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.

Business process management employs process models and event logs to represent the behavior of the information systems under study. Traditional case-centric notions consider the order of activities and events in isolated process instances. The emerging field of object-centric processes challenges this assumption by putting objects in the center. Object-centric process mining and modeling approaches identify the structure of co-evolving data objects that influence the behavior of an information system to provide a comprehensive view of the system behavior. Object-centricity has been investigated independently in process modeling and in process mining, which resulted in the coexistence of seemingly contradictory assumptions and definitions. As a community effort, this research manifesto relates and aligns existing terminologies, definitions, and perspectives to provide a common ground for current and future research in object-centric business process management. Based on the current state of research, we propose a conceptualization that sets process models and event logs in relation to the information system’s behavior and the execution data it generates. The conceptualization aims at aligning different terminologies and, thus, providing a basis to model and analyze behavioral characteristics. Building on this common ground, we identify open research challenges along the most relevant research areas in object-centric process management. For each research area, its current status is investigated and an outline of the most relevant research challenges is presented.

Agent system mining is a recently introduced type of process mining that takes a bottom-up approach to the data-driven analysis of socio-technical systems that execute business processes in organizations. Instead of the top-down approach used in conventional process mining that studies a system in terms of its global state evolution, agent system mining analyzes the system as if it is composed of autonomous agents, each with its local state and behavior, interacting with other agents and the environment to contribute to the emerging global behavior of the business process. Recently, Agent Miner, the first algorithm for discovering agent systems from event data generated by process-aware information systems, has been proposed. The quality of the agent systems discovered by this algorithm depends on the quality of the agent types (or agents), which are identified from the available information about agent instances in the data. In this paper, we study the suitability and benefits of using methods from the organizational mining subarea of process mining for identifying agent types. The experiments we conduct over real-world datasets confirm the usefulness of such methods for discovering simple, modular, and accurate agent systems. These conclusions are grounded in quality metrics such as the size of discovered models (simplicity), Louvain modularity and the Gini coefficient (modularity), and precision and recall (accuracy). The results confirm the benefits of using organizational mining for identifying agent types when discovering agent systems from event data, leading to the construction of models of superior quality in precision, recall, and simplicity compared to models constructed by state-of-the-art conventional process discovery algorithms.

In Process Mining (PM), “high-level” conceptual models of business processes, in the form of directly-follows graphs, Petri nets, and finite-state automata, are discovered from “low-level” event data recorded by information systems. The quality of the discovered models is usually assessed by measures that depend on assumptions made by discovery algorithms; for example, they often assume that sequences of activities recorded in the event data do not interfere. Models produced by recent discovery algorithms consider domain knowledge and relax these assumptions, making traditional PM measures less suitable for evaluating their quality. This paper proposes an ontology-aware framework, called SOLID-M, for analyzing the quality of conceptual models discovered from event data generated by systems. SOLID-M relies on domain knowledge and provides guidelines for introducing quality measures for models constructed by process discovery algorithms that go beyond the traditional PM assumptions. In addition, the paper describes an instantiation of the framework for assessing the quality of Multi-Agent System models discovered using Agent System Mining techniques, hence addressing a growing demand for data-driven analysis of business processes emerging in interactions of human and artificial intelligence agents.

Process mining leverages event logs extracted from information systems to generate insights into the business processes of organizations. These insights are enhanced by explicitly accounting for the frequency of behavior captured in stochastic process models constructed from event logs. Causal nets are an elegant declarative process modeling formalism that relies on a small number of modeling constructs, yet is expressive. In this paper, we extend this formalism to the stochastic setting, that is, to allow the extended nets to capture the likelihoods of the observed process. We also propose a stochastic causal net discovery approach using Markovian abstraction. Our approach begins with a standard causal net model generated by a control flow discovery algorithm, and then employs optimization techniques to determine optimal binding weights. These weights enable the stochastic interpretation of the model to closely approximate the Markovian abstraction of the original event log. Our technique has been implemented and made publicly available. The evaluation based on this implementation demonstrates the feasibility of the technique. Compared to baseline models, the discovered models achieve noticeable improvements in the quality of stochastic conformance.

