ABSTRACT
Evolutionary computation is widely used in real-world applications, such as digital filters design, design and optimization of digital circuits, partitioning of VLSI circuits on subcircuits with minimal number of external connections between them, optimization of placement of 2D elements, training of artificial neural networks, and optimization of parameters in grinding process. In the real-world, engineers often need to solve difficult optimization problems, such as digital filters designed with non-standard characteristics, floor planning of 2D elements, decomposition of digital circuit on subcircuits. The fitness function in evolutionary algorithm is an element linking the considered problem and the population of individuals. The main task of this function is a determination of qualities of particular individuals in the aspect of problem to be solved. The selection, also named as reproduction, is a procedure of choosing given individuals from the population in order to create a new population in the next generation of evolutionary algorithm.
