A fuzzy-set based strategy in dynamic environments

Carlos Cruz Corona, David Pelta,Jose Luis Verdegay.

Many complex real-world optimization problems are dynamic. In order to approach them is necessary to have tools that are able to adapt to the changes that take place in the time. In this work we propose a strategy that jointly use a set of solutions and a set of simple agents. Implicit and explicit memory mechanisms are used and we analyze the behavior of the strategy when coupled with a fuzzy rule to control the updating of the solution’s set. Tests are performed on the moving peaks benchmark problem under different scenarios.

PDF full paper