Hello fellow Camunda users,
I was wondering if there is a possibility for users to exclude tasks from a case based on their wishes. To further clarify what I mean:
Suppose I have a case which applies to person 1, person 2, and person 3. The case I made will serve the purposes of all three people and has the right design, marks, sentries and variables. Unfortunately, 50% of the plan items in the case are completely unusable for person 1. He or she will grow tired of having to disable or look at tasks which will never be in his or her process. The simple solution here would be to delete that 50% of the case and offer them the other 50%. Obviously, deleting plan items shouldn’t disrupt the process’ ability to complete.
Imagine this situation but with 50 or more people and 50 or more variations in desirability. I can offer them all a 100% solution but some of them might only use 20% of the case. It’s not time efficient to review each person separately for their needs. I’m wondering if there’s some way where I can offer the person interested in the CMMN case a questionnaire/form with questions about the plan items and depending on their answers and the criteria appended to these decisions automatically exclude any items they do not want inside the case. This way they can now use their more specific case and still choose their flow in that case. Any insights or advice is appreciated, thanks.
I think entry criterions on sentries (https://docs.camunda.org/manual/7.8/reference/cmmn11/sentry/) may serve for you.
Hope this helps, Ingo
Hello again @Ingo_Richtsmeier ,
Yes, that would serve as a solution if I could automatically attach the filled in questionnaire/form linked to variables which are then put into the conditions in sentries. The down side of this is that I can’t use anymore sentries because the maximum is one for if-parts. It could work, but it’s not optimal since then I would have to work around using other variables in the case.
Yeah, I agree that the condition would be more complex when you add more than one variable.
But these problems can be tackled with decision tables, as you can call a DMN decision from a sentry. And if you use a DRD, then you could model complex decisions with simpler decision tables.
So basically I create a series of DMN tables which will filter the ‘unusable’ plan items based on entry data and add these tables in my case and to the sentries so they are onparts instead of ifparts. This could work if I could use a set profile of input data(variables) the person only has to fill in once and can then add as standard input. Am I correct in understanding you in this way?
Example: I’ve got 5 plan items which are only enabled if the entry criterion is fulfilled. This sentry is connected to a DMN table. The input for the DMN table either is “1”(yes) for complete and enabling the plan item in the case or “2”(no) and terminates/disables the DMN plan item and therefore will not show the connected plan item when users start te case. All of this happens on instantiation.
The “1” and “2” choices only have to be made once and can be automatically added and imported for every instance of the case from here on.
Or perhaps it’s possible to create a DRD with all the DMN tables so that I only have to refer to one DMN in my casefile and this will feed me back several variables to fulfill sentries, maybe that’s a more clean way of doing it? I’m sorry for the many questions/suggestions, I’m just not quite sure how to effectively handle this problem yet.
Something like this: Results from multiple DMN tables
If I use one Rule Task in my case it will then generate the multiple variables generated in the DRD and store it in my case. I can combine sentries and fill them with these variables to enable or disable certain plan items, am I right?