Genetic Algorithm Based on Explicit Memory for Solving Dynamic Problems

Nowadays, it is common to find optimal point of the dynamic problem; dynamicproblems whose optimal point changes over time require algorithms whichdynamically adapt the search space instability. In the most of them, the exploitationof some information from the past allows to quickly adapt after an environmentalchange (some optimal points change). This is the idea underlinin the use ofmemory in the field, which involves key design issues concerning the memorycontent, the process of memory update, and the process of memory retrieval. Withuse of the Aging Best Solution and Keeping Diversity in Population, the speedconvergence of algorithm can be increased. This article presents a geneticalgorithm based on memory for dealing with dynamic optimization problems andfocuses on explicit placement of memory schemes, and performs a comprehensiveanalysis on current design of Moving Peaks Benchmark (MPB) problem. The MPBproblem is the most proper benchmark for simulation of dynamic environments. Theexperimental study show the efficiency of the proposed approach for solvingdynamic optimization problems in comparison with other algorithms presented inthe literature.

Keywords:

Dynamic Optimization , Genetic Algorithm , Explicit Memory , Offine Error