Hi,
Is the purpose of a Process Interface to be used to connect process models into a whole, or to cut it in reasonable portions to promote readability of it?
Maybe it's for both scenarios, but what if it's just for one of them - what is then the recommended way for the other one scenario?
Regards,
Jonas Ekstroem
Dear Jonas,
Did you read already the other discussions tagged with process interface?
There are various aspects discussed.
Cheers
Rune
Yes, I read some good articles about how to use a PI but no one actually touched down on details of when and when not to use a PI.
The closest discussion was from this article excellently answered by Mr. Ralf Angeli. Mr. Ralf states that:
"Process interfaces are used when a process is cut into pieces and the pieces are put into different diagrams."
My question is really what is the purpose of cutting it into pieces? Is it for functional boundary, for a distinction to third party, or is it for cutting the model into pieces that can be printed in reports with readable grafics, all or none of the above. Do you have an idea?
The purpose is up to the modeler and the situation. A process interface links a process to either the preceding process or following process at the same level. It doesn't really matter if you split the process for readability or for some functional boundary. It is really just a shortcut to allow you easy navigation between models which make up an entire process.
Rick, you don't see a reason to a distinction between shapes used for splitting models into readable pieces of the same level and shapes splitting models into reusable components. What I would like to avoid is an assignment from one level X activity to a level X model which is part of a whole level X model and originally just a piece of the whole model. I can compare with how BPMN is making a clear distinction between embedded sub-processes and reusable sub-processes.
Jonas,
Sorry, I didn't pick up on the reusability aspect from your original post. A reusable process in ARIS standard methodology is represented by the Function symbol. The Process Interface is intedned to link 2 processes at the same modeling level. There is no distinction between a sub-process that is re-used vs one that is used only once. I do know of customers that created a distinction either through naming conventions or by modifying the symbol to indicate that a process is reusable.
Rick