Dr.

Yos Sunitiyoso
School of Business & Management
Institut Teknologi Bandung
 AHP basic principles:
• Hierarchy Formulation
• Setting priority
• Testing consistency

Introduced by Thomas L. Saaty (1980) is an effective tool for dealing with complex decision-making,
and may aid the decision maker to set priorities and make the best decision.
By reducing complex decisions to a series of pairwise comparisons, and then synthesizing the
results, the AHP helps to capture both subjective and objective aspects of a decision.

2
(Saaty, 2008)
o The AHP is a theory of measurement through pairwise comparisons
o It relies on the judgements of experts to derive priority scales among alternatives.
o The comparisons are made using a scale of absolute judgements that represents
how much more one element dominates another with respect to a given attribute.
o The priority scales are synthesized by multiplying them by the priority of their parent
nodes and adding for all such nodes.

(Yang & Shi, 2002)

o The AHP approach involves decomposing a complex problem into a set of
components organized in a multilevel hierarchic form.
o A salient feature of the AHP is to quantify decision makers' subjective judgments by
assigning corresponding numerical values based on the relative importance of
criteria.
o A conclusion can be reached by synthesizing the judgments to determine the overall
priorities of alternatives.

3
Selected MCDM Method :
The AHP is the most suitable MCDM method for
the areas of application related to performance-
type problems, resource management, corporate
policy and strategy, public policy, political
strategy, and planning.

AHP’s Pros and Cons :

The advantages are easy to use; scalable; hierarchy
structure can easily adjust to fit many sized problems;
and not data intensive.
The disadvantages are problems due to
interdependence between criteria and alternatives;
may lead to inconsistencies between judgment and
ranking criteria; and rank reversal.
4
(Velasquez & Hester, 2013)
Example of Hierarchy Structure

Goal Selecting business partner

Criteria Intention Technical Capability Non-Technical Capability Experiences in Selling and

support ICT

Financial Hardware (non

Pre-sales
network)

Sub Criteria Project

Logistic/Warehouse
Implementation Software
and spare

Service Support Hardware

Investment for
(network)
resources

Sales Team

Alternatives Company A Company B Company C

Example of Hierarchy Structure
AHP - Identify best renewable energy power plant development in Indonesia
(Pulungan & Sunitiyoso, 2017)

· Investment cost
· O & M cost Geothermal Power
Economical Plant
· fuel cost
· Economic Life
Wind Power Plant
· Energy production capacity
To identify the · capacity factor
best renewable Technical · reliability Biomass Power Plant
resource for · technology maturity
electricity in · distance location
Indonesia Hydro Power Plant
· CO2 emission
Enviromental · impact on ecosystem
· Land use Solar PV Power Plant
· social acceptability
Social · job creation Mycro Hydro Power
· social benefit Plant

GOAL Criteria Sub-Criteria Alternatives

Example of Hierarchy Structure

LEVEL 1: DECISION GOAL

DECISION GOAL (To obta i n project portfol i o compos i ti on wi th ma x. benefi ts to the compa ny goa l s )

LEVEL 2: A B C D
CRITERIA INTERNAL INNOVATION
FINANCIAL CUSTOMER
PROCESS AND LEARNING

LEVEL 3: 1 5 9 STRATEGIC 13
INVESTMENT PROFITABILITY EXTERNAL DEPENDENCIES NEW TECHNOLOGY
SUBCRITERIA ALIGNMENT

2 6 10 14
INVESTMENT RETURNS BUSINESS OPPORTUNITIES INTERNAL DEPENDENCIES LESSON LEARNED

3 7 11 15
INVESTMENT PERIODS LOCAL ECONOMIC IMPACTS RESOURCE DEMAND NONFINANCIAL BENEFITS

4 8 12
FINANCIAL RISKS EXTERNAL RISKS INTERNAL RISKS

LEVEL 4: Hi gh Profi ta bi l i ty 1 Hi gh Dependences 5 Hi gh Al gnment 9 Hi gh Sca l e 13

