The exclusive gateway at the end of the process is correct.
However, I think the upper part of your diagramme might need some clarification. I see two design choices that you might want to rethink:
Does the Search for Candidates
task create a list of candidates to interview or a single candidate ? In your diagramme, it first looks like there is a list of candidates that get interviewed in parallel during the Level 1 Interview
task. However, the subsequent gateway suggests that you decide for each candidate, whether he/she has passed that level. If not, you move back to searching. The same applies for the Level 2 Interview
task.
The inclusive gateways also suggest that you are deciding on individual candidates, whether they have passed each level because you can have a Yes
and a No
at the same time.
If a search for candidates results in a list of candidates that get interviewed at the same time, then I would put the interviews and Assign to project
tasks inside a sub-process. You would loop through the interviews and assigning tasks until all posts are filled or all candidates examined (note the exit condition in the annotation at the top). If one round of interviews has not filled all posts, you would decide whether you need to launch another round.
If you rather interview candidates individually and want to avoid a sub-process, then you could keep your structure but use exclusive instead of inclusive gateways:
Note the data items in both examples that make it explicit whether your search resulted in a list of candidates that get interviewed or a single candidate at a time.