Zachman-frameworks-and-baseball607.pdf

A Zachman Framework Populated with Baseball Models Terry Bahill Systems and Industrial Engineering University of Arizona [email protected] copyright ©, 2004-09, Bahill This file is located at http://www.sie.arizona.edu/sysengr/slides/

References • Bahill, A. T., Botta, R. and Daniels, J., The Zachman Framework Populated with Baseball Models, Journal of Enterprise Architecture, 2(4): 50-68, 2006. This paper is available at http://www.sie.arizona.edu/sysengr/publishedPa pers/ZachmanBaseball.pdf • Bahill, A. T., Botta, R. and Daniels, J., A systems engineering approach to organizing baseball models, keynote address at the AsiaPacific Congress on Sports Technology, Tokyo, September 11-14, 2005.

06/08/10

2

© 2009 Bahill

Definitions • A model is a simplified representation of some view of a real system. • A simulation is an implementation of a model, often on a digital computer.

06/08/10

3

© 2009 Bahill

Example • This is a perfect model for the vertical movement of a major league fastball • [Bahill does a demonstration here]*  That is how long the pitch is in the air  That is how far a fastball drops due to gravity • It is a good model because  The model is simpler than the real system  It models only one aspect of the real system  Vertical movement  It illustrates that all models are not mathematical

06/08/10

4

© 2009 Bahill

Model do not have to be smaller • Models do not have to be smaller than the entity they model. • Geometrical lines are one-dimensional and, therefore, cannot be seen with our eyes. ____________________________ • Things like the above are models of the geometrical concept of a line. • They have a measurable width and are, therefore, not true lines.

06/08/10

5

© 2009 Bahill

Tasks in the modeling process

*

• • • • • • • •

• • • • • 06/08/10

Describe the system to be modeled Determine the purpose of the model Determine the level of the model Gather experimental data describing system behavior Investigate alternative models Select a tool or language for the simulation Make the model Validate the model*  Show that the model behaves like the real system  Emulate something not used in the model’s design  Perform a sensitivity analysis  Show interactions with other models Integrate with models for other systems Analyze the performance of the model Re-evaluate and improve the model Suggest new experiments on the real system State the assumptions 6

© 2009 Bahill

06/08/10

5 HDO 6 \ VW HP

0 RGHOHUV * RRG 0 RGHOHUV

0 RGHORI 5 HDO 6 \ VW HP

7

0 RGHOHUV * RRG 0 RGHOHUV

&RP SXW HU 6 LP XODW LRQ RI0 RGHO

0 DW KHP DW LFLDQV

( [ SHULP HQW DOLVW V

Philosophy

© 2009 Bahill

Kinds of models • • • •

06/08/10

Models of behavior Models of structure Models of physical properties Models for analysis

8

© 2009 Bahill

Purpose of models • Guide decisions • Understand or improve an existing system or organization • Create a new design or system • Control a system • Suggest new experiments • Guide data collection activities • Allocate resources • Identify cost drivers • Increase return on investment • Identify bottlenecks • Help sell the product • Reduce risk 06/08/10

9

© 2009 Bahill

Purpose of frameworks • Organize integrated models of an enterprise • Assess completeness of the descriptive representation of an enterprise • Understand an organization or a system • Assist in identification and categorization • Provide a communication mechanism • Help manage complexity • Identify the flow of money in the enterprise

06/08/10

10

© 2009 Bahill

Frameworks • Frameworks help people organize and assess completeness of integrated models of their enterprises. • There are few public examples where a framework has been completely populated. • This paper will fill in a complete framework for Baseball. • We know of no other enterprise that has  models in as many cells,  where the models are not proprietary,  where the models will be understood by a large number of people without a steep learning curve. 06/08/10

11

© 2009 Bahill

Common frameworks • Zachman, Zachman Enterprise Architecture Framework

• DoDAF, Department of Defense Architecture Framework

06/08/10

12

© 2009 Bahill

An empty Zachman framework 1. What (data)

2. How 3. Where 4. Who 5. When (function) (network) (people) (time)

6. Why (motivation)

1. Scope (context) 2. Business model (concept) 3. System model (logical) 4. Technology model (physical) 5. Detailed representation (component) 6. Real system, i. e. executing the game of baseball

06/08/10

13

© 2009 Bahill

The rows present

*

• different perspectives of the enterprise • different views of the enterprise • different roles in the enterprise

06/08/10

14

© 2009 Bahill

The rows Scope describes the system’s vision, mission, boundaries, architecture and constraints. The scope states what the system is to do. It is called a black box model, because we see the inputs and outputs, but the not the inner workings. Business model shows goals, strategies and processes that are used to support the mission of the organization. System model contains system requirements, objects, activities and functions that implement the business model. The system model states how the system is to perform its functions. It is called a white box model, because we see its inner workings. Technology model considers the constraints of humans, tools, technology and materials. Detailed representation presents individual, independent components that can be allocated to contractors for implementation. Real system depicts the operational system under consideration.

06/08/10

15

© 2009 Bahill

The columns I Keep six honest serving men (They taught me all I knew): Their names are What and Why and When And How and Where and Who. From “The Elephant’s Child,” Rudyard Kipling, 1902.

06/08/10

16

© 2009 Bahill

The columns present various aspects of the enterprise

06/08/10

17

© 2009 Bahill

The columns What (data) describes the entities involved in each perspective of the enterprise. Examples include equipment, business objects and system data. How (functions) shows the functions within each perspective. Where (networks) shows locations and interconnections within the enterprise. This includes major business geographical locations, networks and the playing field. Who (people) represents the people within the enterprise and metrics for assessing their capabilities and performance. The design of the enterprise organization has to do with the allocation of work and the structure of authority and responsibility. When (time) represents time, or the event relationships that establish performance criteria. This is useful for designing schedules, the processing architecture, the control architecture and timing systems. Why (motivation) describes the motivations of the enterprise. This reveals the enterprise goals, objectives, business plan, knowledge architecture, and reasons for thinking, doing things and making decisions.

