Lecture Fundamentals of operations management (4/e): Chapter 12 - Davis, Aquilano, Chase

Chapter 9 Forecasting, after studying this chapter you will be able to: Introduce the basic concepts of forecasting and its importance within an organization, identify several of the more common forecasting methods and how they can improve the performance of both manufacturing and service operations, provide a framework for understanding how forecasts are developed,…

DAVIS AQUILANO CHASE

PowerPoint Presentation by Charlie Cook

F   O   U   R   T   H       E   D   I   T   I   O   N

Forecasting

© The McGraw-Hill Companies, Inc., 2003

chapter 9

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–2

Chapter Objectives

Chapter Objectives

• Introduce the basic concepts of forecasting and its

importance within an organization

• Identify several of the more common forecasting

methods and how they can improve the performance

of both manufacturing and service operations

• Provide a framework for understanding how forecasts are developed

• Demonstrate that errors exist in all forecasts and show how to measure and assess these errors

• Discuss some of the software programs that are

available for developing forecasting models

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–3

Managerial Issues

Managerial Issues

• Recognizing the increased importance of forecasting

in both manufacturing and services

• How to go about implementing forecasting at all levels

in the organization

• Understanding how managers can use the various

forecasting methods to decide when to add

manufacturing capacity and where to locate retail

service outlets for maximum sales

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–4

Types of Forecasting

Types of Forecasting

• Qualitative Techniques

–Nonquantitative forecasting techniques based

on expert opinions and intuition Typically used when there are no data available.

• Time Series Analysis

–Analyzing data by time periods to determine if

trends or patterns occur.

• Causal Relationship Forecasting

–Relating demand to an underlying factor other

than time.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–6

Forecasting Techniques and Common Models

Forecasting Techniques and Common Models

Exhibit 9.2a

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–7

Forecasting Techniques and Common Models

Forecasting Techniques and Common Models

Exhibit 9.2b

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–8

Comparison of Forecasting Techniques

Comparison of Forecasting Techniques

Exhibit 9.3

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–11

Common Types of Trends

Common Types of Trends

Exhibit 9.5a

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–12

Common Types of Trends (cont’d)

Common Types of Trends (cont’d)

Exhibit 9.5b

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–13

Time Series Analysis

Time Series Analysis

• Simple Moving Average

–Average over a given number of time periods

that is updated by replacing the data in the

oldest period with that in the most recent period.

F t = Forecasted sales for the period

At­1 = Actual sales in period t­1

n = Number of periods in the moving average

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–14

Forecast Demand Based on a Three­ andNine­Week Simple Moving Average

Forecast Demand Based on a Three­ andNine­Week Simple Moving Average

Exhibit 9.6

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–16

Time Series Analysis (cont’d)

Time Series Analysis (cont’d)

• Weighted Moving Average

–Simple moving average where weights are

assigned to each time period in the average The sum of all of the weights must equal one.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–17

Time Series Analysis (cont’d)

Time Series Analysis (cont’d)

• Exponential Smoothing

–Times series forecasting technique that does

not require large amounts of historical data.

• Benefits of Using Exponential Models

–Models are surprisingly accurate.

–Model formulation is fairly easy.

–Readily understood by users.

–Little computation is required.

–Limited use of historical data.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–18

Time Series Analysis (cont’d)

Time Series Analysis (cont’d)

• Exponential Smoothing Constant Alpha ( )

–A value between 0 and 1 that is used to minimize

the error between historical demand and

respective forecasts.

–Use small values for if demand is stable,

larger values for if demand is fluctuating.

–Adaptive forecasting

• Two or more predetermined values of alpha

• Computed values of alpha

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

F t 1 t 1 t 1

F A

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

FIT t t t

FIT A

T

T t t 1 t 1 t 1

T F

FIT t 1 t 1 t 1

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–22

Forecasting Errors in  Time Series Analysis

Forecasting Errors in  Time Series Analysis

undetected trends.

–Random errors

• Unexplainable variations (noise) in a forecast that cannot be explained by the forecast model.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–23

Forecasting Errors in  Time Series Analysis (cont’d)

Forecasting Errors in  Time Series Analysis (cont’d)

• Measurement of Error

–MAD (mean absolute deviation)—Average

forecasting error based on the absolute

difference between actual and forecast

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–25

Forecasting Errors in  Time Series Analysis (cont’d)

Forecasting Errors in  Time Series Analysis (cont’d)

• Measurement of Error (cont’d)

–Tracking signal—a measurement of error that

indicates if the forecast is staying within

specified limits of the actual demand.

MAD

RSFE Signal

  Tracking

RSFE = Running sum of forecast errorsMAD = Mean absolute deviation

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–26

Computing the Mean Absolute Deviation (MAD), the Running Sum of Forecast Errors (RSFE), and the Tracking Signal from Forecast and Actual Data

Computing the Mean Absolute Deviation (MAD), the Running Sum of Forecast Errors (RSFE), and the Tracking Signal from Forecast and Actual Data

Exhibit 9.10

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–28

The Percentage of Points Included within the  Control Limits for a Range of 0 to 4 MADs

The Percentage of Points Included within the  Control Limits for a Range of 0 to 4 MADs

Exhibit 9.12

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–29

Forecasting Errors in  Time Series Analysis (cont’d)

Forecasting Errors in  Time Series Analysis (cont’d)

• Mean Absolute Percentage Error (MAPE)

–Used to determine the forecasting errors as a

percentage of the actual demand.

n = number of periods in forecast

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–30

Linear Regression Analysis

Linear Regression Analysis

• Linear Regression Analysis

–A forecasting technique that assumes that the

relationship between the dependent and

independent variables is a straight line.

bX a

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–31

Least Squares Regression Line

Least Squares Regression Line

Exhibit 9.13

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–32

Least  Squares  Regression  Analysis

Least  Squares  Regression  Analysis

Exhibit 9.14A

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–33

Linear Regression Analysis (cont’d)

Linear Regression Analysis (cont’d)

• Standard Error of the Estimate

–A measure of the dispersion of data about a

S

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–34

Standard Error of the Estimate in a Spreadsheet

Standard Error of the Estimate in a Spreadsheet

Exhibit 9.14B

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–35

Causal Relationship Forecasting

Causal Relationship Forecasting

• Leading Indicator

–An event whose occurrence causes, presages or

influences the occurrence of another

subsequent event.

• Warning strips on the highway

• Prerequisites to a college course

• An engagement ring

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–36

Causal Relationship: Sales to Housing Starts

Causal Relationship: Sales to Housing Starts

Exhibit 9.15

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Y

Y

y

y y

y

r

i

i i

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–39

Causal Relationship Forecasting (cont’d)

Causal Relationship Forecasting (cont’d)

• Reliability of Data

–Mean squared error—A measure of the

variability in the data about a regression line.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–40

Causal Relationship Forecasting (cont’d)

Causal Relationship Forecasting (cont’d)

• Multiple Regression Analysis

–Forecasting using more than one independent

variable; measuring the combined effects of

several independent variables on the dependent variable.

bz by

bx a

Y

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

9–41

Causal Relationship Forecasting (cont’d)

Causal Relationship Forecasting (cont’d)

• Neural Networks

–A forecasting technique simulating human

learning that develops complex relationships between the model inputs and outputs.

Management 4e 

© The McGraw­Hill Companies, Inc., 2003

–Point-of-Sale (POS) equipment

–Yield management—attempts to maximize the

revenues of a firm.