RATINGS / GRADES Medi um Profi ta bi l i ty Medi um Dependences Medi um Al i gnment Medi um Sca l e
Low Profi ta bi l i ty Low Dependences Low Al i gnment Low Sca l e

Hi gh Returns 2 Hi gh Opportuni ti es 6 Hi gh Dependences 10 Hi gh Succes s Ra te 14

Medi um Returns Medi um Opportuni ti es Medi um Dependences Medi um Succes s Ra te
Low Returns Low Opportuni ti es Low Dependences Low Succes s Ra te

Long Peri od 3 Hi gh Impa ct 7 Hi gh Dema nd 11 Hi gh Benefi ts 15

Medi um Peri od Medi um Impa ct Medi um Dema nd Medi um Benefi ts
Short Peri od Low Impa ct Low Dema nd Low Benefi ts

Hi gh Ri s k 4 Hi gh Ri s k 8 Hi gh Ri s k 12
Medi um Ri s k Medi um Ri s k Medi um Ri s k
Low Ri s k Low Ri s k Low Ri s k

LEVEL 5:
ALTERNATIVES
PROJECT PORTFOLIO COMPONENTS 7
 AHP Approach in MCDM Model on
Company Investment Portfolio Prioritization (Widagdo & Sunitiyoso, 2020)
Decision Goal Decision Goal

Criteria
(n number)
Criteria #1 Criteria #2 ..... Criteria #n

Sub-criteria #1.1 Sub-criteria #2.1 Sub-criteria #n.1

Sub-criteria Sub-criteria #1.2 Sub-criteria #2.2 Sub-criteria #n.2

(x, y, z number)

...
...

...
Sub-criteria #1.x Sub-criteria #2.y Sub-criteria #n.z

Alternatives
(t number)
Alternative #1 Alternative #2 ..... Alternative #t

The AHP Steps

Model Development Model Implementation Model Outcome

• Goal, criteria, alternatives, • Pairwise comparison, • Selection of the alternative
alignment, prioritization quantitative rating, weight and with the highest priority index
model, expert selection, rating rating multiplication
range 8
 Model Development: Investment Prioritization

Decision Goal To Obtain Investment Portfolio which Maximize Company Value and Benefit to the Stakeholders

Criteria Strategy and Governance Marketing and Sales Project Management Financial Feasibility

Strategic Alignment & Urgency Market Capability Improvements Project-attributed Risks Investment Amount: Capex

Investment Profitability: Net

Sub-criteria Internal Process Improvements Product and Quality Leadership Duration of Execution
Present Value

Investment Returns:
HSE, Legal, and Compliance Corporate Reputation Organizational Readiness
Internal Rate of Return

Investment Periods:
Technological Sufficiency
Pay Back Period

Alternatives
The Molten Iron Project The Black Steel Project The White Steel Project

9
 Scales of pairwise comparison judgement (Saaty,1990)
The purposes of the pairwise comparison survey
for all factors (criteria, subcriteria and ratings) are
to:
o Elicit the pairwise comparison judgements
throughout the hierarchy structure of the AHP,
i.e. among the criteria with respect to the
decision goal; among the subcriteria with
respect to their parent criterion; and among the
ratings or grades with respect to each
associated subcriterion.
Pairwise comparison survey (sample)
o Give feed back to the respondents associated
with consistency of their judgements and agree
with each other to get the improved judgements
within tolerable consistency.
o The pairwise comparison is performed by
comparing the relative importance between
two factors (criteria, subcriteria or ratings) which
are measured according to a numerical scale
from 1 to 9. 10
Example List of Respondents
No. Position Years of Experience Quantity
1 Vice President of Transformation Office 10 1
2 Vice President of Project Development 10 1
3 Downstream Project Integration Manager 10 1
4 Head of Project Contract Preparation 8 1
5 Senior Specialist Project Proposal Analysis 8 1
6 Specialist Project Proposal Analysis 6 2
7 Specialist Project Integration 6 2
8 Specialist Project Contract Preparation 4 1
9 Specialist Project Planning 4 1
10 Specialist Project Inspector 4 1
11 Officer Project Planning 2 1
TOTAL 13

11
The importance level of respondents is to provide
weighting of the respondents in group decision
making by using the formula of weighted geometric
mean for aggregation of individual priorities (AIP).