06/08/10

18

© 2009 Bahill

Goal Jacques Barzun* said, “Whoever wants to know the heart and mind of America had better learn baseball, the rules and realities of the game.” Zachman says the way to learn Baseball is to understand the models in the 36 cells of a Zachman framework. Our goal was to populate a Zachman framework with objective simulatable models for the science of baseball that were published in peer reviewed journals.

06/08/10

19

© 2009 Bahill

Column 1, What (data) The object modeled in column 1 is the baseball bat.

06/08/10

20

© 2009 Bahill

A Zachman framework 1. What

2. How

3. 4. 5. 6. Where Who When Why

1. Scope 2. Business model 3. System model 4. Technology model 5. Detailed representation 6. Real system, Baseball i. e. executing bat the game of baseball

06/08/10

21

© 2009 Bahill

Column 1, Row 5 Many models for the baseball bat explain the sweet spot, moment of inertia (MoI), center of percussion (CoP), etc. (Adair, 1994: Cross, 1998: Nathan, 2000 and 2003); column 1 (what), row 5 (detailed representation). knob

pivot

dknob-pivot

06/08/10

cm

dpivot-cm dpivot-cop

22

cop

dcm-cop

© 2009 Bahill

A Zachman framework 1. What

2. How

3. 4. 5. 6. Where Who When Why

1. Scope 2. Business model 3. System model 4. Technology model 5. Detailed representation

Sweet spot, MoI, CoP 6. Real system, Baseball i. e. executing bat the game of baseball

06/08/10

23

© 2009 Bahill

Column 1, Row 5 The swing of a bat can be modeled with a translation and two rotations, one about the batter’s spine and the other between the two hands (Brancazio, 1987: Watts and Bahill, 2000); column 1 (what), row 5 (detailed representation).

06/08/10

24

© 2009 Bahill

06/08/10

25

© 2009 Bahill

06/08/10

26

© 2009 Bahill

A Zachman framework 1. What

2. How

3. 4. 5. 6. Where Who When Why

1. Scope 2. Business model 3. System model 4. Technology model 5. Detailed representation

A swing, Sweet spot, MoI, CoP 6. Real system, Baseball i. e. executing bat the game of baseball

06/08/10

27

© 2009 Bahill

Column 1, Row 4

*

There is an ideal bat weight and a best weight distribution for each batter (Bahill and Karnavas, 1989 and 1991: Bahill and Morna Freitas, 1995: Bahill, 2004). The team helps the individual select and acquire the right bat; column 1 (what), row 4 (technology model). 90

Batted-ball Speed

80

Speed (mph)

70

Ideal Bat Weight

60 50

Bat Speed

40 30 20 10 0 0

10

20

30

40

50

60

70

Bat Weight (oz) 06/08/10

28

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager) 5. Detailed representation (Individual Baseball Player) 6. Real system, i. e. executing the game of baseball

06/08/10

2. How

Ideal bat weight & weight distribution A swing, Sweet spot, MoI, CoP

Baseball bat

29

© 2009 Bahill

Column 1, Row 3

*

Each organization provides facilities for batting practice, conditioning and skills development; column 1 (what), row 3 (system model).

06/08/10

30

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager) 5. Detailed representation (Individual Baseball Player) 6. Real system, i. e. executing the game of baseball

06/08/10

2. How

Batting practice facilities Ideal bat weight & weight distribution A swing, Sweet spot, MoI, CoP Baseball bat

31

© 2009 Bahill

Column 1, Row 2 The NCAA has created rules governing the allowed dimensions and performance of aluminum bats (Crisco, 1997: Nathan, 2003), For example, the bat shall not weigh less (in ounces) than its length (in inches); column 1 (what), row 2 (business model).

06/08/10

32

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager) 5. Detailed representation (Individual Baseball Player) 6. Real system, i. e. executing the game of baseball

06/08/10

2. How

NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution A swing, Sweet spot, MoI, CoP Baseball bat

33

© 2009 Bahill

Column 1, Row 1

*

The rules of ball and stick games (baseball, softball, cricket, tennis) are written to challenge the physiological limits of the human in many dimensions (Regan, 1992). There are rules for each piece of equipment; column 1 (what), row 1 (scope).

06/08/10

34

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager) 5. Detailed representation (Individual Baseball Player) 6. Real system, i. e. executing the game of baseball

06/08/10

2. How

Equipment rules NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution A swing, Sweet spot, MoI, CoP Baseball bat

35

© 2009 Bahill

Slivers • We cannot present all baseball models in one study. • So we only present slivers. • The sliver used in column 1 (what) was the physical baseball bat. An alternate sliver for column 1 is information.

06/08/10

36

© 2009 Bahill

Information1 • Multiple TV cameras in major league stadiums pick up the flight of the pitch. The TV signals can be used to construct a computer model for the flight of the ball. When these data are used by the TV networks to display to the TV audience the location of the pitch relative to the strike zone, then they are being used as a Detailed Representation: row 5, column 1, role TV audience. • These data could be used to determine if the ball passed through the strike zone and this Technology Model information could be transmitted in real time to the umpire to help him call balls and strikes, this would be row 4, column 1, role umpire. 06/08/10

37

© 2009 Bahill

Information2 • These data could also be put on the Internet for researchers to use to help determine the speed and spin of the ball, to allow them to model the movement of the ball and for sports fans to play fantasy baseball; this would be row 4, column 1, role researcher. http://webusers.npl.uiuc.edu/~anathan/pob/pitchtracker.html • When this information is put on a CD and given to the umpire at the end of the game to give him feedback to improve the consistency of the set of umpires, then it is being used in a System Model (www.QuesTec.com): row 3 column 1, role umpires. 06/08/10

38

© 2009 Bahill

Information3 • When knowledge about the difference in the strike zones of American and National League umpires is used to regulate enforcement of baseball rules, then it is being used in the Business Model: row 2, column 1, role commissioner and owners. • If such information were gathered and analyzed for cricket and tennis, then the derived wisdom would transcend baseball and become row 1, column 1.

