ahp-class

38
Selection of School

Upload: moonyjena1874

Post on 17-Dec-2015

4 views

Category:

Documents


0 download

DESCRIPTION

AHP

TRANSCRIPT

  • Selection of School

  • Selection of School

    Selection of School:

    Academic Reputation, Location,

    Placement, Fees, Infrastructure

  • Analytic

    Hierarchy Process

    by

    G N Patel

  • Admission

    Location PlacementReputation Fees Infrastructure

    YX Z

    CRITERIA

    ALTERNATIVES

    Hierarchic

    Thinking

  • Purchase of apple

  • Purchase of apple

    Purchase of Apple:

    Taste, Aroma, Ripeness, Size, Price

  • Purchase of Oven

  • Purchase of Oven

    Choose best oven:

    function, economy,

    safety and reliability,

    convenience

  • Whom to marry

  • Whom to marry

    Whom to marry:

    flexibility, independence,

    commitment, humor,

    intelligence, physical,

    socio-cultural

  • Selection of Job

  • Selection of Job

    Selection job:

    salary, life quality of the city where

    job is located, job profile, nearness to home, security,

    reputation, flexibility (time, work place), opportunity

    (salary potential, top level position)

  • Evaluation of Employees for Promotion

    Evaluation of employees for promotion:

    dependability, education, experience,

    quality, attitude, leadership

  • Perhaps the most creative task in making

    a decision is to choose the factors that

    are important for the decision

  • Analytic Hierarchy Process (AHP)

    A popular mathematical method for solving MCDM problems .

  • AHP Applications

    It has been used in more than 30 diversified field like

    1. Planning and development

    2. Selection

    3. Resource allocation

    4. Forecasting

    5. Evaluation

    6. Priority and ranking.

    7. Health and related fields

    and many more.

  • Importance of AHP

    It provides a unique way of incorporating qualitative and quantitative judgments.

    It is able to rank choices in their order of effectiveness in realizing their objectives.

    It allows decision makers to include inconsistency in their choices.

    It is versatile and flexible to enable a combination of other mathematical methods not limited to LP, Fuzzy logic

    theories and network models.

  • Analytical Hierarchy Process

    1.A multi-attribute analysis technique that

    organizes complex systems/unstructured

    problems with many elements of different

    influence.

    2.Easy to use, automatable algorithm,

    well-accepted by policy makers.

    3.Uses stakeholder perceptions to find the

    most important elements influencing

    complex decisions (only as good as the

    experts).

  • The Analytic Hierarchy

    Process (AHP)

    Decision Making involves setting priorities and the AHP is the

    methodology for doing that. It is designed for situations in which

    ideas, feelings, and emotions affecting the decision process are

    quantified to provide a numeric scale for prioritizing the

    alternatives.

  • AHP- Methodology

    It has the following steps:

    Step 1.

    Stating the goal, criteria, sub- criteria, alternatives and

    constraints of the problem clearly.

    Step 2.

    Arrange fragmented decision problems into (tree- like)

    hierarchy structure of different levels ( goal, criteria, sub-

    criteria, alternatives.)

  • The Analytic Hierarchy Process (Step 2 contd.)

    Objective

    Criterion 2Criterion 1 Criterion KLevel 2

    Subcriterion 1 Subcriterion 2 Subcriterion K

    Alternative 1 Alternative 2 Alternative N

    Level 3

    Level P

    Hierarchy with P Levels

    Level 1

    Step 2. Decompose the problem into a hierarchy of interrelated

    decision criteria and alternatives

    .

    .

    .

  • GOAL

    CRITERIA

    ALTERNATIVES

  • Relative Measurement

    The Process of Prioritization

    Step 3.

    Develop a pair wise comparison matrix using fundamental scale to

    measure both tangible and intangible criteria.

    In relative measurement a preference, judgment is expressed on

    each pair of elements with respect to a common property they

    share.

    In practice this means that a pair of elements in a level of the

    hierarchy are compared with respect to parent elements to which

    they relate in the level above.

