Abstract
MACBETH (Measuring Attractiveness by a Categorical Based Evaluation Technique) is a multicriteria decision analysis approach that requires only qualitative judgements about differences of value to help an individual or a group quantify the relative attractiveness of options. This chapter presents a new up-to-date survey of the mathematical foundations of MACBETH. Reference is also made to real-world applications and an extensive bibliography, spanning back to the early 1990s, is provided.
Access provided by Autonomous University of Puebla. Download chapter PDF
Similar content being viewed by others
Keywords
- MACBETH
- Questioning procedure
- Qualitative judgements
- Judgmental inconsistency
- Cardinal value measurement
- Interaction
1 Introduction
Let X (with # X = n ≥ 2) be a finite set of elements (alternatives, choice options, courses of action) that an individual or a group, J, wants to compare in terms of their relative attractiveness (desirability, value).
Ordinal value scales (defined on X) are quantitative representations of preferences that reflect, numerically, the order of attractiveness of the elements of X for J. The construction of an ordinal value scale is a straightforward process, provided that J is able to rank the elements of X by order of attractiveness—either directly or through pairwise comparisons of the elements to determine their relative attractiveness. Once the ranking is defined, one needs only to assign a real number v(x) to each element x of X, in such a way that:
-
1.
v(x) = v(y) if and only if J judges the elements x and y to be equally attractive;
-
2.
v(x) > v(y) if and only if J judges x to be more attractive than y.
The problem, however, is that, in a multiple criteria decision analysis, conclusions based on a additive value model may be quantitatively meaningless, because “to be quantitatively meaningful a statement should be unaffected by admissible transformations of all the quantities involved” [126, p. 91]. A necessary condition is that each value scale should be unique up to a positive affine transformation (an interval scale), as it is with a value difference scale. A value difference scale (defined on X) is a quantitative representation of preferences that is used to reflect, not only the order of attractiveness of the elements of X for J, but also the differences of their relative attractiveness , or in other words, the strength of J’s preferences for one element over another. Unfortunately, the construction of an interval value scale is usually a difficult task.
Both numerical and non-numerical techniques have been proposed and used to build a value difference scale (hereafter, simply called a value scale)—see [113] for a survey. Examples of numerical techniques are direct rating and difference methods—see descriptions in [63, 195, 197]. They require J to be able to produce, either directly or indirectly, numerical representations of his or her strengths of preferences, which can be a difficult cognitive task—see [114]. Non-numerical techniques based on indifference judgements, such as the bisection method (also described by the same authors), force J to compare his or her strengths of preferences between two pairs of elements of X, therefore involving at least three different elements in each judgement. This requires J to perform an intensive cognitive task and is prone to be substantively meaningless—“substantive meaningfulness (…) requires that the qualitative relations (…) being modelled should be unambiguously understood by the decision maker” [126, p. 91].
The aforementioned difficulties inspired the development of MACBETH “Measuring Attractiveness by a Categorical Based Evaluation Technique”. The original research on the MACBETH approach was carried out in the early 1990s—see [6, 22, 27]—as a response to the following question:
How can a value scale be built on X, both in a qualitatively and quantitatively meaningful way, without forcing J to produce direct numerical representations of preferences and involving only two elements of X for each judgement required from J?
Using MACBETH, J is asked to provide preferential information about two elements of X at a time, firstly by giving a judgement as to their relative attractiveness (ordinal judgement) and secondly, if the two elements are not deemed to be equally attractive, by expressing a qualitative judgement about the difference of attractiveness between the most attractive of the two elements and the other. Moreover, to ease the judgemental process, six semantic categories of difference of attractiveness, “very weak”, “weak”, “moderate”, “strong”, “very strong” or “extreme”, or a succession of these (in case hesitation or disagreement arises) are offered to J as possible answers. This is somewhat in line with similar ideas previously proposed by Saaty [178] in a ratio measurement framework, or by Freeling [125] and Belton [62] in difference value measurement. By pairwise comparing the elements of X a matrix of qualitative judgements is filled in, with either only a few pairs of elements, or with all of them (in which case \(n \cdot (n - 1)/2\) comparisons would be made by J).
A brief review of the previous research on MACBETH is offered in Sect. 11.2, together with the evolution of its software’s development. It shows that, on a technical level, MACBETH has evolved through the course of theoretical research and also through its extension to the multicriteria value measurement framework in numerous practical applications (see Sect. 11.10). Its essential characteristics, however, have never changed—see [57].
Section 11.3 through 11.9 of this chapter present an up-to-date survey of the mathematical foundations of MACBETH. Section 11.3 describes the two MACBETH modes of questioning mentioned above (both involving only two elements at a time) used to acquire preferential information from J, as well as the types of information that can be deduced from each of them. The subsequent sections are devoted to an up-to-date rigorous survey of the mathematical foundations of MACBETH. Section 11.4 addresses the numerical representation of those different types of information. These numerical representations are only possible if J’s responses satisfy certain rational working hypotheses. Section 11.5 deals with the “consistency/inconsistency” of the preferential information gathered from J and Sect. 11.6 explores the practical problem of testing the consistency of preferential information. How should an inconsistency be dealt with? The answer to this question is the subject of Sect. 11.7. Sections 11.8 and 11.9 present what MACBETH proposes to J once the preference information provided by J is consistent. Finally, Sect. 11.10 lists several real-world applications of multicriteria value analysis in which the MACBETH approach was used.
This chapter will use the following notation:
-
J is an evaluator, either a individual or group.
-
X (with # X = n ≥ 2) is a finite set of elements (alternatives, choice options, courses of action) that J wants to compare in terms of their relative attractiveness (desirability, value).
-
\(\Delta att(x,y)\) is the “difference of attractiveness between x and y for J”, where x and y are elements of X such that x is more attractive than y for J.
-
\(\Delta att(x,y)\, \succ \, \Delta att(z,w)\,\) means that \(\,\Delta att(x,y)\,\) is greater than \(\Delta att\) (z, w).
-
ϕ is an empty set.
-
\(\mathbb{R}\) is the set of real numbers.
-
\(\mathbb{R}_{+} =\{ x \in \mathbb{R}\ \vert \ x \geq 0\}\).
-
\(\mathbb{R}^{{\ast}} = \mathbb{R}\setminus \{0\}\).
-
\(\mathbb{R}_{+}^{{\ast}} = \mathbb{R}_{+}\setminus \{0\}\).
-
\(\mathbb{Z}\) is the set of integer numbers.
-
\(\mathbb{N}\) is the set of non-negative integer numbers.
-
\(\mathbb{N}^{{\ast}} = \mathbb{N}\setminus \{0\}\).
-
\(\mathbb{N}_{s,t} =\{ s,s + 1,\ldots,t\} =\{ x \in \mathbb{N}\ \vert \ s \leq x \leq t\}\) where \(s,t \in \mathbb{N},\ \mbox{ and }s < t\).
-
The transpose of a matrix A will be denoted byt A.
2 Previous Research and Software Evolution
In order to build an interval (value) scale based on the qualitative judgements of difference of attractiveness formulated by J, it is necessary that the six MACBETH categories “very weak”, “weak”, “moderate”, “strong”, “very strong” or “extreme” be represented by non-overlapping (disjoint) intervals of real numbers. The basic idea underlying the initial development of MACBETH was that the limits of these intervals should not be arbitrarily fixed a priori, but determined simultaneously with numerical value scores for the elements of X. Research was then conducted on how to test for the existence of such intervals and how to propose numerical values for the elements of X and for the limits of the intervals—see [6, Chap. IV]. This gave rise to the formulation of a chain of four linear programs—see [22–25]—that, implemented in GAMS, were used in the first real-world applications of MACBETH as a decision aiding tool to derive value scores and criteria weights in the framework of an additive aggregation model—see [27, 30, 76, 77]. Theoretical research conducted at the same time, and first presented in 1994 at the 11th International Conference on MCDM, demonstrated the equivalence of the approach by constant thresholds and the approach by measurement conditions—see [28].
The first MACBETH software was developed in 1994. In it, the objective function used in the GAMS implementation to determine a value scale was modified, on the basis of a simple principle—see [30, 31]—that makes it possible, for simple cases, to determine the scale “by hand” [57]. However, complete procedures to address and manage all cases of inconsistency were not available at that time. Therefore, the software offered its users the possibility of obtaining a compromise scale in the case of inconsistency. This initial software was used in several real world applications—see, for example, [14, 24, 29, 32, 33, 35, 103]. However, it had several important limitations:
-
1.
The determination of suggestions was still heuristic and did not guarantee the minimal number of changes necessary to achieve consistency;
-
2.
It was not possible for the evaluator to hesitate between several semantic categories when expressing judgements. It, therefore, did not enable one to facilitate the management of group judgemental disagreements;
-
3.
It forced the evaluator to first provide all of the judgements before it could run any procedure. Consequently, judgemental inconsistency could only be detected for a full matrix of judgements. As a result, suggestions of changes to resolve inconsistency could only then be discussed, a restriction that did not lend itself to good interaction.
Subsequent theoretical research was therefore concentrated on resolving these problems. Results reported in [95, 159], allowing inconsistencies to be dealt with in a mathematically sound manner, were the turning point in the search for a more interactive formulation. Indeed, it was then possible to implement a procedure that automatically detects “inconsistency”, even for an incomplete matrix of judgements, in a new software called M-MACBETH—see www.m-macbeth.com and [45]—which has been used to produce some of the figures in this paper. The objective of abandoning the suggestion of a compromise scale could also finally be achieved, since the origin of the inconsistency could now be found (detection of elementary incompatible systems) and explained to J. M-MACBETH finds the minimal number of necessary changes and, for any number of changes not greater than five, suggests all of the possible ways in which the inconsistency can be resolved. Furthermore, it is able to provide suggestions of multiple category changes, where a “k categories change” is considered to be equivalent to k “1 category changes”.
Real-world applications in the specific context of bid evaluation (see references in Sect. 11.10) inspired research regarding the concepts of “robustness” [95] and sensitivity [10], the results of which were then included in the software, together with the possibility of addressing potential imprecision (uncertainty) associated with impacts of options, incorporating reference levels for one criterion at any time, and graphically representing comparisons of options on any two groups of criteria [57]. These issues are out of the scope of the present chapter and they are not also included in the version of the software, limited to scoring and weighting, embedded into the HIVIEW3 software in 2003—see [82] and www.catalyze.co.uk.
3 Types of Preferential Information
3.1 Type 1 Information
Type 1 information refers to preferential information obtained from J by means of Questioning Procedure 1.
Let x and y be two different elements of X.
Questioning Procedure 1.
A first question (Q1) is asked of J:
Q1: Is one of the two elements more attractive than the other?
J’s response (R1) can be: “Yes”, or “No”, or “I don’t know”.
If R1 = “Yes”, a second question (Q2) is asked:
Q2: Which of the two elements is the most attractive?
The responses to Questioning Procedure 1 for several pairs of elements of X enable the construction of three binary relations on X:
- P = :
-
{(x, y) ∈ X × X: x is more attractive than y}
- I = :
-
{(x, y) ∈ X × X: x is not more attractive than y and y is not
more attractive than x, or x = y}
- ? = :
-
{(x, y) ∈ X × X: x and y are not comparable in terms of their
attractiveness}.
P is asymmetric, I is reflexive and symmetric, and? is irreflexive and symmetric. Note that \(? = X \times X\setminus (I \cup P \cup P^{-1}),\) with \(P^{-1} =\{ (x,y) \in X \times X\ \vert \ yPx\}\).
Definition 1.
Type 1 information about X is a structure {P, I, ?} where P, I and? are disjoint relations on X, P is asymmetric, I is reflexive and symmetric, and \(? = X \times X\setminus (I \cup P \cup P^{-1})\).
3.2 Type 1+2 Information
Suppose that type 1 information {P, I, ?} about X is available.
Questioning Procedure 2.
The following question (Q3) is asked, for all (x,y) ∈ P:
Q3: How do you judge the difference of attractiveness between x and y?
J’s response (R3) would be provided in the form “d s ” (where d 1 ,d 2 , …,d Q ( \(Q \in \mathbb{N}\setminus \{0,1\}\) ) are semantic categories of difference of attractiveness defined so that, if i < j, the difference of attractiveness d i is weaker than the difference of attractiveness d j ) or in the more general form (possibility of hesitation) “d s to d t ”, with s ≤ t (the response “I don’t know” is assimilated to the response “d 1 to d Q ”).
Remark 1.
When Q = 6 and d 1 = very weak, d 2 = weak, d 3 = moderate, d 4 = strong, d 5 = very strong, d 6 = extreme, Questioning Procedure 1 is the mode of interaction used in the MACBETH approach and its M-MACBETH software.
R3 responses give rise to relations \(C_{st}\ (s,t \in \mathbb{N},1 \leq s \leq t \leq Q)\) where \(C_{st} =\{ (x,y) \in P\ \vert \ \Delta _{att}(x,y)\) is “d s to d t ”}. They enable the construction of an asymmetric relation on P: \(\{((x,y),(z,w)) \in P \times P\ \vert \ \exists \ i,j,s,t \in \mathbb{N}\mbox{ with }1 \leq i \leq j < s \leq t\,\leq \,Q,\) (x, y) ∈ C st , (z, w) ∈ C ij }. Hereafter, C ss will simply be referred to as C s .
Definition 2.
Type 1+2 information about X is a structure {P, I, ?, P e} where {P, I, ?} is type 1 information about X and P e is an asymmetric relation on P, the meaning of which is “\((x,y)P^{e}(z,w)\mbox{ when }\Delta _{att}(x,\) \(y) \succ \Delta _{att}(z,w)\)”.
4 Numerical Representation of the Preferential Information
4.1 Type 1 Scale
Suppose that type 1 information {P, I, ?} about X is available.
Definition 3.
A type 1 scale on X relative to {P, I} is a function \(\mu: X \rightarrow \mathbb{R}\) satisfying Condition 1.