06/08/10

39

© 2009 Bahill

Column 2, How (function) • The activity modeled in column 2 is one pitch and people’s response to it. • One pitch and people’s response to it is called a sliver of Baseball. I cannot present all of the thousands of Baseball models. So to limit the scope, I present only a few slivers. To enlarge the coverage I use a different sliver in each column.

06/08/10

40

© 2009 Bahill

Catcher

Pitcher

Umpire

Batter

Fielder

Predict When and Where

Pitch Ball Umpire is a role that can be filled by one (for the minors in Little League) to six (for World Series games) people.

Decision [no]

Swing Bat? [yes]

[no]

Ball Hit? [yes]

Merge

There could also be decisions for fair or foul and fly or grounder.

Fork

Run to First Base Runner

Field Ball and Throw to Appropriate Base

Ball

{joinSpec = Runner OR Ball} Decide if Ball or Runner Got to the Base First

Fork

Catch Ball & Return it to Pitcher

Call Ball, Strike, Out, Walk or Other

Other could be batter hit by pitch, foul ball with two strikes, catcher interference, time-out called before pitch, etc.

Wait for Batter Give Sign to Pitcher

06/08/10

Signal Ready Join

41

Terry Bahill Summer 2007

© 2009 Bahill

A Zachman framework 1. What

2. How

3. 4. 5. 6. Where Who When Why

1. Scope

Equipment rules 2. Business NCAA bat model rules 3. System Batting model practice facilities 4. Technology Ideal bat model weight & weight distribution 5. Detailed A swing, representation Sweet spot, MoI, CoP 6. Real system, Baseball bat One pitch i. e. executing & the game of responses baseball 06/08/10

42

© 2009 Bahill

Column 2, Row 5

*

The movement of the pitch depends only on gravity, the ball’s velocity, and the spin (Watts and Bahill, 2000: Bahill and Baldwin, 2004), the right-hand rules shows how the ball is deflected by spininduced forces (Bahill and Baldwin, 2007); column 2 (how), row 5 (detailed representation). Magnus force

Ball direction

Ball

Spin

06/08/10

43

Drag force

Gravity force © 2009 Bahill

From the perspective of a right-handed pitcher

SaD Sid

06/08/10

44

© 2009 Bahill

Column 2, Row 5 Two strategies are used by the batter for tracking the pitch using the saccadic and smooth pursuit eye movement systems (Bahill and LaRitz, 1984: McHugh and Bahill, 1985); column 2 (how), row 5 (detailed representation).

06/08/10

45

© 2009 Bahill

Two strategies • The optimal hitting strategy: track the ball with smooth pursuit eye movements and fall behind in the last five feet. • The optimal learning strategy: track the ball over the first part of its trajectory with smooth pursuit eye movements, make a fast saccadic eye movement to the predicted point of bat-ball collision, then let the ball catch up to the eye. The batter observes the ball, makes a prediction of where it will hit his bat, sees the actual position of the ball when it hits the bat, and uses this feedback to learn to predict better next time.

06/08/10

46

© 2009 Bahill

Column 2, Row 5 A neurophysiological model shows how the batter predicts where and when the ball will cross the plate (Karnavas, Bahill and Regan, 1990: Bahill and Baldwin, 2004); column 2 (function, how), row 5 (detailed representation).

06/08/10

47

© 2009 Bahill

Neurophysiological model Other Sensory Inputs Memory

Speed Estimator

Estimated Pitch Speed

Ball Spin Ball's Retinal Image

Primary Visual Processes

^

Time Since Release

Timer

γ dγ/dt

Time Until Contact

D0

-

Σ ^

D ^

zfall

dφ/dt

06/08/10

48

© 2009 Bahill

Column 2, Row 5 Underestimating the pitch speed can induce the perceptual illusion of the rising fastball (Karnavas, Bahill and Regan, 1990: Bahill and. Karnavas, 1993); column 2 (how), row 5 (detailed representation).

06/08/10

49

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager) 5. Detailed representation (Individual Baseball Player) 6. Real system, i. e. executing the game of baseball

06/08/10

2. How

3. Where

Equipment rules NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution A swing, Sweet spot, MoI, CoP

Rising fastball, Eye movement strategies, Speed & spin

Baseball bat

One pitch & responses

50

© 2009 Bahill

Column 2, Row 4 Teamwork and signals allow the manager, the batter and the runners to execute tactics such as hit and run, bunt, steal, take the pitch, swing away, etc.; column 2 (how), row 4 (technology model).

06/08/10

51

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager)

2. How

Equipment rules NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution

5. Detailed representation A swing, Sweet spot, MoI, CoP (Individual Baseball Player) 6. Real system, i. e. executing the Baseball bat game of baseball

06/08/10

3. Where

52

Teamwork & signals for hit and run, bunt, etc. Rising fastball, Eye movement strategies, Speed & spin One pitch & responses

© 2009 Bahill

Column 2, Row 3 Stadiums can be equipped with a variety of optional equipment that can record and playback the pitch, such as the multiple television cameras used to aid the umpires (www.QuesTec.com) or entertain the TV audience (http://www.gueziec.org/kzone.html) and the stadium instant replay screens for the benefit of the players and spectators; column 2 (how), row 3 (system model).

