Extracting process hierarchies by multi-sequence alignment adaptations

Esgin E., Karagöz P.

ENTERPRISE INFORMATION SYSTEMS, vol.16, no.10-11, pp.1617-1652, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 16 Issue: 10-11
  • Publication Date: 2022
  • Doi Number: 10.1080/17517575.2021.1913239
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, INSPEC, Metadex, Civil Engineering Abstracts
  • Page Numbers: pp.1617-1652
  • Keywords: Cross-organisational business processes, dominant behaviour, process clustering, process hierarchies, process mining, multi-sequence alignment, SIMILARITY, PATTERNS
  • Middle East Technical University Affiliated: Yes


Process mining is an active research area that provides a wide range of automated process discovery, conformance checking and process enhancement solutions by extracting process information from event logs. With the emerge of new shared economy models and system architectures, monolithic perspective of process mining, i.e. a single process within a single organisation, is evolved towards cross-organisational business processes. The results can be used to form new collaboration among the organisations and share best practices, business knowledge or a common infrastructure. However, the variability across the cross-organisational process variants constitutes a challenge to deal with, especially at flexible environments. This paper demonstrates a quantitative approach to measure how similar organisations execute the same process. The approach is based on the multi-sequence alignment of dominant behaviour, which is a typical sequence of behaviour recurring at the process cases. Then process clusters are extracted in the form of process hierarchy according to the degree of process similarities. A novel feature of the proposed approach is that branching in process hierarchy reflects how organisations' behaviours differ from each other. Additionally, common fragments at the aligned process variants within the same cluster pinpoint a functional inheritance or a shared process behaviour for the corresponding organisations. The proposed approach is applied on a real-life case and the comparison with previous alignment techniques reveals the advantages of the approach.