How to make sure if it will work or not?
The reason why it's in the "future work" section is that no-one knows because it hasn't yet been done (and it's easily possible the original authors never will try it). "Future work" ideas can range in feasibility from wild speculation to simple, almost-guaranteed-to-work extensions that just need a bit more manpower or computer power to finish.
This is a good topic to discuss with your supervisor as soon as possible. You should be asking questions such as:
- How feasible might it be for these ideas to lead to some useful conclusions?
- What if it doesn't work? How soon will we know, and what alternative paths are there?
Of course, some ideas are much easier and more feasible than others. There's no way to know for sure but to try the idea yourself (after making sure that the problem is not already solved in later literature). Because dead-ends often go unpublished, it might even be worth contacting the relevant authors to ask if they attempted to follow up on their future ideas.
It's a courteous thing to do especially if theirs is a recent publication: if the original authors are still working on the problem, publishing the solution first might annoy the other researchers. Though whether to proceed anyway (and give them notice that you're working on the idea as well), stay silent, or offer to collaborate, is a dilemma that's off-topic for this question.