It is known that, from the point of view of the accuracy of a machine-tool, at its design, the dynamic behaviour of each element of the kinematic chains prevails. Worm-gear drives are widely used in the different machine-tools and robots. Therefore, it is important that during meshing, as far as possible, there are no vibrations, shocks, power losses, noise and low durability. These requirements can be met if, for example, the gear ratio is constant during meshing, without transmission errors, which means that the worm-gear drive should have a high accuracy. The accuracy improvement of the worm-gear drive has long been a focus of attention for machine-tools designers. Thus, this paper presents various approaches to solving such problems, based on modelling and simulation, such as: estimating the load share of worm-gear drives and to calculate the instantaneous tooth meshing stiffness and loaded transmission errors; the desired worm-gear drive design configuration by altering the optimum set of worm-gear drive design parameters which are suitable for the required performance by associating it with SVM (Support Vector Machine); optimization approach for design of worm-gear drive based on Genetic Algorithm; design optimization of worm-gear drive with reduced power loss; etc. The optimization of the worm-gear design is an important problem for the research because the design variables are correlated to each other. An optimal design algorithm developed by the authors of this paper, for worm-gear drive, is also presented.