Efficient Optimal Alignment Between Dynamic Condition Response Graphs and Traces

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Dynamic Condition Response (DCR) Graphs is a popular declarative process modelling notation which is supported by commercial modelling tools and has seen significant industrial adoption. The problem of aligning traces with DCR Graphs, with it’s multitude of applications such as conformance checking and log repair, has surprisingly not been solved yet. In this paper we address this open gap in the research by developing an algorithm for efficiently computing the optimal alignment of a DCR Graph and a trace. We evaluate the algorithm on the PDC 2022 dataset, showing that even for large models and traces alignment problems can be solved within milliseconds, and present a case study based on test-driven modelling.

TitelBusiness Process Management - 21st International Conference, BPM 2023, Proceedings
RedaktørerChiara Di Francescomarino, Andrea Burattin, Christian Janiesch, Shazia Sadiq
Antal sider17
ISBN (Trykt)9783031416194
StatusUdgivet - 2023
BegivenhedProceedings of the 21st International Conference on Business Process Management , BPM 2023 - Utrecht, Holland
Varighed: 11 sep. 202315 sep. 2023


KonferenceProceedings of the 21st International Conference on Business Process Management , BPM 2023
NavnLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vol/bind14159 LNCS

Bibliografisk note

Funding Information:
This work is supported by Digital Research Centre Denmark (DIREC). We acknowledge Søren Debois for sharing his Isabelle/HOL DCR Graph formalization.

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

ID: 369555986