You are on page 1of 21

NED University of Engineering & Technology

NED University of Engineering & Technology Department of Industrial & Manufacturing Engineering Batch: 2005- Final Year

Department of Industrial & Manufacturing Engineering

Batch: 2005-

Final Year Project Presentation

2006

A Supplier Selection Methodology using Analytic Hierarchy Process

Presented By:

Faizan Ahmed

IM-05023

Muhammad Umair Khanzada

IM-05027

Hassan Hameed

IM-05029

Wali ul Islam Hashmi

IM-05042

Project Industry:

Limited

Internal Advisor:

Crescent Steel and Allied Products

Mr. Sayed Muhammad Irfan

Associate Professor

External Advisor:

AM Buying

Mr. Fahham Hasan Qaiser

Crescent Steel And Allied Products Limited

Problem Background

To select the best supplier of Hot Rolled Steel Coils for Crescent Steel & Allied Products Limited using the multi criteria approach of Analytic Hierarchy Process.

The suppliers to be evaluated are:

Shanghai Baosteel Group Corporation Saudi Iron & Steel Company (Hadeed) Taiyuan Iron & Steel (Group) Co., Ltd. (TISCO)

The existing supplier selection method uses nine criteria, having an equal weight. Suppliers are rated using a 1-5 scale and an average score is calculated

The Supplier Selection Problem

It is a comparison of suppliers using a common set of criteria and measures, to identify suppliers with the highest potential for meeting the needs of the firm consistently

Selecting the right supplier significantly reduces the purchasing cost and improve corporate competitiveness

Commonly adopted Supplier Selection Techniques are:

Categorical Weighted Point Cost Ratio Analytic Hierarchy Process (AHP)

Supplier selection problems are often:

Multi criteria Unstructured Involve both quantitative and qualitative factors

What is AHP?

AHP was developed by Thomas L. Saaty in the 1970s, it is designed to solve complex multi-criteria decision problems.

It is based on the human ability to make sound judgments about small problems

It breaks down a problem into its smaller constituent parts and brings in simple pair wise comparison judgment to obtain the priorities in the hierarchy

Advantages of AHP:

AHP can deal with complex and unstructured problem Its simple to use and understand Can handle subjective judgements of individuals

Used by: Xerox, IBM, British Airways, Ford Motor Company, American Army etc

Steps in Implementing AHP

Steps in Implementing AHP  Decomposition: Identify Supplier Selection criteria and sub criteria and structure them

Decomposition: Identify Supplier Selection criteria and sub criteria and structure them into a hierarchy

Comparative judgment: Compare two criteria in a cluster at a time and state their relationship using a numerical score

Synthesis: Derive relative weights of supplier selection criteria from the pair wise comparisons

Evaluate Alternatives (Suppliers) on basis of each lowest level criteria using pair wise comparisons or quantitative values and synthesize relative performance

Perform Sensitivity Analysis Make a decision

Decomposition: The Supplier Selection Hierarchy

Level 1: Objective Level 2: Primary Criteria Level 3: Secondary Criteria Level 4: Baosteel Hadeed TISCO
Level 1:
Objective
Level 2:
Primary
Criteria
Level 3:
Secondary
Criteria
Level 4:
Baosteel
Hadeed
TISCO
Alternatives

Comparative Judgment

Quality

Is strongly more preferred than

Technical Capability

of relationship between two elements

The Fundamental Scale of

 The set of all such judgments for all elements of a cluster is AHP: Verbal
 The set of all such judgments for all elements of a cluster is
AHP:
Verbal statement of
Verbal statement of
Rating
Rating
preference
preference
called a comparison matrix
Equally preferred
Equally preferred
1
1
Weak or slight
2
Moderately preferred
3
Moderately preferred
3
Moderate plus
4
Strongly
5
1
1/3
1/5
1/7
1/2
1/3
Strongly
5
3
1
1
1/5
5
1
Very Strongly
7
Strong plus
6
5
1
Very Strongly
1
1/5
7
5
1
Extremely
9
Very, very strong
7
5
5
1
8
5
4
Reciprocals of above
If activity i has one of the above
Extremely
9
2
non zero numbers assigned to it
1/5
1/5
1/5
1
1/5
when compared with activity j, then
Reciprocals of above
If activity i has one of the above
3
j has the reciprocal value when
1
1
1/4
5
1
non zero numbers assigned to it
compared with i
when compared with activity j, then
j has the reciprocal value when
compared with I
A Supplier Selection Methodology Using Analytic Hierarchy Process
7 of 21

Synthesis: Deriving priorities

To extract numerical values of importance out of the comparisons

Normalization converts ratio scale into absolute scale

The weights can be determined by :

Normalizing the Comparison Matrix

By dividing each element in a column by the sum of that column for whole matrix.

Determine the relative weights

Take average of each row of the Normalized Matrix which gives the priorities of the corresponding elements in the form of a priority vector.

Synthesis: Deriving priorities

(Contd )

..

0.05 1 0.03 1/3 0.02 1/5 0.08 1/7 0.02 1/2 0.04 1/3 0 9 4 3
0.05
1
0.03
1/3
0.02
1/5
0.08
1/7
0.02
1/2
0.04
1/3
0
9
4
3
3
4
0.15
3
0.11
1
0.11
1
0.09
1/5
0.23
5
0.13
1
0
7
9
9
3
3
5
0.25
0.11
1
0.11
1
0.09
1/5
0.23
5
0.13
1
0
7
9
9
3
3
7
5
5
0.30
0.58
0.59
0.49
1
0.23
5
0.53
4
0
6
5
6
3
1
2
1/5
1/5
1/5
1/5
0.10
0.02
0.02
0.09
0.04
1
0.02
0
3
4
9
7
7
0.15
3
0.11
1
0.11
1
0.12
1/4
0.23
5
0.13
1
0
7
9
4
3
3
20
8.53
8.4
2.01
21.5
7.53
3
7
3

Priority

Vector

4.384%

14.179%

15.845%

45.676%

5.325%

14.592%

Consistency of Judgments

..…

.….

