Assume you are writing a Genetic Program. How would you deal with the following issues?

A) Suppose we have the following chromosomes:

(Y / (X - Z)) and A+B+X

After swapping two subtrees we get:

(Y / (X - X)) and A+B+Z

The first chromosome triggers a divide-by-zero. How might this problem be addressed?

B) Suppose we have the following chromosome:

(Y / X)

This does not trigger a divide-by-zero, but if attribute X contains a row with a zero value, then it results in a divide-by-zero. How might this problem be addressed?

C) Suppose within a population there are multiple chromosomes that are syntactally different, but semantically equivalent. A simple example is:

(Y / Z); (Y*3) /(3*Z); and (Y*Z) / (Z*Z)

Obviously this situation leads to redundancy in the population and inefficiency in the whole GP process. How might this problem be addressed?

(Y / X)

This does not trigger a divide-by-zero, but if attribute X contains a row with a zero value, then it results in a divide-by-zero. How might this problem be addressed?

Can someone explain these questions to me