Based on the theory of similarity for describing laser hardening processes, we suggested dimensionless parameters that simultaneously take into account both process parameters of the laser heat processing and the thermophysical characteristics of the metal being processed, and have a clear physical meaning. To build a statistical model of the laser exposure area depth from the laser radiation parameters, we used in our studies the methods of mathematical experiment design, and in order to summarize the results obtained on the basis of applying similarity theory relations, we suggested to use the following dimensionless (generalized) parameters: 1) the laser tun overlap factor S*; 2) the relative power of the laser processing P*; 3) the relative velocity of the laser beam movement V*; 4) the relative depth of the hardened layer Z*. The overlap factor S* describes the effect of subsequent adjacent laser nms on the previous ones: with S* < 1, the runs overlap, and the material structure in the previous laser area changes; usually, with S* > 2, the mutual effect of adjacent runs can be neglected; therefore, this condition is often used in the laser hardening of a metal working process (MWP) tool. The value P* corresponds to the ratio of the effective laser radiation power P-ef = K-abs.P to the power that can be diverted from the surface due to thermal conductivity deep into the metal without melting. The value V* is equal to the ratio of the laser beam velocity to the temperature front propagation velocity in this material. As the relative value Z*, we adopted the ratio of the hardened layer depth to the maximum possible theoretical value , which is achieved when the temperature on the metal surface reaches its melting point. The use of dimensionless parameters allows us to build mathematical models for laser hardening and, based on the same, to develop and to optimize the laser heat processing method.