To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.

Problem with a petri net created in WoPeD for conformance checking

Hi, I am trying to do conformance checking in ProM 6.8 using the multi-perspective process explorer (MPE) plugin.

I have created a sound petri net model using WoPeD. I have imported the .pnml file into ProM and unpacked the accepting petri net. When I replay my event log data over my model the average fitness within the MPE is 3.6%. This is because all the un-named transitions (that should be silent) have 0% fitness. 

This is a problem with the petri net model created in WoPeD, as previously I created a petri net model using a csv file. This was difficult as it had errors that I needed to correct by first creating a process tree, then editing it. I converted the process tree to a petri net. I then edited the petri net file to redirect two of the arcs. This was not a neat solution, though the resulting petri net was correct and I was able to use it for conformance checking. When replaying the data I got 98.5% average fitness.

I could just use the model created from the csv file, though it is for my PhD and I would like a neater solution using WoPeD.

If anybody could help it would be appreciated.
Thank you 
Sam

Answers

  • Hi Sam,

    Based on what you mention, chances are that the un-named transitions are not considered to be invisible by ProM. If you have imported the net, you can check (and fix) this using the Configure Visibility of Transitions plugin by Arya Adriansyah.

    Kind regards,
    Eric.
  • Thank you yet again Eric! This has worked perfectly 98.5% average fitness :)

    Kind regards

    Sam

Sign In or Register to comment.