  • The Principle of the AHP

    This comparison is carried out using the following fundamental scale from 1 to 9 as follows: 1 Equally preferred

    2 Equally to Moderately preferred

    3 Moderately preferred

    4 Moderately to Strongly preferred

    5 Strongly preferred

    6 Strongly to Very Strongly preferred

    7 Very Strongly preferred

    8 Very strongly to Extremely preferred

    9 Extremely preferred

  • AHP- Methodology(Contd)

    The pairwise comparison matrix looks like

    where is the relative importance of criterion i over criterion j number of comparisons required for the pair-wise comparison matrix is

    given by n(n-1)/2 where n = number of elements

    Diagonal elements are 1 and other elements are reciprocals of earlier comparisons that is

  • AHP- Methodology(Contd)

    Step 4

    The priority vector w is obtained by using Additive Normalization as follows:

  • Consistency of the Comparison Matrix

    Step 5 : Testing the consistency of the comparison matrix

    If all the columns of the normalized matrix are identical, then the

    original matrix is said to be consistent

    is a consistent matrix

    Consistency implies coherent judgment on the part of the decision

    maker regarding the pair-wise comparisons.

    R L

    A= R 1 5

    L 1/5 1

  • Consistency of the Comparison Matrix

    Step 5 : Testing the consistency of the comparison

    matrix

    Mathematically, A is consistent if

    Also it can defined as when n is the size of the matrix

    and A is consistent.

    When A is not consistent , than , max being the

    principal eigen vector

  • Consistency of the Comparison Matrix

    (contd.)

    Step 5

    Testing of the consistency

    Where max = principal eigen vector

    CI = consistency index

    RI = the random consistency index.

    .

  • Consistency of the Comparison Matrix

    (contd.)

    It is unusual for all comparison matrices to be consistent.

    Indeed, given that human judgment is the basis of the

    construction of these matrices, some reasonable degreeof inconsistency is expected and tolerated.

    The matrix is accepted as consistent if

  • AHP- Methodology(Contd)

    Step 6: Global weights of alternatives with respect to the goal defined in the AHP model are calculated.

  • Selection of Management Institute for EDP/FDP

    Ms. Y has received information from three Management

    institutions A, B, C to attend EDP / FDP on similar topics. She

    specifies two main criteria location and academic reputation.

    Being a participant believing to enhance knowledge through

    attending such programmes, she judges academic reputation to

    be five times as important as location. Using the symbols R and

    L to represent reputation and location, the associated

    comparison matrix is given as

    R L

    A= R 1 5

    L 1/5 1

  • Example Contd.

    The relative weights of the alternatives of A,B and C

    within each of the L and R criteria, whose elements

    are based on Ms Xs judgment regarding the relative importance are as follows:

    L A B C

    A 1 1/2 1/5

    AL= B 2 1 1/2

    C 5 2 1

    A B C

    A 1 2 3

    AR= B 1/2 1 3/2

    C 1/3 2/3 1

  • Pros and cons of AHP

    Pros

    Hierarchical structuring of a decision problem

    Combining multiple inputs from several persons to a consolidated

    outcome

    People usually agree with the out coming process

    Calculations using excel

  • Pros and cons of AHP

    Cons

    Pair wise comparison is a quite artificial way a set of items. People

    are more use to agree-disagree or ranking

    If consistency index is too high: request to reconsider

  • Recommendations

    When using AHP: Try to structure the model in group of max 4-5

    criteria or sub criteria. If possible introduce additional hierarchical

    levels

    Spend time to explain the use of scale for pair wise comparisons

    without knowledge of AHP. Ask them to use the whole range 1 to 9.

    Even with a consistency value above the recommended limit, the

    results usually reflects the correct ranking and still can be used.

  • Recommendations

    AHP is ideal to get a consolidated result for inputs from several

    participants using the geometric mean

    Once AHP is introduced and used as method for decision making,

    results are generally accepted, as the method is based on

    mathematics, seen as neutral and objective