Duplicated sequence from activity to Exclusive Gateway in BPMN visualization

[jacaballero]

Hi all,

I don't know if this is a problem with the library or if it is accepted in BPMN models, but I have been working for a long time with other frameworks such as ProM, and it's the first time that I found it.

I'm importing a CSV log to show its BPMN model. When I apply a significant noise_threshold (i.e. 0.20), a duplicated sequence with two exclusive gateways is shown as appears in the image below:

The function where I create and save the BPMN is very simple, it is included below:

# BPMN: Discover and save model
def get_bpmn(filtered_log, file_name):
    noise_threshold = 0.0
    noise_threshold = float(input("Insert the ratio (0-100) to filter infrequent paths: "))
    bpmn_model = pm4py.discover_bpmn_inductive(filtered_log, noise_threshold/100)
    pm4py.save_vis_bpmn(bpmn_model, file_name)

When I get the BPMN model without filtering infrequent paths (or with scarce filters as 0.01), this doesn't happen.

Could you help me with this, please?

Thanks

[fit-alessandro-berti]

Dear @jacaballero

It's not technically wrong, but we'll have a look to avoid these double arcs in the conversion

[jacaballero]

Dear @fit-alessandro-berti

That's great.

Thank you for your quick response.

[fit-alessandro-berti]

By the way, could you attach a small example log where this happens?

Cheers

[jacaballero]

The log that I'm using has 240000 lines in XES format. I'll try to test if with a sample of the traces it happens as well.

[jacaballero]

Dear @fit-alessandro-berti

I attached a CSV sample of where this happens. I loaded it as below:

log = pm4py.format_dataframe(pandas.read_csv('filtered_file.csv', sep=';'), case_id='case_id',activity_key='concept:name', timestamp_key='time:timestamp')

And then applied the function that I included in a previous comment.

Hope this helps to debug it.

Regards

filtered_file.csv

[fit-alessandro-berti]

Thanks, we identified and resolved the issue :-)

[jacaballero]

I checked the last version and now it works. Thank you!!