Page 368 - Biofuels for a More Sustainable Future
P. 368

324   Biofuels for a More Sustainable Future


          Table 11.2 Linguistic variable for assigning weights to evaluation criteria and rating
          alternative

          Number    Linguistics for weights  Linguistics for performance  Fuzzy scale
          1         Lowest (LT)          Worst (WT)              (0,0,0.1)
          2         Lower (LR)           Worse (WE)              (0,0.1,0.3)
          3         Low (L)              Bad (B)                 (0.1,0.3,0.5)
          4         Medium (M)           Medium (M)              (0.3,0.5,0.7)
          5         High (H)             Good (G)                (0.5,0.7,0.9)
          6         Higher (HR)          Better (BR)             (0.7,0.9,1.0)
          7         Highest (HT)         Best (BT)               (0.9,1.0,1.0)


          production in the work of Ren et al. (2013). This FMCDM method was
          presented as follows (Li, 2003; Ren et al., 2013):
             Step 1: Linguistic assessment.
             Assume that a set of the stakeholders and decision-makers have been
          invited to participate in the decision-making process, M alternatives have
          been assessed, and N criteria have been used to assess the alternatives.
          The decision-makers are asked to assign the importance of the criteria
          and rate the alternatives using the linguistic variables (see Table 11.2).
             Step 2: Transformation.
             Transfer the linguistic assessment into fuzzy triangular numbers according
          to Table 11.2. Let ω j be the weight of the ( j)th criterion by the stakeholders
          and decision-makers and e ij be the assessment on the performance of the ith
                                x
          alternative with respect to the jth criterion. Assume the fuzzy decision-
          makingmatrixdeterminedbythedecision-makersispresentedinEq.(11.11).

                                     C 1 C 2 ⋯ C n
                                      e ω 1 ω 2 ⋯ e ω n
                                          e
                                          x
                                                 x
                                     x
                                 A 1 e 11 e 12 ⋯ e 1n
                                                                     (11.11)
                                     x
                                          x
                                                 x
                                 A 2 e 21 e 22 ⋯ e 2n
                                  ⋮   ⋮   ⋮   ⋮   ⋮
                                 A m ex m1 ex m2 ⋯ ex mn
          where A i represents the ith alternative, C j represents the jth criterion,

          e ω j ¼ ω ω M  ω U  is the weight of the jth criterion, and e ij represents
                 L
                                                               x
                 j   j  j
          the performance of the ith alternative with respect to the ith.
             Step 3: Determining the ranking matrix.
             Rank the alternatives corresponding to each criterion according to
          Eq. (11.10), for the ( j)th criterion, the ranking matrix can be obtained with
          the following method.
   363   364   365   366   367   368   369   370   371   372   373