Last week I presented a webinar on BPMN Modeling with Method and Style that explains what it’s all about. You can see the recording by clicking here. Or click here to download the slides.
What Is BPMN Method and Style?
About the Author: Bruce Silver
Related Posts
3 Comments
Leave A Comment
You must be logged in to post a comment.
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Hello Bruce,
I have a question on your BPMN rules #’s:
19: Don’t use an XOR gateway to merge alternative paths, unless into another gateway. Just connect the sequence flows directly.
20: Don’t use an AND gateway to join parallel paths into a None end event. A join is always implied at a None end event.
Why is this so? It doesn’t explicitly explain why tis is not a “good” practice. I’ve been taught from my Masters in BPM (QUT) that whenever you use an XOR or AND split gateways, as much as possible, close it with a JOIN. The reasoning behind it is that when we do a “token game” for logical and simulation analysis, the model explicitly provides the necessary information needed. However, I would like to understand why this is not the case fro your end and the reasoning behind it. Please kindly advise. Thanks muh.
Gerald
Today I am more accepting of those things. The merge gateways are redundant to the direct merge semantics, which is why I say leave them out. The join into a None end event at least has the merit of emphasizing that both paths must complete . I suspect the QUT style has to do with mapping to Petri nets- not sure. In the BPMN 1.1/1.2 era we suffered through block-structured gateways in order to accommodate BPEL. BPMN 2.0 was an explicit rejection of that.
Thank you for your response Bruce. Indeed, QUT map the models to Petri Nets.