Process mining (PM)-based goal recognition (GR) techniques, which infer goals or targets based on sequences of observed actions, have shown efficacy in real-world engineering applications. This study explores the applicability of PM-based GR in identifying target poses for users employing powered transhumeral prosthetics. These prosthetics are designed to restore missing anatomical segments below the shoulder, including the hand. In this article, we aim to apply the GR techniques to identify the intended movements of users, enabling the motors on the powered transhumeral prosthesis to execute the desired motions precisely. In this way, a powered transhumeral prosthesis can assist individuals with disabilities in completing movement tasks. PM-based GR techniques were initially designed to infer goals from sequences of observed actions, where discrete event names represent actions. However, the electromyography electrodes and kinematic sensors on powered transhumeral prosthetic devices register sequences of continuous, real-valued data measurements. Therefore, we rely on methods to transform sensor data into discrete events and integrate these methods with the PM-based GR system to develop target pose recognition approaches. Two data transformation approaches are introduced. The first approach relies on the clustering of data measurements collected before the target pose is reached (the clustering approach). The second approach uses the time series of measurements collected while the dynamic user movement to perform linear discriminant analysis (LDA) classification and identify discrete events (the dynamic LDA approach). These methods are evaluated through offline and human-in-the-loop (online) experiments and compared with established techniques, such as static LDA, an LDA classification based on data collected at static target poses, and GR approaches based on neural networks. Real-time human-in-the-loop experiments further validate the effectiveness of the proposed methods, demonstrating that PM-based GR using the dynamic LDA classifier achieves superior F 1 score and balanced accuracy compared to state-of-the-art techniques.

Abstract Process mining aids organisations in improving their operational processes by providing visualisations and algorithms that turn event data into insights. How often behaviour occurs in a process—the stochastic perspective—is important for simulation, recommendation, enhancement and other types of analysis. Although the stochastic perspective is important, the focus is often on control flow. Stochastic conformance checking techniques assess the quality of stochastic process models and/or event logs with one another. In this paper, we address three limitations of existing stochastic conformance checking techniques: inability to handle uncertain event data (e.g. events having only a date), exponential blow-up in computation time due to the analysis of all interleavings of concurrent behaviour and the problem that loops that can be unfolded infinitely often. To address these challenges, we provide bounds for conformance measures and use partial orders to encode behaviour. An open-source implementation is provided, which we use to illustrate and evaluate the practical feasibility of the approach.

Process discovery involves the construction of process models to describe real-world systems, allowing study and improvement of systems based on their data footprints. One quality criterion of discovered models is model-system generalization, which assesses how well a model describes both seen and unseen processes of the system. When the system itself is unknown, event logs must be used to determine model-system relationships, such as generalization. Here we investigate event log representativeness, which measures how well an event log represents its generative system, exploring the extent to which representativeness affects the accuracy of generalization estimation. Our focus is on a bootstrap approach, adopting a simple approximation for log representativeness that correlates strongly with previous measures. Extensive experiments show that log representativeness substantially affects generalization estimation accuracy: highly representative logs can directly represent the system for measuring model generalization, while less representative logs require additional estimations. We also provide insights into bootstrap generalization estimation: reasonable assumptions on the process discovery technique allow the bootstrap method to yield more accurate estimates of generalization for model-system precision than for model-system recall.

A bioprocess is a series of biological, chemical, and physical operations used to produce a product using living cells or their components. Bioprocesses are often used for the production of monoclonal antibodies (mAbs). The first step of the mAb production bioprocess is to take a vial containing a small amount of the selected cell line and grow those cells until they are of sufficient quantity. This step is known as the seed train in bioprocess development. During the seed train phase, it is essential to monitor the stability of the cells and their growth due to challenges such as variations in cell behaviour, batch-to-batch differences, and potential changes in cultivation conditions. In this paper, we present a case study where process mining is used to analyse the stability of cell lines during the seed train phase at a large pharmaceutical company in Australia. In order to do so, first it was necessary to transform the collected seed train data into an event log. Next, process models were discovered for high- and low-growth seed trains. We then derived insights into the performance of the seed train growth rate whereby characteristics of cell cultures in early stages can be associated with growth rate performance in later stages. Finally, we showed how the discovered models can be used to predict the growth performance of new seed trains.

Organizations can benefit from the use of practices, techniques, and tools from the area of business process management. Through the focus on processes, they create process models that require management, including support for versioning, refactoring and querying. Querying thus far has primarily focused on structural properties of models rather than on exploiting behavioral properties capturing aspects of model execution. While the latter is more challenging, it is also more effective, especially when models are used for auditing or process automation. The focus of this paper is to overcome the challenges associated with behavioral querying of process models in order to unlock its benefits. The first challenge concerns determining decidability of the building blocks of the query language, which are the possible behavioral relations between process tasks. The second challenge concerns achieving acceptable performance of query evaluation. The evaluation of a query may require expensive checks in all process models, of which there may be thousands. In light of these challenges, this paper proposes a special-purpose programming language, namely Process Query Language (PQL) for behavioral querying of process model collections. The language relies on a set of behavioral predicates between process tasks, whose usefulness has been empirically evaluated with a pool of process model stakeholders. This study resulted in a selection of the predicates to be implemented in PQL, whose decidability has also been formally proven. The computational performance of the language has been extensively evaluated through a set of experiments against two large process model collections.

