Shields Capstone Project

Decision Maker Profile Determination and Decision Modeling to Project Most Likely

Decision Outcomes

April 15, 2011

Matt ShieldsJoe Hartman advising

2

Introduction• Have you ever made a bad first impression at a business

meeting?• Did you say something that the customer took completely

out of context?• Did you glance over the topic that the customer found

most important?

How can you avoid this mistake the next time…?

3

Purpose

• The purpose of this project is to– Develop a methodology to determine decision profiles – Develop a simulation model to forecast decision

outcomes using decision profile elements– Interpret results to understand decision tendencies and

improve decision-making

4

Project Overview

• Most decision support tools determine how a decision-maker should decide -- this project seeks to determine how a decision-maker will decide

• By determining a person’s decision profile and modeling the decision calculus, one may be able to forecast decision outcomes by a person, a group of people, or an organization

6

Key Tasks

• Develop methodology to determine a person’s decision profile using biographical, cultural, and behavioral information.

• Assign decision element values for information types as probabilities for simulation and modeling.

• Develop an ARENA Monte Carlo simulation that generates decision element values based upon a decision profile and forecasts outcomes by comparing decision element values to constraints from a decision scenario.

• Interpret the results to determine decision-making tendencies and recommendations.

Methodology

• Capability – Capable of making a rational decision?• Method – How does entity make decisions?• Bias – Does the entity make rational decisions?

MethodologyCapability - Are you able to make rational decisions?

Capacity Responsibility Achievement Power Status

Intelligence Dependability Proven success Reward

Verbal faculty Initiative Productivity Coercive

Originality Persistence Work ethic Expert

Self Confidence Aggressiveness Charismatic

Judgment Self Confidence Legitimate Authority

Cognition Desire to Excel

Methodology

Method - How do you make decisions?

Risk Aversion Efficiency Deliberation Term Participation Involvement

Uncertainty Avoidance

Satisficing Reactivity Favor Near Term

Active

Risk shift for group decision

Prioritization Fully researched Favor Long Term

Sociable

Stability Decentralize Adaptor

Security Cooperation

Conflict avoidance

Need for consensus

Methodology

Biases - Do you make rational decisions?

Conflict Reaction

Values Disposition Personality Emotional Intelligence

Perceptions Judgments

Avoiding Moral Affectivity Extroversion Experience Race

Accommodating Political Pragmatism Tolerance Historical perspective

Sex

Competing Religious Intuition Conscientious Neuroticism Lifestyles

Compromising Philosophy Power needs Openness to experience

Tension National origin

Collaborating Self interest Cognition SecurityStability

Age

Ethical Self Confidence Framing

11

Decision Element Assignment

• Derive values for the decision elements to determine how that information group performs in that decision element relative to the general populous

• More “art” than “science”• Based upon research on information groups

– Myers Briggs Type Analysis– Hofstede’s Dimensional Analysis– Sex– Age– Achievement (Education and Work Status)

12

Decision Element AssignmentPersonality

MBTI ISTJ ISFJ ESTJ ESFJ ISTP ISFP ESTP ESFP INFJ INFP ENFJ ENFP INTJ INTP ENTJ ENTPCapability Capacity 0.43 0.43 0.55 0.55 0.33 0.33 0.53 0.53 0.60 0.35 0.60 0.55 0.68 0.55 0.68 0.63

Intelligence Verbal faculty 0.30 0.30 0.70 0.70 0.30 0.20 0.80 0.80 0.50 0.30 0.80 0.60 0.20 0.70 0.40 0.80Originality 0.20 0.20 0.20 0.20 0.40 0.50 0.50 0.50 0.80 0.50 0.60 0.80 0.80 0.70 0.80 0.80

Self Confidence 0.40 0.40 0.50 0.50 0.40 0.40 0.60 0.60 0.30 0.40 0.40 0.60 0.90 0.50 0.70 0.70Judgement 0.80 0.80 0.80 0.80 0.20 0.20 0.20 0.20 0.80 0.20 0.60 0.20 0.80 0.30 0.80 0.20Cognition

Responsibility 0.52 0.50 0.55 0.58 0.40 0.22 0.63 0.40 0.50 0.63 0.55 0.65 0.68 0.53 0.68 0.58Dependability 0.90 0.80 0.70 0.70 0.30 0.10 0.30 0.20 0.60 0.70 0.70 0.40 0.70 0.50 0.70 0.50