In this case, the respondents are grouped by four

level of importance, i.e. Level 4 (the most importance)
until Level 1 (the lowest importance), as defined
below:

o Level 4: President/ General Manager

o Level 3: Vice Presidents

o Level 2: Managers

o Level 1: Project Leaders, Team Leaders, and

Senior Analysts/ Engineers.
12
 Model Implementation: Pairwise Comparison
1 Survey to fill the form
9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9
Criteria
Strategy and Governance 1 2 Marketing and Sales
5 Repeat to All Sub-criteria
Strategy and Governance 1 3 Project Management
Strategy and Governance 1 4 Financial Feasibility
Marketing and Sales 2 3 Project Management
Marketing and Sales 2 4 Financial Feasibility
Project Management 3 4 Financial Feasibility

5 matrices
2 Conversion to Matrix
Strategy and Governance Marketing and Sales Project Management Financial Feasibility
Strategy and Governance 1
Marketing and Sales 1
Project Management 1
Financial Feasibility 1
Obtaining Global Weights
6 for Each Sub-criteria

3 Apply Matrix Algebra 4 Obtain the Weight Index

Square Eigenvalue
Normalization Eigenvector
14 sub-criteria 13
 Model Development: Investment Prioritization

Decision Goal To Obtain Investment Portfolio which Maximize Company Value and Benefit to the Stakeholders

Criteria Strategy and Governance Marketing and Sales Project Management Financial Feasibility

Strategic Alignment & Urgency Market Capability Improvements Project-attributed Risks Investment Amount: Capex

Investment Profitability: Net

Sub-criteria Internal Process Improvements Product and Quality Leadership Duration of Execution
Present Value

Investment Returns:
HSE, Legal, and Compliance Corporate Reputation Organizational Readiness
Internal Rate of Return

Investment Periods:
Technological Sufficiency
Pay Back Period

Alternatives
The Molten Iron Project The Black Steel Project The White Steel Project

14
Pairwise Comparison: Concept Detail

(6)
(4)
Multiply the (7)
(1) (2) (3) Calculating (5)
weight of the The final weight
Survey of conversion the Applying eigenvector as Repeat (1) to (4)
each criteria for each sub-
weight to the survey results to multiplication the weight and for each criteria
with its criteria is
experts the matrix form matrix algebra check the and sub-criteria
respective sub- obtained
consistency ratio
criteria

15
Pairwise Comparison: The Form (1)
Pairwise Comparison on PTKS Investment Portfolio Prioritization
9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9
Criteria
Strategy and Governance 1 2 Marketing and Sales
Strategy and Governance 1 3 Project Management
Strategy and Governance 1 4 Financial Feasibility
Marketing and Sales 2 3 Project Management
Marketing and Sales 2 4 Financial Feasibility
Project Management 3 4 Financial Feasibility

Sub-criteria 1: Strategy and Governance

Strategic Alignment & Internal Process
1.1 1.2
Urgency Improvements
Strategic Alignment & HSE, Legal, and
1.1 1.3
Urgency Compliance
Strategic Alignment &
1.1 1.4 Technological Sufficiency
Urgency
Internal Process HSE, Legal, and
1.2 1.3
Improvements Compliance
Internal Process
1.2 1.4 Technological Sufficiency
Improvements
HSE, Legal, and
1.3 1.4 Technological Sufficiency
Compliance

Sub-criteria 2: Marketing and Sales

16
 Urgency
Internal Process HSE, Legal, and
1.2 1.3

Pairwise Comparison: The Form (2)

Improvements Compliance
Internal Process
1.2 1.4 Technological Sufficiency
Improvements
HSE, Legal, and
1.3 1.4 Technological Sufficiency
Compliance

