OPTIMASI FUZZY LOGIC MENGGUNAKAN ALGORITMA GENETIKA UNTUK PENENTUAN KELAYAKAN KREDIT

Abstract

 One of many factors in the bad debt are errors in the decision of a credit-worthiness of credit applicants. Required a decision support system to assist decision makers in taking a decision. Fuzzy Logic is used as a decision support system that can help in decision-making in the member's credit worthiness CU XYZ. This research uses Fuzzy Logic inference method Tsukamoto and optimized using Genetic Algorithms. Genetic Algorithms used for generating membership functions. This optimization can optimize the value of a variable threshold membership value changing following the environmental conditions. In case of testing optimization from this method, comparing fuzzy with and without using Genetic Algorithms and fuzzy with Modified Particle Swarm optimization (MPSO) and Genetic Algorithms (GA). The evaluation result of comparing between GA, MPSO, and without optimazion, indicates that the application of GA has total misunderstanding comparison with actual data by 18 percent, while compared with MPSO by 20 percent, and by 32 percent without optimization. Fuzzy Logic and optimized using Genetic Algorithms can be help credits dicision a member at CU XYZ.

Keywords—Fuzzy Logic, Genetic Algorithms, Credit-Worthines