Scribd
Upload
155 views

Week6 - Multi-Attribute Decision Making (MADM)

This document discusses multi-attribute decision making (MADM), which refers to making decisions with multiple conflicting criteria. MADM involves selecting from a finite number of alternatives based on attributes. It requires weighting attributes based on their importance and evaluating alternative performances on attributes. The document outlines common MADM methods like dominance, max-min, max-max, conjunctive constraint, and lexicographic, as well as compensatory methods like simple additive weighting and analytical hierarchy process. Decision matrices in MADM contain alternatives, attributes, attribute weights, and alternative performances on attributes.

Uploaded by

Amalia Anjani
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
155 views

Week6 - Multi-Attribute Decision Making (MADM)

This document discusses multi-attribute decision making (MADM), which refers to making decisions with multiple conflicting criteria. MADM involves selecting from a finite number of alternatives based on attributes. It requires weighting attributes based on their importance and evaluating alternative performances on attributes. The document outlines common MADM methods like dominance, max-min, max-max, conjunctive constraint, and lexicographic, as well as compensatory methods like simple additive weighting and analytical hierarchy process. Decision matrices in MADM contain alternatives, attributes, attribute weights, and alternative performances on attributes.

Uploaded by

Amalia Anjani
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Multi-Attribute Decision Making

(MADM)
Decission Support Methods

Amalia Anjani Arifiyanti, M.Kom


Sistem Informasi – Universitas Pembangunan Nasional “Veteran” Jawa Timur
Multi Criteria Decision Making (MCDM)
refers to making decisions in the presence of multiple, usually conflicting criteria

MADM : Multi Attribute Decision Making MODM : Multi Objective Decision Making

MADM methods, are generally discrete, with a MODM methods have decision variable values
limited number of predetermined alternatives.
that are determined in a continuous or integer

MADM is an approach employed to solve domain, with either an infinitive or a large


problems involving selection from among a finite number of choices, the best of which should
number of alternatives. satisfy the decision maker’s constraints and
preference priorities
An MADM method specifies how attribute information
is to be processed in order to arrive at a choice.

MADM methods require both inter- and intra-


attribute comparisons, and involve appropriate
explicit tradeoffs
MADM Methods
Non-compensatory Methods Compensatory Methods

• Dominance method: Eliminate all dominated alternatives. • Scoring methods, The scoring method selects or evaluates
There could be more than one solutions generated by this analternative according to its score (or utility). Utility or score is
method. used to express the decision maker’s preference.

• Max-min method: Find the weakest attribute value (min) of Example: Simple additive weighting (SAW) and analytical hierarchy
each alternative and then choose the alternative with the best process (AHP)
(max) weakest attribute value. The logic is that a chain is as • Compromising methods, The compromising method selects an
strong as its weakest link. This method is applicable only when alternative that is closest to the ideal best and ideal worst
attribute values are comparable with one another, either solutions obtained from the available alternatives.
measured in the same unit or transformed to a common scale. Example: TOPSIS (The Technique for Order of Preference by Similarity
• Max-max method: In contrast to the Max-min method, the to Ideal Solution) and VIKOR (Visekriterijumsko KOmpromisno
Max-max method selects an alternative by its best attribute Rangiranje)
value. It is also applicable only when attributes are comparable • Concordance methods
• Conjunctive constraint method Example: ELECTRE (Elimination et Choice Translating Reality) and
• Lexicographic method PROMETHEE (Preference Ranking Organization Method for
Enrichment Evaluation)
MADM : Multi Attribute Decision Making

Each decision matrix in MADM methods has four main parts:

(a) Alternatives,

(b) Attributes,

(c) Weight or relative importance of each attribute (i.e., weight),

(d) Measures of performance of alternatives with respect to the attributes.


Multi-Attribute Decision Making (MADM)

Masalah MADM adalah mengevaluasi m alternatif Ai (i=1,2,...,m) terhadap


sekumpulan atribut atau kriteria Cj (j=1,2,...,n), dimana setiap atribut saling
tidak bergantung satu dengan yang lainnya.

Kriteria atau atribut dapat dibagi menjadi dua kategori, yaitu:

• Kriteria keuntungan (benefit) adalah kriteria yang nilainya akan dimaksimumkan,


misalnya: keuntungan, IPK (untuk kasus pemilihan mahasiswa berprestasi), dll.

• Kriteria biaya (cost) adalah kriteria yang nilainya akan diminimumkan, misalnya:
harga produk yang akan dibeli, biaya produksi, dll.
Multi-Attribute Decision Making (MADM)
Pada MADM, matriks keputusan setiap alternatif terhadap setiap atribut, X, maka:

 x 11 x 12  x 1n 
x x 22  x 2 n 
X  21

    
 
 x m1 x m 2  x mn 

dengan xij merupakan rating kinerja alternatif ke-i terhadap atribut ke-j.
Nilai bobot yang menunjukkan tingkat kepentingan relatif setiap atribut, diberikan
sebagai, W:
W = {w1, w2, ..., wn}
Multi-Attribute Decision Making (MADM)
Rating kinerja (X), dan nilai bobot (W) merupakan nilai utama yang
merepresentasikan preferensi absolut dari pengambil keputusan.
Masalah MADM diakhiri dengan proses perankingan untuk
mendapatkan alternatif terbaik yang diperoleh berdasarkan nilai
keseluruhan preferensi yang diberikan (Yeh, 2002).
Pada MADM, umumnya akan dicari solusi ideal.
Pada solusi ideal akan memaksimumkan semua kriteria keuntungan dan
meminimumkan semua kriteria biaya.
Multi-Attribute Decision Making (MADM)

Masalah

Kriteria-1 Kriteria-2 ... Kriteria-m


(C1) (C2) (Cm)

Alternatif-1 Alternatif-2 ... Alternatif-n


(A1) (A2) (An)
SAW can be considered the most intuition

Simple Additive and easy way to deal with MCDM problems,


because the linear additive function can
Weighting (SAW) represent the preferences of decision
makers (DM)

Find out more at next slides

You might also like