A framework for the description of evolutionary algorithms

Evolutionary algorithms (EA) are optimisation techniques inspired from natural evolution processes. They handle a population of individuals that evolve with the help of information exchange procedures. Each individual may also evolve independently. Periods of co-operation alternate with periods of self-adaptation. We define a terminology and give a general framework for the description of the main features of any particular evolutionary algorithm. Such a description does not provide, nor does it replace, algorithm pseudo-codes. The aim is to develop tools that may help understanding the "philosophy" of such methods. (C) 2000 Elsevier Science B.V. All rights reserved.