Process Mining Approach to Formal Business Process Modelling and Verification: A Case Study

@article{44043,
   author = {Ito, Sohei and Vymětal, Dominik and Šperka, Roman},
   article_number = {2},
   doi = {http://dx.doi.org/10.1108/JM2-03-2020-0077},
   keywords = {Business process modelling; Computing; Formal method; Modelling; Process mining; Timed automaton},
   language = {eng},
   issn = {1746-5664},
   journal = {Journal of Modelling in Management},
   title = {Process Mining Approach to Formal Business Process Modelling and Verification: A Case Study},
   url = {https://www.emerald.com/insight/content/doi/10.1108/JM2-03-2020-0077/full/html},
   volume = {16},
   year = {2020}
}

TY  - JOUR
ID  - 44043
AU  - Ito, Sohei - Vymětal, Dominik - Šperka, Roman
PY  - 2020
TI  - Process Mining Approach to Formal Business Process Modelling and Verification: A Case Study
JF  - Journal of Modelling in Management
VL  - 16
IS  - 2
SP  - 1-21
EP  - 1-21
PB  - Emerald Group Publishing Ltd.
SN  - 17465664
KW  - Business process modelling
KW  - Computing
KW  - Formal method
KW  - Modelling
KW  - Process mining
KW  - Timed automaton
UR  - https://www.emerald.com/insight/content/doi/10.1108/JM2-03-2020-0077/full/html
L2  - https://www.emerald.com/insight/content/doi/10.1108/JM2-03-2020-0077/full/html
N2  - Purpose: The need for assuring correctness of business processes in enterprises is widely recognised in terms of business process re-engineering and improvement. Formal methods are a promising approach to this issue. The challenge in business process verification is to create a formal model that is well-aligned to the reality. Process mining is a well-known technique to discover a model of a process based on facts. However, no studies exist that apply it to formal verification. This study aims to propose a methodology for formal business process verification by means of process mining, and attempts to clarify the challenges and necessary technologies in this approach using a case study. Design/methodology/approach: A trading company simulation model is used as a case study. A workflow model is discovered from an event log produced by a simulation tool and manually complemented to a formal model. Correctness requirements of both domain-dependent and domain-independent types of the model are checked by means of model-checking. Findings: For business process verification with both domain-dependent and domain-independent correctness requirements, more advanced process mining techniques that discover data-related aspects of processes are desirable. The choice of a formal modelling language is also crucial. It depends on the correctness requirements and the characteristics of the business process. Originality/value: Formal verification of business processes starting with creating its formal model is quite new. Furthermore, domain-dependent and domain-independent correctness properties are considered in the same framework, which is also new. This study revealed necessary technologies for this approach with process mining.
ER  - 

ITO, Sohei, Dominik VYMĚTAL a Roman ŠPERKA. Process Mining Approach to Formal Business Process Modelling and Verification: A Case Study. \textit{Journal of Modelling in Management}. Emerald Group Publishing Ltd., 2020, roč.~16, č.~2, s.~1-21. ISSN~1746-5664. Dostupné z: https://dx.doi.org/10.1108/JM2-03-2020-0077.