Greetings:
In a model we want to include the relation between a cluster and the process that produces it. So we have come up with option A to show that relation. What we usually do is not to attach the cluster to any interface but the event with the interface (option B). However, as you can see it is not actually accurate since the catalog is created once per year and then it doesn't change whereas the requirements are received at any time.
Since ARIS allows it I suppose option A would be better. But I'd like to hear other opinions or solutions about it.
Regards
Holger
Hello Holger,
surely it's a matter of taste, but I would vote for Option B.
If it is important to show the source of the cluster object, show another occurrence copy of the cluster next to the interface in option B and draw the connection "Interface has as output Cluster" there. I think that was the intention of allowing that connection.
With option A the reader may get confused about the nature of the process flow.
Regards, M. Zschuckelt
Hi M Zschuckelt:
Thanks for your input. That is a good option, but the problem with option B is the AND condition. The process is triggered when the requirement is received and it will have to wait the event "catalog created", which could only happens once a year in order to proceed to the activity. What actually happens is that the activity "register requirement" uses the cluster "Good and services catalog". Like you said what we wanted to show is the origin of that cluster which is the product of the "Making good and service catalog" process.
Best regards
Holger Oviedo
Hi Holger,
this depends, how you look at it. If you are running simulations with your processes, the cardinality between your process instances indeed may be a problem. If you just look at those processes from a business perspective, you may be happy to read, that you must have a completed catalogue and a requirement as prerequisite for your process, regardless of the cardinality.
If you look at it strictly, your catalogue is some product of an entirely different (once-a-year) process, represented by the general availability of the catalogue cluster object. On VACD level you just chart out, that you have some "product development" process, which incidentally produces a catalogue once a year (maybe you state that in the FAD of "product development"). Then you would leave out the AND branch altogether in your example and just use the catalogue as a given resource. Further information on the origin of the catalogue then can be looked up in the properties of the cluster object, where the "product development" process (or some function inside it) will produce it as output. Then it may also be a good idea to have your process represented by a value added chain in the VACD process landscape and assign the catalogue as an input to it in the corresponding FAD.
Was that clear or too confusing?
Regards, M. Zschuckelt
Greetings M. Zschuckelt
I have another question, perhaps you or someone else at software AG can help me with that. Here is the link:
http://www.ariscommunity.com/users/hoviedo/2015-07-14-use-logic-rules
Regards
Holger