Reducer selection using artificial intelligence

In this study, an artificial intelligence method has been planned to be select the most convenient reducer from among 1200 commercially available ones in a firm inventory considering the design variables. The catalogue of the `Yılmaz Gearbox? has been taken as a database for the analysis. Among the possible artificial intelligence methods, the neural network has been exploited with the help of an artificial neural network-based program called Qwicknet 2.23. The parameters used in this program have been transferred to excel environment, which helps to generate the database for the operation constraints of the reducer. The learning ability of the artificial intelligence method has been determined with the use of the artificial neural network program, which uses the parameters from the excel database. Finally, the reducer has been chosen by controlling the degree of learning.