4.6.2 MCDA Method
In the evaluation, the Non-Additive Robust Ordinal Regression (NAROR) has been
applied (Angilella et al. 2010a, b, 2016). NAROR is a recent extension of the Robust
Ordinal Regression family of MCDA methods (Greco et al. 2008b; Figueira et al.
2010; Corrente et al. 2013, 2014) to the Choquet integral preference model which
permits to represent interaction between considered criteria through the use of a set
of non-additive weights called capacity or fuzzy measure.
For evaluating the alternatives under examination, a family of six criteria has been
constructed which includes location of the building, current destination, age, flexibility of the estate to the transformation, estimated value of the asset and surface. A
panel of experts has been created for simulating the decision-making process. The
panelists provided preference information in terms of preferences between pairs of
alternatives and relative importance and interaction of considered criteria.
Figure 6.9 represents the exploitation of the preference model obtained from the
work with the experts’ panel.
Even if this information is very meaningful, in general the DM would like to have
one ranking of all the considered sites. For this reason, the most representative value
function (MRVF) has been computed considering the results of the NAROR and
giving a total order of all the assets at hand.
4.6.3 Results and Role Played by the MCDA
According to the results of the calculation, the asset 2 turned out to be the best
performing solution. In the lights of these results, the arguments at disposal of the
Fig. 6.8 Location of the considered real estate properties
110
I. M. Lami et al.
In the evaluation, the Non-Additive Robust Ordinal Regression (NAROR) has been
applied (Angilella et al. 2010a, b, 2016). NAROR is a recent extension of the Robust
Ordinal Regression family of MCDA methods (Greco et al. 2008b; Figueira et al.
2010; Corrente et al. 2013, 2014) to the Choquet integral preference model which
permits to represent interaction between considered criteria through the use of a set
of non-additive weights called capacity or fuzzy measure.
For evaluating the alternatives under examination, a family of six criteria has been
constructed which includes location of the building, current destination, age, flexibility of the estate to the transformation, estimated value of the asset and surface. A
panel of experts has been created for simulating the decision-making process. The
panelists provided preference information in terms of preferences between pairs of
alternatives and relative importance and interaction of considered criteria.
Figure 6.9 represents the exploitation of the preference model obtained from the
work with the experts’ panel.
Even if this information is very meaningful, in general the DM would like to have
one ranking of all the considered sites. For this reason, the most representative value
function (MRVF) has been computed considering the results of the NAROR and
giving a total order of all the assets at hand.
4.6.3 Results and Role Played by the MCDA
According to the results of the calculation, the asset 2 turned out to be the best
performing solution. In the lights of these results, the arguments at disposal of the
Fig. 6.8 Location of the considered real estate properties
110
I. M. Lami et al.