Marlon Dumas

David Chapela-Campa, Ismail Benchekroun, Opher Baron et al.

Business Process Simulation (BPS) is an approach to analyze the performance of business processes under different scenarios. For example, BPS allows us to estimate what would be the cycle time of a process if one or more resources became unavailable. The starting point of BPS is a process model annotated with simulation parameters (a BPS model). BPS models may be manually designed, based on information collected from stakeholders and empirical observations, or automatically discovered from execution data. Regardless of its origin, a key question when using a BPS model is how to assess its quality. In this paper, we propose a collection of measures to evaluate the quality of a BPS model w.r.t. its ability to replicate the observed behavior of the process. We advocate an approach whereby different measures tackle different process perspectives. We evaluate the ability of the proposed measures to discern the impact of modifications to a BPS model, and their ability to uncover the relative strengths and weaknesses of two approaches for automated discovery of BPS models. The evaluation shows that the measures not only capture how close a BPS model is to the observed behavior, but they also help us to identify sources of discrepancies.

Mahmoud Shoush, Marlon Dumas

Prescriptive process monitoring approaches leverage historical data to prescribe runtime interventions that will likely prevent negative case outcomes or improve a process's performance. A centerpiece of a prescriptive process monitoring method is its intervention policy: a decision function determining if and when to trigger an intervention on an ongoing case. Previous proposals in this field rely on intervention policies that consider only the current state of a given case. These approaches do not consider the tradeoff between triggering an intervention in the current state, given the level of uncertainty of the underlying predictive models, versus delaying the intervention to a later state. Moreover, they assume that a resource is always available to perform an intervention (infinite capacity). This paper addresses these gaps by introducing a prescriptive process monitoring method that filters and ranks ongoing cases based on prediction scores, prediction uncertainty, and causal effect of the intervention, and triggers interventions to maximize a gain function, considering the available resources. The proposal is evaluated using a real-life event log. The results show that the proposed method outperforms existing baselines regarding total gain.

Zahra Dasht Bozorgi, Marlon Dumas, Marcello La Rosa et al.

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.

David Chapela-Campa, Marlon Dumas

Business Process Simulation (BPS) is a common approach to estimate the impact of changes to a business process on its performance measures. For example, it allows us to estimate what would be the cycle time of a process if we automated one of its activities, or if some resources become unavailable. The starting point of BPS is a business process model annotated with simulation parameters (a BPS model). In traditional approaches, BPS models are manually designed by modeling specialists. This approach is time-consuming and error-prone. To address this shortcoming, several studies have proposed methods to automatically discover BPS models from event logs via process mining techniques. However, current techniques in this space discover BPS models that only capture waiting times caused by resource contention or resource unavailability. Oftentimes, a considerable portion of the waiting time in a business process corresponds to extraneous delays, e.g., a resource waits for the customer to return a phone call. This article proposes a method that discovers extraneous delays from event logs of business process executions. The proposed approach computes, for each pair of causally consecutive activity instances in the event log, the time when the target activity instance should theoretically have started, given the availability of the relevant resource. Based on the difference between the theoretical and the actual start times, the approach estimates the distribution of extraneous delays, and it enhances the BPS model with timer events to capture these delays. An empirical evaluation involving synthetic and real-life logs shows that the approach produces BPS models that better reflect the temporal dynamics of the process, relative to BPS models that do not capture extraneous delays.

Diego Calvanese, Angelo Casciani, Giuseppe De Giacomo et al. · oxford

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 process view of the business process, driven by the realization of process awareness and an agent-oriented abstraction, where software and human agents act as primary functional entities that perceive, reason, and act within explicit process frames. This perspective marks a shift from traditional, automation-oriented BPM toward systems in which autonomy is constrained, aligned, and made operational through process awareness. We introduce the core abstractions and architectural elements required to realize APM systems and elaborate on four key capabilities that such APM agents 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.

