Beyond Control-Flow: Integrating the Resource Perspective into Multi-Collaborative Process Modeling from Text

Researchers have introduced a resource-aware generation pipeline that produces formal BPMN 2.0 collaboration diagrams from natural-language descriptions, moving beyond the control-flow focus of current text-to-model systems [1]. The approach, detailed in a paper submitted on 23 May 2026, addresses a gap in automated process modeling. While Large Language Models can now translate text into process models, existing methods concentrate on ordering activities without capturing the collaborative dimension of business operations [1]. Business Process Management, or BPM, is a discipline centered on translating process artifacts into formal models, a task that has traditionally demanded substantial human expertise in both modeling notations and the specific business context [2]. The new pipeline produces BPMN 2.0 collaboration diagrams by employing a compact, executable intermediate language that mandates resource details. These details define both the organization, represented as a pool, and the role, represented as a lane [1]. A business model describes how an organization creates, delivers, and captures value, and its construction can involve formal descriptions of organizational structures, operational processes, and infrastructure [3]. The pipeline materializes cross-organization dependencies using message events, the standard formal notation for such interactions, while an orthogonal layout routine automatically handles the spatial arrangement of elements within pools and lanes [1]. Experiments conducted on ten business processes using nine different LLMs demonstrated strong resource discovery. The tests showed that control-flow quality was preserved, and the method added only marginal runtime overhead [1]. The discipline of managing technology resources within a firm, known as information technology management, involves functions such as budgeting, staffing, and organizing, and extends to the strategic oversight of technology in alignment with organizational objectives [4]. This new generative modeling approach aims to provide a more comprehensive, multi-collaborative representation of business operations [1].