Pitfalls of genetic algorithms

The reader may study about the basics of genetic algorithm here.

There algorithms are generally used for optimization i.e. finding the global maxima instead of regional ones.

Although such algorithms borrow from evolutionary biology they violate one fundamental principle. No one thought in 1000 A.D that we could send a mission to Mars. This is exactly where genetic algorithms fail. The endgame has to be decided at the start i.e. you need to set your expected maxima even before you are running the algorithm.

What if the management makes a mistake in their initial estimate ?

I think linear programming based optimizations are still better.