Condition 1
\(\forall \ x,y \in X,\ [xPy \Rightarrow \mu (x) >\mu (y)]\) and \([xIy \Rightarrow \mu (x) =\mu (y)].\)
Let \(Sc_{1}(X,P,I) =\{\mu: X \rightarrow \mathbb{R}\ \vert \ \mu\) is a type 1 scale on X relative to {P, I}}. When X, P and I are well determined, Sc 1(X, P, I) will be noted Sc 1.
When? = ϕ and Sc 1(X, P, I) ≠ ϕ, each element of Sc 1(X, P, I) is an ordinal scale on X.
4.2 Type 1+2 Scale
Suppose that type 1+2 information {P, I, ?, P e} about X is available.
Definition 4.
A type 1+2 scale on X relative to {P, I, ?, P e} is a function \(\mu: X \rightarrow \mathbb{R}\) satisfying Conditions 1 and 2.
Condition 2
\(\forall \ x,y,z,w \in X\) , \([(x,y)P^{e}(z,w) \Rightarrow \mu (x) -\mu (y) >\mu (z) -\mu (w)].\)
Let \(Sc_{1+2}(X,P,I,P^{e}) =\{\mu: X \rightarrow \mathbb{R}\ \vert \ \mu\) is a type 1+2 scale on X relative to {P, I, P e}}. When X, P, I and P e are well determined, Sc 1+2(X, P, I, P e) will be noted Sc 1+2.
5 Consistency: Inconsistency
Definition 5.
Type 1 information {P, I, ?} about X is
-
consistent when Sc 1(X, P, I) ≠ ϕ
-
inconsistent when Sc 1(X, P, I) = ϕ.
Definition 6.
Type 1+2 information {P, I, ?, P e} about X is
-
consistent when \(Sc_{1+2}(X,P,I,P^{e})\not =\phi\)
-
inconsistent when \(Sc_{1+2}(X,P,I,P^{e}) =\phi\).
When \(Sc_{1+2}(X,P,I,P^{e}) =\phi\), one can have Sc 1(X, P, I) = ϕ or Sc 1(X, P, I) ≠ ϕ. In the first case, the message “no ranking” will appear in M-MACBETH; it occurs namely when J declares, in regards to elements x, y and z of X, that [xIy, yIz and xPz] or [xPy, yPz and z P x]. In the second case, the message “inconsistent judgement” will appear in M-MACBETH.
Although this is the only difference between the types of inconsistency introduced in M-MACBETH, it is interesting to mention, from a theoretical perspective, that one could further distinguish two sub-types of inconsistency (sub-type a and sub-type b) when \(Sc_{1+2}(X,P,I,P^{e}) =\phi\) and S c 1(X, P, I) ≠ ϕ.
Sub-type a inconsistency arises when there is a conflict between type 1 information and P e that makes the simultaneous satisfaction of conditions 1 and 2 impossible. These kinds of conflicts are found essentially in four types of situations; namely when x, y, z ∈ X exist such that
Sub-type b inconsistency arises when there is no conflict between type 1 information and P e but at least one conflict exists inside P e that makes satisfying Condition 2 impossible. An example of this type of conflict is (see Fig. 11.1):
In such a case, Condition 2 cannot be respected, because one should have
which is impossible.
On the other hand, it is easily shown that the following two systems are compatible, that is, there is no conflict between type 1 information and P e:
For a detailed study of inconsistency, see [95].
6 Consistency Test for Preferential Information
6.1 Testing Procedures
Suppose that X = { a 1, a 2, …, a n }.
During the interactive questioning process conducted with J, each time that a new judgement is obtained, the consistency of all the responses already provided is tested. This consistency test begins with a pre-test aimed at detecting the (potential) presence of cycles within the relation P and, if no such cycle exists, making a permutation of the elements of X in such a way that, in the matrix of judgements, all of the cells P or C i j will be located above the main diagonal.
When there is no cycle in P, the consistency of type 1 information {P, I, ?} is tested as follows:
-
If? ≠ ϕ, a linear program named LP-test1 is used;
-
if? = ϕ, rather than linear programming, a method named DIR-test1 is used, which has the advantage of being easily associated with a very simple visualization of an eventual ranking within the matrix of judgements.
When {P, I, ?} is consistent, the consistency of type 1+2 information {P, I, ?, P e} is tested with the help of a linear program named LP\(\sigma\)-test1+2.
6.2 Pre-test of the Preferential Information
The pre-test of the preferential information is based on Property 1. (Evident because # X is finite.)
Property 1.
Let \(X^{{\ast}}\subset X\); if \(\forall \ x \in X^{{\ast}},\exists \ y \in X^{{\ast}}\) such that x P y, then \(\exists \ x_{1},x_{2},\ldots,x_{p} \in X^{{\ast}}\) such that x 1 P x 2 P … P x p P x 1 (cycle).
The pre-test consists of seeking a permutation \(\varphi: \mathbb{N}_{1,n} \rightarrow \mathbb{N}_{1,n}\) such that
The permutation of the elements of X is made by the algorithm PRETEST, that detects cycles within P and sorts the elements(s) of X.
PRETEST:
-
1.
\(s \leftarrow n\);
-
2.
among a 1, a 2, …, a s find a i which is not preferred over any other:
-
if a i exists, go to 3.;
-
if not, return FALSE (S c 1 = ϕ, according to Property 1); finish.
-
-
3.
permute a i and a s ;
-
4.
\(s \leftarrow s - 1\):
-
if s = 1, return TRUE; finish.
-
If not, go to 2.
-
6.3 Consistency Test for Type 1 Information
Suppose that PRETEST detected no cycle within P and that the elements of X were renumbered as follows (to avoid the introduction of a permutation in the notation):
6.3.1 Consistency Test for Incomplete (? ≠ ϕ) Type 1 Information
Consider the linear program LP-test1 with variables x 1, x 2, …, x n :
where d min is a positive constant, and the variables x 1, x 2, …, x n represent the numbers μ(a 1), μ(a 2), …, μ(a n ) that should satisfy Condition 1 so that μ is a type 1 scale.
The objective function min x 1 of LP-test1 is obviously arbitrary. It is trivial that \(Sc_{1}\not =\phi \Leftrightarrow \) LP-test1 is feasible.
6.3.2 Consistency Test for Complete (? = ϕ) Type 1 Information
When? = ϕ and the elements of X have been renumbered (after the application of PRETEST), another simple test (DIR-test1) allows one to verify if \(P \cup I\) is a complete preorder on X. DIR-test1 is based on Proposition 1 (Proved in [95]).
Proposition 1.
If \([\ \forall \ i,j \in \mathbb{N}_{1,n}\) with i < j, \((a_{i},a_{j}) \in P \cup I\ ]\) then \(P\,\cup \,I\) is a complete preorder on X if and only if \(\forall \ i,j \in \mathbb{N}_{1,n}\mbox{ with }i < j: \left [a_{i}Pa_{j} \Rightarrow \left \{\begin{array}{l} \forall \ s \leq i,\forall \ t \geq j,a_{s}Pa_{t} \\ \exists \ s: i \leq s \leq j - 1\mbox{ and }a_{s}Pa_{s+1}\\ \end{array} \right.\right ].\)
Proposition 1 means that when the “P cases” of the matrix of judgements forms a “staircase”, a ranking exists such that each step of the “staircase” rests, at least partly, on the principal diagonal of the matrix.
6.4 Consistency Test for Type 1+2 Information
It would be possible to test the consistency of type 1+2 information with a linear program based on Conditions 1 and 2. However, the more efficient linear program LP-test1+2, which includes “thresholds conditions” equivalent to Conditions 1 and 2, is used instead. LP-test1+2 is based on Lemma 1 (Proved in [95]).
Lemma 1.
Let \(\mu: X \rightarrow \mathbb{R}.\) μ satisfies Conditions 1 and 2 if and only if there exist Q “thresholds” \(0 <\sigma _{1} <\sigma _{2} <\ldots <\sigma _{Q}\) that satisfy Conditions 3 , 4 and 5 .
Condition 3
\(\forall \ (x,y) \in I,\ \mu (x) =\mu (y)\) .
Condition 4
\(\forall \ i,j \in \mathbb{N}_{1,Q}\) with \(i \leq j,\ \forall \ (x,y) \in C_{ij},\ \sigma _{i} <\mu (x) -\mu (y)\) .
Condition 5
\(\forall \ i,j \in \mathbb{N}_{1,Q-1}\) with \(i \leq j,\ \forall \ (x,y) \in C_{ij},\ \mu (x) -\mu (y) <\sigma _{j+1}\) .
Program LP-test1+2 has variables \(x_{1}(=\mu (a_{1})),\ldots,x_{n}(=\mu (a_{n})),\sigma _{1},\ldots,\) \(\sigma _{Q}\):
Taking into account Lemma 1, it is trivial that \(Sc_{1+2}\not =\phi\) if and only if the linear program LP-test1+2, which is based on Conditions 3– 5, is feasible.
7 Dealing with Inconsistency
When a type 1+2 information {P, I, ?, P e} about X is inconsistent, it is convenient to be able to show J systems of constraints that render his or her judgements inconsistent and modifications of these judgements that would render LP\(\sigma\)-test1+2 feasible.
7.1 Systems of Incompatible Constraints
Suppose that LP-test1+2 is not feasible or, in other words, that the following system is incompatible (variables \(x_{1}(=\mu (a_{1})),\ldots,x_{n}(=\mu (a_{n}))\), \(\sigma _{1},\ldots,\sigma _{Q}\) nonnegative):
Conventions:
-
\(\mathbb{R}^{m\times n}\) is the set of the real matrices with m lines and n columns.
-
Matrix \(M \in \mathbb{R}^{m\times n}\) is “non-zero” (M ≠ 0) if at least one of its elements is not null.
-
Matrix \(M \in \mathbb{R}^{m\times n}\) is positive or null (M ≥ 0) if all of its elements are positive or null.
The system of incompatible constraints can be written in the matrix format as follows:
where
Note: if r = 0, one could consider that \(B = 0 \in \mathbb{R}^{1\times (n+Q)}\) without losing generality.
Let A be the matrix \(\left [\begin{array}{c} C\\ D \\ E\\ \end{array} \right ]\) \(\in \mathbb{R}^{p\times (n+Q)}\ \ \ (p = p_{1} + p_{2} + p_{3})\). The system of incompatible constraints can be written more simply as
In order to detect incompatibilities between the constraints (t1), (t2), (t3), (t4) and (t5) and propose eventual corrections, we apply Proposition 2 (Proved in [95]), which is a corollary of Mangasarian’s [150] version of the Theorem of the Alternative.
Proposition 2.
The system \(S\;\;\left \{A \cdot Z > 0;B \cdot Z = 0\right \}\) admits a solution Z ∈ \(\mathbb{R}^{(n+Q)\times 1}\) or there exists \(Y \in \mathbb{R}^{p\times 1},V,W \in \mathbb{R}^{r\times 1}\) with Y ≠0,Y ≥ 0,V ≥ 0,W ≥ 0 such that \(^{t}A \cdot Y + ^{t}B \cdot (V - W) = 0\) and \(\forall i \in \mathbb{N}_{1,r},\ V _{i} \cdot W_{i} = 0\) but never both.
The interest of Proposition 2 is that vectors Y, V and W have positive or null components, thus making it compatible with linear programming (see Sects. 11.7.3 and 11.7.4)
7.2 Example 1
Suppose that X = { a 1, a 2, a 3, a 4} and that J has formulated the following judgements:
-
P = { (a 1, a 2), (a 1, a 3), (a 2, a 3), (a 3, a 4)}
-
(a 1, a 2) ∈ C 1, (a 1, a 3) ∈ C 4, (a 2, a 3) ∈ C 2, (a 3, a 4) ∈ C 2.
Suppose that J also judges that \(a_{2}Pa_{4}\) and that (a 2, a 4) ∈ C 3. LP-test1 is feasible: the judgements are compatible with a ranking. LP-test1+2 is not feasible: the software informs J that his or her judgements are “inconsistent”.
Suppose now that J confirms his or her judgements. One must then have:
or, in matrix format (which one can denote as A ⋅ Z > 0):
Since it is known, according to Proposition 2, that the system has no solution, there necessarily exists \(Y \in \mathbb{R}^{16\times 1}(Y \not =0,Y \geq 0)\) such thatt A ⋅ Y = 0. Thus, positive or null (but not all null) real numbers y 1, y 2, …, y 16 exist such that \(\sum _{i=1}^{16}y_{i} \cdot Col_{i}\,=\,0\) (where C o l i is the column i of the matrixt A).
In this simple example, one can see that it is enough to make \(y_{2} = y_{5} = y_{8} = y_{9} = 1\) and \(y_{1} = y_{3} = y_{4} = y_{6} = y_{7} = y_{10} = y_{11} = y_{12} = y_{13} = y_{14} = y_{15} = y_{16} = 0\):
These four vectors correspond to the four constraints (2), (5), (8) and (9) above:
( *) and ( **) bring to the contradiction \(x_{1} - x_{4} > x_{1} - x_{4}\). The incompatibility between ( *) and ( **) is presented in M-MACBETH as shown in Fig. 11.2.
Note that the problem disappears if
Note also that the inconsistency would not be eliminated for any modification of the judgement “(a 2, a 3) ∈ C 2”.
If J confirms the judgement “(a 2, a 4) ∈ C 3”, M-MACBETH calculates the different possibilities (four in example 1) that J can follow to make his or her judgements consistent with a “minimal” number of changes of category (one in Example 1). (We will specify in Sect. 11.7.4 the meaning of this notion).
In M-MACBETH, the “suggestions” of changes are presented (graphically) in the matrix of judgements. They are:
-
to replace the judgement (a 1, a 3) ∈ C 4 with the judgement (a 1, a 3) ∈ C 3
-
or to replace the judgement (a 2, a 4) ∈ C 3 with the judgement (a 2, a 4) ∈ C 4
-
or to replace the judgement (a 3, a 4) ∈ C 2 with the judgement (a 3, a 4) ∈ C 1
-
or to replace the judgement (a 1, a 2) ∈ C 1 with the judgement (a 1, a 2) ∈ C 2.
