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.

Issues with PlugIn Replay a Log on Petri Net for Performance / Conformance Analysis

Dear ProM Community,

I am working on my Master Thesis at the TUE. I am conducting a performance analysis on an Event Log from a software developing company. The event log contains a lot of noise, which is the reason why the general petri net, that I have created has three process steps in parallel. To run the performance analysis I use the PlugIn 'Replay a Log on Petri Net for Performance / Conformance Analysis'.

Due to the parallelism in the model, the Waiting times for the process steps Estimation, Code Review and Test Rework differ from my manual calculation that I took on a small sample. My guess is that in the plugin in ProM, the Waiting Times are the time difference between the end of the previous step and the start of the current step. However, in the given model, the previous step for, e.g. the third step "estimation", is not the second step "Technical Design" but the first step "Functional Design". Therefore, the waiting time for the "estimation" include the sojourn time of the previous step, "Technical Design" which is not correct.

I have also created a 'To-Be' petri net, which has the sequence that is described by the case company. However, it has a very low fitness for the event log and I cant use, because in most cases, the sequence is different. E.G. the third process step, takes place before the first. In this case, the values for those steps can't be calculated, see attachement to_be_performance analysis.

Does anyone has experience with this issue?

Best regards

Malte Fritsch




Comments

  • Dear Malte,

    As far as I know, this is work done by Arya Adriansyah, see https://research.tue.nl/en/publications/6eac4d54-1d66-4161-80c3-6e95696ea87f for his PhD thesis on this topic. On pages 202-203 there is a short description of how these times are calculated. As far as I can tell, the waiting time for Estimation is supposed to be the time between the completion of Functional Design and the start of Estimation.

    Kind regards,

    Eric.

  • Dear Eric,

    thank you very much! I also use the inductive visual miner for the performance / Waiting time analysis. Do you know other Plug-Ins in ProM or tools that I can use for this?


    Best regards

    Malte

  • hverbeek
    edited February 10

    Dear Malte,

    The only one I know by heart is the Transition System Analyzer plugin. This plugin computes a duration for transitions and a sojourn time, an elapsed time and a remaining time for states. Of course, you first will have to mine a transition system from the event log using, for example, the Mine Transition System plugin, and then provide both the event log and the mined transition system as inputs to the analyzer.

    Kind regards,

    Eric.

Sign In or Register to comment.