gap-aware-alignments

Zenodo (CERN European Organization for Nuclear Research)

Abstract

Gap-Aware Alignments is a Python implementation of a conformance-checking technique for diagnosing process deviations. Instead of penalising every deviating step individually, it groups consecutive deviations into gaps and scores them with dedicated cost functions. The core library computes cost-optimal alignments via shortest-path search on a Petri net's reachability graph (using r4pm and NetworkX), with an evaluation framework that benchmarks all cost models across event logs.The input event logs and Petri-net models are standard, publicly available datasets and are not part of this repository. Running the evaluation reproduces the results reported in the associated paper.

Authors 1

1. Christopher T. Schwanen Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University