How to set the elitism preservation strategy in NSGA-I

[hjh-2001]

How to directly preserve 10% of the elite individuals in the population to the next generation? Then, 80% of the individuals are obtained through crossover, and 20% of the worst individuals are mutated, that is, reinitialized.

[blankjul]

Basically this comes back to writing your own algorithm. pymoo is very modular and it seems like you only want to write your own survival operator.

A good example for you might be to look at RVEA
https://raw.githubusercontent.com/anyoptimization/pymoo/refs/heads/main/pymoo/algorithms/moo/rvea.py
where I have implemented also a survival.