The Art and Science of Forecasting

Business Forecasting Summit | March 6, 2013Copyright © 2013 The Goodyear Tire & Rubber Company

Copyright © 2013 The Goodyear Tire & Rubber Company

Michael Ryan, The Goodyear Tire & Rubber Company

Business Forecasting: Predictive Intelligence Summit, San Francisco, CA

March 6, 2013

The Art & Science of Good Forecasting:

Combining Math & Analysis with Good Instincts and Sound Judgment


Replacement, 70%

OE, 30%

Company

2012 Sales = $20.992 Billion1

1 http://www.goodyear.com/investor/pdf/slides/4qtr12_slides.pdf2 http://www.goodyear.com/investor/pdf/10k/2012_10K_w-exhibits.pdf

Segmentation2

(% of 2012 Units of 164 million)

Consumer, 54%

Commercial, 20%

Other, 13%

Retail, 7%Chemical, 6%

2


Consensus Sales Plan

Agreed Sales Plan

18 Month Forward Horizon

E x t e r n a l I n t e r n a l

3

Desired OutcomeScience Art

What is the outcome you would like to achieve?

Art & Science of Forecasting

In Range

of Forecast

Undersold

Forecast

Oversold

Forecast

Profit

(Increases)

(Decreases)

Manage Organizational Knowledge

Forecast

Execution

Forecast vs.

Actual

History

Statistical

Forecast

External

Input

Discussion &

Consensus

Internal

InputScenario

Planning

Feedback

Loop


"Forecasting Isn't Rocket Science"

In Rocket Science, parameters are well understood

Processes in place to accurately model impact of variables

Laws of Physics:

• Immutable. Constant.

Within Control:

• Vehicle Mass

• Engine Thrust

• Launch Timing

Model Effects of:

• Wind

• Temperature

• Interaction of Systems

4

Photo: http://history.nasa.gov/ap11ann/kippsphotos/39961.jpg


More Variables Than Rocket Science

In Forecasting, variables can be both an art and a science

Law of Supply & Demand:

• Challenging to Fine Tune

Within Control:

• Price

• Supply

Model Effects of:

• Demand

• Supply

• Quantity Sold

5

Illustration: http://en.wikipedia.org/wiki/Supply_and_demand

In some regards, Rocket Science is easier than forecasting


Value of Forecast Accuracy

Profit generally improves as forecast accuracy increases

Opportunity to translate the value of forecasting to your business

Undersold:

• Cash

• Inventory

• Capacity

• Utilization

Oversold:

• Lost Sales

• Backorders

• Overtime

• Expediting

In Range

of Forecast

Undersold

Forecast

Oversold

Forecast

Profit

(Increases)

(Decreases)

6


“Don't be afraid to give up the good to go

for the great.”

-John D. Rockefeller

7


What is a Process?

Having a process to begin with helps focus improvements

Beginning Middle End

Input Transformation Output

Start And Then Some Magic Happens Finish

Definable: Can draw a picture of it…

Predictable: Know what result will be…

Repeatable: Outcome is assured & repeated…

=+ +=

8


Create the plan

Measure performance

Incorporate feedback

To optimize effectiveness, make forecasting a closed loop

Good forecasting incorporates a thorough gap analysis

Forecasting Cycle

Forecast

Execution

Forecast vs.

Actual

History

Statistical

Forecast

External

Input

Discussion &

Consensus

Internal

InputScenario

Planning

Feedback

Loop

9


Art vs. Science

Part of the science is in understanding what is art

Science: Objective

• Measurable

• Cause and effect

• Selective few

• Quantitative

Art: Subjective

• How to tell “Good” from “Bad”

• What information is useful

• Random many

• Qualitative

500,000

600,000

700,000

800,000

900,000

1,000,000

1,100,000

1,200,000

1,300,000

1,400,000

1,500,000

1,600,000

1,700,000

1,800,000

1,900,000

2,000,000

Jan

-10

Feb

-10

Mar

-10

Ap

r-1

0

May

-10

Jun

-10

Jul-

10

Au

g-1

0

Sep

-10

Oct

-10

No

v-1

0

De

c-1

0

Jan

-11

Feb

-11

Mar

-11

Ap

r-1

1

May

-11

Jun

-11

Jul-

11

Au

g-1

1

Sep

-11

Oct

-11

No

v-1

1

De

c-1

1

Jan

-12

Feb

-12

Mar

-12

Ap

r-1

2

10


Manage Organizational Knowledge

External inputs help shape long term forecast

Integrating multiple sources of information

Consensus Sales Plan

Agreed Sales Plan

18 Month Forward Horizon

E x t e r n a l I n t e r n a l

11


“Good judgment comes from

experience, and a lot of that comes from

bad judgment”

-Will Rogers

12


Qualitative vs. Quantitative

Beware of qualitative masquerading as quantitative

Different ways to describe a simple cup of coffee

Qualitative:

Descriptions

• Big Cup

• Rich Color

• Strong Brew

• Very Hot

Quantitative:

Measurements

• 20 oz. Serving

• Index of Refraction

• X mg Caffeine / Y oz.

