©2012 Emirates. All Rights Reserved.

Reflecting Stakeholder Perspectives in Architecture Reviews

Simon Field

Emirates Group

University of Glamorgan Business School

About Emirates GroupAbout Emirates GroupAbout Emirates GroupAbout Emirates Group

In 2010/11 financial year, the Emirates Group announced record net profits of US$ 1.5 billion

The Emirates Group’s extensive and diverse international portfolio includes Emirates airline, Dubai's 52 year-old household name dnata, Emirates SkyCargo, Emirates Destination and Leisure Management and Skywards. Together with a number of joint venture businesses, the Group comprises more than 50 brands, employing 57,000 people.

Architecture Reviews

Stakeholder Model

Quality Model Risk Model

Almost Certain 5 5 10 15 25 30

Likely 4 4 8 12 20 24

Possible 3 3 6 9 15 18

Unlikely 2 2 4 6 10 12

Rare 1 1 2 3 5 6

1 2 3 5 6

Negligible Minor Moderate Major Catastrophic

Impact

Likelihood

Impact x Likelihood = Exposure

David van Dantzig

Delta Works Project 1953

RiskRiskRiskRisk

Why risk?Why risk?Why risk?Why risk?

A common business language

Impact

of failing to achieve a scenario

Likelihood

of a solution option failing to achieve a scenario

Stakeholders can consider mitigation options

QualityQualityQualityQuality

Power

Legitimacy

Urgency

Dormant

Stakeholder

Latent

Discretionary

Stakeholder

Latent

Dominant

Stakeholder

Expectant

Definitive

StakeholderDangerous

Stakeholder

Expectant

Dependent

Stakeholder

Expectant

Demanding

Stakeholder

Latent

source: Mitchell, Agle & Wood, Academy of Management Review Vo. 22, No. 4, p874

StakeholdersStakeholdersStakeholdersStakeholders

Stakeholder Analysis

Scenario Development

Classify according to Quality Model

For each scenario:

Agree Impact for each Scenario

Assess level of stakeholder interest

Validate using ‘heat map’

Tradeoff Analysis

By Quality Characteristic

By Stakeholder

Explore by Scenario

Cost Benefit Analysis

Architecture Reviews at EmiratesArchitecture Reviews at EmiratesArchitecture Reviews at EmiratesArchitecture Reviews at Emirates

Power

Legitimacy

Urgency

Dormant

Stakeholder

Latent

Discretionary

Stakeholder

Latent

Dominant

Stakeholder

Expectant

Definitive

StakeholderDangerous

Stakeholder

Expectant

Dependent

Stakeholder

Expectant

Demanding

Stakeholder

Latent

source: Mitchell, Agle & Wood, Academy of Management Review Vo. 22, No. 4, p874

1 2

3

4

5

6

7

8

9

10

11

12

13

14

StakeholdersStakeholdersStakeholdersStakeholders

ScenariosScenariosScenariosScenarios

Functionality Reliability Usability Efficiency Maintainability AdaptabilityOverall Average Rank

Number of scenarios -> 31 2 9 5 2 3 52

NCC Cmd & Ctrl 0.50 0.25 0.28 0.50 0.25 0.33 0.43 4

NCC Ops 0.71 0.25 0.33 0.40 0.25 0.33 0.56 1

Govt Cmd & Cntrl 0.18 0.25 0.00 0.20 0.00 0.00 0.13 9

DOCC Cmd & Ctrl 0.61 0.25 0.28 0.40 0.25 0.33 0.49 3

DOCC Ops 0.65 0.25 0.39 0.40 0.50 0.33 0.54 2

External suppliers 0.11 0.25 0.00 0.20 0.00 0.00 0.10 11

Govt Ops 0.06 0.25 0.00 0.20 0.25 0.00 0.08 12

Front line service 0.32 0.25 0.39 0.40 0.50 0.33 0.35 5

Front line service (mob) 0.27 0.25 0.50 0.40 0.50 0.33 0.34 6

Application Owners 0.10 0.25 0.00 0.40 0.00 0.00 0.11 10

EG-IT 0.15 1.00 0.11 0.70 0.25 0.17 0.23 8

Steering Group 0.42 0.00 0.17 0.10 0.00 0.00 0.29 7

Does it look right?Does it look right?Does it look right?Does it look right?

Now we know what we wantNow we know what we wantNow we know what we wantNow we know what we want

Reuse, Buy, Build?

How?