Initiative 0.40 0.50 0.50 0.60 0.50 0.20 0.80 0.50 0.60 0.70 0.60 0.80 0.40 0.50 0.50 0.50Persistence 0.70 0.60 0.60 0.60 0.40 0.20 0.70 0.30 0.40 0.80 0.50 0.60 0.70 0.60 0.70 0.60

Aggressiveness 0.30 0.20 0.60 0.60 0.30 0.20 0.80 0.40 0.40 0.50 0.40 0.80 0.70 0.50 0.80 0.60Self Confidence 0.40 0.40 0.50 0.50 0.40 0.40 0.60 0.60 0.30 0.40 0.40 0.60 0.90 0.50 0.70 0.70Desire to Excel 0.40 0.50 0.40 0.50 0.50 0.20 0.60 0.40 0.70 0.70 0.70 0.70 0.70 0.60 0.70 0.60Achievement 0.60 0.53 0.57 0.63 0.40 0.20 0.57 0.33 0.50 0.60 0.57 0.53 0.70 0.57 0.67 0.53Proven success 0.40 0.40 0.60 0.60 0.40 0.20 0.70 0.40 0.50 0.40 0.60 0.60 0.60 0.50 0.70 0.60Productivity 0.70 0.60 0.50 0.70 0.40 0.20 0.30 0.30 0.60 0.60 0.60 0.40 0.80 0.60 0.60 0.40Work ethic 0.70 0.60 0.60 0.60 0.40 0.20 0.70 0.30 0.40 0.80 0.50 0.60 0.70 0.60 0.70 0.60

PowerStatus 0.44 0.40 0.64 0.48 0.38 0.20 0.64 0.60 0.46 0.40 0.58 0.64 0.44 0.38 0.58 0.54Reward 0.50 0.60 0.50 0.50 0.30 0.20 0.60 0.80 0.80 0.60 0.80 0.60 0.20 0.20 0.20 0.20Coercive 0.50 0.20 0.60 0.20 0.30 0.20 0.80 0.50 0.20 0.40 0.20 0.60 0.50 0.50 0.80 0.50Expert 0.60 0.70 0.50 0.30 0.80 0.20 0.30 0.30 0.50 0.50 0.50 0.50 0.80 0.80 0.50 0.60

Charasmatic 0.20 0.20 0.80 0.80 0.20 0.20 0.90 0.90 0.50 0.30 0.80 0.80 0.20 0.20 0.60 0.80Legitimate Authority 0.40 0.30 0.80 0.60 0.30 0.20 0.60 0.50 0.30 0.20 0.60 0.70 0.50 0.20 0.80 0.60

13

Decision Element AssignmentCulture

• Hofstede compiled large database of cultural information and determined patterns in five dimensions– Uncertainty avoidance– Power Distance– Collectivism– Masculinity / femininity– Short or long term

• Many decision elements relate directly to these dimensions• Hofstede’s dimension analysis on scale of 0 – 100

– Divide by 100 to obtain value as a probability

14

Decision Element AssignmentGender / Age / Achievement

• Little “uncontested” research on gender, but most agree that, in general– Men make decisions more efficiently than women– Women are more risk averse and involve more people in decisions– 0.4 or 0.6

• Age – As age goes up, so does experience, deliberation, risk aversion and term orientation

• Achievement – Education level and work force position relate directly to levels of capacity, responsibility, achievement and power status

The ProblemYou have learned of an opportunity to manufacture two new products, a pressure sensor and a dual pressure / temperature sensor. The market for each product is known if the products can be successfully developed. However, there is some possibility that your R&D department will not be able to successfully develop them. Production profit of $600,000 would be realized from selling the dual sensor and of $400,000 from selling the pressure sensor. A production profit of $800,000 would be realized from selling both (full profits not realized due to capacity constraints). All profits account for production cost but do not include development cost. If development is unsuccessful for a product, then there will be no sales, and the development cost will be totally lost. Development cost would be $300,000 for the dual sensor and $100,000 for the pressure sensor. You are the production manager and must decide whether to develop the pressure sensor, the dual sensor, both or neither. The probability of development success is somewhat uncertain, although pressure sensor development success is at least 50%. Dual sensor development relies on successful development of the pressure sensor and is at least 50% of the development success of the pressure sensor.