7.3 Identifying Constraints which Cause Inconsistency
Let us detail the various stages of our search for “suggestions”. The first step consists of determining the constraints (t1), (t2) and (t3) which are “the origin of the incompatibilities” present in the system
We consider that a constraint is “at the origin of an incompatibility” when it is part of a system S′ that
-
is a “sub-system” of S,
-
is incompatible,
-
does not contain any incompatible “sub-system”.
Mathematically, this idea can be represented by Definition 7.
Definition 7.
An incompatible elementary system (SEI) is a system
such that
-
1.
\(A' \in \mathbb{R}^{p'\times (n+Q)}\) is a sub-matrix of A, and \(B' \in \mathbb{R}^{r'\times (n+Q)}\) is a sub-matrix of B;
-
2.
S′ is incompatible;
-
3.
If \(\left \{\begin{array}{l} A'' \in \mathbb{R}^{p''\times (n+Q)}\mbox{ is a sub-matrix of }A', \\ B'' \in \mathbb{R}^{r''\times (n+Q)}\mbox{ is a sub-matrix of }B', \\ p'' + r'' < p' + r'\end{array} \right.\) then \(\left \{\begin{array}{l} A'' \cdot Z > 0\\ B'' \cdot Z = 0\\ \end{array} \right.\) is compatible.
However, our goal is not to determine all the SEI that could be extracted from the constraints using LP\(\sigma\)-test1+2. We just want to find all of the judgements of the type (a s , a t ) ∈ C i j that “generate” an incompatibility. In Sect. 11.7.4.3, we will explain how we use these judgements.
We know that an inconsistency occurs when the system
is incompatible; that is, \(\exists Y \in \mathbb{R}^{p}\) and \(V,W \in \mathbb{R}^{r}\) such that
In such a case, if i 0 ≤ p 1 + p 2, where p 1 is the number of constraints (t2) and p 2 is the number of constraints (t3) (see Sect. 11.7.1), a constraint of the type \(x_{s} - x_{t} <\sigma _{j}\) or \(x_{s} - x_{t} >\sigma _{j}\) will correspond to S.
Consider, then, the system (with i ≤ p 1 + p 2):
If Syst-Y i is compatible, for one of its solutions it corresponds to a system of incompatible constraints (t1), (t2), (t3), (t4) and (t5) where at least one constraint (that which corresponds to Y i = 1) is of the type \(x_{s} - x_{t} <\sigma _{j}\) or \(x_{s} - x_{t} >\sigma _{j}\) and is part of a SEI. If Syst-Y i is incompatible, the constraint that corresponds to Y i is not part of any SEI.
To find all of the constraints (t2) and (t3) which are part of a SEI, it is sufficient to study the compatibility of all of the systems Syst-Y i , for \(i = 1,2,\ldots,p_{1} + p_{2}\).
We will proceed in a similar way, using the systems Syst-V i and Syst-W i , to find all of the constraints (t1) which are part of a SEI:
and
It is not necessary to examine all of the systems Syst-Y i , Syst-V i and Syst-W i :
-
If Syst-Y i is compatible and has the solution Y, V, W, then
-
\(\forall \ j > i\) such that Y j ≠ 0, Syst-Y i is compatible;
-
\(\forall \ j \in \mathbb{N}_{1,r}\) such that V j ≠ 0, Syst-V i is compatible;
-
\(\forall \ j \in \mathbb{N}_{1,r}\) such that W j ≠ 0, Syst-W i is compatible.
-
-
If Syst-V i is compatible and has the solution Y, V, W, then
-
\(\forall \ j > i\) such that V j ≠ 0, Syst-V i is compatible;
-
\(\forall \ j \in \mathbb{N}_{1,r}\) such that W j ≠ 0, Syst-W i is compatible.
-
-
If Syst-W i is compatible and has the solution Y, V, W, then
-
\(\forall \ j > i\) such that W j ≠ 0, Syst-W i is compatible.
-
It is for this reason that a “witness-vector” \(T \in \mathbb{N}^{p_{1}+p_{2}+2\cdot r}\) must be used, initially null, updated as follows:
-
For any solution Y, V, W of a system Syst-Y i , Syst-V i or Syst-W i do
-
\(\forall \ j \in \mathbb{N}_{1,p_{1}+p_{2}},\ [\ Y _{j}\not =0 \Rightarrow T_{j} = 1\ ]\)
-
\(\forall \ j \in \mathbb{N}_{1,r},\ [\ V _{j}\not =0 \Rightarrow T_{p_{1}+p_{2}+j} = 1\ ]\)
-
and \([\ W_{j}\not =0 \Rightarrow T_{p_{1}+p_{2}+r+j} = 1\ ]\).
-
To find the interesting pairs, the compatibility of at most \(p_{1} + p_{2} + 2r\) systems should be studied. The general algorithm to seek equations (t1) and inequalities (t2) and (t3) that are part of a SEI is the following:
-
T = (0, 0, …, 0)
-
for \(i = 1,2,\ldots,p_{1} + p_{2}\) do:
-
T i = 0,
-
then if Syst-Y i compatible and Y, V, W solution of Syst-Y i
-
then update T
-
-
-
for i = 1, 2, …, r do:
-
if \(T_{p_{1}+p_{2}+i} = 0\),
-
then if Syst-V i compatible and Y, V, W solution of Syst-V i
-
then update T
-
-
-
for i = 1, 2, …, r do:
-
if \(T_{p_{1}+p_{2}+r+i} = 0\),
-
then if Syst-W i compatible and Y, V, W solution of Syst-W i
-
then update T.
-
-
In this way one obtains the set of all of the equations and inequalities that make up the SEI.
7.4 Augmentation: Reduction in a Judgement with p Categories
7.4.1 Preliminaries
Notation:
-
Judgement (x, y) ∈ C i j will be represented by element (x, y, i, j) of \(X \times X \times \mathbb{N}_{1,Q} \times \mathbb{N}_{1,Q}\).
-
Judgement (x, y) ∈ I will be represented by element (x, y, 0, 0) of \(X \times X \times \mathbb{N} \times \mathbb{N}\).
Definition 8.
A reduction in judgement (s, t, i, j) with p categories (1 ≤ p ≤ Q + i) is the replacement of this judgement
-
by the judgement \((s,t,i - p,i - p)\) if i ≥ p
-
by the judgement \((t,s,p - i,p - i)\) if i < p.
Definition 9.
An augmentation of the judgement (s, t, i, j) with p categories (1 ≤ p ≤ Q − j) is the replacement of this judgement by the judgement \((s,t,j + p,j + p)\).
Definition 10.
A change of judgement (s, t, i, j) with p categories is an augmentation or a reduction of the judgement with p categories.
Comment: It is evident that one obtains the same final judgement as a result of “1 reduction of a judgement with p categories” or the “p successive reductions of a category of 1 judgement”.
Convention: A “change in judgement (s, t, i, j) with p categories” will be represented by \((s,t,i,j,p) \in X \times X \times \mathbb{N}_{1,Q} \times \mathbb{N}_{1,Q} \times \mathbb{Z}\) (augmentation if p > 0, reduction if p < 0).
7.4.2 Exploitation of the Constraints of SEI
Let us recall from end of Sect. 11.7.3 (T is the “witness-vector”) that
-
if T i > 0, it has a corresponding constraint (t2) or (t3) or (t1) that is part of an SEI;
-
if T i = 0, it has no corresponding constraint that is part of an SEI.
These variables, then, provide us with an indication as to the future “modification” to be made to the judgements associated with these constraints. Indeed, suppose that T i > 0:
-
(a)
if 1 ≤ i ≤ p 1, a constraint \(\sigma _{u} < x_{s} - x_{t}\) which is part of an SEI corresponds to variable T i ; if a change in its judgement (s, t, …, …) can help to eliminate the SEI, it ensures that it will be a “reduction” (evident).
-
(b)
if \(p_{1} + 1 \leq i \leq p_{1} + p_{2}\), a constraint \(x_{s} - x_{t} <\sigma _{u}\) which is part of an SEI corresponds to variable T i ; if a change in its judgement (s, t, …, …) can help to eliminate the SEI, it ensures that it will be an “augmentation” (evident).
-
(c)
if \(p_{1} + p_{2} + 1 \leq i \leq p_{1} + p_{2} + r\), a constraint \(x_{s} - x_{t} = 0\) which is part of an SEI corresponds to variable T i ; if a change in its judgement (s, t, 0, 0) can help to eliminate the SEI, it ensures that it will be a “reduction”.
-
(d)
if \(p_{1} + p_{2} + r + 1 \leq i \leq p_{1} + p_{2} + 2r\), a constraint \(x_{s} - x_{t} = 0\) which is part of an SEI corresponds to variable T i ; if a change in its judgement (s, t, 0, 0) can help to eliminate the SEI, it ensures that it will be an “augmentation” (proof similar to that of (c)).
Proof of (c):
Being \(h = i - (p_{1} + p_{2})\), one knows (by the definition of T i ) that \(\exists \ Y \in \mathbb{R}^{p},\ \exists \ V,W \in \mathbb{R}^{r}\) with \(Y \geq 0,\ V \geq 0,\ W \geq 0,\ Y \not =0,\ V _{h}\not =0\) and W h = 0 such that \(^{t}(A') \cdot Y + ^{t}(B') \cdot (V - W) = 0\) or, if one notes L i n e B j the jth line of B′,
(because W h = 0).
The corresponding SEI \(\left \{\begin{array}{l} A' \cdot Z > 0\\ B' \cdot Z = 0\\ \end{array} \right.\) can be written \(\left \{\begin{array}{l} A' \cdot Z > 0\\ x_{ s} - x_{t} = 0\\ B'' \cdot Z = 0, \\ \end{array} \right.\) where \(B'' = \left [\begin{array}{c} LineB_{1}\\ \vdots \\ LineB_{h-1} \\ LineB_{h+1}\\ \vdots \\ LineB_{r} \end{array} \right ]\) (the matrix B′ without line L i n e B h ).
If one considers an “augmentation” of judgement (s, t, 0, 0), the constraint \(x_{s} - x_{t} = 0\) would be replaced by the constraint x s − x t > 0. The new system \(\left \{\begin{array}{l} A' \cdot Z > 0\\ x_{ s} - x_{t} > 0\\ B'' \cdot Z = 0\\ \end{array} \right.\) can be written \(\left \{\begin{array}{l} A'' \cdot Z > 0\\ B'' \cdot Z = 0,\\ \end{array} \right.\) where \(A'' = \left [\begin{array}{c} A'\\ LineB_{h}\\ \end{array} \right ]\) (the matrix A′ “augmented” with line L i n e B h ).
The system is still incompatible; indeed, if one poses
-
\(Y ' = (Y _{1},Y _{2},\ldots,Y _{p},V _{h}) \in \mathbb{N}^{p+1}\)
-
\(V ' = (V _{1},\ldots,V _{h-1},V _{h+1},\ldots,V _{r}) \in \mathbb{N}^{r-1}\)
-
\(W' = (W_{1},\ldots,W_{h-1},W_{h+1},\ldots,W_{r}) \in \mathbb{N}^{r-1}.\)
can be written: \(^{t}(A'') \cdot Y ' + ^{t}(B'') \cdot (V ' - W') = 0\), where Y ′ ≠ 0 (since Y ≠ 0), which proves the incompatibility of the system.
Each “suggestion” of a potential change (T i = 1) of a judgement (s, t, …, …) can thus be stored in a vector S of \(\mathbb{N}^{4}\) where
We will denote by P r e S u g g the set of these “pre-suggestions”. In the case of example 1 (see Sect. 11.7.3) one has
7.4.3 Search for Suggestions
Definition 11.
Changing judgements by m categories is any set M o d i f m of the form \(Modif_{m} =\{ (s_{1},t_{1},i_{1},j_{1},p_{1}),(s_{2},t_{2},i_{2},j_{2},p_{2}),\ldots,(s_{u},t_{u},i_{u},\) \(j_{u},p_{u})\ \vert \forall v \in \mathbb{N}_{1,u},(s_{v},t_{v},i_{v},j_{v},p_{v})\) is a change of judgement (s v , t v , i v , j v ) with p v categories} such that \(\sum _{v=1}^{u}\vert p_{v}\vert = m\)
Within Example 1, {(a 1, a 2, 1, 1, 2), (a 3, a 4, 2, 2, −1)} is a “change of judgements with 3 categories”, which consists of
-
to replace the judgement (a 1, a 2) ∈ C 1 with the judgement (a 1, a 2) ∈ C 3 (augmentation of 2 categories)
-
to replace the judgement (a 3, a 4) ∈ C 2 with the judgement (a 3, a 4) ∈ C 1 (reduction of 1 category)
Notation: the set of “judgement changes with m categories” which renders the judgements consistent will be denoted by S u g g m .
Within Example 1,
-
{(a 1, a 2, 1, 1, 2), (a 3, a 4, 2, 2, −1)} ∈ S u g g 3
-
{(a 1, a 3, 4, 4, −1)}, {(a 3, a 4, 2, 2, −1)}, {(a 1, a 2, 1, 1, 1)} and {(a 2, a 4, 3, 3, 1)} ∈ S u g g 1,
these are the 4 changes suggested in Sect. 11.7.3.
Once the PreSugg group is determined, the third step is to:
-
determine the “minimum number of changes” (some possibly successive) necessary to render the judgements consistent;
-
determine all of the combinations of such “minimal” changes.
More rigorously, this means
-
find \(m_{0} =\min \{m \in \mathbb{N}^{{\ast}}\vert Sugg_{m}\not =\emptyset \}\)
-
clarify S u g g m
In Example 1, we have already seen that m 0 = 1 (since \(Sugg_{1}\not =\emptyset\)).
We will proceed as follows for all cases of inconsistency (see Fig. 11.3).
At each step i,
-
the set of all “judgement changes of i categories”, built on the basis of element PreSugg are considered;
-
for each of the elements in this group:
-
carry out the modifications included in the selected item;
-
test the consistency of the new matrix of judgements; if it is consistent, store the element in S u g g i ;
-
restore the matrix to the initial judgements.