06/08/10

53

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager)

2. How

Equipment rules NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution

5. Detailed representation A swing, Sweet spot, MoI, CoP (Individual Baseball Player) 6. Real system, i. e. executing the Baseball bat game of baseball

06/08/10

3. Where

54

Pitch trackers, Instant replay Teamwork & signals for hit and run, bunt, etc. Rising fastball, Eye movement strategies, Speed & spin One pitch & responses

© 2009 Bahill

Column 2, Row 2 Major League Baseball Inc. defines the strike zone and manages umpires; column 2 (how), row 2 (business model).

06/08/10

55

© 2009 Bahill

A Zachman framework 1. What 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager)

2. How

Equipment rules NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution

5. Detailed representation A swing, Sweet spot, MoI, CoP (Individual Baseball Player) 6. Real system, i. e. executing the Baseball bat game of baseball

06/08/10

3. Where

56

Define the strike zone Pitch trackers; Instant replay Teamwork & signals for hit and run, bunt, etc. Rising fastball, Eye movement strategies, Speed & spin One pitch & responses

© 2009 Bahill

Column 2, Row 1 The rules of baseball evolved over its first 50 years, but have been relatively stable over the last century (Gould, 2003); column 2 (how), row 1 (scope).

06/08/10

57

© 2009 Bahill

A Zachman framework 1. Scope (Commissioner of Baseball) 2. Business model (Team Owner) 3. System model (General Manager) 4. Technology model (Team Manager)

1. What

2. How

Equipment rules

Rules of baseball

NCAA bat rules Batting practice facilities Ideal bat weight & weight distribution

5. Detailed representation A swing, Sweet spot, MoI, CoP (Individual Baseball Player) 6. Real system, i. e. executing the Baseball bat game of baseball

06/08/10

58

3. Where

Define the strike zone Pitch trackers; Instant replay Teamwork & signals for hit and run, bunt, etc. Rising fastball, Eye movement strategies, Speed & spin One pitch & responses

© 2009 Bahill

Column 3, Where (network) • The topic of column 3 is the baseball field; column 3 (where), row 6 (real system). • The human brain does not have x, y and z coordinates of objects. Humans must track objects using neurophysiological parameters. As a result outfielders run a curved path when tracking down fly balls (McBeath, Shaffer and Kaiser, 1995); column 3 (where), row 5 (detailed representation). • Batters must predict where and when the ball will cross the plate (Karnavas, Bahill and Regan, 1990: Bahill and Baldwin, 2004); column 3 (where), row 5 (detailed representation). • Before every pitch all fielder rehearse where they will throw the ball if they receive a ground ball or a fly ball; column 3 (where), row 4 (technology model). 06/08/10

59

© 2009 Bahill

Column 3, Where (continued) • Where home plate is placed in the stadium effects the area behind the plate, the design of protective netting, the orientation to the sun, the distance to the fences and therefore safety and playing performance; column 3 (where), row 3 (system model). • Stadiums can be designed for baseball only or they may be shared by baseball, football and other events; column 3 (where), row 2 (business model). • Baseball is played in stadiums and broadcast on television. The teams are organized into leagues and divisions according to geography; column 3 (where), row 1 (scope).

06/08/10

60

© 2009 Bahill

Column 4, Who (people) • The people modeled in column 4 are major league baseball players; column 4 (who), row 6 (real system). • Physiological state of individual players determines whether and how well they play or if they are on the disabled list; column 4 (who), row 5 (detailed representation). • The count is the model for how the batter is doing during an at-bat; column 4 (who), row 4 (technology model). • Defining and locating the sweet spot of the bat is a humanmachine interface: the teams help individuals understand this issue ; column 4 (who), row 4 (technology model). • Individual player performances are published daily in the box scores in the sports sections of newspapers. Player average performances are published weekly; column 4 (who), row 4 (technology model).

06/08/10

61

© 2009 Bahill

Column 4, Who (continued) • Individual player statistics, Markov models and manager decisions (such as batting order) are used to simulate games and seasons: this is called fantasy baseball (Burkiet, Harold and Palacios, 1997; www.stats.com). Scouts make observations and evaluations of players’ performances and report this information back to their organizations; column 4 (who), row 3 (system model). • The General Manager (GM) creates the 25-man roster and trades players to improve it. It must contain a balance of players at each position, including short relievers, long relievers, etc.; column 4 (who), row 3 (system model). • Billy Beane, General Manager of the Oakland Athletics, evaluates the worth of players with innovative high-level metrics and he has the most successful low-salary team in the major leagues (Lewis, 2003). The GM must consider player positions, salaries, performance, etc.; column 4 (who), row 3 (system model).

06/08/10

62

© 2009 Bahill

Column 4, Who (continued) • George Steinbrenner, owner of the New York Yankees, evaluates the worth of his players with traditional metrics and he has the most successful high-salary team in the major leagues. The team owner must consider the team salary cap and return on investment; column 4 (who), row 2 (business model). • Before the 2001 season, the owners of the Texas Rangers created a ten year $252 million contract for Alex Rodriguez (Cohen and Wallace, 2003); column 4 (who), row 2 (business model). • The Commissioner of Baseball coordinates the teams; the Major League Baseball Players Association and players’ agents orchestrate the activities of the players. The Commissioner must consider drug testing, salary caps, retirement plans, profit sharing and the reserve clause; column 4 (who), row 1 (scope).