15

Decision Tree

16

-$400K

-$300K

-$100K

$0

Alt 4 – Neither. E(x) = $0$0

Success ($800K) P >25%

Failure ($0)


Failure ($0)


Failure ($0)

$400K

-$400K

$300K

-$300K

$200K

-$100K

Alt 1 – Both. E(x) > -$100K

Alt 2 – Dual. E(x) > -$150K

Alt 3 – Pressure. E(x)>$100K

Figure 1 – Decision Tree

Scenario 1 (Routine Decision)• You are the production manager for a profitable plant that is doing better

than its competitors. • It is uncertain if your plant has a sustainable competitive advantage to

continue this trend in the long term.• This development opportunity will be available for the next several

months. (Time availability is not a factor).• The future market for these products is unknown. • Your production team recommends manufacturing both products since they

are both profitable. (Decision point for participation)• Your previous development decisions have all been profitable.• Your plant manufactures 100’s of products – you make these types of

decisions on a weekly basis.

17

Scenario 2 (Important decision)• You are the production manager for a profitable plant that is lagging its

competitors. • It is uncertain if your plant has a sustainable competitive advantage.• This development opportunity will be available for the next several months.

(Time availability is not a factor).• The development cycle is one year, and you will not realize these profits

until next year. However, the profits should be sustainable for several years.

• Your production team recommends manufacturing both products since they are both profitable. (Decision point for participation)

• Your previous development decisions have all been profitable.• Your plant manufactures 100’s of products – you make these types of decisions

on a weekly basis.

18

Scenario 3 (Important decision)

• You are the production manager for a profitable plant that is lagging its competitors.

• It is uncertain if your plant has a sustainable competitive advantage.• This development opportunity is only available if you make the

decision today. • The future market for these products is unknown. • Your production team recommends manufacturing both products since

they are both profitable. (Decision point for participation)• This is your first production decision.

19

Scenario 1 (Routine Decision)

20

MakerEnt er Decision

Prof ileAssign Decision Tr ue

False

De c i s i o n ?DM Ca p a b l e o f Ra t i o n a lTr ue

False

De c i s i o n ?DM Ne e d to M a k e Tr ue

False

d e c i s i o n ?Do e s DM m a k e

sensorRecord Pressure

Dispose Pressure

Tr ue

False

Gro u p De c i s i o n ?

do not hingRecord delay or Dispose Not hing

Tr ue

False

Gro u p De c i s i o n ? ? Tr ue

False

d e c i s i o n ? ?Do e s DM m a k e

Record both Dispose bot h

d e c i s io n ?DM wa n t to m a k e Tr ue

False

Tr ue

False

d e c i s i o n s ? ? ?Do e s DM m a k e Record dual Dispose dualTr ue

False

Is h e re a l l y th i s d u m b ?

Tr ue

False

Ra n d o m c h o i c e

0

0

0

0

0

0

00

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0


21

M a k e rEn te r De c i s i o n

Pro f i l eAs s i g n De c i s i o n Tr ue

False

De c i s io n ?DM Ca p a b le o f Ratio n a l Tr ue

False

Do e s DM m a k e d e c is i on ?

s e n s o rRe c o rd Pre s s u re

Di s p o s e Pre s s u re

Tr ue

False

Grou p Dec is io n ?

n o th i n gRe c o rd d e l a y o r d o Di s p o s e No th i n g

Tr ue

False

Gro u p De c is io n? ? Tr ue

False

de c is io n ? ?Do es DM m a k e

Re c o rd b o th Di s p o s e b o th

Tr ue

False

de c is io n ?? ?Do es DM m a k e

Re c o rd d u a l Di s p o s e d u a l

Tr ue

False

Lo n g te rm ?

M id te rm ?Tr ue

False

Tr ue

False

d e c is io n?Do e s DM _ m i d m a k e

Tr ue

False

d e c is io n?Do es DM _ s h o rt m ak e

Tr ue

False

Lo ng te rm ? ?