-
It is worth mentioning that we consider the possibility of changing a judgement by several categories.
This algorithm is always convergent since one can always give consistent judgements in a finite number of changes.
We emphasize that in practice, the cases of inconsistency that require more than 2 “changes of 1 category” are almost non-existent. The main reason being that any change in judgement that generates an inconsistency is immediately announced to J, who must then confirm or cancel his or her judgement.
This procedure allows one to avoid
-
coarse errors of distraction (by cancelling the judgement);
-
the “accumulation” of inconsistencies since, if J confirms his or her judgement, suggestions of changes that will eliminate the inconsistency are made.
7.5 Example 2
Suppose that X = { a 1, a 2, a 3, a 4} and that J has formulated the following consistent judgements:
-
P = { (a 1, a 2), (a 1, a 3), (a 2, a 3), (a 3, a 4)}
-
(a 1, a 2) ∈ C 1, (a 1, a 3) ∈ C 4, (a 2, a 3) ∈ C 2, (a 3, a 4) ∈ C 3
Suppose that J adds that \(a_{2}Pa_{4}\) and that (a 2, a 4) ∈ C 3: M-MACBETH informs J that his or her judgements are “inconsistent”.
If J confirms the judgement (a 2, a 4) ∈ C 3, M-MACBETH will display the message: “Inconsistent judgements: MACBETH has found 6 ways to render the judgements matrix consistent with 2 category changes.”
This time, it will be necessary to make at least 2 “changes of 1 category” to render the judgements consistent; there are 6 distinct combinations of such changes. Each of these 6 suggestions is presented graphically (see Fig. 11.4) within the table of judgements, accompanied by SEI which, moreover, shows why the suggestions made eliminate this incompatibility: Fig. 11.4 presents the first of six suggestions.
8 The MACBETH Scale
8.1 Definition of the MACBETH Scale
Suppose that \(Sc_{1+2}\not =\phi\) and \(a_{1}(P \cup I)a_{2}\ldots a_{n-1}(P \cup I)a_{n}\). The linear program LP-MACBETH with variables \(x_{1},\ldots,x_{n},\sigma _{1},\ldots,\) \(\sigma _{Q}\) is therefore feasible:
Definition 12.
Any function \(EchMac: X \rightarrow \mathbb{R}\) such that \(\forall \ i \in \mathbb{N}_{1,n},\) \(EchMac(a_{i}) = x_{i}^{{\ast}}\)—where \((x_{1}^{{\ast}},\ldots,x_{n}^{{\ast}})\) is an optimal solution of LP-MACBETH—is called a basic MACBETH scale.
Definition 13.
\(\forall \ a \in \mathbb{R}_{+}^{{\ast}},\forall \ b \in \mathbb{R}\) with \((a,b)\not =(1,0),\ \ a \cdot EchMac + b\) is a transformed MACBETH scale.
8.2 Discussing the Uniqueness of the Basic MACBETH Scale
Nothing guarantees that a LP-MACBETH optimal solution is unique. For example, consider the matrix of judgements and the basic MACBETH scale shown is Fig. 11.5.
One can verify that, \(\forall \ x \in [6,7],(8,x,5,2,1,0)\) is still an optimal solution of LP-MACBETH. Thus, a basic MACBETH scale is not necessarily unique. As long as the MACBETH scale is interpreted as a technical aid whose purpose is to provide the foundation for a discussion with J, this does not constitute a true problem. However, we have observed that in practice decision makers often adopt the MACBETH scale as the final scale. It is, therefore, convenient to guarantee the uniqueness of the MACBETH scale. This is obtained technically, as follows (where S m a c is the group of the constraints of LP-MACBETH):
-
Step (1) solution of LP-MACBETH
-
\(\rightarrow \) optimal solution x 1, x 2, …, x n
-
\(\rightarrow \mu (a_{1}) = x_{1},\mu (a_{n}) = x_{n} = 0\) (remark: μ(a 1) is unique)
-
-
Step (2) for i = 2 to n − 1
-
to solve \(\max x_{i}\) under \(\left \{\begin{array}{l} S_{mac} \\ x_{1} =\mu (a_{1}),\ldots,x_{i-1} =\mu (a_{i-1})\end{array} \right.\)
-
\(\rightarrow \) optimal solution x 1, x 2, …, x n
-
\(\rightarrow xmax = x_{i}\)
-
to solve \(\min x_{i}\) under \(\left \{\begin{array}{l} S_{mac} \\ x_{1} =\mu (a_{1}),\ldots,x_{i-1} =\mu (a_{i-1})\end{array} \right.\)
-
\(\rightarrow \) optimal solution x 1, x 2, …, x n
-
\(\rightarrow xmin = x_{i}\)
$$\displaystyle{\mu (a_{i}) = \frac{xmin + xmax} {2} }$$
-
Thus,
-
to calculate μ(a 2), the variable x 1 is “fixed” to the value μ(a 1), the minimum and maximum values of x 2 are calculated and the average of the two results is taken as the value of μ(a 2);
-
to calculate μ(a 3), the variable x 1 is “fixed” to the value of μ(a 1), the variable x 2 is “fixed” to the value of μ(a 2), the minimum and maximum values of x 3 are calculated and the average of the two values is taken as the value of μ(a 3);
-
etc.
This method guarantees that μ(a 1), μ(a 2), … μ(a n ) are unique for a given preferential information {P, I, ? = ϕ, P e}. It permits us to speak of “the” basic MACBETH scale, instead of “one” MACBETH scale.
8.3 Presentation of the MACBETH Scale
The MACBETH scale that corresponds to {P, I, ? = ϕ, P e} consistent information is represented in two ways in M-MACBETH: a table and a “thermometer”. In the example in Fig. 11.6, the transformed MACBETH scale represented in the thermometer was obtained by imposing the values of the elements d and c as 100 and 0 respectively.
Even though the values attributed to c and d are fixed, in general an infinite number of scales that satisfy Conditions 1 and 2 exist. It is, thus, necessary to allow J to, should he or she want to, modify the values suggested. This is the subject of the Sect. 11.9.
8.4 Determining by Hand the Basic MACBETH Scale
In the case of small matrices of judgements, the basic MACBETH scale can be determined by hand.
Let us firstly present two alternative and equivalent formulations of the linear program LP-MACBETH (see Sect. 11.8.1): LP-MACBETH2008 [50] and LP-MACBETH2011 [55, 57].
-
1.
LP-MACBETH2008 with variables x 1, ⋯ , x n , \(\sigma _{1},\cdots \,,\sigma _{Q}\)
$$\displaystyle\begin{array}{rcl} & & \min \ \ (x_{1} - x_{n}) {}\\ & & \mbox{ s. t. }\,x_{p} - x_{r} = 0\qquad \qquad \forall (a_{p},a_{r}) \in \mathbb{I}\mbox{ with }p < r\qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{1}$)} {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} \sigma _{i} + \frac{1} {2} \leq x_{p} - x_{r}& & \qquad \begin{array}{r} \forall i,j \in \mathbb{N}_{1,Q}\mbox{ with }i \leq j, \\ \forall (a_{p},a_{r}) \in C_{ij} \end{array} \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{2}^{{\prime}}$)}{}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} x_{p} - x_{r} \leq \sigma _{j+1} -\frac{1} {2}& & \qquad \begin{array}{r} \forall i,j \in \mathbb{N}_{1,Q-1}\mbox{ with }i \leq j, \\ \forall (a_{p},a_{r}) \in C_{ij} \end{array} \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{3}^{{\prime}}$)}{}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} \sigma _{1} = \frac{1} {2}\qquad \qquad \qquad \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{4}^{{\prime}}$)}& & {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} \sigma _{i-1} + 1 \leq \sigma _{i}& & \qquad \forall i \in \mathbb{N}_{2,Q}\qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{5}^{{\prime}}$)} {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} x_{n} = 0\qquad \qquad \qquad \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{6}$)}& & {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} & x_{i} \geq 0\qquad \forall i \in \mathbb{N}_{1,n}& {}\\ & \sigma _{i} \geq 0\qquad \forall i \in \mathbb{N}_{1,Q} & {}\\ \end{array}$$ -
2.
LP-MACBETH2011 with variables x 1, ⋯ , x n
$$\displaystyle\begin{array}{rcl} & & \min \ \ (x_{1} - x_{n}) {}\\ & & \mbox{ s. t. }\,x_{p} - x_{r} = 0\qquad \qquad \forall (a_{p},a_{r}) \in \mathbb{I}\mbox{ with }p < r\qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{1}$)} {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} & & x_{n} = 0\qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{6}$)} {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} x_{p} - x_{r} \geq i& & \qquad \begin{array}{r} \forall i,j \in \mathbb{N}_{1,Q}\mbox{ with }i \leq j, \\ \forall (a_{p},a_{r}) \in C_{ij} \end{array} \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{7}$)}{}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} x_{p} - x_{r} \geq x_{k} - x_{m} + i - j^{{\prime}}&&\qquad \begin{array}{r} \forall i,j,i^{{\prime}},j^{{\prime}}\in \mathbb{N}_{1,Q}\mbox{ with } \\ i \leq j,i^{{\prime}}\leq j^{{\prime}}\mbox{ and } \\ i > j^{{\prime}},\forall (a_{p},a_{r}) \in C_{ij}, \\ \forall (a_{k},a_{m}) \in C_{i^{{\prime}},j^{{\prime}}} \end{array} \qquad \qquad \qquad \qquad \qquad \mbox{ ($t_{8}$)} {}\\ & & {}\\ \end{array}$$$$\displaystyle\begin{array}{rcl} x_{i} \geq 0& & \qquad \forall i \in \mathbb{N}_{1,n} {}\\ \end{array}$$
Indeed, it is easy to prove that:
If \([\forall i,j \in \mathbb{N}_{1,Q}\text{ with }i < j,Cij =\emptyset ]\) (i.e., there is no hesitation between two categories for any of the judgements elicited), the constraints (t 7) and (t 8) can be written simply as:
Consider the consistent matrix of judgements (with n = 5 and Q = 6) in Fig. 11.7. On the basis of constraint (t 9), the corresponding basic MACBETH scale can be determined by hand as follows:
As x 5 = 0 [constraint (t 6)], one only needs to determine the four “elementary differences”
-
\(\forall (a_{i},a_{i+1}) \in C_{1}\text{ with }i \in \{ 1,2,3,4\}\text{, take }x_{i} - x_{i+1} = 1\)
Here: \(x_{3} - x_{4} = 1\)
-
\(\forall i,j \in \{ 1,2,3,4,5\}\) with i < j and j − i ≥ 2, calculate the difference x i − x j whenever it is possible.
Here, i.e. for the matrix in Fig. 11.7 no difference x i − x j with i < j and j − i ≥ 2 can be calculated at this stage.
-
If constraint (t 9) is not respected, modify the values of the elementary differences \(x_{i} - x_{i+1}\).
Here, the constraint is obviously respected at this stage.
-
\(\forall (a_{i},a_{i+1}) \in C_{2}\) with i ∈ { 1, 2, 3, 4}, take \(x_{i} - x_{i+1} =\alpha _{1} + 1\) where \(\alpha _{1} =\max \{ x_{i} - x_{j}\vert i < j\text{ and }(a_{i},a_{j}) \in C_{1}\}\).
Here, α 1 = 1 and we take: \(x_{1} - x_{2} = 2\) and \(x_{4} - x_{5} = 2\).
-
\(\forall i,j \in \{ 1,2,3,4,5\}\) with i < j and j − i ≥ 2, calculate the difference x i − x j whenever it is possible.
Here, we have: \(x_{3} - x_{5} = (x_{3} - x_{4}) + (x_{4} - x_{5}) = 3\).
-
If constraint (t 9) is not respected, modify the values of the elementary differences \(x_{i} - x_{i+1}\).
Here, the constraint is respected at this stage.
-
\(\forall (a_{i},a_{i+1}) \in C_{3}\) with i ∈ { 1, 2, 3, 4}, take \(x_{i} - x_{i+1} =\alpha _{2} + 1\) where \(\alpha _{2} =\max \{ x_{i} - x_{j}\vert i < j\text{ and }(a_{i},a_{j}) \in C_{2}\}\).
Here, α 2 = 3 and we take: \(x_{2} - x_{3} = 4\).
-
\(\forall i,j \in \{ 1,2,3,4,5\}\) with i < j and j − i ≥ 2, calculate the difference x i − x j whenever it is possible.
$$\displaystyle{\begin{array}{ll} \text{Here, we have:}&x_{1} - x_{3}\; = (x_{1} - x_{2}) + (x_{2} - x_{3}) = 6 \\ &x_{1} - x_{4}\: = (x_{1} - x_{2}) + (x_{2} - x_{3}) + (x_{3} - x_{4}) = 7 \\ &\begin{array}{r} x_{1} - x_{5}\; = (x_{1} - x_{2}) + (x_{2} - x_{3}) + (x_{3} - x_{4}) + (x_{4} \\ - x_{5}) = 9\end{array} \\ &x_{2} - x_{4}\; = (x_{2} - x_{3}) + (x_{3} - x_{4}) = 5 \\ &x_{2} - x_{5}\; = (x_{2} - x_{3}) + (x_{3} - x_{4}) + (x_{4} - x_{5}) = 7 \\ &x_{3} - x_{5}\; = (x_{3} - x_{4}) + (x_{4} - x_{5}) = 3.\end{array} }$$ -
If constraint (t 9) is not respected, modify the values of the elementary differences \(x_{i} - x_{i+1}\).
As can be observed in Fig. 11.8, there is a problem here: as (a 2, a 5) ∈ C 5 and (a 1, a 4) ∈ C 4, one must have \(x_{2} - x_{5} \geq x_{1} - x_{4} + 1\) that is \((x_{2} - x_{3}) + (x_{3} - x_{4}) + (x_{4} - x_{5}) \geq (x_{1} - x_{2}) + (x_{2} - x_{3}) + (x_{3} - x_{4}) + 1\). So, in terms of the elementary differences (x 1 − x 2), (x 2 − x 3), (x 3 − x 4) and (x 4 − x 5), one must have \((x_{4} - x_{5}) \geq (x_{1} - x_{2}) + 1\). which implies, in our case: (x 4 − x 5) ≥ 3 and one takes the smallest possible value, that is \(x_{4} - x_{5} = 3\), and consequently \(x_{3} - x_{5} = 4\).
