In an increasingly digital world, collecting, processing, and exchanging personal data are critical drivers for enacting enterprise business processes. However, the long-term retention and access of personal data expose organizations to data breaches, in which sensitive and protected data are disclosed and exploited unauthorizedly. To mitigate the damage that data breaches can cause, in the European Union (EU), the right to data privacy is enforced through the General Data Protection Regulation (GDPR), which defines how organizations must store and manage EU citizens’ data. GDPR is highly influencing how organizations approach data privacy, forcing them to rethink and upgrade their business processes to become GDPR compliant, which can be daunting. In this paper, in line with the privacy-by-design principles of GDPR, we propose a methodology that shows how to capture the main privacy GDPR constraints in the form of design patterns and integrate them into business process models specified in BPMN (Business Process Model and Notation). This allows us to achieve full transparency of privacy constraints in business processes, making it possible to ensure their compliance with GDPR at design-time. We adopt a design science research approach to present our methodology and make design decisions explicit. We also introduce GDPR-Pilot, a BPMN editor that assists process designers and Data Controllers in integrating GDPR patterns into existing models. The methodology is evaluated through real-world use cases against structural, usage, and environmental requirements.

Organizations increasingly rely on Process Performance Indicators (PPIs) to evaluate how effectively their business processes achieve strategic and operational goals. Recent approaches compute PPIs directly from event logs leveraging the increasing availability of event data in organizations, but they assume that PPIs have been previously defined using well-established reference models and that their linkage to the event log has been explicitly specified. However, this definition of appropriate PPIs remains challenging, as managers often lack detailed insights into real process executions and must adapt generic frameworks to organizational contexts. To address this gap, this paper introduces PPIPilot, an approach and tool that supports managers in defining measurable PPIs by suggesting a set of relevant PPIs and their computations directly from a business process description, an associated event log, and a specified organizational goal. PPIPilot leverages large language models for contextual PPI suggestion and structured translation, operating through a three-component pipeline of suggestion, translation, and computation, with a fallback mechanism to address computation errors. The quantitative evaluation shows that PPIPilot has a strong ability to suggest and compute PPIs, and that the fallback mechanism reduces failures, increasing the share of PPIs that can be computed without errors. Complementing these findings, an in-depth qualitative study indicates that users find the tool intuitive, transparent and helpful for easing both the conceptual and technical challenges of defining PPIs.

Declarative process models allow us to capture the behavior of a business process through temporal constraints on the evolution of process activities. In process mining, declarative process discovery focuses on deriving these constraints from event logs. Although the semantic aspects of declarative processes have been extensively investigated, there has been less focus on designing declarative visual notations that enhance model understanding and support analysts in solving process mining tasks. To improve the human understandability of declarative process models, in this paper, we present easyDeclare , a novel visual notation to specify declarative process models using the Declare language. easyDeclare was developed with consideration of the well-established Moody’s design principles. We conducted extensive user experiments to demonstrate that easyDeclare , when compared with the original graphical representation of Declare , reduces the cognitive load required to interpret Declare models of increasing complexity, making it a promising alternative to enhancing overall comprehension of declarative process discovery tasks.

This paper presents a manifesto that articulates the conceptual foundations of Agentic Business Process Management (APM), an extension of Business Process Management (BPM) for governing autonomous agents executing processes in organizations. From a management perspective, APM represents a paradigm shift from the traditional view on business processes. This shift is driven by the realization of process awareness by agent-oriented abstractions: software and human agents act as primary functional entities that perceive, reason, and act within explicit process frames. Thus, APM moves away from automation-oriented BPM towards systems in which autonomy is constrained, aligned, and made operational through process aware agents. We introduce the core abstractions and architectural elements required to realize APM systems and elaborate on four key capabilities that agents in APM systems must support: framed autonomy , explainability , conversational actionability , and self-modification . These capabilities jointly ensure that agents’ goals are aligned with organizational goals and that agents behave in a framed yet proactive manner in pursuing those goals. We discuss the extent to which the capabilities can be realized and identify research challenges whose resolution requires further advances in BPM, AI, and multi-agent systems. The manifesto thus serves as a roadmap for bridging these communities and for guiding the development of APM systems in practice.

The Business Process Modeling Notation (BPMN) is the de facto standard for business process modeling. While widely adopted for its intuitive graphical notation, its execution semantics described in natural language lacks a commonly agreed formal foundation, leading to variability in execution across different BPM systems (BPMSs) and increasing the risk of creating models with semantic errors costly to correct at runtime. Although many formalisms have been used to model portions of BPMN, their reasoning capabilities are mostly restricted to control-flow, making them unsuitable for semantic analysis where data and global exception handling play a central role in execution. To address this, we propose a formalization from BPMN to ConGolog, a logical concurrent processes language based on the Situation Calculus, for representing and reasoning about dynamic domains. A major innovation is using ConGolog to rigorously capture the semantics of BPMN global exceptions. Our framework supports advanced reasoning, allowing for semantic analysis of BPMN models before execution to predict runtime errors within a safe simulation setting, while laying the foundation for reasoning layers in next-generation AI-augmented BPMSs. We validate the approach through a prototype and comprehensive evaluation, demonstrating the computational feasibility of the translation and the semantic correctness of reasoning tasks.

