How do you Ask Answerable Questions

I think it is so easy to make design problems that are esentially impossible to act upon. I mean, problems that have requirements that do not take into consideration reality, the issue of course is that it is hard to tell how far something is from reality when it is conceived. I guess sometimes instances like this are considered over ambitious however I think it is actually an issue in many situations where evaluating the complexity of a problem is difficult. For instance in operating systems, the complexity of the systems involved is so high that working out what effect a requirement will have on the rest of the system is really not easy. In some cases, I would say problems like this can be considered wicked however in many cases I think it is not the problem that its self is complex it is the evaluation method that returns data on the problem that is too complicated. For example. The problem of going to the moon is not really that wicked. I mean it involves a lot of complexity and a lot of very high rigour decision making but it is really not a problem that continuously changes and in which any answer is only a very temporary answer. The issue of difficulty in deciding weather the question is answerable is really high. 

Q: Let's go to the moon (can we go to the moon)? 

A: Yes lets, yay (we have no idea if we can or not).

 I wonder, is there a good way to work out early on that a design problem is like this. I think it wastes a lot of time and energy when people persue unsolvable problems. And I really mean is there a way that humans can look on the problem level, not on the implementation level, and establish an informed point of view either way. 

If anybody knows anything about methods for this I would be quite interested. Please comment.