-
\(\forall (a_{i},a_{i+1}) \in C_{3}\) with i ∈ { 1, 2, 3, 4}, take \(x_{i} - x_{i+1} =\alpha _{ 2}^{{\ast}} + 1\) where \(\alpha _{2}^{{\ast}} =\max \{ x_{i} - x_{j}\vert i < j\text{ and }(a_{i},a_{j}) \in C_{2}\}\).
Here, \(\alpha _{2}^{{\ast}} = 4\) and one takes: \(x_{2} - x_{3} = 5\).
-
\(\forall i,j \in \{ 1,2,3,4,5\}\) with i < j and j − i ≥ 2, calculate the difference x i − x j whenever it is possible.
$$\displaystyle\begin{array}{rcl} \text{Here, we have: }& & x_{1} - x_{3}\; = (x_{1} - x_{2}) + (x_{2} - x_{3}) = 7 {}\\ & & x_{1} - x_{4}\; = (x_{1} - x_{2}) + (x_{2} - x_{3}) + (x_{3} - x_{4}) = 8 {}\\ & & \begin{array}{r} x_{1} - x_{5} = (x_{1} - x_{2}) + (x_{2} - x_{3}) + (x_{3} - x_{4}) + (x_{4} \\ - x_{5}) = 11 \end{array} {}\\ & & x_{2} - x_{4}\; = (x_{2} - x_{3}) + (x_{3} - x_{4}) = 6 {}\\ & & x_{2} - x_{5}\; = (x_{2} - x_{3}) + (x_{3} - x_{4}) + (x_{4} - x_{5}) = 9 {}\\ & & x_{3} - x_{5}\; = (x_{3} - x_{4}) + (x_{4} - x_{5}) = 4. {}\\ \end{array}$$
As all the elementary differences are determined and the constraint (t 9) is respected (see Fig. 11.9), the basic MACBETH scale can be obtained:
9 Discussion About a Scale
Suppose that, in the example in Fig. 11.6, J considers that the element a is badly positioned when compared to elements c and d and therefore J wants to redefine the value of a. It is then interesting to show J the limits within which the value of a can vary without violating the preferential information provided by J. Let us suppose in this section that we have a type 1+2 information about X which is consistent and that? = ϕ.
Let μ 0 be a particular scale of S c 1+2, L and H be two fixed elements of X with H P L (H more attractive than L) and a be an element of X (not indifferent to L and not indifferent to H) that J would like to have repositioned.
Let
-
\(Sc_{(\mu _{0},H,L)} =\{\mu \in Sc_{1+2}\ \vert \ \mu (H) =\mu _{0}(H)\) and μ(L) = μ 0(L)} (scales for which values associated with H and L have been fixed)
-
\(Sc_{(\mu _{0},\hat{a})} =\{\mu \in Sc_{1+2}\ \vert \ \forall \ y \in X\) with y not indifferent to a: μ(y) = μ 0(y)} (scales for which the values of all of the elements of X except a and its eventual equals have been fixed).
We call free interval associated to interval a :
We call dependent interval associated to interval a :
In the example in Fig. 11.6, if one selects a, two intervals are presented to J (see Fig. 11.10) which should be interpreted as follows:
The closed intervals (in the example [66. 69, 99. 98] and [72. 74, 90. 9]) that have been chosen to present to J are not the precise free and dependent intervals associated to a (which, by definition, are open); however, by taking a precision of 0. 01 into account, they can be regarded as the “greatest” closed intervals included in the free and dependent intervals.
M-MACBETH permits the movement of element a with the mouse but, obviously, only inside of the dependent interval associated to a.
If J wants to give element a a value that is outside of the dependent interval (but still inside the free interval), the software points out that the values of the other elements must be modified. If J confirms the new value of a, a new MACBETH scale is calculated, taking into account the additional constraint that fix the new value of a.
The (“closed”) free interval is calculated by integer linear programming. The (“closed”) dependent interval could be also calculated in the same manner. However, M-MACBETH computes it by “direct” calculation formulas which make the determination of these intervals extremely fast—for details, see [95].
10 MACBETH and MCDA
The MACBETH approach and the M-MACBETH software have been used to build value functions and scoring and weighting scales, in the process of developing multicriteria decision aid models, in particular many simple additive value models. In this framework, the MACBETH weighting procedure is presented in detail in [57]. A classification of applications of MACBETH reported in the literature is presented hereafter (further references to applications reported in Portuguese can be found in [58]).
- Agriculture, Manufacturing & Services: :
-
Finance: [19, 20, 42, 49, 70, 72, 73, 78, 105, 116–119, 134, 185];
Information systems: [57, 124, 188];
Performance measurement: [64–67, 69, 71, 83, 88, 101, 107, 108, 127–129, 138, 142–145, 151, 189, 192];
Production & service planning: [2, 3, 29, 36, 79, 89, 90, 131, 139, 157, 171, 172, 174, 179];
Quality management: [12, 56, 80, 81, 112];
R&D project selection: [96];
Risk management: [84, 121, 161];
Strategy & resource allocation: [37, 38, 170];
Supply chain and logistics: [97–99, 133, 140, 141, 165, 183, 198];
- Energy: :
-
Project prioritization and selection: [50, 100];
Technology choice: [60, 61, 75, 110, 111, 115, 120, 155, 156, 196];
- Environment: :
-
Landscape management: [186, 190];
Risk management: [13, 51, 87, 135, 166, 184];
- Medical: :
- Military: :
- Public Sector: :
-
Conflict analysis and management: [8, 41, 68, 102, 177, 187];
Procurement: [7, 9, 11, 43, 57] [14, 15, 18, 24, 33, 40, 48, 50, 149, 173];
Project prioritization & resource allocation: [16, 26, 30, 32, 35, 47, 76, 77, 152, 167–169, 180, 181, 193];
Strategic planning & development: [34, 44, 52, 53, 59, 130, 160, 194];
- Others: :
-
Human resource management: [17, 54, 103, 104, 106, 109, 122, 123, 132, 146, 147, 191];
Job selection: [10];
Sports: [74].
References
Abtahi, M.S.: The effectiveness in distance education for Iranian higher education. In: Procedia – Social and Behavioral Sciences, vol. 47, pp. 1315–1319, North Cyprus, 8–10 February 2012. Cyprus International Conference on Educational Research (CY-ICER) (2012)
Amora Silva, M.B.F., Bana e Costa, C.A.: Modelo multicritério de avaliação de capacidade empreendedora em empresas de base tecnológica. Engevista 10(1), 4–14 (2008)
Andrade, A.C.Z.B., Bana e Costa, C.A.: Measuring security value: a multi-criteria decision analysis approach. Working paper cegist 20/2001, CEG-IST, Technical University of Lisbon (2001)
Antão da Silva, P., Bana e Costa, C.A., Correia, F.N.: Avaliação multicritério das incidências ambientais de medidas de controlo de cheias: Aplicação à bacia hidrográfica da Ribeira do Livramento. In: Trabalhos Técnicos do 1o Congresso sobre Aproveitamento e Gestão de Recursos Hídricos em Países de Idioma Português, pp. 210–217, Rio de Janeiro (2000). Associação Brasileira de Engenharia Sanitária e Ambiental, CD-ROM edition
Ba, F., Bouchard, C., Abi-zeid, I.: Analyse multicritère pour la priorisation des interventions en matière d’approvisionnement en eau en milieu rural au Sénégal: cas de la région de Diourbel. Rev. Sci. Eau 24(1), 9–22 (2011)
Bana e Costa, C.A.: Structuration, Construction et Exploitation d’un Modèle Multicritère d’Aide à la Décision. Ph.D. thesis, Technical University of Lisbon, Lisbon (1992)
Bana e Costa, C.A.: O modelo de apoio à avaliação de propostas nos concursos do Metro do Porto. FER XXI XVIII, 111, 113–115 (1999)
Bana e Costa, C.A.: The use of multicriteria decision analysis to support the search for less conflicting policy options in a multi-actor context: case-study. J. Multi-Criteria Decis. Anal. 10(2), 111–125 (2001)
Bana e Costa, C.A.: Issues in facilitating bid evaluation in public call for tenders. In: Khosrowshahi, F. (ed.) Proceedings of the 3rd International Conference on Decision Making in Urban and Civil Engineering, CD-ROM edition, pp. 703–709. SOAS, London (2002)
Bana e Costa, C.A., Chagas, M.P.: A career choice problem: an example of how to use MACBETH to build a quantitative value model based on qualitative value judgments. Eur. J. Oper. Res. 153(2), 323–331 (2004)
Bana e Costa, C.A., Corrêa, E.C.: Framing public calls for tenders through a multicriteria approach. In: Despotis, D.K., Zopounidis, C. (eds.) Proceedings of the 5th International Conference of the Decision Sciences Institute, vol. II, pp. 1353–1355, Athens University of Business and Economics, Athens (1999)
Bana e Costa, C.A., Corrêa, E.C.: Construction of a total quality index using a multicriteria approach: the case of Lisbon Gas Company. Research paper 10/2000, CEG-IST, Technical University of Lisbon (2000)
Bana e Costa, C.A., Corrêa, E.C.: O processo de construção do índice FUNCEME de vulnerabilidade à seca. Research paper 16/2000, CEG-IST, Technical University of Lisbon (2000)
Bana e Costa, C.A., Ferreira, J.A.: Concursos de obras públicas: Metodologia de avaliação de propostas. Aplicação a uma grande empreitada de infraestruturas de transportes. In: Actas do Congresso 94, Engenharia Portuguesa na Viragem do Século, pp. 47–w/9. Ordem dos Engenheiros, Lisbon (1994)
Bana e Costa, C.A., Marques, M.S.: Procurement decision support for the Portuguese MoD: The MACBETH approach and the acquisition of 8 × 8 AWV. In: Decision Support Methodologies for Acquisition of Military Equipment, NATO Research and Technology Organisation, Brussels (1-1, 1-14) (2009)
Bana e Costa, C.A., Oliveira, R.C.: Assigning priorities for maintenance, repair and refurbishment in managing a municipal housing stock. Eur. J. Oper. Res. 138(2), 380–391 (2002)
Bana e Costa, C.A., Oliveira, M.D.: A multicriteria decision analysis model for faculty evaluation. OMEGA Int. J. Manag. Sci. 40(4), 424–436 (2011)
Bana e Costa, C.A., Silva, F.N.: Concepção de uma ‘boa’ alternativa de ligação ferroviária ao porto de Lisboa: Uma aplicação da metodologia multicritério de apoio à decisão e à negociação. Investigação Operacional 14(2), 115–131 (1994)
Bana e Costa, C.A., Soares, J.O.: Multicriteria approaches for portfolio selection: an overview. Rev. Financ. Mark. IV(1), 19–26 (2001)
Bana e Costa, C.A., Soares, J.O.: A multicriteria model for portfolio management. Eur. J. Financ. 10(3), 198–211 (2004)
Bana e Costa, C.A., Thomaz, J.F.: Locating centres of information and recruitment of volunteers for the Portuguese armed forces: a decision-analysis case-study. In: Proceedings of the 42nd Annual Conference of the International Military Testing Association, CD-ROM edition, pp. 173–180. International Military Testing Association, Edinburgh (2000)
Bana e Costa, C.A., Vansnick, J.C.: Sur la quantification des jugements de valeur: L’approche MACBETH. Cahiers du LAMSADE, 117, Université Paris-Dauphine, Paris (1993)
Bana e Costa, C.A., Vansnick, J.C.: MACBETH – an interactive path towards the construction of cardinal value functions. Int. Trans. Oper. Res. 1(4), 489–500 (1994)
Bana e Costa, C.A., Vansnick, J.C.: Uma nova abordagem ao problema de construção de uma função de valor cardinal: MACBETH. Investigação Operacional 15(1), 15–35 (1995)
Bana e Costa, C.A., Vansnick, J.C.: General overview of the MACBETH approach. In: Pardalos, P.M., Siskos, Y., Zopounidis, C. (eds.) Advances in Multicriteria Analysis, pp. 93–100. Kluwer, Dordrecht (1995)
Bana e Costa, C.A., Vansnick, J.C.: Applications of the MACBETH approach in the framework of an additive aggregation model. J. Multi-Criteria Decis. Anal. 6(2), 107–114 (1997)
Bana e Costa, C.A., Vansnick, J.C.: The MACBETH approach: Basic ideas. In: Proceedings of the International Conference on Methods and Applications of Multicriteria Decision Making, pp. 86–88. FUCAM, Facultés Universitaires Catholiques de Mons (1997)
Bana e Costa, C.A., Vansnick, J.C.: A theoretical framework for measuring attractiveness by a categorical based evaluation technique (MACBETH). In: Clímaco, J. (ed.) Multicriteria Analysis, pp. 15–24. Springer, Berlin (1997)
Bana e Costa, C.A., Vansnick, A.C.: Construção multicritério da ‘matriz de custos’ de um modelo de afectação para gestão de stands de aeronaves. Investigação Operacional 17(2), 179–199 (1997)
Bana e Costa, C.A., Vansnick, J.C.: The MACBETH approach: basic ideas, software and an application. In: Meskens, N., Roubens, M. (eds.) Advances in Decision Analysis, pp. 131–157. Kluwer, Dordrecht (1999)
Bana e Costa, C.A., Vansnick, J.C.: Cardinal value measurement with MACBETH. In: Zanakis, S.H., Doukidis, G., Zopounidis, C. (eds.) Decision Making: Recent Developments and Worldwide Applications, pp. 317–329. Kluwer, Dordrecht (2000)
Bana e Costa, C.A., Oliveira, R.C., Alvarez, M.L.: Concepção de um sistema multicritério de definição de prioridades de intervenção/conservação do património habitacional da Câmara Municipal de Lisboa. In: Comunicações ao 2o Encontro sobre Conservação e Reabilitação de Edifícios, vol. II, pp. 925–934. LNEC, Laboratório Nacional de Engenharia Civil, Lisbon (1994)
Bana e Costa, C.A., Ferreira, J.A., Vansnick, J.C.: Avaliação multicritério de propostas: O caso de uma nova linha do Metropolitano de Lisboa. Revista de Transporte e Tecnologia VII(14), 31–65 (1995)
Bana e Costa, C.A., Costa-Lobo, M.L., Ramos, I.A.: Contributo da metodologia multicritério na elaboração do Plano Estratégico de Barcelos. Sociedade e Território 24, 102–115 (1997)
Bana e Costa, C.A., Oliveira, R.C., Alvarez, M.L.: SMAI: Sistema multicritério de avaliação de intervenções num parque habitacional municipal. In: Bezelga, A.A., Leitão, A.B. (eds.) A Avaliação do Imobiliário, pp. 109–122. Associação Portuguesa dos Avaliadores de Engenharia, Lisbon (1997)
