Purpose – The work approaches the finding of a good sorting combination, starting from the traditional method of selective assembly, so that the control over the precision classes of the bearings is more precise and obtaining management procedures to minimize the resulting waste. Methodology/approach - Systemic optimization through a selective assembly method, regarding the control on the radial play of the radial ball bearings. Findings – Following the application of the selective assembly algorithm, some classifications of the radial clearance in the bearing have resulted, with the help of which some management procedures can be obtained for the reduction of bearing rejects. Research limitations/implications – Limitations regarding the application of the method over a wide range of dimensions. Practical implications – Being a wide range of bearing sizes, the proposed algorithm requires a C +++ programming language, for the automatic calculation and sorting of the bearing components. Originality/value – optimization algorithm by the method of selective assembly.