Sub-criteria 2: Marketing and Sales

Market Capability Product and Quality
2.1 2.2
Improvements Leadership
Market Capability
2.1 2.3 Corporate Reputation
Improvements
Product and Quality
2.2 2.3 Corporate Reputation
Leadership

Sub-criteria 3: Project Management

Project-attributed Risk 3.1 3.2 Duration of Execution
Project-attributed Risk 3.1 3.3 Organizational Readiness
Duration of Execution 3.2 3.3 Organizational Readiness

Sub-criteria 4: Financial Feasibility

Investment Amount: Investment Profitability:
4.1 4.2
Capex NPV
Investment Amount:
4.1 4.3 Investment Returns: IRR
Capex
Investment Amount:
4.1 4.4 Investment Periods: PBP
Capex
Investment Profitability:
4.2 4.3 Investment Returns: IRR
NPV
Investment Profitability:
4.2 4.4 Investment Periods: PBP
NPV
Investment Returns: IRR 4.3 4.4 Investment Periods: PBP

17
The method to calculate a priority vector is proposed by The consistency index (CI) is obtained by first computing λmax
Thomas L. Saaty, where the method stems from the following as the principal eigen value. λmax for consistent reciprocal
observation. matrix is equal to the number of comparisons, λmax = m.
Taking a matrix A (m x m ), m is number of criteria, whose Saaty (2008) has proposed a consistency index (CI) which
entries are obtained as ratios between weights & multiplying it by is related to the eigenvalue method:
w obtains:

A perfectly consistent decision maker always obtain CI = 0,

but small values of inconsistency may be tolerated. If the
We know that a formulation of A w = n w implies that n is an consistency ratio (CR):
eigenvalue and w is an eigenvector of matrix A.
Vector w (priority vector) can be obtained from any pairwise
comparison matrix A as the solution of the following equation
system,

where RI is random index, the consistency of judgement is

tolerable. If CR is greater than 0.10, the judgment shall be
where max is the maximum eigenvalue of A. improved.

18
Pairwise Comparison: Consistency Ratio

N 1 2 3 4 5 6 7 8 9 10
RI 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.49

Matrix CR Target Max

No. Aspect N RI CR Actual Conclusion
Order (Max) Eigenvalue
1. Main Criteria 4 4 0.90 0.10 4.09203 0.03067 Consistent

Sub-Criteria Strategic
2. 4 4 0.90 0.10 4.04768 0.01589 Consistent
Alignment and Urgency

Sub-Criteria Marketing
3. 3 3 0.58 0.42 3.01459 0.00729 Consistent
and Sales

Sub-Criteria Project
4. 3 3 0.58 0.42 3.00005 0.00002 Consistent
Management
Sub-Criteria Financial
5. 4 4 0.90 0.10 4.01426 0.00475 Consistent
Feasibility
Remarks:
N = number of criteria; RI = Random consistency index (Saaty, 2005), λmax = maximum eigenvalue; CI = Consistency Index; CR = Consistency ratio 19
Summary: AHP Method Scheme
Model the decision
problem as a hierarchy
Goal
Compare criteria in
same level

Pairwise comparison Criteria 1 Criteria 2 Criteria 3

matrix

Check
inconsistency

Calculate local Sub- Sub- Sub-

priorities Criteria A Criteria B Criteria C

Synthesize global
priorities
Summary: AHP Method Scheme
Model the decision
problem as a hierarchy
Goal
Compare criteria in
same level

Pairwise comparison Criteria 1 Criteria 2 Criteria 3

matrix

Check
inconsistency

Calculate local Sub- Sub- Sub-

priorities Criteria A Criteria B Criteria C

Synthesize global
priorities
Summary: AHP Method Scheme
Model the decision
problem as a hierarchy

Compare criteria in
same level

Pairwise comparison
matrix

Check
inconsistency

Calculate local
priorities

Synthesize global
priorities
Summary: AHP Method Scheme
Model the decision
problem as a hierarchy

Compare criteria in
same level
Consistency Ratio (CR) is computed to check the level of
consistency in the decision makers’ judgments.
Pairwise comparison
matrix
• CR ≤ 0.1 is considered having an acceptable consistency
• CR > 0.1 suggests the judgments in matrix A to be
Check
inconsistency reviewed