Bana e Costa, C.A., Ensslin, L., Costa, A.P.: Structuring the process of choosing rice varieties at the South of Brazil. In: Multicriteria Analysis for Land-Use Management, vol. 9: Environment and Management, pp. 33–45. Kluwer, Dordrecht (1998)
Bana e Costa, C.A., Ensslin, L., Zanella, I.J.: A real-world MCDA application in cellular telephony systems. In: Stewart, T.J., van den Honert, R.C. (eds.) Trends in Multicriteria Decision Making. Lecture Notes in Economics and Mathematical Systems, vol. 465, pp. 412–423. Springer, Berlin (1998)
Bana e Costa, C.A., Corrêa, E.C., Ensslin, L., Vansnick, J.C.: Decision support systems in action: integrated application in a multicriteria decision aid process. Eur. J. Oper. Res. 113(2), 315–335 (1999)
Bana e Costa, C.A., Ensslin, L., Holz, E.: Avaliação ambiental em situação de conflito de interesses – Um estudo de caso. In: IX SILUBESA, Simpósio Luso-Brasileiro de Engenharia Sanitária e Ambiental, CD-ROM edition, pp. 1914–1923. Associação Brasileira de Engenharia Sanitária e Ambiental and Associação Portuguesa de Recursos Hídricos, Porto Seguro (BA) (2000)
Bana e Costa, C.A., Ferreira, J.A., Corrêa, E.C.: Metodologia multicritério de apoio à avaliação de propostas em concursos públicos. In: Antunes, C.H., Tavares, L.V. (eds.) Casos de Aplicação da Investigação Operacional, pp. 336–363. McGraw-Hill Portugal, Lisbon (2000)
Bana e Costa, C.A., Silva, F.N., Vansnick, J.C.: Conflict dissolution in the public sector: a case-study. Eur. J. Oper. Res. 130(2), 388–401 (2001)
Bana e Costa, C.A., Barroso, L., Soares, J.O.: Qualitative modelling of credit scoring: a case study in banking. J. Eur. Res. Stud. 5(1–2), 37–51 (2002)
Bana e Costa, C.A., Corrêa, E.C., De Corte, J.M., Vansnick, J.C.: Facilitating bid evaluation in public call for tenders: a socio-technical approach. OMEGA Int. J. Manag. Sci. 30(3), 227–242 (2002)
Bana e Costa, C.A., da Costa-Lobo, M.L., Ramos, I.A., Vansnick, J.C.: Multicriteria approach for strategic town planning: the case of Barcelos. In: Bouyssou, D., Jacquet-Lagrèze, E., Perny, P., Sĺowiński, R., Vanderpooten, D., Vincke, Ph. (eds.) Aiding Decisions with Multiple Criteria. Essays in Honor of Bernard Roy, pp. 429–456. Kluwer, Boston (2002)
Bana e Costa, C.A., De Corte, J.M., Vansnick, J.C.: MACBETH. Working paper 03.56, London School of Economics (2003)
Bana e Costa, C.A., Antão da Silva, P., Correia, F.N.: Multicriteria evaluation of flood control measures: the case of Ribeira do Livramento. Water Resour. Manag. 18(3), 263–283 (2004)
Bana e Costa, C.A., Fernandes, T.G., Correia, P.V.D.: Prioritisation of public investments in social infrastructures using multicriteria value analysis and decision conferencing: a case study. Int. Trans. Oper. Res. 13(4), 279–297 (2006)
Bana e Costa, C.A., Lourenço, J.C., Bana e Costa, J.C.: Building MACBETH value-function models for bid evaluation at the Portuguese electric transmission company (REN). In: Antunes, C.H., Gomes, A. (eds.) Proceedings of the 19th Mini EURO Conference on Operational Research Models and Methods in the Energy Sector, CD-ROM edition. INESC Coimbra (2006)
Bana e Costa, C.A., Lourenço, J.C., Soares, J.O.: An interval weighting assignment model for credit analysis. J. Financ. Decis. Mak. 3(2), 1–9 (2007)
Bana e Costa, C.A., Lourenço, J.C., Chagas, M.P., Bana e Costa, J.C.: Development of reusable bid evaluation models for the Portuguese electric transmission company. Decis. Anal. 5(1), 22–42 (2008). doi:10.1287/deca.1080.0104
Bana e Costa, C.A., Oliveira, C.S., Vieira, V.: Prioritization of bridges and tunnels in earthquake risk mitigation using multicriteria decision analysis: application to Lisbon. Omega 36(3), 442–450 (2008)
Bana e Costa, C.A., Lourenço, J.C., Bana e Costa, J.C.: Concepção de uma estratégia de desenvolvimento a médio prazo para Pernambuco. In: Actas do 15o Congresso da APDR – Associação Portuguesa para o Desenvolvimneto Regional Cabo Verde (2009) APDR (1959–1681)
Bana e Costa, C.A., Craveiro, M.T., Rocha, A.S., Bana e Costa, J.C., Santos, P.S.: Avaliação multicritério na elaboração da matriz estratégica do programa local de habitação de Lisboa. In: Proceedings do 54o IFHP World Congress, CD-ROM edition, Porto Alegre, 14–17 November 2010
Bana e Costa, C.A., Martins, P.A.F., Oliveira, M.D., Sernadas, A., Mota Soares, C.A.: Faculty evaluation using multicriteria value measurement. In: Rogozea, L. (ed.) Advances in Mathematical and Computational Methods – 12th WSEAS International Conference on Mathematical and Computational Methods in Science and Engineering (MACMESE ’10), University of Algarve, Faro. WSEAS Press (2010)
Bana e Costa, C.A., De Corte, J.M., Vansnick, J.C.: MACBETH (measuring attractiveness by a categorical-based evaluation technique). In: Cochrane, J.J. (ed.) Wiley Encyclopedia in Operational Research and Management Science, pp. 2945–2950. Wiley, New York (2011)
Bana e Costa, C.A., Carnero, M.C., Oliveira, M.D.: A multi-criteria model for auditing a predictive maintenance programme. Eur. J. Oper. Res. 217(2), 381–393 (2012)
Bana e Costa, C.A., De Corte, J.M., Vansnick, J.C.: MACBETH. Int. J. Inf. Technol. Decis. Mak. 11(2), 359–387 (2012)
Bana e Costa, C.A., Angulo-Meza, L., Oliveira, M.D.: O método MACBETH e aplicação no Brasil. Engevista 15(1), 3–27 (2013)
Bana e Costa, C.A., Lourenço, J.C., Oliveira, M.D., Bana e Costa, J.C.: A social–technical approach for group decision support in public strategic planning: the Pernambuco PPA case. Group Decis. Negot. 23(1), 5–29 (2013)
Barin, A., Canha, L.N., Abaide, A.R., Orling, R.B., Martins, L.F.G.: Selection of hybrid renewable energy systems in landfills. In: 9th International Conference on the European Energy Market, pp. 1–5, Florence, 10–12 May 2012. IEEE, New York (2012)
Barin, A., Martins, L.F., Canha, L.N., Abaide, A., Oriling, R.B.: Selection of renewable energy technologies in landfills: the m-macbeth decision support. In: Mendez-Vilas, A. (ed.) Fuelling the Future: Advances in Science and Technologies for Energy Generation, Transmission and Storage, pp. 98–102. Universal-Publishers (2012)
Belton, V.: A Comparative Study of Methods for Multiple Criteria Decision Aiding. Ph.D. thesis, University of Cambridge, Cambridge (1986)
Belton, V., Stewart, T.J.: Multiple Criteria Decision Analysis: An Integrated Approach. Kluwer, Boston (2002)
Berrah, L., Clivillé, V.: Towards an aggregation performance measurement system model in a supply chain context. Comput. Ind. 58(7), 709–719 (2007)
Berrah, L., Mauris, G., Vernadat, F.: Industrial performance measurement: an approach based on the aggregation of unipolar or bipolar expressions. Int. J. Prod. Res. 44(18–19), 4145–4158 (2006)
Berrah, L., Mauris, G., Montmain, J.: Monitoring the improvement of an overall industrial performance based on a Choquet integral aggregation. Omega 36(3), 340–351 (2008)
Berrah, L., Mauris, G., Montmain, J., Clivillé, V.: Efficacy and efficiency indexes for a multi-criteria industrial performance synthesized by Choquet integral aggregation. Int. J. Comput. Integr. Manuf. 21(4), 415–425 (2008)
Bollinger, D., Pictet, J.: Potential use of e-democracy in MCDA processes. Analysis on the basis of a Swiss case. J. Multi-Criteria Decis. Anal. 12(2–3), 65–76 (2003)
Bortuluzzi, S.C., Ensslin, S.R., Lyrio, M.V.L., Ensslin, L.: Proposta de um modelo multicritério de avaliação de desempenho econômico-financeiro para apoiar decisões de investimentos em empresas de capital aberto. CAP Account. Manag. 3(3), 100–110 (2009)
Bortuluzzi, S.C., Ensslin, S.R., Ensslin, L.: Construção de um modelo de avaliacão de desempenho para a gestão financeira de uma empresa de informática. CAP Account. Manag. 4(4), 12–22 (2010)
Bortuluzzi, S.C., Ensslin, S.R., Ensslin, L.: Avaliação de desempenho dos aspectos tangíveis e intangíveis da área de mercado: Estudo de caso em uma média empresa industrial. Revista Brasileira de Gestão de Negócios 12(37), 425–446 (2010)
Bortuluzzi, S.C., Ensslin, S.R., Ensslin, L.: Avaliação de desempenho das variáveis financeiras e não financeiras que respondem pelo desempenho de uma indústria de móveis. Revista Gestão Industria 7(2), 24–47 (2011)
Bortuluzzi, S.C., Ensslin, S.R., Lyrio, M.V.L., Ensslin, L.: Avaliação de desempenho econômico-financeiro: Uma proposta de integração de indicadores contábeis tradicionais por meio da metodologia multicritério de apoio à decisão construtivista (MCDA-C). Revista Alcance (Online) 18(2), 200–218 (2011)
Brandão, L.C., Andrade, F.V.S., Soares de Mello, J.C.C.B.: 2012 UEFA Euro efficiency evaluation based on market expectations. In: 4th International Conference on Mathematics in Sport, pp. 32–37, Leuven, 5–7 June 2013
Burton, J., Hubacek, K.: Is small beautiful? A multicriteria assessment of small-scale energy technology applications in local governments. Energy Policy 35(12), 6402–6412 (2007)
C3E. (ed.): Applying the Multi-criteria Method to the Evaluation of Structural Programmes. MEANS Handbook, vol. 4. European Commission, DG XVI/02, Brussels (1995)
C3E. (ed.): Evaluation Pilote Multicritère du Hainaut. Rapport d’Expérience. European Commission, DG XVI/02, Brussels (1995)
Cabello, J.M., Ruiz, F., Pérez-Gladish, B., Méndez-Rodríguez, P.: Synthetic indicators of mutual funds’ environmental responsibility: an application of the reference point method. Eur. J. Oper. Res. (2014). doi:10.1016/j.ejor.2013.11.031
Cardoso, F.J.P., Gomes, L.F.A.M., Sanglard Júnior, M.: Administração das operações de telecomunicação: Uma análise de decisão. Revista Portuguesa e Brasileira de Gestão 2(2), 91–103 (2003)
Carnero, M.C.: Evaluating a maintenance department in a service company. Int. J. Math. Models Methods Appl. Sci. 3(3), 230–237 (2009)
Carnero, M.C.: Maintenance audit by means of an additive multicriteria model. In: Zafiris, V., Benavides, M., Gaoet, K. (eds.) Advances in Marketing, Management and Finances, pp. 147–152. WSEAS Press, Prague (2009)
Catalyze Ltd.: HIVIEW3 Starter Guide. Catalyze Ltd., London (2003)
Clivillé, V., Berrah, L., Mauris, G.: Quantitative expression and aggregation of performance measurements based on the MACBETH multi-criteria method. Int. J. Prod. Econ. 105(1), 171–189 (2007)
Costa, H.R., Barros, M.D.O., Travassos, G.: Evaluating software project portfolio risks. J. Syst. Softw. 80(1), 16–31 (2007)
Cox, R., Sanchez, J., Revie, C.W.: Multi-criteria decision analysis tools for prioritising emerging or re-emerging infectious diseases associated with climate change in Canada. PLoS ONE 8(8), 1–16 (2013)
Cuadrado, M.R., Fernández, M.G.: Methodology to select the best business game in higher education. Am. J. Ind. Bus. Manag. 3, 589–594 (2013)
Dall’Osso, F., Gonella, M., Gabbianelli, G., Withycombe, G., Dominey-Howes, D.: A revised (PTVA) model for assessing the vulnerability of buildings to tsunami damage. Nat. Hazards Earth Syst. Sci. 9(5), 1557–1565 (2009)
de Azevedo, R., de Oliveira Lacerda, R., Ensslin, L., Jungles, A., Ensslin, S.: Performance measurement to aid decision making in the budgeting process for apartment-building construction: case study using MCDA-C. J. Constr. Eng. Manag. 139(2), 225–235 (2013)
De Castro, A.K.A., Pinheiro, P.R., Souza, G.G.C.: A scheduling process applied to newspaper production. In: Proceedings of the 3rd International Conference on Service Systems and Service Management, Troyes, pp. 1245–1250. IEEE, New York (2006)
de Castro, A.K.A., Pinheiro, P.R., de Souza, G.G.C.: Application multicriteria decision analysis on tv digital. In: Sobh, T.M. (ed.) Advances in Computer and Information Sciences and Engineering, pp. 39–44. Springer, New York (2008) [ISBN 978-1-4020-8740-0]
de Castro, A.K.A., Pinheiro, P.R., de Souza, G.G.C.: A multicriteria model applied in the diagnosis of alzheimer’s disease: a Bayesian network. In: 11th IEEE International Conference on Computational Science and Engineering, São Paulo, pp. 15–22 (2008)
de Castro, A., Pinheiro, P.R., Pinheiro, M.: An approach for the neuropsychological diagnosis of Alzheimer’s disease: a hybrid model in decision making. In: Wen, P., Li, Y., Polkowski, L., et al. (eds.) 4th International Conference on Rough Sets and Knowledge Technology, Gold Coast, pp. 216–223. Springer, Berlin (2009)
de Castro, A., Pinheiro, P.R., Pinheiro, M.: Towards the neuropsychological diagnosis of Alzheimer’s disease: a hybrid model in decision making. In: Lytras, M.D., DePablos, P.O., Damiani, E., et al. (eds.) 2nd World Summit on the Knowledge Society, Chania, pp. 522–531. Springer, Berlin (2009)
de Castro, A., Pinheiro, P.R., Pinheiro, M., Tamanini, I.: Towards the applied hybrid model in decision making: a neuropsychological diagnosis of Alzheimer’s disease study case. Int. J. Comput. Intell. Syst. 4(1), 89–99 (2011)
de Corte, J.M.: Un Logiciel d’Exploitation d’Informations Préférentielles pour l’Aide à la Décision. Bases Mathématiques et Algorithmiques. Ph.D. thesis, Presses Universitaires Montoises, University of Mons-Hainaut, Mons (2002)
de Lima, A.S., Damiani, J.H.S.: A proposed method for modeling research and development (R&D) project prioritization criteria. In: Kocaoglu, D.F., Anderson, T.R., Daim, T.U.; et al. (eds.) Conference of the Portland International Center for Management of Engineering and Technology. (PICMET’09), Portland, pp. 589–598. IEEE, New York (2009).