06/08/10

63

© 2009 Bahill

Column 5, When (time) • The fundamental unit of time in a baseball game is one pitch; column 5 (when), row 6 (real system). • The batter’s mental model for the pitch is based on the last one or two pitches or perhaps on the last 20 seconds (Bahill and Baldwin, 2004: Gray, 2003); column 5 (when), row 5 (detailed representation). • A successful tactic of pitchers is “work fast and change speed on every pitch” (Bahill and Baldwin, 2004); column 5 (when), row 5 (detailed representation). • The pitcher pitches the ball. The batter swings and hits the ball. He runs toward first base, etc. Our activity diagrams show one pitch and subsequent activities; column 5 (when), row 5 (detailed representation).

06/08/10

64

© 2009 Bahill

Column 5, When (continued) • Pitch count -- pitchers are often removed after, say, 120 pitches; column 5 (when), row 4 (technology model). • Pitching rotations, e.g. “Spahn and Sain and pray for rain;” column 5 (when), row 3 (system model) • Television networks determine the starting times of many games; column 5 (when), row 3 (system model). • Season schedules for all of the teams are complex because of the many constraints; column 5 (when), row 2 (business model) • Decisions must be made about interleague play, playoff structure, expansion teams, etc.; column 5 (when), row 1 (scope).

06/08/10

65

© 2009 Bahill

Column 6, Why (motivation) • Column 6 concerns why baseball people think the way they do and make the decisions they do; column 6 (why), row 6 (real system). • Why does the pitcher decide to throw a fastball, a slider, a curveball or a changeup? column 6 (why), row 5 (detailed representation). • Critical flicker fusion frequency explains why baseball players think there is a difference between the two-seam and the fourseam fastballs although Physics shows no difference (Bahill and Baldwin, 2004); column 6 (why), row 5 (detailed representation). • Batters have many heuristics. Among them is, with a 3-0 count expect a fastball because the pitcher will have the greatest confidence in throwing it for a strike (Williams and Underwood, 1982: Bahill and Baldwin, 2004); column 6 (why), row 4 (technical model). • The manager motivates his players by knowing when blame and when not to (Baldwin, 2001); column 6 (why), row 4 (technical model). 06/08/10

66

© 2009 Bahill

The four seam fastball

06/08/10

67

© 2009 Bahill

06/08/10

68

© 2009 Bahill

Four seam orientation

06/08/10

69

© 2009 Bahill

The two seam fastball

06/08/10

70

© 2009 Bahill

Two seam orientation

06/08/10

71

© 2009 Bahill

The effect is greater in real life • The four-seam fastball will present 80 seams per second. • There will be no flicker. • The batter will have fewer clues about the spin. • Hence he will predict less accurately, and may perceive more rising fastballs. • The two-seam fastball might flicker between 20 and 40 Hz. • If the batter could perceive this flicker, then he could tune his mental model and predict more accurately.

06/08/10

72

© 2009 Bahill

Column 6, Why (continued) • Why does a manager decide to pitch to a famous slugger or intentionally walk him? In making this decision the manager considers the score, runners on base, batting average, slugging average, etc. (BRJ 2004); column 6 (why), row 4 (technical model). • General Managers trade players in order to have a winning season within their constraints. They are motivated by their drive for success; column 6 (why), row 3 (system model). • What motivates major league baseball players? Money, prestige, pride; column 6 (motivation), row 2 (business model) • What motivates baseball team owners? Power, ego, money; column 6 (motivation), row 2 (business model) • Easton Sports and Hillerich & Bradsby give wooden bats to major league players for free. Why? To build brand image so that they can sell more of their regular sports equipment; column 6 (motivation), row 2 (business model)

06/08/10

73

© 2009 Bahill

Column 6, Why (continued) • The purpose of baseball is entertainment with two major subdivisions: television and baseball stadiums; column 6 (why), row 1 (scope). • Gould (2003) explains the intellectual complexity that causes sagacious Americans to be fascinated with baseball; column 6 (why), row 1 (scope).

06/08/10

74

© 2009 Bahill

A Zachman Framework Populated with Baseball Models Baseball 1. What 2. How 3. Where 1. Scope Equipment (Commissioner rules of Baseball)

Rules of baseball

Stadiums & TV, Leagues & divisions

2. Business model (Team Owner)

Define the strike zone

Shared use of stadiums

NCAA bat rules

Pitch tracker, Batting practice Stadium instant facilities replay Teamwork 4. Technology Ideal bat & signals model weight & for hit & (Team weight run, bunt, Manager) distribution etc. Rising fastball, 5. Detailed representation A swing, Eye CoP, MoI, movement (Scientist, Engineer, CoR strategies, Speed & Coach) spin 6. Real system One pitch Baseball (Baseball & bat Player) responses 3. System model (General Manager)

06/08/10

4. Who Revenue sharing, Reserve clause A-Rod, Olympic teams Fantasy baseball, Player contracts, 25man roster,

5. When

6. Why

Entertainment, Chronology Intellectual stimulation Season schedules

Money, Power, Pride; Free Bats

Pitching rotations

Wants a winning season

Weekly Mental statistics, Box rehearsal & scores, Sweet understandings spot

Pitch count

Intentional walks, The blame game

Predict where & when, Fielders run curved paths

Players’ physiological state, The count

Mental models, CFFF, Expect Work fast & fastball with 3change 0 count speeds

Baseball field

Major league baseball players

Pitch interval

Placement of home plate

75

Motivation for decisions

© 2009 Bahill

Lessons learned

06/08/10

76

© 2009 Bahill

Increasing detail1 • In going from top to bottom more and more detail is introduced. • The number and size of the models in each cell increases from top to bottom, perhaps 3 times per row. Thus a row 5 cell could contain 100 times the mass (money, effort, pages of documentation, lines of code, number of diagrams, etc.) of a row 1 cell. • Our tables do not show this mushrooming.

