain, clean and analyze data”
Step 3 “Select a forecasting technique”
Step 2 “Establish a time horizon”
Step 1 “Determine purpose of forecast”
Forecasting Process
1. Identify the purpose 2. Collect historical data 3. Plot data and identify
of forecast patterns

6. Check forecast 5. Develop/compute 4. Select a forecast

accuracy with one or forecast for period of model that seems
more measures historical data appropriate for data

7.
Is accuracy of No 8b. Select new forecast
forecast
model or adjust
acceptable?
parameters of existing
model

Yes

9. Adjust forecast based on 10. Monitor results and

8a. Forecast over planning
additional qualitative measure forecast
horizon
information and insight accuracy
 Forecast Accuracy
 Error= Actual Demand-Forecasted Demand (Et=At-Ft)
 Positive errors result when the forecast is too low
 Negative errors result when the forecast is too high
 Forecast errors influence decisions in 2 different
ways-
 Making a choice between forecasting alternatives
 Evaluating the success or failure of a technique in use
 Mean Forecast Error (MFE)
 MFE > 0, model tends to under-forecast
MFE < 0, model tends to over-forecast
Forecast Accuracy
 Mean Absolute Deviation (MAD)
 Average absolute error
 Weights all errors linearly/evenly
 While MFE is a measure of forecast model bias, MAD indicates
the absolute size of the errors
 Mean Squared Error (MSE)
 Average of squared error
 Gives more weight to larger errors, which typically cause
more problems
 Mean Absolute Percent Error (MAPE)
 Average absolute percent error
 MAPE should be used when there is a need to put errors in
perspective
MAD, MSE, and MAPE

 Actual  forecast
MAD =
n

 ( Actual  forecast)
2
MSE =
n -1

 ( Actual  forecast / Actual*100)

MAPE =
n
Forecast Accuracy (Example 1)
Types of Forecasts
1. Quantitative Forecasting Approaches
2. Qualitative Forecasting Approaches

 Judgmental - uses subjective inputs

 Time series - uses historical data
assuming the future will be like the past
 Associative models - uses explanatory
variables to predict the future
Judgmental Forecasts

 Executive opinions
 Sales force opinions
 Consumer surveys
 Outside opinion
 Delphi method
 Opinions of managers and staff
 Achieves a consensus forecast
Time Series Forecasts

 Trend - long-term upward of

downward movement in data
 Seasonality - short-term regular
variations in data
 Cycle – wavelike variations of more
than one year’s duration
 Irregular variations - caused by
unusual circumstances
 Random variations – residual
variation, caused by chance
Forecast Variations (Figure 3.1)
Irregular
variation

Trend

Cycles

Seasonal variations
Naive Forecasts

Uh, give me a minute....

We sold 250 wheels last
week.... Now, next week we
should sell....
Naïve Forecasts

 Simple to use
 Virtually no cost as data analysis is non-existent
 Quick and easy to prepare
 Easily understandable
 Cannot provide high accuracy
 Can be a standard for accuracy

IS THE INCREASED ACCURACY OF ANOTHER METHOD

WORTH THE ADDITIONAL RESOURCES REQUIRED TO
ACHIEVE THAT ACCURACY?
Techniques for Averaging
 Moving average: A technique that averages a
number of recent actual values, updated as
new values become available
 Weighted moving average: More recent values
in a series are given more weight in
computing the forecast.
 Exponential smoothing: a weighted averaging
method based on previous forecast plus a
percentage of the forecast error.
Simple Moving Average

At-n + … At-2 + At-1

Ft = MAn=
n
Moving Average (Example 2)
Calculate a three-period moving average forecast for demand in period 6

If the actual demand in period 6 is 38, then the moving average forecast for period
7 is-
Weighted Moving Average (Example 3)

Period Demand Weight

1 42
2 40 10%
3 43 20%
4 40 30%
5 41 40%
 Exponential Smoothing

Ft = Ft-1 + (At-1 - Ft-1)

• Weighted averaging method based on previous
forecast plus a percentage of the forecast error
• The most recent observations might have the
highest predictive value.
 Therefore, we should give more weight to the more
recent time periods when forecasting.
 Uses most recent period’s actual and forecast data
• A-F is the forecast error,  is the % feedback
Exponential Smoothing Example

