Genetic algorithms for adaptation to dynamic environments - a survey

Adaptation to dynamic environments is an important application of genetic algorithms (GAs). In management applications, such as job scheduling, flexibility to cope with changes of orders, due dates and malfunctions of machines is required by problem solvers. In experimental and simulation-based optimization, algorithms to cope with random fluctuations of fitness values are needed. Further, GAs are also applied to online and on-site optimization and adaptation to various problems. Recently, interest in GAs for dynamic environments has been growing, and a number of novel approaches have been proposed. This paper surveys the methods for applying the GAs to adaptation in dynamic environments. First, the frameworks of dynamic environment adaptation problems are categorized. Then GA techniques for these problems are overviewed. Benchmark problems are also reviewed. Finally, the authors touch on practical applications