A parallelization of the genetic algorithm is planned and
along with this the implementation of a strategy adaptation [6].
There are many mutation operators, which are applied with
different frequencies.
It might be ingenious to exchange or move large parts of
the layout during the early stage of optimization, and
doing only minor changes when the population converges to an optimum.
This can become possible by adapting the frequencies of the
different mutation operators during the optimization.
Further a gene-pool recombination operator [1,4] will be implemented which might replace the current crossover operator. For a combinatorial optimization problem like the layout generation, the biologically motivated crossing of two parent chromosomes is likely to be less efficient. The construction of an offspring out of a pool of good building-blocks seems to be more suitable.