A guard on an activity edge leaving an activity


  I would like to ask whether this is allowed. If the answer is yes, why do we need a decision node as all?

It is indeed allowed; UML isn’t as restrictive as some people might think. this might give you a good example as to why its not uncommon to combine the two.

In short; you don’t have to but using decision points points makes your diagrams a lot easier to navigate. However, in the case of the previous example it makes sense to “embed” the decision point since it makes the diagram a whole lot easier to read (2 decisions points right after each other can be kinda icky).

Hope this helps.