06/08/10

77

© 2009 Bahill

&RO XP Q 5 RZ

&RO XP Q 5 RZ

& 5RZ & 5

06/08/10

& 5

&RO XP Q 5 RZ

& 5 RZ & 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

&RO XP Q 5RZ

& 5RZ

& 5 RZ

& 5 RZ

& 5 RZ

& 5RZ

& 5RZ

& 5 RZ

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

& 5

78

& 5 & 5

© 2009 Bahill

Other slivers • Instead of focusing on the bat, the lower rows of column 1 could have included other equipment, such as the ball, gloves, bases, masks, and the scoreboard. • They could have included*  system state  training equipment  actions of the pitcher  teamwork  finances

06/08/10

79

© 2009 Bahill

Hierarchy • A Zachman framework is hierarchical: each cell can contain a framework of its own. We could make a framework for  Baseball  the major leagues  the Arizona Diamondbacks  Brandon Webb, etc.

06/08/10

80

© 2009 Bahill

Granularity, row 6, real system • • • • • •

What: the baseball bat How: one pitch and responses Where: the baseball field Who: major league baseball players Time: the pitch interval* Why: motivation for decisions

• Are these six things at equivalent levels of detail?

06/08/10

81

© 2009 Bahill

Granularity, row 1, scope

*

• What: Equipment rules • How: Rules of baseball, in particular the strike zone • Where: Baseball is played in stadiums and broadcast on television • Who: Commissioner of Baseball coordinates the teams; Major League Baseball Players Association and players’ agents orchestrate players’ activities • When: Chronology; decisions about interleague play, playoff structure, world series dates • Why: The purpose of baseball is entertainment

06/08/10

82

© 2009 Bahill

To help with granularity • you should consider time scales as well as perspectives and roles, e. g. • row 1 -- a few years • row 2 -- seasonal events • row 3 -- one game • row 4 -- one at-bat • row 5 -- one swing

06/08/10

83

© 2009 Bahill

Modeling generalities • All components in a model should be at the same level. • Models should only exchange inputs and outputs with other models of the same level, or maybe one level higher or lower. • Level means level of detail or level of abstraction.  This is not synonymous with Zachman’s rows. Reference: A. T. Bahill, R. Botta and E. Smith, “What Are Levels?” Proceedings of the 15th Annual International Symposium of the International Council on Systems Engineering (INCOSE), July 10-15, 2005, Rochester, NY

06/08/10

84

© 2009 Bahill

Catcher

Pitcher

Umpire

Batter

Fielder

Predict When and Where

Pitch Ball Umpire is a role that can be filled by one (for the minors in Little League) to six (for World Series games) people.

Decision

Swing Bat? [yes]

[no] [no]

Ball Hit? [yes]

Merge

There could also be decisions for fair or foul and fly or grounder.

Fork

Run to First Base Runner

Field Ball and Throw to Appropriate Base

Ball

{joinSpec = Runner OR Ball} Decide if Ball or Runner Got to the Base First

Fork

Catch Ball & Return it to Pitcher

Call Ball, Strike, Out, Walk or Other

Other could be batter hit by pitch, foul ball with two strikes, catcher interference, time-out called before pitch, etc.

Wait for Batter Give Sign to Pitcher

06/08/10

Signal Ready Join

85

Terry Bahill Summer 2007

© 2009 Bahill

Levels1 • Consider the Batter object in the activity diagram of the previous slide • We could model the state of his mind with the following attributes and states  Experience: rookie, veteran, imminent free agent  Salary: considered too low, considered too high, commensurate with earned respect  Physiology: age, health, on disabled list  Competition: other players at his position • Would this be easy to understand?

06/08/10

86

© 2009 Bahill

Levels2 • Again consider the Batter object in that activity diagram • We could model the state of his mind with the following attributes and states  Balls, strikes, outs  Speed of last pitch  Runners on base  Last signal from coach • Is this easier to understand?

06/08/10

87

© 2009 Bahill

For the batter in the activity diagram Here is a class diagram at the appropriate level

Here is a class diagram at the wrong level Batter (high level) Experience Salary Physiology Competition talkToPress talkWithAgent dealWithOwner takeDrugTest

06/08/10

Batter (low level) Count, Outs MentalModels GameSituation Signals predictWhen predictWhere swing takePitch bunt 88

© 2009 Bahill

0

1

1.1

1.2

No skip level links 1.3.1.1

06/08/10

2

No skip level links

1.3

1.3.1

1.3.2

1.3.1.2

1.3.1.3

89

1.3.3

© 2009 Bahill

Interoperability • Company models should interact with each other. • ISO AP233 is an application protocol that is creating an information model to capture the semantics needed for the interchange of information between tools. • Interoperability standards will help with reuse.

06/08/10

90

© 2009 Bahill

The right tool There is no correct modeling technique for any particular cell. For each cell, you should use which ever modeling tool is most appropriate, e. g., physical analogs, analytic equations, state machines, functional flow block diagrams, block diagrams of linear systems theory, transfer functions, state space models, differential or difference equations, object-oriented models, UML diagrams, SysML, Monte Carlo, statistical distributions, animations, mathematical programming, Markov processes, time-series models, financial models, Pert charts, Gantt charts, computer programs, use cases, mental models, tradeoff studies ... 06/08/10

91

© 2009 Bahill

Top-down or bottom-up? • We discussed the models from bottom to top.  That is the way they were derived: in basic

research, new models build on previous research. • In designing systems, better results will usually be obtained with a top-down approach. • Architects and Systems Engineers start at the top and work down. • Discipline engineers usually start at the bottom and work up. • Software engineers often start in the middle with use cases, objects and class diagrams. • This means that different people might be working at different levels. 06/08/10

92

© 2009 Bahill

Column order • A newspaper article should start with who, what, when, where, why and sometimes how, usually in that order. • A Zachman framework has a different purpose and therefore a different column order.

