Genetic algorithms for engineering optimization: theory and practice

The genetic algorithms are heuristics and thus they do not ensure an optimal solution. We propose to use a fuzzy controller for an improvement of genetic algorithms. The speed of natural evolution is changeable. Genetic algorithms can be classified into three main categories: a basic GA, evolution strategies, and a mobile GA. The mobile GA has a variable chromosome structure. The aim of this paper is to consider an efficiency of various GAs. The paper explores the utility of the recently developed GA paradigm for model fitting using sets of empirical data. To support this work, the real-world problems were explored. Examples of real-world problems are telecommunication networks traffic optimization and the task of elements placement on plane. In the case of telecommunication network traffic optimization, the fitting model is a fuzzy rule based system. In this paper, the concept of fuzzy probabilistic variable is introduced.