1. Modeling of neuro-fuzzy system as a support in decision-making processes
- Author
-
Dragan Marinković, Duško Tešić, Darko Bozanic, and Aleksandar Milić
- Subjects
Logarithm ,Neuro-fuzzy ,business.industry ,Computer science ,TOPSIS ,General Medicine ,Systems modeling ,Multiple-criteria decision analysis ,Machine learning ,computer.software_genre ,Loader ,Key (cryptography) ,Artificial intelligence ,business ,computer ,Selection (genetic algorithm) - Abstract
In the paper is presented Neuro-Fuzzy System as a decision-making support in the selection of construction machines (the example of selecting a loader is provided). Construction characteristics of a loader make the basis for selection, but also other elements of importance. The data for Neuro-Fuzzy System modeling are prepared using the Multi-Criteria Decision Making (MCDM) methods: Logarithm Methodology of Additive Weights (LMAW), VIKOR, TOPSIS, MOORA and SAW. The paper also presents the method of aggregation of weights of rules premises (AWRP), which defines the key rules of Neuro-Fuzzy System. Finally, the training of the model is tested. The data for the selection of input variables and for model training are obtained by engaging experts.
- Published
- 2021
- Full Text
- View/download PDF