06/08/10

93

© 2009 Bahill

Which cells are most important? • To increase the performance of baseball players, the lower-left cells are the most important and the upper-right cells are least important.  UML diagrams are also most useful in the lower-left corner. • A CEO would be most interested in the top rows and in particular the upper-right corner*

06/08/10

94

© 2009 Bahill

On which rows should you focus? • Rows 1 and 2 are the domain of the President, CEO, board of directors and rule making organizations. • Rows 3 and 4 are owned by managers. • Row 5 tasks are performed by the worker bees.

06/08/10

95

© 2009 Bahill

Four collections of papers Number of Papers with Primary Emphasis in Each Cell 1. 2. 3. 4. 5. What How Where Who When 1. Scope 1 1 2. Business 5 1 model 3. System 5 9 4 model 4. Technology 12 34 8 model 5. Detailed 135 70 8 8 2 representation 6. Real system

06/08/10

96

6. Why

1 2 1

© 2009 Bahill

Classify this talk

*

• Given that we are studying Modeling and Simulation in the Enterprise, in which row and column should this talk be placed? • Who would pay for developing a framework such as this?

06/08/10

97

© 2009 Bahill

Purpose? • What is the purpose of a framework?  Understand an organization or a system • What is the purpose of models?  Understand or improve an existing system or organization  Create a new design or system  Control a system  Suggest new experiments  Guide data collection activities  Allocate resources  Identify cost drivers  Increase return on investment 06/08/10

98

© 2009 Bahill

To understand simulation in a particular company • The company could create a database and answer these questions for each model and simulation they have created.  Purpose? Owner? Architecture? Inputs? Outputs? Functions? Interfaces? Interacts with? Cost? Business case? Level? Who? What? When? Where? Why? How? Standard (e.g. UML, RUP, CORBA)? Zachman row? Zachman column?

• Reduce the number of classes using affinity analysis. • Abstract this into a metamodel that shows how that company does modeling and simulation.

06/08/10

99

© 2009 Bahill

Zachman and DoDAF • The Zachman framework is very useful for understanding existing enterprises. • DoDAF is intended to help document new systems of complex systems.  It is not requirements based.  It is based on operational capabilities, so it might be related to use cases.

06/08/10

100

© 2009 Bahill

DoDAF’s four views • Operational View (OV), a description of the tasks, activities, operational elements and information exchanges required to accomplish DoD missions. • Systems and Services View (SV), a description of systems and interconnections supporting DoD functions. • Technical Standards View (TV), rules governing arrangement, interaction and interdependence of system parts or elements, whose purpose is to ensure that a conformant system satisfies a specified set of requirements. • All View (AV), information pertinent to the entire architecture. AV products set the scope and context of the architecture. 06/08/10

101

© 2009 Bahill

All View (AV) products • All View (AV) products provide overarching descriptions of the entire architecture and define the scope and context of the architecture. The AV products are • AV-1 Overview and Summary Information* • AV-2 Integrated Dictionary

06/08/10

102

© 2009 Bahill

Operational View (OV) products • • • • • • • • •

06/08/10

OV-1 High Level Operational Concept Graphic OV-2 Operational Node Connectivity Description OV-3 Operational Information Exchange Matrix OV-4 Organizational Relationships Chart OV-5 Operational Activity Model OV-6a Operational Rules Model OV-6b Operational State Transition Description OV-6c Operational Event-Trace Description OV-7 Logical Data Model

103

© 2009 Bahill

Product descriptions

Table 2-1: List of Products, DoDAF version 1.5 Applicable View, Product, Product Name, General Description All View AV-1 Overview and Summary Information Scope, purpose, intended users, environment depicted, analytical findings All View AV-2 Integrated Dictionary Architecture data repository with definitions of all terms used in all products Operational OV-1 High-Level Operational Concept Graphic High-level graphical/textual description of operational concept Operational OV-2 Operational Node Connectivity Description Operational nodes, connectivity, and information exchange need lines between nodes Operational OV-3 Operational Information Exchange Matrix Information exchanged between nodes and the relevant attributes of that exchange Operational OV-4 Organizational Relationships Chart Organizational, role, or other relationships among organizations Operational OV-5 Operational Activity Model Capabilities, operational activities, relationships among activities, inputs, and outputs; overlays can show cost, performing nodes, or other pertinent information Operational OV-6a Operational Rules Model One of three products used to describe operational activity — identifies business rules that constrain operation Operational OV-6b Operational State Transition Description One of three products used to describe operational activity — identifies business process responses to events Operational OV-6c Operational Event-Trace Description One of three products used to describe operational activity — traces actions in a scenario or sequence of events Operational OV-7 Logical Data Model Documentation of the system data requirements and structural business process rules of the Operational View Systems and Services SV-1 Systems Interface Description Identification of systems nodes, systems, system items, services, and service items and their interconnections, within and between nodes Systems and Services SV-2 Systems Communications Description Systems nodes, systems, system items, services, and service items and their related communications laydowns Systems and Services SV-3 Systems-Systems Matrix, Services-Systems Matrix, Services-Services Matrix Relationships among systems and services in a given architecture; can be designed to show relationships of interest, e.g., system-type interfaces, planned vs. existing interfaces, etc. Systems and Services SV-4a Systems Functionality Description Functions performed by systems and the system data flows among system functions Systems and Services SV-4b Services Functionality Description Functions performed by services and the service data flow among service functions Systems and Services SV-5a Operational Activity to Systems Function Traceability Matrix Mapping of system functions back to operational activities Systems and Services SV-5b Operational Activity to Systems Traceability Matrix Mapping of systems back to capabilities or operational activities Systems and Services SV-5c Operational Activity to Services Traceability Matrix Mapping of services back to operational activities Systems and Services SV-6 Systems Data Exchange Matrix, Services Data Exchange Matrix Provides details of system or service data elements being exchanged between systems or services and the attributes of that exchange Systems and Services SV-7 Systems Performance Parameters Matrix, Services Performance Parameters Matrix Performance characteristics of Systems and Services View elements for the appropriate time frame(s) Systems and Services SV-8 Systems Evolution Description, Services Evolution Description Planned incremental steps toward migrating a suite of systems or services to a more efficient suite, or toward evolving a current system to a future implementation Systems and Services SV-9 Systems Technology Forecast, Services Technology Forecast Emerging technologies and software/hardware products that are expected to be available in a given set of time frames and that will affect future development of the architecture Systems and Services SV-10a Systems Rules Model, Services Rules Model One of three products used to describe system and service functionality — identifies constraints that are imposed on systems/services functionality due to some aspect of systems design or implementation Systems and Services SV-10b Systems State Transition Description, Services State Transition Description One of three products used to describe system and service functionality—identifies responses of a system/service to events Systems and Services SV-10c Systems Event-Trace Description, Services Event-Trace Description One of three products used to describe system or service functionality — identifies system/service-specific refinements of critical sequences of events described in the Operational View Systems and Services SV-11 Physical Schema Physical implementation of the Logical Data Model entities, e.g., message formats, file structures, physical schema Technical Standards TV-1 Technical Standards Profile Listing of standards that apply to Systems and Services View elements in a given architecture Technical Standards TV-2 Technical Standards Forecast Description of emerging standards and potential impact on current Systems and Services View elements, within a set of time frames

