In last decades, the interest to solve dynamic combinatorial optimization problems has increased. Metaheuristics have been used to find good solutions in a reasonably low time, and the use of self-adaptive strategies has increased considerably due to these kind of mechanism proved to be a good alternative to improve performance in these algorithms. On this research, the performance of a genetic algorithm is improved through a self-adaptive mechanism to solve dynamic combinatorial problems: 3-SAT, One-Max and TSP, using the genotype-phenotype mapping strategy and probabilistic distributions to define parameters in the algorithm. The mechanism demonstrates the capability to adapt algorithms in dynamic environments.

This paper presents a brief review of techniques used to allow evolutionary algorithms to adapt to optimization problems in dynamic environments, through exploration of the control parameters of genetic algorithms as well as genotypic interpreters. A description of some of the most used evolutionary techniques is included, with major emphasis on genetic algorithms and their relationship with the problem of adaptation to the environment. The article also discusses state used models to tackle these kinds of problems, including self-adaptation and genotype-phenotype mapping.