Chain of Events: Modular Process Models for the Law

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

In this paper, we take technical and practical steps towards the modularisation of compliant-by-design executable declarative process models. First, we demonstrate by example how the specific language of timed DCR graphs is capable of modelling complex legislation, with examples from laws regulating the functioning of local governments in Denmark. We then identify examples of law paragraphs that are beyond these modelling capabilities. This incompatibility arises from subtle and—from a computer science perspective—non-standard interactions between distinct paragraphs of the law, which must then become similar interactions between model fragments. To encompass these situations, we propose a notion of networks of processes, where the processes are allowed to interact and regulate their interaction through the novel mechanisms of exclusion and linking. Networks are parametric in the underlying process formalism, allowing interactions between processes specified in arbitrary and possibly distinct trace-language semantics formalisms as the individual models. Technically, we provide a sufficient condition for a good class of network compositions to realise refinement of the constituent processes. Finally, parts of the theoretical framework (networks and exclusion) have been implemented by our industry partners, and we report on a preliminary evaluation suggesting that inter-model synchronisation is indeed both necessary and helpful in practical modelling scenarios.
TitelIntegrated Formal Methods: 16th International Conference, IFM 2020, Lugano, Switzerland, November 16–20, 2020, Proceedings
ISBN (Trykt)978-3-030-63460-5
ISBN (Elektronisk)978-3-030-63461-2
StatusUdgivet - 2020
Begivenhed16th International Conference on Integrated Formal Methods - IFM 2020 - Lugano, Schweiz
Varighed: 16 nov. 202020 nov. 2020


Konference16th International Conference on Integrated Formal Methods - IFM 2020
NavnLecture Notes in Computer Science

ID: 252766228