Robust sustainable development assessment with composite indices aggregating interacting dimensions: The hierarchical-SMAA-Choquet integral approach

As simple as possible but not simpler in Multiple Criteria Decision Aiding: the robust-stochastic level dependent Choquet integral approach

“…We believe that the good properties of the methodology we have presented can be very useful in many real life complex decision problems in the main domains of interest such as sustainable development [1], wellbeing and happiness economics [19,20], ranking of universities [11], country competitiveness [10], and so on.…”

“…The DM provided also the following preference information on the universities which performances are shown in To represent this preference information and, in particular, the fact that the importance of criteria as well as the interactions between criteria are dependent on the considered performances we have to use the level dependent Choquet integral. As already explained above, we consider the partition a 0 = 1, a 2 = 3, a 2 = 5 of the interval [1,5]. This means that we are taking into account two capacities, one on the interval [1,3] and one on the interval ]3, 5].…”

“…This enforces the conviction that U 16 is the best among the universities at hand. On the basis of the rank acceptability indices, as done also in [1], for each university we compute the best and the worst possible positions (Table 7(a)) and the most frequent ones (Table 7(b)). In this way, we have a clearer view of the goodness of the considered universities.…”

“…Looking for an amendment of the weighted sum able to represent preferences that cannot be expressed by using (1), one could imagine that weights depend on the level of the notes. For example, one can imagine that Mathematics has a weight w 1 M for notes up to 25, that is M 1 (X) = min{M(X), 25}, and weight w 2 M for the possible excess to 25, that is, M 2 (X) = max{M(X) −25, 0}.…”

A stochastic multicriteria acceptability analysis–evidential reasoning method for uncertain multiattribute decision‐making problems

“…Depending on the DM's setting, SMAA‐2 computes the probability of each most preferred alternative by inverse exploration of the feasible weight space for full ranking (Lahdelma & Salminen, ). In recent years, SMAA‐2 and its variants have been applied to many fields, such as decision analysis (Angilella et al, ; Pelissari, Oliveira, Amor, Kandakoglu, & Helleno, ), market segmentation (Liu, Liao, Huang, & Liao, ), sustainable energy evaluation (Loikkanen, Lahdelma, & Salminen, ), and reservoir flood control operation (Zhu, Zhong, & Sun, ).…”