Deriving parameters and preferential model for a total order in ELECTRE III

The paper describes a procedure constructed with a Genetic Algorithm (GA) to derive the intercriteria parameters for the outranking method ELECTRE III. The present paper deals with the total order problem as reference set from expert. An aggregation/disaggregation approach is implemented to derive intercriteria parameters from reference set. The intercriteria parameters solutions generated by GA constructs a preferential model, which is exploited by net flow deriving an order of alternatives at least as similar as input order from expert.