Next activity prediction is one of the main tasks of Predictive Process Monitoring (PPM), enabling organizations to forecast the execution of business processes and respond accordingly. Deep learning models are effective at predictions, but with the price of intensive training and feature engineering, rendering them less generalizable across domains. Large Language Models (LLMs) have been recently suggested as an alternative, but their capabilities in Process Mining tasks are still to be extensively investigated. This work introduces a framework leveraging LLMs and Retrieval-Augmented Generation to enhance their capabilities for predicting next activities. By leveraging sequential information and data attributes from past execution traces, our framework enables LLMs to make more accurate predictions without additional training. We evaluate the approach on a wide range of event logs and compare it with state-of-the-art techniques. Findings show that our framework achieves competitive performance while being more adaptable across domains. Moreover, we assess early prediction capabilities, validate the significance of observed differences through statistical testing, and explore the impact of fine-tuning. Despite these advantages, we also report the framework’s limitations, mainly related to interleaving activity sensitivity and concept drifts. Our findings highlight the potential of retrieval-augmented LLMs in PPM while identifying the need for future research into handling evolving process behaviors and the development of standard benchmarks.

AI-augmented Business Process Management Systems (ABPMSs) are an emerging class of process-aware information systems empowered by AI technology for autonomously unfolding and adapting the execution flow of business processes (BPs) within a set of potentially conflicting procedural and declarative constraints, called process framing . In this respect, framed autonomy enables an ABPMS to autonomously decide how to progress the execution of a BP, as long as the boundaries imposed by the frame are respected. Among these constraints, there could be a partial BP execution that needs to be completed, activating a different near-optimal framing that enables the BP to progress its execution. In this paper, we present an automata-based technique that pairs constraint-based framing with automated planning in AI to recommend, given a partial BP execution trace, the continuation of that trace that minimizes the violation cost of the conforming space defined by the process frame. We report on the results of experiments of increasing complexity to showcase our technique’s performance and scalability.

Abstract Traditionally, process-aware Decision Support Systems (DSSs) have been enhanced with AI functionalities to facilitate quick and informed decision-making. In this context, AI-Augmented Business Process Management Systems have emerged as innovative human-centric information systems, blending flexibility, autonomy, and conversational capability. Large Language Models (LLMs) have significantly boosted such systems, showcasing remarkable natural language processing capabilities across various tasks. Despite the potential of LLMs to support human decisions in business contexts, empirical validations of their effectiveness for process-aware decision support are scarce in the literature. In this paper, we propose the Business Process Large Language Model (BPLLM) framework, a novel approach for enacting actionable conversations with human workers. BPLLM couples Retrieval-Augmented Generation with fine-tuning, to enrich process-specific knowledge. Additionally, a process-aware chunking approach is incorporated to enhance the BPLLM pipeline. We evaluated the approach in various experimental scenarios to assess its ability to generate accurate and contextually relevant answers to users’ questions. The empirical study shows the promising performance of the framework in identifying the presence of particular activities and sequence flows within the considered process model, offering insights into its potential for enhancing process-aware DSSs.

Within the process mining field, Deterministic Finite State Automata (DFAs) are largely employed as foundation mechanisms to perform formal reasoning tasks over the information contained in the event logs, such as conformance checking, compliance monitoring and cross-organization process analysis, just to name a few. To support the above use cases, in this paper, we investigate how to leverage Model Learning (ML) algorithms for the automated discovery of DFAs from event logs. DFAs can be used as a fundamental building block to support not only the development of process analysis techniques, but also the implementation of instruments to support other phases of the Business Process Management (BPM) lifecycle such as business process design and enactment. The quality of the discovered DFAs is assessed wrt customized definitions of fitness, precision, generalization, and a standard notion of DFA simplicity. Finally, we use these metrics to benchmark ML algorithms against real-life and synthetically generated datasets, with the aim of studying their performance and investigate their suitability to be used for the development of BPM tools.