F3  F2   ( A2  F2 )
  0.10, F3  42  0.10( 40  42)  42  0.10  ( 2)  42  0.2  41.8

F4  F3   ( A3  F3 )
  0.10, F4  41.8  0.10( 43  41.8)  41.8  0.10  1.2  41.8  0.12  41.92
Picking a Smoothing Constant
• Selecting a smoothing constant is a matter of
Judgment or trial and error.
• Commonly used values of  range from .05 to .50
• Low values of are used when the underlying average
tends to be stable.
• Higher values of  are used when the underlying
average is susceptible to change.
• Closer the  is to 0, the greater the smoothing
(Figure 3.4B)
Techniques for trend
• Analysis of trend involves developing an equation
that will describe trend.
• The trend component may be linear, or it may not
be.
• There are 2 important techniques that can be used
to develop forecasts when trend is present
 Use of trend equation
 An extension of exponential smoothing.
Linear Trend Equation
Ft

Ft = a + bt

0 1 2 3 4 5 t

 Ft = Forecast for period t

 t = Specified number of time periods
 a = Value of Ft at t = 0
 b = Slope of the line
Calculating a and b

n  (ty) -  t y
b =
2
n  t - (  t) 2

 y - b t
a =
n
Linear Trend Equation Example
t y
2
W eek t Sales ty
1 1 150 150
2 4 157 314
3 9 162 486
4 16 166 664
5 25 177 885

 t = 15 t  y = 812  ty = 2499

2
= 55
2
(t) = 225

3-27
 Linear Trend Calculation
5 (2499) - 15(812) 12495 -12180
b = = = 6.3
5(55) - 225 275 -225

812 - 6.3(15)
a = = 143.5
5

y = 143.5 + 6.3t
Associative Forecasting

 Predictor variables- used to predict values of

variable interest
 Regression- technique for fitting a line to a
set of points
 Least squares line- minimizes sum of squared
deviations around the line
Linear Model Seems Reasonable (Example 9)
X Y Computed
7 15
relationship
2 10
6 13
4 15
14 25
15 27
16 24
12 20
14 27
20 44
15 34 A straight line is fitted to a set of
7 17 sample points.
Linear Regression Assumptions
 Variations around the line are random
 Deviations around the line normally distributed
 Predictions are being made only within the range
of observed values
 For best results:
 Always plot the data to verify linearity
 Check for data being time-dependent
 Small correlation may imply that other variables
are important
Regression Methods
 Linear regression
 mathematical technique that relates a
dependent variable to an independent
variable in the form of a linear equation
 Primary method for Associative Forecasting

 Correlation
 a measure of the strength of the
relationship between independent and
dependent variables

12-32
Correlation and Coefficient of Determination

 Correlation, r
 Measure of strength of relationship
between the dependent variable
(demand) and the independent variable
 Varies between -1.00 and +1.00
 Coefficient of determination, r2
 Percentage of variation in dependent
variable resulting from changes in the
independent variable

12-33
Linear Regression Example
x y
(WINS) (ATTENDANCE) xy x2
4 36.3 145.2 16
6 40.1 240.6 36
6 41.2 247.2 36
8 53.0 424.0 64
6 44.0 264.0 36
7 45.6 319.2 49
5 39.0 195.0 25
7 47.5 332.5 49
49 346.7 2167.7 311

12-34
 Computing Correlation

n xy -  x y
r=
[n x2 - ( x)2] [n y2 - ( y)2]

(8)(2,167.7) - (49)(346.9)
r=
[(8)(311) - (49)2] [(8)(15,224.7) - (346.9)2]

r = 0.947

Coefficient of determination
r2 = (0.947)2 = 0.897

12-35
Choosing a Forecasting Technique
 No single technique works in every situation
 Two most important factors
 Cost
 Accuracy
 Other factors include the availability of:
 Historical data
 Computers
 Time needed to gather and analyze the data
 Forecast horizon

Sources of Forecast errors

 Model may be inadequate
 Irregular variations
 Incorrect use of forecasting technique
Operations Strategy
 Sharing forecasts with supply can
 Improve forecast quality in the supply chain
 Lower costs
 Shorter lead times
 Forecasts are the basis for many decisions
 Work to improve short-term forecasts
 Accurate short-term forecasts improve
 Profits
 Lower inventory levels
 Reduce inventory shortages
 Improve customer service levels
 Enhance forecasting credibility