On the Hybrid Nature of ABPMS Process Frames and its Implications on Automated Process Discovery Journal: Information Systems

Anti Alman, Izack Cohen, Avigdor Gal, Fabrizio Maria Maggi, Marco Montali · 2026

Read On the Hybrid Nature of ABPMS Process Frames and its Implications on Automated Process Discovery Journal: Information Systems by Anti Alman, Izack Cohen, Avigdor Gal, Fabrizio Maria Maggi, Marco Montali (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-22455.txt.

Introduction

Alman, Cohen, Gal, Maggi and Montali present the ABPMS frame as a hybrid form. The frame sets outer limits for the system, more permissive than old models, enabling framed autonomy. The authors model it as procedural and declarative models side by side, extending the open-world assumption across types. Non-clashing models can combine for execution, but discovery from event data is hard for models relying on directly-follows relations. They analyse how one type's behaviours appear as Declare constraints matching eventually-follows relations, revealing overlaps for future discovery.

Catalog record

CREATORAnti Alman, Izack Cohen, Avigdor Gal, Fabrizio Maria Maggi, Marco Montali
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE76832
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.22455

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On the Hybrid Nature of ABPMS Process Frames and its Implications on Automated Process Discovery Journal: Information Systems Anti Alman, Izack Cohen, Avigdor Gal, Fabrizio Maria Maggi, Marco Montali Abstract A core component of any AI-Augmented Business Process Management System (ABPMS) is the process frame , which gives the system process-awareness and defines the boundaries in which the system must operate. Compared to traditional process models, the process frame should, in principle, provide a somewhat more permissive representation of the managed processes, such that the (semi) autonomous behavior of an ABPMS, referred to as framed autonomy , could emerge. At the same time, it is not limited to a single linguistic or symbolic formalism and may incorporate heterogeneous knowledge ranging from predefined procedures to commonsense rules and best practices. In this paper, we conceptualize the notion of an ABPMS process frame as a hybrid business process representation, consisting of semi-concurrently executed procedural and declarative process models. We rely on our earlier works to outline the execution semantics of this type of process frame, arguing in favor of adopting the open-world assumption of the declarative paradigm also for procedural process models. The latter leads to a constraint-like interpretation, where each procedural model is considered to constrain the activities within that model, without imposing explicit execution requirements nor limitations on activities that may be present in other models. This is analogous to existing declarative languages, such as Declare , where each constraint has a direct effect only on the specific activities being constrained. Given this similarity, we propose mapping subsets of discovered declarative constraints into equivalent semi-concurrently executed procedural fragments, thus laying the foundation for a corresponding process (frame) discovery approach. 1 Introduction The research manifesto on AI-Augmented Business Process Management Systems (ABPMS) outlines a vision in which the traditional BPMS lifecycle is augmented with AI technologies, enabling the system to “reason about the current state of the process (or across several processes) to determine a course of action that improves the performance of the process” [ 34 ] . A core component of any ABPMS is the process frame , which provides the system with process awareness and defines the boundaries in which the system must operate to achieve its goals. The latter represents a fundamental shift from a system that supports the execution of business processes, potentially through automating some tasks, towards a system that exhibits a degree of autonomy over how the processes are executed, which decisions are taken therein, and what optimizations are implemented over time, with the end-goal of uplifting business process management systems to a level of autonomous decision-making comparable to human actors. In this paper, we propose an approach for representing a business process (or a set of processes) within the process frame, as this was left open in [ 34 ] . We take inspiration from hybrid business process representations [ 9 ] and the recent call for holistic business process management [ 11 ] , combining both with our earlier work on the enactment and monitoring of interacting business processes [ 6 , 5 , 2 , 80 ] and on developing the multi-model paradigm [ 8 , 51 ] . In particular, we conceptualize the notion of an ABPMS process frame as a hybrid business process representation, consisting of semi-concurrently executed procedural and declarative models, moving beyond a single modeling formalism and supporting the incorporation of heterogeneous knowledge. In this context, we argue for adopting the open-world assumption of the declarative modeling paradigm also for procedural models, such that, each model of the process frame, declarative or procedural, would impose explicit execution requirements and/or limitations only on the activities within that model. As in our earlier works, we rely on Petri nets [ 58 , 77 ] and Declare [ 62 ] , however, with a focus on their automated discovery in the context of the our conceptualization of the process frame, and without advance knowledge of how many models there, what behaviors they represent, nor which activities belong to which models. In that direction, we highlight how adopting the open-world assumption for procedural process models, expressed as Petri nets, complicates their discovery by causing directly-follows relations to become “unreliable”, and we explore the feasibility of mapping specific subsets of discovered Declare constraints, predominantly representing various eventually-follows relations, into behaviorally equivalent Petri nets under the open-world assumption. This paper builds on our previous workshop paper on the discovery of an ABPMS process frame in a multi-process setting [ 3 ] . On the one hand, we deepen the positioning, motivation, and the underlying rationale for pursuing automated process discovery on the basis of eventually-follows relations. On the other hand, we demosntrate the manifestation of common procedural behaviors through declarative modeling constructs, including in the presence of activity repetitions, which were not considered in the workshop paper. In addition, this paper coins the term “pockets of rigidity”, which is effectively a reversal of the well-known “pockets of flexibility” approach [ 63 ] , such that, an overall declarative process model is simplified by selectively using procedural constructs for the more rigid sub-behaviors within it. The rest of this paper is structured as follows. Section 2 provides our conceptualization of the process frame, including the corresponding formalization, and outlines the resulting implications on process discovery. Section 3 explores the feasibility of using discovered Declare process models, for the discovery of common procedural behaviors, which would form the procedural components of the ABPMS process frame, while also simplifying the Declare model. Finally, Sections 4 and 5 conclude the paper by discussing the related works and the future research directions, respectively. 2 The ABPMS Process Frame Figure 1: ABPMS lifecycle (taken from [ 34 ] ). Rounded rectangles represent lifecycle steps, while arrows denote sequencing. The operational lifecycle of an ABPMS (cf. Fig. 1 ) involves two actors: the ABPMS itself and one or more (not necessarily human) agents interacting with the ABPMS. Classification of the ABPMS as one of the actors is a fundamental difference from traditional business process management systems, reflecting that an ABPMS should have some level of autonomy over how the process (or processes) are executed. The exact level of autonomy can vary, but it should be absolute. Instead, the boundaries that would apply to any human actor, in terms of process executions and modifications, should also apply to the autonomy of the ABPMS. This may include, for example, legal requirements (e.g., obtaining a permit before starting construction), internal operating procedures (e.g., mandatory review of some decisions), organizational strategy (e.g., maintaining production costs below some threshold), and other similar knowledge that is relevant for the process execution. The question of how to express such knowledge, as a process frame, is left open in [ 34 ] , with procedural and declarative process modeling languages, including their hybrid combinations, being suggested as possible options. Reusing existing process modeling languages, especially ones with precise execution semantics, is indeed a natural starting point, given the importance of process-aware execution in the ABPMS vision (cf. Fig. 1 ). However, this entails a conceptual shift from explicit modeling of complete process executions to defining the behavioral boundaries of an ABPMS based on heterogeneous sources of process knowledge,

Read the complete text