Abstract Increasing the success rate of a process, i.e. the percentage of cases that end in a positive outcome, is a recurrent process improvement goal. At runtime, there are often certain actions (a.k.a. treatments) that workers may execute to lift the probability that a case ends in a positive outcome. For example, in a loan origination process, a possible treatment is to issue multiple loan offers to increase the probability that the customer takes a loan. Each treatment has a cost. Thus, when defining policies for prescribing treatments to cases, managers need to consider the net gain of the treatments. Also, the effect of a treatment varies over time: treating a case earlier may be more effective than later in a case. This paper presents a prescriptive monitoring method that automates this decision-making task. The method combines causal inference and reinforcement learning to learn treatment policies that maximize the net gain. The method leverages a conformal prediction technique to speed up the convergence of the reinforcement learning mechanism by separating cases that are likely to end up in a positive or negative outcome, from uncertain cases. An evaluation on two real-life datasets shows that the proposed method outperforms a state-of-the-art baseline.

A transhumeral prosthesis restores missing anatomical segments below the shoulder, including the hand. Active prostheses utilize real-valued, continuous sensor data to recognize patient target poses, or goals, and proactively move the artificial limb. Previous studies have examined how well the data collected in stationary poses, without considering the time steps, can help discriminate the goals. In this case study paper, we focus on using time series data from surface electromyography electrodes and kinematic sensors to sequentially recognize patients' goals. Our approach involves transforming the data into discrete events and training an existing process mining-based goal recognition system. Results from data collected in a virtual reality setting with ten subjects demonstrate the effectiveness of our proposed goal recognition approach, which achieves significantly better precision and recall than the state-of-the-art machine learning techniques and is less confident when wrong, which is beneficial when approximating smoother movements of prostheses.

Process analytics is a collection of data-driven techniques for, among others, making predictions for individual process instances or overall process models. At the instance level, various novel techniques have been recently devised, tackling analytical tasks such as next activity, remaining time, or outcome prediction. However, there is a notable void regarding predictions at the process model level. It is the ambition of this article to fill this gap. More specifically, we develop a technique to forecast the entire process model from historical event data. A forecasted model is a will-be process model representing a probable description of the overall process for a given period in the future. Such a forecast helps, for instance, to anticipate and prepare for the consequences of upcoming process drifts and emerging bottlenecks. Our technique builds on a representation of event data as multiple time series, each capturing the evolution of a behavioural aspect of the process model, such that corresponding time series forecasting techniques can be applied. Our implementation demonstrates the feasibility of process model forecasting using real-world event data. A user study using our Process Change Exploration tool confirms the usefulness and ease of use of the produced process model forecasts.

A process discovery algorithm aims to construct a process model that represents the real-world process stored in event data well; it is precise, generalizes the data correctly, and is simple. At the same time, it is reasonable to expect that better quality input event data should lead to constructed process models of better quality. However, existing process discovery algorithms omit the discussion of this relationship between the inputs and outputs and, as it turns out, often do not guarantee it. We demonstrate the latter claim using several quality measures for event data and discovered process models. Consequently, this paper requests for more rigor in the design of process discovery algorithms, including properties that relate the qualities of the inputs and outputs of these algorithms. We present four incremental maturity stages for process discovery algorithms, along with concrete guidelines for formulating relevant properties and experimental validation. We then use these stages to review several state of the art process discovery algorithms to confirm the need to reflect on how we perform algorithmic process discovery.

Through the application of process mining, organisations can improve their business processes by leveraging data recorded as a result of the performance of these processes. Over the past two decades, the field of process mining evolved considerably, offering a rich collection of analysis techniques with different objectives and characteristics. Despite the advances in this field, a solid statistical foundation is still lacking. Such a foundation would allow analysis outcomes to be found or judged using the notion of statistical significance, thus providing a more objective way to assess these outcomes. This paper contributes several statistical tests and association measures that treat process behaviour as a variable. The sensitivity of these tests to their parameters is evaluated and their applicability is illustrated through the use of real-life event logs. The presented tests and measures constitute a key contribution to a statistical foundation for process mining.

Reducing cycle time is a recurrent concern in the field of business process management. Depending on the process, various interventions may be triggered to reduce the cycle time of a case, for example, using a faster shipping service in an order-to-delivery process or calling a customer to obtain missing information rather than waiting passively. However, each of these interventions comes with a cost. This paper tackles the problem of determining if and when to trigger a time-reducing intervention in a way that maximizes a net gain function. The paper proposes a prescriptive monitoring method that uses orthogonal random forests to estimate the causal effect of triggering a time-reducing intervention for each ongoing case of a process. Based on this estimate, the method triggers interventions according to a user-defined policy. The method is evaluated on two real-life datasets.