..…

..…

Human judgment is inevitably inconsistent, approximate and intransitive.

Forcing perfect consistency can reduce accuracy.

Given that w 1 , w 2 … w n are weights of the criteria, we can recover the scale w using the following equation:

A

1

A 1

1

A 2

w

1 /w 2

A

2

w

2

/w 1

1

A

w

Or, Aw = nw

n

n

/w 1

w

n /w 2

..… ..… ..… ….. A n ….. w /w w nw 1 n 1 1 w
..…
..…
..…
…..
A n
…..
w
/w
w
nw
1
n
1
1
w
nw
2
2
…..
w
/w
x
=
2
n
w
nw
n
n
.....
1

This is analogous to the Eigen value problem: Ax = λx

Thus the Eigen vector is equal to the priority vector and the Eigen value is equal to the order of the matrix

Consistency of Judgments

(Contd )

..

For slightly inconsistent comparison matrix, the value of maximum positive Eigen value will deviate from the order of the matrix

Mathematically, the consistency of a comparison matrix is its Consistency Ratio (C.R.)

Consistency of Judgments (Contd ) ..  For slightly inconsistent comparison matrix, the value of maximum

C.I. is the consistency index of the comparison matrix

Consistency of Judgments (Contd ) ..  For slightly inconsistent comparison matrix, the value of maximum

R.I. is the Random Consistency Index

Consistency of Judgments (Contd ) ..  For slightly inconsistent comparison matrix, the value of maximum

Values of C.R. less than or equal to 0.1 are acceptable

Consistency of Judgments

(Contd )

..

Comparison

1

1/3

1/5

Matrix

1/7

1/2

1/3

3

1

1

1/5

5

1

5

1

1

1/5

5

1

7

5

5

1

5

4

2

1/5

1/5

1/5

1

1/5

3

1

1

1/4

5

1

C I =

.

.

6.5828 6

= 0.1164

 

6

1

R I =

.

.

1.98(6 2)

= 1.32

6

x

Priorit

y

   

0.0438

Vector

0.1418

=

0.1584

0.4568

max

 

0.0532

=

0.1459

0.2741

0.9352

1.0229

3.0708

0.3215

0.9580

6.582

8

C.R. =

  • 0.1164 = 0.0883

Since CR<0.1,

  • 1.32 Therefore Comparisons are consistent

Development of MS Excel Spread

Sheets

MS Excel spread sheets were developed to facilitate decision maker for performing pair wise comparisons

Consistency Ratio for each comparison matrix is calculated using real time inputs

Result of comparisons: Relative Weights of Criteria

Result of comparisons: Relative Weights of Criteria A Supplier Selection Methodology Using Analytic Hierarchy Process

Supplier Evaluation

It is the process in which each supplier is accessed and evaluated on the basis of each lowest level criteria

Different evaluation techniques are used, based on the nature of each criteria:

Pair wise Comparisons Pair wise Comparisons using Equivalent Scale Direct Input of Quantitative Data

Supplier Evaluation forms were developed and data and comparisons were provided by the CSAPL buying personnel

Result of Supplier Evaluation

Result of Supplier Evaluation A Supplier Selection Methodology Using Analytic Hierarchy Process

Sensitivity Analysis

Identifies the impact of changes in the priority of criteria on the suppliers' performance.

Decision maker might want to know the consequences of variation in weights of criteria, because:

Some influence could have been underestimated

A slight variation in one criteria weight could lead to a different decision

Increase in the weight of one criterion n the cluster results in a proportional decrease in all other criteria of the cluster

Excel spread sheets were developed to perform the sensitivity Analysis Two sensitivity analysis techniques were used:

Performance Sensitivity Gradient Sensitivity Sensitivity Analysis.xlsm

Recommendation

Supplier evaluation and sensitivity analysis showed that Shanghai Baosteel Group Corporation was the best supplier out of the three suppliers of Hot Rolled Coils and should be selected

Visual Basic Application for Supplier Selection

Visual Basic Application for Supplier Selection  Visual Basic Application designed to provide a flexible and

Visual Basic Application designed to provide a flexible and structured solution to the supplier selection problem 24216 pure lines of coding using Visual Basic 6 User can define Hierarchies, perform comparisons and evaluate suppliers. Following software were used in the development of the application

Microsoft Visual Basic 6 Microsoft Excel 2007 IcoFX 1.6.4 Inno Setup Compiler

Contd ..

Some of the features are:

Up to four different Hierarchies can be defined and edited Each Hierarchy can be structured into three levels of criteria. Altogether a hierarchy can have up to 360 criteria Calculation of Consistency Ratio, most inconsistent judgments and proposed modifications in those judgments Each hierarchy can have up to 10 alternatives (suppliers). Different supplier evaluation methods can be used for supplier evaluation (Pair wise, Quantitative, Factor rating) Generation of an automated MS EXCEL spread sheet containing summary of Vendor Ranking

Conclusion

The Supplier Selection Model using AHP improves Supplier

Selection Process:

Both tangible and intangible factors can be included in the process

Using pair wise comparisons reduced dependency of the system on human judgment

The organizational preferences and policies are reflected in the relative weights of the criteria

Analysis of several “what-if” scenarios is possible using the sensitivity analysis techniques

The Visual Basic application, designed specifically for the

supplier selection problem using AHP approach, provides

flexibility and ease in model formulation for future

purchasing requirements and to accommodate varying

needs and requirements