Process Mining is a family of techniques for analyzing business process execution data recorded in event logs. Process models can be obtained as output of automated process discovery techniques or can be used as input of techniques for conformance checking or model enhancement. In Declarative Process Mining, process models are represented as sets of temporal constraints (instead of procedural descriptions where all control-flow details are explicitly modeled). An open research direction in Declarative Process Mining is whether multi-perspective specifications can be supported, i.e., specifications that not only describe the process behavior from the control-flow point of view, but also from other perspectives like data or time. In this article, we address this question by considering SAT (Propositional Satisfiability Problem) as a solving technology for a number of classical problems in Declarative Process Mining, namely, log generation, conformance checking, and temporal query checking. To do so, we first express each problem as a suitable FO (First-Order) theory whose bounded models represent solutions to the problem, and then find a bounded model of such theory by compilation into SAT.

Trace alignment is the problem of finding the best possible execution sequence of a business process (BP) model that reproduces an (observed) execution trace of the same BP by pinpointing where it deviates. One limiting assumption that governs the state-of-the-art alignment algorithms relies in a static cost function assigning fixed costs to all the possible types of deviations related to a BP activity, thus neglecting the specific context in which the deviation takes place and flattening the analysis of its potential impact. In this paper, we relax this assumption by providing a technique based on theoretic manipulations of deterministic finite state automata (DFAs) to build optimal alignments driven by dedicated cost models that assign context-dependent variable costs to the deviations. We show how the algorithm can be implemented relying on automated planning in Artificial Intelligence (AI), which is proven to be an effective tool to address the alignment task in the case of BP models and event logs of remarkable size. Finally, we report on the results of experiments conducted in a real-life case study on incident management and on larger synthetic ones performed through three well-known planning systems to showcase the performance, scalability and versatility of our technique.

Declarative business process (BP) models define the behavior of BPs as a set of temporal constraints, which can be summarized as a deterministic finite state automaton (DFA). Declarative BP discovery aims at inferring such constraints from event logs. To this aim, it requires as additional input the set of candidate constraints to be verified with respect to the event log. Intuitively, this restricts the discovery task to a conformance checking activity between a predefined set of constraint templates and an event log, preventing to learn any observed behavior that is not captured by those templates. In this paper, we investigate how to leverage Model Learning (ML) for the automated discovery of the DFA underlying the behavior of a declarative BP model, without using any further a-priori information in addition to the event log. To assess the quality of the discovered DFA, we introduce a novel definition of the standard process mining quality metrics, i.e., precision, generalization and simplicity, tailored to DFAs. Finally, a preliminary evaluation performed with real-life logs shows that ML enables to generate extremely simpler DFAs than state-of-the-art BP declarative discovery techniques, keeping similar values of precision and generalization.

Process mining allows analysts to exploit logs of historical executions of business processes to extract insights regarding the actual performance of these processes. One of the most widely studied process mining operations is automated process discovery. An automated process discovery method takes as input an event log, and produces as output a business process model that captures the control-flow relations between tasks that are observed in or implied by the event log. Various automated process discovery methods have been proposed in the past two decades, striking different tradeoffs between scalability, accuracy, and complexity of the resulting models. However, these methods have been evaluated in an ad-hoc manner, employing different datasets, experimental setups, evaluation measures, and baselines, often leading to incomparable conclusions and sometimes unreproducible results due to the use of closed datasets. This article provides a systematic review and comparative evaluation of automated process discovery methods, using an open-source benchmark and covering 12 publicly-available real-life event logs, 12 proprietary real-life event logs, and nine quality metrics. The results highlight gaps and unexplored tradeoffs in the field, including the lack of scalability of some methods and a strong divergence in their performance with respect to the different quality metrics used.

The increasing application of process-oriented approaches in new challenging dynamic domains beyond business computing (e.g., healthcare, emergency management, factories of the future, home automation, etc.) has led to reconsider the level of flexibility and support required to manage complex knowledge-intensive processes in such domains. A knowledge-intensive process is influenced by user decision making and coupled with contextual data and knowledge production, and involves performing complex tasks in the “physical” real world to achieve a common goal. The physical world, however, is not entirely predictable, and knowledge-intensive processes must be robust to unexpected conditions and adaptable to unanticipated exceptions, recognizing that in real-world environments it is not adequate to assume that all possible recovery activities can be predefined for dealing with the exceptions that can ensue. To tackle this issue, in this paper we present SmartPM, a model and a prototype Process Management System featuring a set of techniques providing support for automated adaptation of knowledge-intensive processes at runtime. Such techniques are able to automatically adapt process instances when unanticipated exceptions occur, without explicitly defining policies to recover from exceptions and without the intervention of domain experts at runtime, aiming at reducing error-prone and costly manual ad-hoc changes, and thus at relieving users from complex adaptations tasks. To accomplish this, we make use of well-established techniques and frameworks from Artificial Intelligence, such as situation calculus, IndiGolog and classical planning. The approach, which is backed by a formal model, has been implemented and validated with a case study based on real knowledge-intensive processes coming from an emergency management domain.