Calculate local (note: this is for 4 x 4 matrix)

priorities

Synthesize global
priorities
Summary: AHP Method Scheme
Model the decision
problem as a hierarchy
Goal
Compare criteria in
same level

Pairwise comparison Criteria 1 Criteria 2 Criteria 3

matrix 0.4

Check
inconsistency

Calculate local Sub-Criteria A Sub-Criteria B Sub-Criteria C

priorities 0.1 0.5 Sum = 1
Local priorities 0.4

Synthesize global Global priorities 0.04 0.16 0.20 Sum =0.4

priorities
Weights of Criteria & Sub-Criteria

Decision Goal To Obtain Investment Portfolio which Maximize Company Value and Benefit to the Stakeholders

Criteria Strategy and Governance Marketing and Sales Project Management Financial Feasibility
9.11% 33.95% 15.65% 41.28%

Strategic Alignment & Urgency Market Capability Improvements Project-attributed Risks Investment Amount: Capex
2.85% 15.11% 7.13% 8.83%
Investment Profitability: Net
Sub-criteria Internal Process Improvements Product and Quality Leadership Duration of Execution
Present Value
2.35% 12.65% 4.05% 8.05%
Investment Returns: Internal Rate
HSE, Legal, and Compliance Corporate Reputation Organizational Readiness
of Return
1.95% 6.19% 4.47% 14.45%
Investment Periods:
Technological Sufficiency
Pay Back Period
1.97% 9.97%

Alternatives
The Molten Iron Project The Black Steel Project The White Steel Project

25
 Alternatives Quantitative Rating
Set the Rating Range
1 of Sub-criteria Repeat to All Project
5 Alternatives

2 Define the Description

for Each Rating Range 3 x 14 rated
properties

3 Mapping the Project Properties with its Associated Sub-criteria

Normalized the All Project
6 Alternatives’ Rating on each
4 Determine the Rating of Each Properties based on Its Associated Range respective Sub-criteria

3 x 14 normalized rated properties

26
 Model Outcome: Result of Calculation
Global Weight of Sub-Criteria Priority Index for Each Criteria
15,03%
Market Capability Improvements
14,29%
13,85%
Investment Returns: IRR
12,77%
Product and Quality Leadership 12,41%

Investment Periods: PBP

Investment Amount: Capex
Investment Profitability: NPV
Project-attributed Risk
6,90%
Corporate Reputation 6,30%
5,28%
Organizational Readiness
4,07%
Duration of Execution 3,75%
3,30%
Strategic Alignment & Urgency
2,07%
Internal Process Improvements
Technological Sufficiency
HSE, Legal, and Compliance Strategy and Governance Marketing and Sales Project Management Financial Feasibility

0% 2% 4% 6% 8% 10% 12% 14% 16% The Molten Iron Project The Black Steel Project The White Steel Project

The Final Prioritization Result

37,16% 37,40%
25,44%

The Molten Iron Project The Black Steel Project The White Steel Project 27
Final Results

Decision Goal To Obtain Investment Portfolio which Maximize Company Value and Benefit to the Stakeholders

Criteria Strategy and Governance Marketing and Sales Project Management Financial Feasibility
9.11% 33.95% 15.65% 41.28%

Strategic Alignment & Urgency Market Capability Improvements Project-attributed Risks Investment Amount: Capex
2.85% 15.11% 7.13% 8.83%
Investment Profitability: Net
Sub-criteria Internal Process Improvements Product and Quality Leadership Duration of Execution
Present Value
2.35% 12.65% 4.05% 8.05%
Investment Returns: Internal Rate
HSE, Legal, and Compliance Corporate Reputation Organizational Readiness
of Return
1.95% 6.19% 4.47% 14.45%
Investment Periods:
Technological Sufficiency
Pay Back Period
1.97% 9.97%

Alternatives
The Molten Iron Project The Black Steel Project The White Steel Project
25.44% 37.16% 37.40%

28