The conventional analytical treatment of actual systems can in fact only be successfully performed in the simplest cases. It fails when a more or less realistic model of multiphase alloys in considered. As a matter of fact, computer simulation is a new approach to the probelm and is applicable to real alloys. The computer makes it possible to solve problems that have been considered unsolvable. The use of high-speed computers for modeling the processes occurring in alloys cannot be obviated, and I believe that computer-simulated research will form a new field in the material sciences.
Even the first attempts to simulate, with the help of computers, the martensitic transformation and the strain-induced coarsening in decomposed alloys using but idealized crude models gave an encouragingly good description of those processes, and the results obtained are sometimes in excellent agreement with electron microscopic observations. We may hope that the use of more realistic (and therefore more elaborate) models will make it possible to come close to theoretically predicting the structures of two-phasealloys applied in industry.
There is another problem that should be mentioned. The physical processes occurring in alloys are affected by varied factors that sometimes conceal the important characteristics of the phenomenon. Eliminating them may be a difficult problem which can hardly be solved in all cases. From this standpoint computer simulation serves as an “experiment” carried out under ideal conditions in the absence of any interfering factors.
— A G Khachaturyan, Theory of structural transformations in solids, 1983.