I use them interchangeably, however I'd like to know when one is better or more appropriate than the other.
We use them both to refer to future events.
Going to is used to refer to a planned or previously considered decision
Will is used to refer to a spontaneous decision.
If the phone rings, you would say
I'll get it -not- I'm going to get it
as the latter would suggest that you had plans to answer the phone, even though it has only just started ringing.
If someone asks about your plans for the summer holiday which begins next week, you would answer
I'm going to tour Europe -not- I'll tour Europe
as the latter would suggest that you have just decided on the spot that you are going to travel around Europe, a fairly large decision that requires considerably more planning.
With regards to @Rbustos answer above, which at first glance seems like it might contradict mine:
It's going to blow is correct. Here we can clearly see from the situation that an explosion is imminent (a ticking clock or a bubbling chemical, for example). So our observation is based on a calculation, hence 'going to' even though it was a relatively on the spot deduction.
It will blow suggests a consequence, i.e. 'it will blow if...' or in answer to the question 'what will happen if I...?' - here we are dealing with a prediction, not a clear observation.
Hope that helps.
I created the following method to help understanding, might be not precise but works as a guideline.
Here 'decision' and 'action' belongs to usage "be going to" while "Decision" and "Action" fall in the usage of "will".
There is more of a sense of immediacy to "is/are going to" in most cases.
For example, if talking about an overheating boiler, you might say:
This means get out of the building now.
This means get a team in there to fix the situation.