Abstract In this study a mathematical–computational model of tool wear of PCBN (polycrystalline cubic boron nitride) was developed in turning of quenched and tempered AISI D6 steel (57 HRC) using experimental planning and statistic techniques. On the experimental trials many parameters are important such as: surface roughness, cutting force and tool wear. These parameters were evaluated according to their statistical significance using Statistica ® and Matlab ® softwares. Through a multiple-regression analysis, it was possible to establish a mathematical model for estimating tool wear as a function of the cutting parameters. This model enhanced estimation of the ideal cutting conditions for turning hardened steel, i.e., those that generate minimum damage on the PCBN tool without compromising productivity.