Bozarth ch09

You might also like

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 48

Forecasting

Why Forecast?
Assess long-term capacity needs

Develop budgets, hiring plans, etc.


Plan production or order materials Get agreement within firm and across supply chain partners

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 2

Forecast Characteristics
Almost always wrong by some amount More accurate for groups or families More accurate for shorter time periods No substitute for calculated demand.

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 3

Forecasting Approaches
Qualitative Methods
Used when situation is vague and little data exists
New products New technology

Quantitative Methods
Used when situation is stable and historical data exists
Existing products Current technology

Involves intuition, experience


***************************** E.g., forecasting sales to a new market

Heavy use of mathematical techniques ******************************* E.g., forecasting sales of a mature product
Chapter 9, Slide 4

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Q2 Forecasting
Quantitative, then qualitative factors to filter the answer

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 10, Slide 5

Qualitative Forecasting
Executive opinions

Sales force composite


Consumer surveys

Outside opinions
Delphi method

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 6

Demand Forecasting
Basic time series models

Linear regression
For time series or causal modeling

Measuring forecast accuracy Mini-case: Northcutt Bikes (A)


2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 7

Time Series Models


Period 1 2 3 4 5 6 7 8 Demand 12 15 11 9 10 8 14 12

What assumptions must we make to use this data to forecast?

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 8

Time Series Components of Demand . . .


Demand

. . . randomness

Time
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 9

Time Series with . . .


Demand

. . . randomness and trend

Time
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 10

Time series with . . .


Demand

. . . randomness, trend and seasonality

May

May

May

May
Chapter 9, Slide 11

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Idea Behind Time Series Models


Distinguish between random fluctuations and true changes in underlying demand patterns.
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 10, Slide 12

Moving Average Models


Period 1 2 3 4 5 6 7 8 Demand 12 15 11 9 10 8 14 12

Ft 1

i 1

Dt 1i

n
(14 + 8 + 10) / 3 10.67
Chapter 9, Slide 13

3-period moving average forecast for Period 8:

= =

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Weighted Moving Averages


Ft 1 i 1
Wt 1i Dt 1i
i 1 n

Wt 1i

Forecast for Period 8 = [(0.5 14) + (0.3 8) + (0.2 10)] / (0.5 + 0.3 + 0.1) = 11.4

What are the advantages? What do the weights add up to? Could we use different weights? Compare with a simple 3-period moving average.
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 14

Table of Forecasts and Demand Values . . .


Period
1 2 3 4 5 6 Actual Demand 12 15 11 9 10 8 13.5 13 10 9.5 12.4 10.8 9.9 Two-Period Moving Average Forecast Three-Period Weighted Moving Average Forecast Weights = 0.5, 0.3, 0.2

7
8 9

14
12

9
11 13

8.8
11.4 11.8
Chapter 9, Slide 15

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

. . . and Resulting Graph


20 15 Demand 2-Period Avg 3-Period Wt. Avg.

Volume

10 5 0 1 2 3 4 5 Period 6 7 8 9

Note how the forecasts smooth out variations


2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 16

Exponential Smoothing I
Sophisticated weight averaging model Needs only three numbers: Ft Dt
= Forecast for the current period t = Actual demand for the current period t = Weight between 0 and 1

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 17

Exponential Smoothing II
Formula

Ft+1 = Ft + a (Dt Ft) = a Dt + (1 a) Ft


Where did the current forecast come from? What happens as a gets closer to 0 or 1? Where does the very first forecast come from?
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 18

Exponential Smoothing Forecast with a = 0.3


Period 1 Actual Demand 12 Exponential Smoothing Forecast 11.00

2
3 4 5

15
11 9 10

11.30
12.41 11.99 11.09

F2 = 0.312 + 0.711 = 3.6 + 7.7 = 11.3

6
7 8 9

8
14 12

10.76
9.93 11.15 11.41

F3 = 0.315 + 0.711.3 = 12.41

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 19

Resulting Graph
16 14 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 Period
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Demand

Demand Forecast

Chapter 9, Slide 20

Trends

What do you think will happen to a moving average or exponential smoothing model when there is a trend in the data?
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 10, Slide 21

Same Exponential Smoothing Model as Before:


Period
1 2 3 4 5 6 7

Actual Demand
11 12 13 14 15 16 17

Exponential Smoothing Forecast


11.00 11.00 11.30 11.81 12.47 13.23 14.06

Since the model is based on historical demand, it always lags the obvious upward trend
Chapter 9, Slide 22

8
9

18

14.94
15.86

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Adjusting Exponential Smoothing for Trend


Add trend factor and adjust using exponential smoothing Needs only two more numbers: Tt = Trend factor for the current period t = Weight between 0 and 1 Then: Tt+1 = (Ft+1 Ft) + (1 ) Tt And the Ft+1 adjusted for trend is = Ft+1 + Tt+1

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 23

Simple Linear Regression


Time series OR causal model Assumes a linear relationship:
y

y = a + b(x)

x
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 24

Definitions
Y = a + b(X) Y = predicted variable (i.e., demand)

X = predictor variable
X can be the time period or some other type of variable (examples?)
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 25