M id te rm ? ?Tr ue

False

Tr ue

False

de c is io n ??Do e s DM _ m id m a k e

de c is io n ??Do es DM _ s h o rt m a k eTr ue

False

Is he re a l l y th i s d u m b ?Tr ue

False

Tr ue

False

Ra nd om s e le c tio n

Lo ng Te rm ? ? ?Tr ue

False

Tr ue

False

M id Te rm ? ? ? Tr ue

False

de c is ion ?? ?Do e s DM _ m id m a k e

Tr ue

False

de c is ion ?? ?Do es DM _ s ho rt m a k e

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0


22

M a k e rEn te r De c i s i o n

Pro f i l eAs s ig n De c i s i o n Tr ue

False

De c is i on?DM Ca p ab le o f Ra ti o na lTr ue

False

dec is i on?DM Tim e to m a k e Tr ue

False

dec is i on?Does DM m a k e

s e n s o rRe c o rd Pre s s u re

Di s p o s e Pre s s u re

Tr ue

False

Gro up Dec is ion ?

d o n o th i n gRe c o rd d e l a y o r Di s p o s e No th i n g

Tr ue

False

Group De c is ion ?? Tr ue

False

de c is io n??Does DM m a k e

Re c o rd b o th Di s p o s e b o th

Is h e re a l l y th i s du m b?Tr ue

False

Tr ue

False

de c is io n s ? ? ?Doe s DM m ak e

Re c o rd d u a l Di s p o s e d u a l

Tr ue

False

Ran dom c ho ic e

0

0

0

0

0

0

00

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Results

23

AttributesThe Field Marshall

The Free Spirit

Chinese Field

Marshall

Female Field

MarshallYoung Field

MarshallPersonality ENTJ ISFP ENTJ ENTJ ENTJCulture USA USA China USA USAGender Male Male Male Female MaleAge 40 40 40 40 18Education MBA MBA MBA MBA HS

Scenario 1Both 30.3% 13.9% 26.5% 47.3% 33.4%Dual Sensor 0.2% 0.5% 0.3% 0.1% 1.6%Pressure Sensor 45.2% 7.8% 55.8% 28.8% 20.1%Delay or do nothing 24.3% 77.8% 17.4% 23.8% 45.0%



Key Findings• For routine decisions, proposals should emphasize why the manager has a

fiduciary responsibility to continue process improvement since the manager may be reluctant to make a decision.

• For important, but routine decisions, the manager is more likely to choose the rational outcome. Proposals should emphasize why that rational outcome is the best outcome based upon analysis.

• For urgent decisions, managers are more likely to side with the group. Proposals to the production manager should discredit the group logic (if wrong), then provide analysis for the optimal outcome.

• For less rational managers, proposals need to be tailored to discredit the group logic (if wrong) and why the manager should continue process improvement using qualitative (less technical) reasoning.

24

Conclusion

• This project provides an initial framework to determine decision profiles and forecast outcomes.

• The results of the simulation runs are useful and can be interpreted to influence senior leader decision.

• More detailed analysis is needed to relate cognitive and behavioral tendencies within the information groups to discrete decision elements for decision profile determination.

• No simulation model fits all – an operations researcher must analyze each decision scenario to tailor the model to the decision space.

25

Backups

26

Methodology

27

Arrival

Entity arrives with decision profile•MBTI•Hofstede’s Dimension•Sex•Age•Achievement

DP1

Decision Point 1Is entity capable of making any rational decisions?

No

Yes DP2

Decision Point 2Is entity capable of making a rational decision given the decision space?

No

Yes DP3

Decision Point 3Will entity make a rational decision given problem dynamics and decision profile?

No

Yes

Strong bias?

Yes

No

Rational Outcome

Biased Outcome

Alternate Outcome

Alternate Outcome

P=.25 P=.25 P=.25 P=.25

28

Applicable Theories• Hofstede’s Dimensional Analysis• Myers Briggs Type Indicators• Utility Theory• Decision Trees• Probabilistic Risk Analysis• Monte Carlo Simulation

29

Challenges• Translating subjective decision-making processes into

measureable operations research.– Marketing research and Hofstede’s dimensional analysis– Additional criteria for the individual are harder to measure

(personality / achievement)

• Decision calculus database creation and validation– 12,800 individual profiles– City-sized (Houston) distribution needed to validate

30

Timeline• January 9-10, 2010: Presentation of proposal and

plan; Written project proposal and plan due to advisor

• February 15, 2010: Progress Report #1 due to advisor

• March 15, 2010: Progress Report #2 due to advisor

• April 17-18, 2010: Written final project report due; Final project presentation

Documents

Shields Capstone Project