Zekeng Ndadji surveys 34 studies on formal grammars in BPM, identifies 7 research streams

This paper from arXiv (cs.SE, June 2026) presents a comprehensive systematic literature review of 34 studies that apply formal grammars — mathematical structures for defining language syntax — to Business Process Management. The author, Milliam Maxime Zekeng Ndadji, categorizes the existing work into seven distinct research streams: (i) process grammars that help design organizational processes; (ii) evaluations of process modeling languages using the Bunge-Wand-Wever ontological framework; (iii) production-rule grammars for specifying process structures and managing variants; (iv) attribute grammars that enable declarative specification and distributed execution of workflows; (v) graph grammars for transforming, generating, and semantically analyzing process models; (vi) grammatical inference techniques for mining processes from event logs; and (vii) process algebras as grammar-like frameworks for compositional behavioral specification and verification.

For each stream, the review synthesizes key contributions, the formalisms used, and their limitations. A critical finding is that these seven streams have evolved largely in isolation, without cross-pollination of ideas. The author identifies five corpus-grounded open challenges — ranging from integrating grammatical inference with graph grammars to unifying verification and design — and argues that a deeper, unified exploitation of grammatical theory could significantly advance BPM, especially in automating lifecycle phases like design, execution, monitoring, and improvement. For practitioners, this means formal grammars offer a rigorous, math-backed approach to process automation, but current fragmented efforts limit practical adoption. The paper serves as a roadmap for researchers and engineers looking to bridge these communities.