Define competing solution options

A sufficiency of architecture

���� 13.50 ���� 4.50 7.50 ���� 4.50

���� 6.00 ���� 6.00 ���� 3.00 4.50

���� 3.00 ���� 15.00 9.00 13.50

���� 6.50 ���� 12.00 8.00 8.50

���� 3.00 ���� 9.00 6.00 ���� 9.00

� 0.00 � 0.00 � 0.00 � 0.00

Aggregate Risk Burden 32.00 46.50 33.50 40.00

Efficiency

Maintainability

Adaptability

Functionality

Reliability

Usability

By characteristic:

���� 6.86 ���� 9.57 6.93 8.20

���� 6.75 ���� 6.75 ���� 4.50 5.25

���� 6.00 ���� 10.50 6.75 8.25

Demanding

Discretionary

Dormant

Definitive

Dependent

Dangerous

Dominant

By stakeholder type:

Risk Tradeoff AnalysisRisk Tradeoff AnalysisRisk Tradeoff AnalysisRisk Tradeoff Analysis

Ability to develop stateless

application with no state

maintenance in the servers side

beyond web application Major Almost certain 5 Almost certain 5 Almost certain 5 Almost certain 5

Ability to support Externalization

for native transports Major Almost certain 5 Almost certain 5 Possible 15 Possible 15

Ability to support user

management out of the box Major Almost certain 5 Almost certain 5 Possible 15 Possible 15

Ability to support chat kind of

application requirements Moderate Almost certain 3 Almost certain 3 Possible 9 Likely 6

Ability to support code master

Major Almost certain 5 Almost certain 5 Possible 15 Possible 15

Ability to support list of values

Major Almost certain 5 Almost certain 5 Possible 15 Possible 15

Ability to shutdown, start , stop

Major Possible 15 Likely 10 Likely 10 Possible 15

Ability to integrate , use or

availability of String, Date

Moderate Almost certain 3 Almost certain 3 Almost certain 3 Almost certain 3

Ability to support procedures,

functions and ref cursors Moderate Almost certain 3 Almost certain 3 Almost certain 3 Almost certain 3

Ability to shutdown functions,

channels and so on. Ability to

force user out of the system Moderate Almost certain 3 Almost certain 3 Possible 9 Possible 9

Ability to support contract

(interface) based development.

Ability to separate interfaces

from implementation Major Almost certain 5 Almost certain 5 Almost certain 5 Almost certain 5

Ability to support two factor

Moderate Possible 9 Possible 9 Possible 9 Possible 9

Support for Thick client

Major Possible 15 Almost certain 5 Almost certain 5 Possible 15

Support IT security policies Catastrophic Almost certain 6 Almost certain 6 Likely 12 Likely 12

Major Almost certain 5 Almost certain 5 Possible 15 Possible 15

ability to support java standalone

Moderate Almost certain 3 Likely 6 Unlikely 12 Almost certain 3

By individual scenario:

One-off benefit 0

Annual benefit 30,000

Number of years 1

Maximum benefit 30,000

Development cost

Annual cost

Exit difficulty

Total

Estimated Benefit

Net Return

Scenario Impact Weight Maximum Benefit Likelihood Estimated Benefit Likelihood Estimated Benefit Likelihood Estimated Benefit Likelihood Estimated Benefit

Quick Journey Major 5 4839 Almost certain 4839 Unlikely 484 Likely 3871 Possible 2419

Predictable journey to work - minimal

delays Moderate 3 2903 Likely 2323 Likely 2323 Almost certain 2903 Almost certain 2903

Comfortable journey to work Moderate 3 2903 Almost certain 2903 Rare 0 Possible 1452 Unlikely 290

Minimise cost Minor 2 1935 Unlikely 194 Likely 1548 Possible 968 Almost certain 1935

No accidents Catastrophic 6 5806 Likely 4645 Almost certain 5806 Almost certain 5806 Almost certain 5806

Flexibility Moderate 3 2903 Almost certain 2903 Possible 1452 Likely 2323 Possible 1452

Ability to share car with family Moderate 3 2903 Rare 0 Almost certain 2903 Possible 1452 Almost certain 2903

Predictable journey home - minimal

delays Moderate 3 2903 Likely 2323 Likely 2323 Almost certain 2903 Likely 2323

Comfortable journey home Moderate 3 2903 Almost certain 2903 Rare 0 Possible 1452 Rare 0

Total Benefits

619.8% 547.6%

16,839 23,129

513.5% 853.9%

20,032

19,832 14,239 19,359 17,932

23,032

0

3,770

0

2,100

Low Low

3,770 2,100

3,200

Low

3,200

0

2,600

Low

2,600

Car Bus / Metro Car / Metro Lift / Metro / Bus

Cost Benefit AnalysisCost Benefit AnalysisCost Benefit AnalysisCost Benefit Analysis

Architecture Reviews

Stakeholder Model

Quality Model Risk Model

©2012 Emirates. All Rights Reserved.

Questions

simon.field@emirates.com