Business Math 101 - pt3

Management and Business in Engineering
A Short Primer on
Demand Forecasting
Dr. Abbott Weiss

MIT
Types of Forecasting
◆ Qualitative Techniques
– Non-quantitative 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.
Non-quantitative techniques
6
*image taken from user Stephanie Barber on Pinterest.com
Forecasting Techniques and Common Models
Exhibit 9.2a
© The McGraw-Hill Companies, Inc., 2003 9–7

Types of Forecasting
◆ Qualitative Techniques
– Non-quantitative 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.
Components of Demand
◆ Average Demand for the Period

◆ Trends
◆ Seasonal Influence
◆ Cyclical Elements
◆ Random Variation
F O U R T H E D I T I O N DAVIS
AQUILANO
CHASE
chapter 9 Forecasting
© The McGraw-Hill Companies, Inc., 2003
11
PowerPoint
*image taken from Fundamentals of Operations Management
Presentation (c)
Historical Monthly Product Demand Consisting of
a Growth Trend, Cyclical Factor, and Seasonal Demand
Exhibit 9.4

Comparison of Forecasting Techniques
Exhibit 9.3

Common Types of Trends
Exhibit 9.5a

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.
= At -1 + At - 2 ! + At - n
Ft n
Ft = Forecasted sales for the period
At-1 = Actual sales in period t-1
n = Number of periods in the moving average

Forecast Demand Based on a Three- and
Nine-Week Simple Moving Average
Exhibit 9.6

Moving Average Forecast of Three- and
Nine-Week Periods versus Actual Demand
Exhibit 9.7

HARD TO FORECAST
No matter how much evidence exists that seers do not exist,
suckers will pay for the existence of seers.
Forecasting Techniques and Common Models
Exhibit 9.2b

Forecasting Errors in Time Series Analysis
◆ Sources of Error
– Projection of past trends into the future when conditions change
– Bias errors
» Consistent mistakes causing a forecast to be too high or too low:
wrong relationships, wrong trend line, errors in shifting seasonal
demand, undetected trends.
– Random errors
» Unexplainable variations (noise) in a forecast that cannot be
explained by the forecast model.

So what? What I’ve learned doing this for 40 years...
1. First, always graph the data

2. Look for patterns
3. Think about causal factors, and data validity
4. Choose appropriate time frames
5. Aggregate where possible
6. Use simplest model you can (but no simpler)
7. Know that forecasts will be wrong
8. Examine forecast errors, learn and improve from them
9. Shorten process lead times, so you don’t need to forecast
24
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
» Summer heat and beer sales

Causal Relationship: Sales to Housing Starts
Exhibit 9.15

Linear Regression Analysis
◆ Linear Regression Analysis

– A forecasting technique that assumes that the relationship between
the dependent and independent variables is a straight line.
Y = a + bX
Y = Dependent variable to be solved for
a = Y intercept
b = Slope of the XY relationship
X = Independent variable (e.g., units of time)

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.
Y = a + bx + by + bz
Examples:
• advertising media mix
• macro econometric models

Examples in Practice
29
Five years of sales, by week: what’s going on here?
10000
9000
8000
7000
Sales in 1
thousand 6000 9
units 5000
4000
3000
2000
1000
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Week of the Year
30
Quarterly Cycle
Actual Weekly Sales vs. 13-Week Moving Average

10,000
9,000
8,000 13-week…
7,000
6,000
Sales in Units
5,000
4,000
3,000
2,000
1,000
0
1 3 5 7 9 11 13 15 17 19 21 23 25 of
Week 27the
29Year
31 -33 35 37 39 41 43 45 47 49 51
1994
31
Long term trend?
10,000 Actual Weekly Sales vs. 13-Week Moving

Average
8,000 13-week …
6,000
Sales in Units
4,000
2,000
0
1 9 17 25 33 41 49 57 65 73 81 89 97105113121-129
Weeks 137145
1994 153161169177
through 185193201209217225233241249
1998
32

Business Math 101 - pt3

Uploaded by

Copyright:

Available Formats

Business Math 101 - pt3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Business Math 101 - pt3

Uploaded by

Copyright:

Available Formats

Management and Business in Engineering

Dr. Abbott Weiss

© The McGraw-Hill Companies, Inc., 2003 9–7

◆ Average Demand for the Period

© The McGraw-Hill Companies, Inc., 2003

© The McGraw-Hill Companies, Inc., 2003 9–12

© The McGraw-Hill Companies, Inc., 2003 9–13

© The McGraw-Hill Companies, Inc., 2003 9–14

◆ Simple Moving Average

© The McGraw-Hill Companies, Inc., 2003 9–15

© The McGraw-Hill Companies, Inc., 2003 9–16

© The McGraw-Hill Companies, Inc., 2003 9–17

© The McGraw-Hill Companies, Inc., 2003 9–21

© The McGraw-Hill Companies, Inc., 2003 9–22

1. First, always graph the data

© The McGraw-Hill Companies, Inc., 2003 9–25

© The McGraw-Hill Companies, Inc., 2003 9–26

◆ Linear Regression Analysis

© The McGraw-Hill Companies, Inc., 2003 9–27

◆ Multiple Regression Analysis

© The McGraw-Hill Companies, Inc., 2003 9–28

Actual Weekly Sales vs. 13-Week Moving Average

10,000 Actual Weekly Sales vs. 13-Week Moving

You might also like