Fuzzy-based evaluation model for testing the machinability of alloy steels

Abstract

This study presents the development of a Mamdani-type fuzzy inference system for evaluating the machinability of 1.2379 tool steel. The aim of the model is to handle uncertainties and nonlinear relationships arising in machining processes in a more flexible and realistic manner than traditional, simple ratio-based evaluation methods. Machining experiments were carried out using three milling tools with different numbers of teeth and helix angles, while the feed per tooth was varied at four levels (0.03, 0.05, 0.07, and 0.09 mm/tooth). The experiments provided measured surface roughness values (Ra), whereas the tensile strength (Rm) values were incorporated into the fuzzy model based on catalog data. The fuzzy system consists of two submodules: the first aggregates the influence of the alloying elements into an Alloying Effect Index (AEI), while the second calculates the Machinability Index (MI) using the AEI, Ra, and Rm parameters. The results of the fuzzy-based evaluation were compared with an assessment model constructed by us based on traditional linear calculations, which exhibits lower sensitivity and produces discrete, sharp transitions. In contrast, the fuzzy-based approach provides smoother and more realistic transitions, making it a more effective tool for the predictive analysis of machinability and for supporting decision-making.