This paper proposes an approach to analyze an event log of a business process in order to generate case-level recommendations of treatments that maximize the probability of a given outcome. Users classify the attributes in the event log into controllable and non-controllable, where the former correspond to attributes that can be altered during an execution of the process (the possible treatments). We use an action rule mining technique to identify treatments that co-occur with the outcome under some conditions. Since action rules are generated based on correlation rather than causation, we then use a causal machine learning technique, specifically uplift trees, to discover subgroups of cases for which a treatment has a high causal effect on the outcome after adjusting for confounding variables. We test the relevance of this approach using an event log of a loan application process and compare our findings with recommendations manually produced by process mining experts.

Robotic Process Automation (RPA) is a technology to develop software bots that automate repetitive sequences of interactions between users and software applications (a.k. a. routines). To take full advantage of this technology, organizations need to identify and to scope their routines. This is a challenging endeavor in large organizations, as routines are usually not concentrated in a handful of processes, but rather scattered across the process landscape. Accordingly, the identification of routines from User Interaction (UI) logs has received significant attention. Existing approaches to this problem assume that the UI log is segmented, meaning that it consists of traces of a task that is presupposed to contain one or more routines. However, a UI log usually takes the form of a single unsegmented sequence of events. This paper presents an approach to discover candidate routines from unsegmented UI logs in the presence of noise, i.e. events within or between routine instances that do not belong to any routine. The approach is implemented as an open-source tool and evaluated using synthetic and real-life UI logs.

Given an event log as a collection of recorded real-world process traces, process mining aims to automatically construct a process model that is both simple and provides a useful explanation of the traces. Conformance checking techniques are then employed to characterize and quantify commonalities and discrepancies between the log’s traces and the candidate models. Recent approaches to conformance checking acknowledge that the elements being compared are inherently stochastic – for example, some traces occur frequently and others infrequently – and seek to incorporate this knowledge in their analyses.Here we present an entropic relevance measure for stochastic conformance checking, computed as the average number of bits required to compress each of the log’s traces, based on the structure and information about relative likelihoods provided by the model. The measure penalizes traces from the event log not captured by the model and traces described by the model but absent in the event log, thus addressing both precision and recall quality criteria at the same time. We further show that entropic relevance is computable in time linear in the size of the log, and provide evaluation outcomes that demonstrate the feasibility of using the new approach in industrial settings.

Automated process discovery techniques enable users to generate business process models from event logs extracted from enterprise information systems.Traditional techniques in this field generate procedural process models (e.g., in the BPMN notation).When dealing with highly variable processes, the resulting procedural models are often too complex to be practically usable.An alternative approach is to discover declarative process models, which represent the behavior of the process as a set of constraints.Declarative process discovery techniques have been shown to produce simpler models than procedural ones, particularly for processes with high variability.However, the bulk of approaches for automated discovery of declarative process models focus on the control-flow perspective, ignoring the data perspective.This paper addresses the problem of discovering declarative process models with data conditions.Specifically, the paper tackles the problem of discovering constraints that involve two activities of the process such that each of these two activities is associated with a condition that must hold when the activity occurs.The paper presents and compares two approaches to the problem of discovering such conditions.The first approach uses clustering techniques in conjunction with a rule mining technique, while the second approach relies on redescription mining techniques.The two approaches (and their variants) are empirically compared using a combination of synthetic and real-life event logs.The experimental results show that the former approach outperforms the latter when it comes to re-discovering constraints artificially injected in a log.Also, the former approach is in most of the cases more computationally efficient.On the other hand, redescription mining discovers rules with higher confidence (and lower support) suggesting that it may be used to discover constraints that hold for smaller subsets of cases of a process.

Conformance checking is a subarea of process mining that studies relations between designed processes, also called process models, and records of observed processes, also called event logs. In the last decade, research in conformance checking has proposed a plethora of techniques for characterizing the discrepancies between process models and event logs. Often, these techniques are also applied to measure the quality of process models automatically discovered from event logs. Recently, the process mining community has initiated a discussion on the desired properties of such measures. This discussion witnesses the lack of measures with the desired properties and the lack of properties intended for measures that support partially matching processes, i.e., processes that are not identical but differ in some steps. The paper at hand addresses these limitations. Firstly, it extends the recently introduced precision and recall conformance measures between process models and event logs that possess the desired property of monotonicity with the support of partially matching processes. Secondly, it introduces new intuitively desired properties of conformance measures that support partially matching processes and shows that our measures indeed possess them. The new measures have been implemented in a publicly available tool. The reported qualitative and quantitative evaluations based on our implementation demonstrate the feasibility of using the proposed measures in industrial settings.