Della Bruna E. Jr., Ensslin, L., Ensslin, S.R.: Supply chain performance evaluation: a case study in a company of equipment for refrigeration. In: Proceedings of the 2011 IEEE International Technology Management Conference, San Jose (2011)
Dhouib, D.: Fuzzy Macbeth method to analyze alternatives in automobile tire wastes reverse logistics. In: International Conference on Advanced Logistics and Transport, pp. 321–326, Sousse, 29–31 May 2013
Dhouib, D.: An extension of MACBETH method for a fuzzy environment to analyze alternatives in reverse logistics for automobile tire wastes. OMEGA Int. J. Manag. Sci. 42(1), 25–32 (2014)
Dias-Sardinha, I., Carolino, J., Matos, P.V., Mendes, I., Milheiras, S., Craveiro, D., Marques, M., Primdahl, J., Krutwagen, B.: Rehmine project - redevelopment of so domingos mine. Technical report, School of Economics & Management, Mértola, Portugal, June 2013
Dutra, A., Ensslin, S.R., Ensslin, L., Lima, M.V.A.: A incorporação da dimensão integrativa nos processos de avaliação do desempenho organizacional: Um estudo de caso. Revista Contemporânea de Contabilidade 6(11), 109–136 (2009)
Eklund, P., Rusinowska, A., Swart, H.D.: A consensus model of political decision-making. Ann. Oper. Res. 158(1), 5–20 (2008)
Ensslin, L., Ensslin, S.: Elaboração de um modelo construtivista para identificação de oportunidades de aperfeiçoamento de docentes do EPS-UFSC. Revista Produto & Produção 2(3), 143–149 (1998)
Ensslin, L., Dutra, A., Ensslin, S.: MCDA: a constructivist approach to the management of human resources at a governmental agency. Int. Trans. Oper. Res. 7(1), 79–100 (2000)
Ensslin, L., Montibeller, G.N., Lima, M.V.A.: Constructing and implementing a DSS to help evaluate perceived risk of accounts receivable. In: Haimes, Y.Y., Steuer, R.E. (eds.) Research and Practice in Multiple Criteria Decision Making. Lecture Notes in Economics and Mathematical Systems, vol. 487, pp. 245–259. Springer, Berlin (2000)
Ensslin, L., Montibeller, G.N., Noronha, S.: Apoio à Decisão: Metodologias para Estruturação de Problemas e Avaliação Multicritério de Alternativas. Editora Insular, Florianópolis (SC) (2001)
Ensslin, L., Giffhorn, E., Ensslin, S., Petri, S., Vianna, W.: Avaliação do desempenho de empresas terceirizadas com o uso da metodologia multicritério de apoio à decisão-construtivista. Revista Pesquisa Operacional 30(1), 125–152 (2010)
Ensslin, L., Scheid, L.C.M., Ensslin, S.R., de Oliveira Lacerda, R.T.: Software process assessment and improvement using multicriteria decision aiding-constructivist. JISTEM: J. Inf. Syst. Technol. Manag. 9(3), 473–496 (2012)
Ensslin, S.R., Ensslin, L., Back, F., de Oliveira Lacerda, R.T.: Improved decision aiding in human resource management: a case using constructivist multi-criteria decision aiding. Int. J. Product. Perform. Manag. 62(7), 735–757 (2013)
Ertay, T., Kahraman, C.: Evaluation of renewable energy alternatives using MACBETH multicriteria method. In: 9th International FLINS Conference on Foundations and Applications of Computational Intelligence, Chengdu, pp. 937–943 (2010)
Ertay, T., Kahraman, C., Kaya, Äř: Evaluation of renewable energy alternatives using MACBETH and fuzzy AHP multicriteria methods: the case of Turkey. Technol. Econ. Dev. Econ. 19(1), 38–62 (2013)
Fakhfakh, N., Pourraz, F., Verjus, H., Moreaux, P.: Client-oriented preferences model for QoS aggregation in service-based applications. In: Obaidat, M.S., Sevillano, J.L., Filipe, J. (eds.) E-Business and Telecommunications, vol. 314, pp. 141–155, Seville, 18–21 July 2011. International Joint Conference, ICETE. Springer, Berlin/Heidelberg (2011)
Farquhar, P.H., Keller, L.R.: Preference intensity measurement. Ann. Oper. Res. 19(1), 205–217 (1989)
Fasolo, B., Bana e Costa, C.A.: Tailoring value elicitation to decision makers’ numeracy and fluency: expressing value judgments in numbers or words. OMEGA Int. J. Manag. Sci. 44, 83–90 (2014)
Fernandes, M.B., Almeida, M.C., Henriques, A.G.: Assessing desalination as a sustainable alternative using a multiple criteria decision support model. Water Pract. Technol. 3(3), 1–8 (2008)
Ferreira, F.A.F.: Bank branches’ performance evaluation. In: IADIS International Conference e-Society, Dublin, pp. 392–394 (2006)
Ferreira, F.A.F., Santos, S.P., Rodrigues, P.M.M.: Adding value to bank branch performance evaluation using cognitive maps and MCDA: a case study. J. Oper. Res. Soc. 62(7), 1320–1333 (2011)
Ferreira, F.A.F., Spahr, R.W., Santos, S.P., Rodrigues, P.M.M.: A multiple criteria framework to evaluate bank branch potential attractiveness. Int. J. Strateg. Prop. Manag. 16(3), 254–276 (2012)
Ferreira, F., Santos, S.P., Marques, C., Ferreira, J.: Assessing credit risk of mortgage lending using MACBETH: a methodological framework. Manag. Decis. 52(2), 1–36 (2014)
Ferreira, J., Pinheiro, M.D., de Brito, J.: Portuguese sustainable construction assessment tools benchmarked with BREEAM and LEED: an energy analysis. Energy Build. 69, 451–463 (2014)
Figueiredo, M.S.M., Oliveira, M.D.: Prioritizing risks based on multicriteria decision aid methodology: Development of methods applied to ALSTOM power. In: IEEE International Conference on Industrial Engineering and Engineering Management, Hong Kong, pp. 1568–1572 (2009)
Filho, A.B., Marçal, A.S., Costa, G., Pinheiro, P.R.: Staff scheduling optimization in information technology projects. In: The Fourth IEEE Conference on Service Systems and Service Management, pp. 1–6, University of Electronic Science and Technology of China, Chengdu (2007)
Filho, A.B., Marçal, A.S., Costa, G., Pinheiro, P.R., Pinheiro, R.F.: A novel approach based on staff scheduling optimization in information technology projects. Int. J. Comput. Sci. Netw. Secur. 9(9), 277–286 (2009)
Flix, B., Vasconcelos, A., Tribolet, J.: SimplexIS: evaluating the impact of e-Gov simplification measures in the information system architecture. World Acad. Sci. Eng. Technol. 63, 636–641 (2012)
Freeling, A.N.S.: Belief and Decision Aiding. Ph.D. thesis, University of Cambridge (1983)
French, S.: Decision Theory: An Introduction to the Mathematics of Rationality. Ellis Horwood Limited, Chichester (1988)
Gallon, A.V., Ensslin, S.R., Ensslin, L.: Avaliação de desempenho organizacional em incubadoras de empresas por meio da metodologia multicritério de apoio à decisão construtivista (MCDA-C): a experiência do midi tecnológico. RAI: Revista de Administração e Inovação 8(1), 37–63 (2011)
Giffhorn, E., Ensslin, L., Ensslin, S.R., Vianna, W.B.: Proposal of a multicriteria performance evaluation for outsourced project providers. In: Performance Measurement Association Conference 2009, University of Otago, New Zealand, 18 pp (2009)
Giffhorn, E., Ensslin, L., Ensslin, S.R., Petri, S.M., Vianna, W.B.: Avaliação do desempenho de empresas terceirizadas com o uso da metodologia multicritério em apoio à decisão - construtivista. Pesquisa Operacional 30(1), 125–152 (2010)
Gomes, L.F.A.M., Rangel, L.A.D., Mello, J., Gomes, E., Fuks, S.: Apoio à administração universitária: Aplicação dos métodos Macbeth e Uta na universidade federal fluminense. Revista de Administração Mackenzie 3(2), 155–174 (2002)
Grzebieluckas, C., Busson, M.A., Queiroz, S.G., Ensslin, L., Ensslin, S., Nickel, E., Balbim, A.J.: Instrumento para identificação das necessidades do consumidor no processo de desenvolvimento do design: Ilustrado ao caso de projeto de um automóvel. Gestão & produção 18(2), 335–350 (2011)
Gurbuz, T.: Multiple criteria human performance evaluation using Choquet integral. Int. J. Computat. Intell. Syst. 3(3), 290–300 (2010)
Gurbuz, T., Alptekin, S.E., Alptekin, G.I.: A hybrid MCDM methodology for ERP selection problem with interacting criteria. Decis. Support. Syst. 54(1), 206–214 (2012)
Hurson, C., Mastorakis, K., Siskos, Y.: Application of a synergy of MACBETH and MAUT multicriteria methods to portfolio selection in Athens stock exchange. Int. J. Multicriteria Dec. Mak. 2(2), 113–127 (2012)
Joerin, F., Cool, G., Rodriguez, M.J., Gignac, M., Bouchard, C.: Using multi-criteria decision analysis to assess the vulnerability of drinking water utilities.
