genome evaluation

[G.Spagnuolo]

Hi galib-ers,
I am using a non stationary penalty function for solving a constrained 
optimization problem.
I've implemented a steady state ga with, we say for example, 
gaNpReplacement=0.7.
Nevertheless, as the penalty function changes its value, the individuals 
survived from one generation to the subsequent have to be re-evaluated, 
while the "new" individuals does not need to be re-evaluated.
Now, my question is the following: is it enough to update the first 30% of 
the population after the ga-step is completed?
I think this is not right, because after the ga.step the individuals are 
re-ordered on the basis of objective functions that refers to two different 
generations.
This problem occurs whenever the objective function change with the 
generation index (from 0 to the maximum number of generations allowed).
Thanks in advance for any suggestion.

Best regards.

G.Spagnuolo


--
Dr. Giovanni Spagnuolo, Ph.D.
Assistant Professor
University of Salerno
Dipartimento di Ingegneria dell'Informazione ed Ingegneria Elettrica
D.I.I.I.E.
Via Ponte Don Melillo
84084 Fisciano - Salerno - Italy
phone: +39 089 964258
fax: +39 089 964218
e-mail: spanish at ieee.org

"You can tell whether a man is clever by his answers.
You can tell whether a man is wise by his questions."
Naguib Mahfouz