06/08/10

104

© 2009 Bahill

Systems & Services View products • • • • • • • • • • • • • 06/08/10

SV-1 System Interface Description SV-2 Systems Communications Description SV-3 Systems-Systems Matrix SV-4 Systems Functionality Description SV-5 Operational Activity to Systems Functionality Traceability Matrix* SV-6 Systems Data Exchange Matrix SV-7 Systems Performance Parameters Matrix SV-8 Systems Evolution Description SV-9 Systems Technology Forecast SV-10a Systems Rules Model SV-10b Systems State Transition Description SV-10c Systems Event-Trace Description SV-11 Physical Schema 105

© 2009 Bahill

Technical Standards View (TV) • The TV products define technical standards, implementation conventions, business rules and criteria that govern the architecture. The TV products are • TV-1 Technical Standards Profile • TV-2 Technical Standards Forecast

06/08/10

106

© 2009 Bahill

DoD Framework Products AV-1

OV-1

OV-4

OV-5

OV-6c OV-6b

OV-2

OV-6a OV-3 OV-7

Legend Documents

SV-11

Models

SV-1

SV-6

Matrices Forecast

TV-1

SV-2

Scenario Data

TV-2 SV-10c

Rules

SV-10b SV-3

SV-4

SV-5

SV-7

SV-8

SV-9

SV-10a AV-2

06/08/10

107

© 2009 Bahill

Compared to our design process, DoDAF is missing • • • • • • • •

06/08/10

Document 2: Customer Requirements Document 3: Derived Requirements Document 4: System Test and Validation Document 5: Concept Exploration Tradeoff Matrix and Sensitivity Analysis Schedule and Budget Technical Analysis Risk Management

108

© 2009 Bahill

Compared to DoDAF, our design process is missing • Operational OV-6a Operational Rules Model • Operational OV-6b Operational State Transition Description • Operational OV-6c Operational Event-Trace Description • Technical Standards TV-1 Technical Standards Profile • Technical Standards TV-2 Technical Standards Forecast • Wymore includes technical standards in the Technology Requirement. 06/08/10

109

© 2009 Bahill

Purpose1 • DoDAF: ensue that architectural descriptions can be compared and related across organizational boundaries  In a net-centric warfighting environment, there is a great need for integration and interoperability in order to achieve warfighting capabilities.  Describe the operation of interrelated systems • Zachman: provide a basic structure that supports organization, integration, development, and management of a set of architectural representations

06/08/10

110

© 2009 Bahill

Purpose2 • Neither DoDAF nor Zachman are design processes. • They can help document existing designs. • DoDAF is intended for unprecedented large systems of complex systems. • Zachman is intended for enterprises.

06/08/10

111

© 2009 Bahill

Strengths • DoDAF:  organized  descriptive  extensive product relationships  identifies duplicate functions  accepted DoD standard • Zachman  organized  intuitive  breath of coverage

06/08/10

112

© 2009 Bahill

Drawbacks • DoDAF  Cumbersome  Inflexible  Ponderous and hard to learn • Zachman  Elementary  Does not prescribe  design tradeoffs  design rationale  documentation of architecture decisions

06/08/10

113

© 2009 Bahill

Zachman summary1 • To understand an enterprise, you should have a model in every cell of a Zachman framework • For each cell, use which ever modeling tool is most appropriate • Because a framework is hierarchical, cells in the lower rows will have many models • Cells in the top rows should trace to the organization’s vision and mission statements • Cells in a row should be at similar levels of detail (granularity)

06/08/10

114

© 2009 Bahill

Zachman summary2 • Filling in a Zachman framework will help convince your customer that you understand the system to be designed and built. • Frameworks help promote integration of models. • Frameworks help show the flow of money.

06/08/10

115

© 2009 Bahill

06/08/10

116

© 2009 Bahill

Seminar materials • A ball • Perhaps the two-seam and four-seam video

06/08/10

117

© 2009 Bahill

How to print (in old MS Office) • • • • • • •

06/08/10

To print this file, do this one time. Pick the View menu Chose Color/grayscale Grayscale Settings Light grayscale Close grayscale view

118

© 2009 Bahill