Junior, H.V., Kienitz, K.H., Belderrain, M.C.N.: Metodologia de apoio a decisão para os processos de seleção de alvos e armamentos. Spectrum - Revista do Comando-Geral de Operações Aéreas 12, 25–27 (2009) Environ. Monit. Assess. 166(1–4), 313–330 (2010)
Junior, A.G.M., Gonçalves, T.J.M., Belderrain, M.C.N.: MACBETH aplicado ao cálculo da pena base do direito penal militar. INGEPRO - Inovação, Gestão e Produção 3(1), 56–66 (2011)
Junior, A.G.M., Junior, M.M.C., Belderrain, M.C.N., Correia, A.R., Schwanz, S.H.: Multicriteria and multivariate analysis for port performance evaluation. Int. J. Prod. Econ. 140(1), 450–456 (2012)
Karande, P., Chakraborty, S.: Evaluation and selection of flexible manufacturing systems using MACBETH method. Int. J. Ser. Oper. Manag. 16(1), 123–144 (2013)
Karande, P., Chakraborty, S.: Using MACBETH method for supplier selection in manufacturing environment. Int. J. Ind. Eng. Comput. 4, 259–272 (2013)
Khaled, A., Idrissi, M.A.J.: A semi-structured tailoring-driven approach for ERP selection. Int. J. Comput. Sci. Issues 9(5), 71–80 (2012)
Lacerda, R.T.O., Ensslin, L., Ensslin, S.R.: Um estudo de caso sobre gerenciamento de portfólio de projectos e apoio à decisão multicritério. Revista Gestão Industrial 6(1), 1–29 (2010)
Lacerda, R.T.O., Ensslin, L., Ensslin, S.R.: A performance measurement framework in portfolio management: a constructivist case. Manag. Decis. 49(4), 648–668 (2011)
Lacerda, R.T.O., Ensslin, L., Ensslin, S.R.: A performance measurement view of IT project management. Int. J. Product. Perform. Manag. 60(2), 132–151 (2011)
Lauras, M., Marques, G., Gourc, D.: Towards a multi-dimensional project performance measurement system. Decis. Support. Syst. 48(2), 342–353 (2010)
Leitão, M.A.L.S., Thomaz, J.F., Bana e Costa, C.A.: Multimethodology in practice: an application in the Portuguese army performance appraisal model. In: Proceedings of the 44th Annual Conference of the International Military Testing Association, IMTA, pp. 511–527, Ottawa (2002)
Leitão, M.A.L.S., Thomaz, J.F., Freire, F.V., Bana e Costa, C.A.: Using decision conferencing and process consultation to facilitate the structuring of a performance appraisal model in the Portuguese army. Research paper 5/2002, CEG-IST, Technical University of Lisbon (2002)
Lopes, D.F.: Development of a multicriteria decision aiding model for monitoring and evaluating the performance of health care units. In: Oliveira, J.F., Vaz, C.B. (eds.) Atas do XVI Congresso da Associação Portuguesa de Investigação Operacional, CD-ROM edition, pp. 212–222, Bragança. Associação Portuguesa de Investigação Operacional (2013)
Lourenço, J.C.: Apoio à avaliação de propostas no concurso limitado por prévia qualificação para a realização dos projectos de especialidades (ao nível de projecto de execução) para a construção do parque de estacionamento subterrâneo na Praça D. Dinis em Coimbra. Working paper cegist n.14/2002, CEG-IST, Technical University of Lisbon (2002)
Mangasarian, O.L.: Nonlinear Programming. McGraw-Hill, New York (1969)
Marques, G., Gourc, D., Lauras, M.: Multi-criteria performance analysis for decision making in project management. Int. J. Proj. Manag. 29(8), 1057–1069 (2010)
Mateus, R., Ferreira, J.A., Carreira, J.: Multicriteria decision analysis (MCDA): Central Porto high-speed railway station. Eur. J. Oper. Res. 187(1), 1–18 (2008)
Menezes, A.C., Pinheiro, P.R., Pinheiro, M.C.D., Cavalcante, T.P.: A hybrid model to support the diagnosis of disease: a case study for diabetes. In: 26th IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), Regina (2013)
Merad, M., Dechy, N., Serir, L., Grabisch, M., Marcel, F.: Using a multi-criteria decision aid methodology to implement sustainable development principles within an organization. Eur. J. Oper. Res. 224(3), 603–613 (2013)
Montignac, F.: An MCDA approach for evaluating hydrogen storage systems for future vehicles. In: 2nd Decision Deck Workshop. Université Paris Dauphine, 2 pp (2008)
Montignac, F., Noirot, I., Chaurdoune, S.: Multi-criteria evaluation of on-board hydrogen storage technologies using the MACBETH approach. Int. J. Hydrog. Energy 34(10), 4561–4568 (2009)
Montmain, J., Sanchez, C., Vinches, M.: Multi criteria analyses for managing motorway company facilities: the decision support system SINERGIE. Adv. Eng. Inf. 23(3), 265–287 (2009)
Moraes, L., Garcia, R., Ensslin, L., Conceição, M.J.: The multicriteria analysis for construction of benchmarkers to support the clinical engineering in the healthcare technology management. Eur. J. Oper. Res. 200(3), 607–615 (2010)
Mousset, C., Vansnick, J.C.: About the representation of a precardinal information. Technical report, Université de Mons-Hainaut (2002)
Neves, A.R., Lourenço, J.C., Leal, V.: A multi-criteria approach to local energy planning - the case of Barreiro municipality. In: Procedia - Social and Behavioral Sciences, pp. 313–320, Vilamoura, Algarve, Portugal, 4–6 February 2012. 1st International Conference on Operations Research and Enterprise Systems (ICORES), SciTePress (2012)
Ntouskas, T., Polemi, N.: STORM-RM: a collaborative and multicriteria risk management methodology. Int. J. Multicriteria Dec. Mak. 2(2), 159–177 (2012)
Nunes, L.C., Pinheiro, P.R., Pequeno, T.C.: An expert system applied to the diagnosis of psychological disorders. In: IEEE International Conference on Intelligent Computing and Intelligent Systems, Shanghai, pp. 363–367 (2009)
Nunes, L.C., Pinheiro, P.R., Pequeno, T.C.: Support tool in the diagnosis of major depressive disorder. In: 3rd World Summit on the Knowledge Society, WSKS (2) 2010, Corfu, pp. 136–145 (2010)
Nunes, L.C., Pinheiro, P.R., Pequeno, T.C.: Toward an application to psychological disorders diagnosis. Softw. Tools Algoritm. Biol. Syst. 696, 573–580 (2011)
Oliveira, R.C., Lourenço, J.C.: A multicriteria model for assigning new orders to service suppliers. Eur. J. Oper. Res. 139(2), 390–399 (2002)
Oliveira, C.S., Ferreira, M.A., Mota de Sá, F.: Seismic vulnerability and impact analysis: elements for mitigation policies. In: XI Congresso Nazionale “L’ingegneria Sismica in Italia, Genova, 29 pp (2004)
Oliveira, M.D., Rodrigues, T.C., Bana e Costa, C.A., Brito de Sá, A.: Prioritizing health care interventions: A multicriteria resource allocation model to inform the choice of community care programmes. In: 15o Congresso da Associação Portuguesa de Investigação Operacional, Coimbra, pp. 177–186. INESC, Coimbra (2011)
Oliveira, M.D., Rodrigues, T.C., Bana e Costa, C.A., Brito de Sá, A.: Prioritizing health care interventions: A multicriteria resource allocation model to inform the choice of community care programmes. In: 37th International Conference on Operational Research Applied to Health Services, Cardiff, pp. 255–265 (2011)
Oliveira, M.D., Rodrigues, T.C., Bana e Costa, C.A., Brito de Sá, A.: Prioritizing health care interventions: A multicriteria resource allocation model to inform the choice of community care programmes. In: Tanfani, E., Testi, A. (eds.) Advanced Decision Making Methods Applied to Health Care, pp. 139–152. Springer, New York (2012)
Omann, I.: Product service systems and their impacts on sustainable development: a multi-criteria evaluation for Austrian companies. Frontiers 16, 1–34 (2003)
Palacios, O.Y.S., Rincón, P.C.N., Corriou, J.P., Pardo, M.C., Fonteix, C.: Multicriteria optimization of production conditions for a new phthalate-free PVC plasticizer. J. Ind. Eng. Chem. (2013). doi:10.1016/j.jiec.2013.09.021
Pedro, M.I., Cabral, F.L.: A new innovative model using RFID: a system design and its implementation. Int. J. Latest Trends Financ. Econ. Sci. 3(2), 508–219 (2013)
Porto, J.L.: Analyse multicritère dans le cadre des appels d’offres pour la constrution de travaux publics et prives: Le cas du métro de Porto au Portugal. Newsletter of the European Working Group ‘Multiple Criteria Decision Aiding’ Serie II(15), 1–2 (1999)
Rodrigues, A., Pinheiro, P.R., Rodrigues, M.M., Albuquerque, A.B., Gonçalves, F.M.: Towards the selection of testable use cases and a real experience. In: WSKS (2), pp. 513–521 (2009)
Rodrigues, T., Oliveira, M.D., Brito de Sá, A.: Multicriteria model to allocate human resources in community care programmes. In: 1s t Portuguese Meeting in Bioengineering (ENBENG), Lisbon, Portugal, pp. 1–6, Lisbon, 1–4 March 2011. Bioengineering (ENBENG). doi:10.1109/ENBENG.2011.6026087
Romeiro, A.R., Bernasconi, P., Puga, B.P., Andrade, D.C., Sobrinho, R.P.: Assessment of existing and proposed policy instruments for biodiversity conservation in São Paulo - Brazil: a coarse grain analysis. Brazil Mata Atlantice Coarse Grain Case Study Report 3, POLICYMIX (2012)
Roubens, M., Rusinowska, A., de Swart, H.: Using MACBETH to determine utilities of governments to parties in coalition formation. Eur. J. Oper. Res. 172(2), 588–603 (2006)
Saaty, T.L.: The Analytic Hierarchy Process. McGraw-Hill, New York (1980)
Sanchez, C., Montmain, J., Vinches, M., Mahieu, B.: Planning of maintenance operations for a motorway operator based upon multicriteria evaluations over a finite scale and sensitivity analyses. In: Conference on Informatics in Control, Automation and Robotics, pp. 23–35 (2007)
Sanchez-Lopez, R., Bana e Costa, C.A.: El enfoque macbeth para la incorporación de temas transversales en la evaluación de proyectos de desarrollo. In: Leyva López, J.C., Avilés Ochoa, E., Zepeda Rodriguez, J.J. (eds.) Herramientas Operativas para el Análisis Multicriterio del Desarrollo Económico Local, pp. 335–356. Plaza y Valdés, México (2009)
Sanchez-Lopez, R., Bana e Costa, C.A., De Baets, B.: The MACBETH approach for multicriteria evaluation of development projects on cross-cutting issues. Ann. Oper. Res. 199(1), 393–408 (2012)
Santos, F.A., Margotti, A.E., Garcia, R.: Multi-Criteria Decision Aid (MCDA) as a tool to support health technology incorporation process. In: Long, M. (ed.) World Congress on Medical Physics and Biomedical Engineering, vol. 39, pp. 692–695, Beijing, 26–31 May 2012. IFMBE Proceedings, Springer, Berlin/Heidelberg (2012)
Shah, L.A., Vernadat, F., Etienne, A., Siadat, A.: Value-Risk Graph: a decision-making tool for supply chain and industrial system engineering. In: 6th IFAC Conference on Management and Control of Production and Logistics, vol. 6, pp. 414–419. Center for Information Technology Renato Archer, Fortaleza (2013)
Silva, J.S.V., Bana e Costa, C.A.: Especificação de uma ferramenta de apoio à decisão para gestão pública em regiões semi-áridas. Research paper 23/2001, CEG-IST, Technical University of Lisbon (2001)
Soares, J.O., Bana e Costa, C.A.: Value judgements in portfolio management. Working paper cegist 28/2001, CEG-IST, Technical University of Lisbon (2001)
Soguel, N., Martin, M.J., Tangerini, A.: The impact of housing market segmentation between tourists and residents on the hedonic price for landscape quality. Swiss J. Econ. Stat. 144(IV), 655–678 (2008)
Soguel, N.C., Chatagny, F.: Analysing Municipalities Amalgamation and Collaboration through the Principle of Fiscal Equivalence and the Principal-Agent Model. Contributions á L’Action Publique. J.L. Chappelet, PPUR & Haupt Verlag, 18 pp (2006)
Sousa, J.V.M., de v Silva, J.S., Fernandez, M.P.: MAC-OLSR: improving OLSR protocol in mesh networks using multiple metrics. In: 6th Euro American Conference on Telematics and Information Systems, pp. 1–8, Valencia, 23–25 May 2012
Stolt, R., Ensslin, L.: Avaliação de pilotos very ligth jets utilizando-se a MCDA-C. Conexão Sipaer 1(1), 85–103 (2009)
Tangerini, A., Pictet, J., Soguel, N.: Using a MultiPLE criteria decision analysis approach for landscape quality assessment. Working paper de l’IDHEAP 17/2006, Chavannes-près-Renens, Chaire de Finances publiques (2006)
Thomaz, J.F., Bana e Costa, C.A.: Decision conferencing within a multimethodological framework: Developing a performance appraisal model for the Portuguese army officers. In: 46th Annual Conference of the International Military Testing Association (IMTA) & NATO’s Research Task Group on Recruiting and Retention of Military Personnel, IMTA, Brussels, 13 pp (2004)
Vernadat, F., Shah, L., Etienne, A., Siadat, A.: VR-PMS: a new approach for performance measurement and management of industrial systems. Int. J. Prod. Res. 51(23–24), 7420–7438 (2013)
Vieira, V., Oliveira, C.S., Bana e Costa, C.A.: A methodology to evaluate the strategic importance of bridges and tunnels considering seismic vulnerability: Application to Lisbon. In: EuroConference on Global Change and Catastrophe Risk Management: Earthquake Risks in Europe, Laxenburg, IIASA, 10 pp (2000)
Vindigni, G., Caniglia, E., D’Amico, M., di Vita, G.: Supporting small producers in mexico’s economic integration process. A case study on rural michoacan based on an MCDA approach. In: 119th EAAE Seminar ‘Sustainability in the Food Sector: Rethinking the Relationship between the Agro-Food System and the Natural, Social, Economic and Institutional Environments’, Capri (2010)
von Winterfeldt, D., Edwards, W.: Decision Analysis and Behavioral Research. Cambridge University Press, Cambridge (1986)
Wancura, H., Mubbala, R., Steinberger-Wilckens, R., Novak, A.: A technology monitoring and assessment tool for technology road mapping. In: 4th European PEFC and H2 Forum, Lucerne, pp. 1–5 (2013)
Watson, S.R., Buede, D.M.: Decision Synthesis: The Principles and Practice of Decision Analysis. Cambridge University Press, Cambridge (1987)
Zamcopé, C.F., Ensslin, L., Ensslin, S.R., Dutra, A.: Modelo para avaliar o desempenho de operadores logísticos - Um estudo de caso na indústria têxtil. Gestão & Produção 17(4), 693–705 (2010)
Acknowledgements
The authors thank José Ferrão, Manuel Pedro Chagas, Mónica Oliveira, Teresa Rodrigues and two anonymous referees for their valuable comments on earlier versions of this chapter and Diana Lopes and Teresa Rodrigues for their invaluable collaboration in updating Sect. 11.10.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer Science+Business Media New York
About this chapter
Cite this chapter
Bana e Costa, C.A., De Corte, JM., Vansnick, JC. (2016). On the Mathematical Foundations of MACBETH. In: Greco, S., Ehrgott, M., Figueira, J. (eds) Multiple Criteria Decision Analysis. International Series in Operations Research & Management Science, vol 233. Springer, New York, NY. https://doi.org/10.1007/978-1-4939-3094-4_11
Download citation
DOI: https://doi.org/10.1007/978-1-4939-3094-4_11
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4939-3093-7
Online ISBN: 978-1-4939-3094-4
eBook Packages: Business and ManagementBusiness and Management (R0)