• 150º F

13


Forecasting as a Collaborative Process

All working to - and collaborating on - the same plan

5/29/2012 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Q1 Q2 Q3 Q4 Full Year

ProductionPlant A 686,533 654,888 809,606 741,601 801,591 945,080 813,117 919,690 925,298 853,779 775,512 721,038 2,151,026 2,488,272 2,658,105 2,350,329 9,647,733

Plant B 1,040,201 992,254 1,226,676 1,123,639 1,214,532 1,431,939 1,231,995 1,393,470 1,401,967 1,293,604 1,175,019 1,092,482 3,259,131 3,770,110 4,027,432 3,561,105 14,617,778

Plant C 416,080 396,902 490,670 449,455 485,813 572,776 492,798 557,388 560,787 517,442 470,007 436,993 1,303,652 1,508,044 1,610,973 1,424,442 5,847,111

Domestic Production 2,142,814 2,044,044 2,526,952 2,314,695 2,501,936 2,949,794 2,537,909 2,870,549 2,888,053 2,664,825 2,420,538 2,250,512 6,713,809 7,766,426 8,296,511 7,335,876 30,112,622

Imports 142,250 135,693 167,751 153,660 166,090 195,821 168,478 190,560 191,722 176,903 160,686 149,399 445,693 515,571 550,760 486,989 1,999,012

Total Production 2,285,064 2,179,737 2,694,702 2,468,355 2,668,026 3,145,615 2,706,387 3,061,109 3,079,775 2,841,728 2,581,225 2,399,911 7,159,503 8,281,996 8,847,271 7,822,864 32,111,634

ShipmentsRegion A 555,663 530,050 655,275 600,234 648,788 764,925 658,117 744,375 748,914 691,028 627,681 583,591 1,740,989 2,013,947 2,151,406 1,902,300 7,808,642

Region B 333,398 318,030 393,165 360,141 389,273 458,955 394,870 446,625 449,349 414,617 376,609 350,154 1,044,593 1,208,368 1,290,844 1,141,380 4,685,185

Region C 689,022 657,262 812,542 744,291 804,498 948,507 816,065 923,025 928,654 856,875 778,324 723,652 2,158,826 2,497,295 2,667,744 2,358,851 9,682,716

Region D 466,757 445,242 550,431 504,197 544,982 642,537 552,818 625,275 629,088 580,464 527,252 490,216 1,462,430 1,691,716 1,807,181 1,597,932 6,559,259

Domestic Shipments 2,044,839 1,950,585 2,411,414 2,208,862 2,387,541 2,814,923 2,421,870 2,739,300 2,756,004 2,542,983 2,309,866 2,147,614 6,406,838 7,411,326 7,917,175 7,000,462 28,735,802

Export 177,812 169,616 209,688 192,075 207,612 244,776 210,597 238,200 239,653 221,129 200,858 186,749 557,116 644,463 688,450 608,736 2,498,765

Total Shipments 2,222,651 2,120,201 2,621,102 2,400,937 2,595,154 3,059,699 2,632,468 2,977,500 2,995,657 2,764,112 2,510,724 2,334,363 6,963,955 8,055,790 8,605,625 7,609,198 31,234,567

Inventory Inv End Q1 Inv End Q2 Inv End Q3 Inv End Q4 Y/E Inv

Warehouse 1 566,776 576,897 589,409 600,870 613,259 627,864 640,431 654,644 668,944 682,139 694,124 705,267 589,409 627,864 668,944 705,267 705,267

Warehouse 2 700,135 712,638 728,094 742,252 757,555 775,597 791,120 808,678 826,343 842,642 857,447 871,213 728,094 775,597 826,343 871,213 871,213

Warehouse 3 533,436 542,962 554,738 565,525 577,185 590,931 602,758 616,136 629,594 642,013 653,293 663,781 554,738 590,931 629,594 663,781 663,781

Warehouse 4 733,475 746,573 762,765 777,597 793,629 812,530 828,793 847,186 865,692 882,768 898,278 912,699 762,765 812,530 865,692 912,699 912,699

Warehouse 5 333,398 339,351 346,711 353,453 360,740 369,332 376,724 385,085 393,497 401,258 408,308 414,863 346,711 369,332 393,497 414,863 414,863

Warehouse 6 466,757 475,092 485,396 494,834 505,036 517,065 527,413 539,119 550,895 561,761 571,632 580,808 485,396 517,065 550,895 580,808 580,808

Total Inventory 3,333,977 3,393,512 3,467,113 3,534,531 3,607,403 3,693,320 3,767,239 3,850,847 3,934,966 4,012,582 4,083,083 4,148,632 3,467,113 3,693,320 3,934,966 4,148,632 4,148,632

SAP/DP

Supply

Planning

Finance

Consensus

Process

Parallel

Views

Internal Views

External Views

Control

Document

Sales & Demand Plan Flow

(Actual)

(Plan)

14


Performance Measurement

Incorporate measurement as feedback to improve the forecast

15

If it can’t be measured, it can’t be (proven) fixed

Direct Measurements:

• Accuracy

• Bias

• MAPE (Mean Average Percent Error)

Indirect Measurements:

• Customer Service

• Inventory Levels

• Supply Chain Costs

Make Measurements Matter:

• Tie Accuracy to Compensation

• Pay for Performance

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

80%

0

1

2

3

4

5

6

7

8

9

10

Jan-12 Feb-12 Mar-12 Apr-12 May-12 Jun-12 Jul-12 Aug-12 Sep-12 Oct-12 Nov-12 Dec-12

Bia

s (%

)

Vo

lum

e

Forecast Bias (%) Forecast Actual


Advancing Forecasting Practices

Use SMART goals to drive improvement

Measuring performance highlights collaboration opportunities

16


“The measure of success is not whether

you have a tough problem to deal with, but

whether it is the same problem you had last

year.”

-John Foster Dulles

17


NAT Segment Operating Income1

($ in Millions)

($305)

$18

$276

$514 $450

2009 2010 2011 2012 Prior Target

Results

Strong process facilitates the path to profitability

S&OP Helps Drive:

Risk Mitigation

Forecast Accuracy

Inventory Turns

Customer Satisfaction

18

North American Tire exceeded its 2013 SOI target

one year ahead of plan

1 http://www.goodyear.com/investor/pdf/slides/4qtr12_slides.pdf


For More Information

[email protected]

www.Linkedin.com/in/isolveproblems

19