Multiobjective fuzzy linear regression analysis for fuzzy input-output data

Abstract

Fuzzy linear regression models, where both input data and output data are fuzzy numbers, are introduced by using three indices for equalities between fuzzy numbers. By considering the conflict between the fuzzy threshold for the three indices and the fuzziness of the fuzzy linear regression model, three types of multiobjective programming problems for obtaining fuzzy linear regression models are formulated corresponding to the three indices. Then a linear programming based interactive decision making method to derive the satisficing solution of the decision maker for the formulated multiobjective programming problems is developed. A numerical example demonstrates the appropriateness and efficiency of the proposed method.