Immune Evolutionary Algorithm for Determining Weights in Interval Judgments

摘要

In a Multiple Criteria Decision Analysis (MCDA) problem, determination of weights is an important aspect of the Analytic Hierarchy Process (AHP). In this paper, the problem to computer the weight vector of the interval number matrix is regarded as a nonlinear optimization model with constraints. An immune evolutionary algorithm is proposed to solve the problem with high speed by the matrix based coding technique and several immune operators including clonal proliferation, hypermutation, clonal elimination and immune memory. The experiments show the algorithm has the promising performance of convergence and accuracy.