The Trick is Determining a and b:


xi y i
n 2 n

( x i )( y i
i 1 i 1

i 1

i 1

xi

n n 2 ( xi )
i 1

a y bx
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 26

Example:
Regression Used for Time Series

Period (X) 1 2 3 4

Demand (Y) 110 190 320 410

X2 1 4 9 16

XY 110 380 960 1640

15 1520 5540 5 b 98 2 15 55 5 1520 15 a 98 10 5 5


Column Sums
Chapter 9, Slide 27

5
15

490
1520

25
55

2450
5540

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Resulting Regression Model: Forecast = 10 + 98Period


600 500 400

300 200 100 0 1 2 3 X 4 5

Demand Regression

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 28

Example:
Simplified Regression I
If we redefine the X values so that their sum adds up to zero, regression becomes much simpler
a now equals the average of the y values b simplifies to the sum of the xy products divided by the sum of the x2 values

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 29

Example: Simplified Regression II


Period (X) 1 2 3 4 5 Period (X)' -2 -1 0 1 2 Demand (Y) 110 190 320 410 490 X2 4 1 0 1 4 XY -220 -190 0 410 980

0 1520 980 5 b 98 2 0 10 5 1520 0 a 98 304 5 5

1520

10

980
Chapter 9, Slide 30

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Dealing with Seasonality


Quarter
Winter 02 Spring Summer Fall Winter 03 Spring Summer Fall

Period
1 2 3 4 5 6 7 8

Demand
80 240 300 440 400 720 700 880
Chapter 9, Slide 31

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

What Do You Notice?


Forecasted Demand = 18.57 + 108.57 x Period Actual Demand
80
240 300 440

Period
Winter 02
Spring Summer Fall

Regression Forecast
90
198.6 307.1 415.7

Forecast Error
-10
41.4 -7.1 24.3

1
2 3 4

Winter 03
Spring Summer Fall

5
6 7 8

400
720 700 880

524.3
632.9 741.4 850

-124.3
87.2 -41.4 30
Chapter 9, Slide 32

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Regression picks up trend, but not seasonality effect


1000 800 600 400 200 0 1 2 3 4 5 6 7 8 Demand Forecast

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 33

Calculating Seasonal Index: Winter Quarter


(Actual / Forecast) for Winter Quarters: Winter 02: Winter 03: (80 / 90) = 0.89 (400 / 524.3) = 0.76

Average of these two = 0.83 Interpret!

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 34

Seasonally adjusted forecast model


For Winter Quarter [ 18.57 + 108.57Period ] 0.83 Or more generally:
[ 18.57 + 108.57 Period ] Seasonal Index

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 35

Seasonally adjusted forecasts


Forecasted Demand = 18.57 + 108.57 x Period Seasonally Adjusted Forecast

Period

Actual Demand

Regression Forecast

Demand/ Forecast

Seasonal Index

Forecast Error

Winter 02
Spring Summer Fall Winter 03 Spring Summer Fall

1
2 3 4 5 6 7 8

80
240 300 440 400 720 700 880

90
198.6 307.1 415.7 524.3 632.9 741.4 850

0.89
1.21 0.98 1.06 0.76 1.14 0.94 1.04

0.83
1.17 0.96 1.05 0.83 1.17 0.96 1.05

74.33
232.97 294.98 435.19 433.02 742.42 712.13 889.84

5.67
7.03 5.02 4.81 -33.02 -22.42 -12.13 -9.84

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 36

Would You Expect the Forecast Model to Perform This Well With Future Data?
1000 800 600 400 200 0 1 2 3 4 5 6 7 8 Demand forecast

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 37

More Regression Models I


Non-linear models
Example: y = a + b ln(x)

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 38

More Regression Models II


Multiple regression
More than one independent variable
y

y = a + b1 x + b2 z

x z
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 39

Causal Models
Time series assume that demand is a function of time. This is not always true. 1. Pounds of BBQ eaten at party. 2. Dollars spent on drought relief. 3. Lumber sales. Linear regression can be used in these situations as well.
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 40

Measuring Forecast Accuracy


How do we know:
If a forecast model is best? If a forecast model is still working? What types of errors a particular forecasting model is prone to make? Need measures of forecast accuracy
2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 41

Measures of Forecast Accuracy


Error = Actual demand Forecast or

Et = Dt Ft

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 42

Mean Forecast Error (MFE)


For n time periods where we have actual demand and forecast values:

MFE

i 1

Ei ) n
Chapter 9, Slide 43

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Mean Absolute Deviation (MAD)


For n time periods where we have actual demand and forecast values:

MAD

i 1

Ei n
Chapter 9, Slide 44

What does this tell us that MFE doesnt?


2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Example
Period Demand Forecast 3 4 5 6 7 8 11 9 10 8 14 12 13.5 13 10 9.5 9 11 Error -2.5 -4.0 0 -1.5 5.0 1.0 Absolute Error 2.5 4.0 0.0 1.5 5.0 1.0

What is the MFE? The MAD? Interpret!


2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 45

MFE and MAD: A Dartboard Analogy

Low MFE and MAD: The forecast errors are small and unbiased

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 46

An Analogy (continued)
Low MFE, but high MAD: On average, the arrows hit the bulls eye (so much for averages!)

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 47

An Analogy (concluded)

High MFE and MAD: The forecasts are inaccurate and biased

2006 Pearson Prentice Hall Introduction to Operations and Supply Chain Management Bozarth & Handfield

Chapter 9, Slide 48

You might also like