Mahmoud Shoush, Marlon Dumas

Prescriptive process monitoring methods seek to optimize the performance of business processes by triggering interventions at runtime, thereby increasing the probability of positive case outcomes. These interventions are triggered according to an intervention policy. Reinforcement learning has been put forward as an approach to learning intervention policies through trial and error. Existing approaches in this space assume that the number of resources available to perform interventions in a process is unlimited, an unrealistic assumption in practice. This paper argues that, in the presence of resource constraints, a key dilemma in the field of prescriptive process monitoring is to trigger interventions based not only on predictions of their necessity, timeliness, or effect but also on the uncertainty of these predictions and the level of resource utilization. Indeed, committing scarce resources to an intervention when the necessity or effects of this intervention are highly uncertain may intuitively lead to suboptimal intervention effects. Accordingly, the paper proposes a reinforcement learning approach for prescriptive process monitoring that leverages conformal prediction techniques to consider the uncertainty of the predictions upon which an intervention decision is based. An evaluation using real-life datasets demonstrates that explicitly modeling uncertainty using conformal predictions helps reinforcement learning agents converge towards policies with higher net intervention gain

Mahmoud Shoush, Marlon Dumas

Prescriptive process monitoring methods seek to improve the performance of a process by selectively triggering interventions at runtime (e.g., offering a discount to a customer) to increase the probability of a desired case outcome (e.g., a customer making a purchase). The backbone of a prescriptive process monitoring method is an intervention policy, which determines for which cases and when an intervention should be executed. Existing methods in this field rely on predictive models to define intervention policies; specifically, they consider policies that trigger an intervention when the estimated probability of a negative outcome exceeds a threshold. However, the probabilities computed by a predictive model may come with a high level of uncertainty (low confidence), leading to unnecessary interventions and, thus, wasted effort. This waste is particularly problematic when the resources available to execute interventions are limited. To tackle this shortcoming, this paper proposes an approach to extend existing prescriptive process monitoring methods with so-called conformal predictions, i.e., predictions with confidence guarantees. An empirical evaluation using real-life public datasets shows that conformal predictions enhance the net gain of prescriptive process monitoring methods under limited resources.

Alessandro Padella, Massimiliano de Leoni, Marlon Dumas

Predictive Process Monitoring is a branch of process mining that aims to predict the outcome of an ongoing process. Recently, it leveraged machine-and-deep learning architectures. In this paper, we extend our prior LLM-based Predictive Process Monitoring framework, which was initially focused on total time prediction via prompting. The extension consists of comprehensively evaluating its generality, semantic leverage, and reasoning mechanisms, also across multiple Key Performance Indicators. Empirical evaluations conducted on three distinct event logs and across the Key Performance Indicators of Total Time and Activity Occurrence prediction indicate that, in data-scarce settings with only 100 traces, the LLM surpasses the benchmark methods. Furthermore, the experiments also show that the LLM exploits both its embodied prior knowledge and the internal correlations among training traces. Finally, we examine the reasoning strategies employed by the model, demonstrating that the LLM does not merely replicate existing predictive methods but performs higher-order reasoning to generate the predictions.

Orlenys López-Pintado, Serhii Murashko, Marlon Dumas

Simulation is a common approach to predict the effect of business process changes on quantitative performance. The starting point of Business Process Simulation (BPS) is a process model enriched with simulation parameters. To cope with the typically large parameter spaces of BPS models, several methods have been proposed to automatically discover BPS models from event logs. Virtually all these approaches neglect the data perspective of business processes. Yet, the data attributes manipulated by a business process often determine which activities are performed, how many times, and when. This paper addresses this gap by introducing a data-aware BPS modeling approach and a method to discover data-aware BPS models from event logs. The BPS modeling approach supports three types of data attributes (global, case-level, and event-level) as well as deterministic and stochastic attribute update rules and data-aware branching conditions. An empirical evaluation shows that the proposed method accurately discovers the type of each data attribute and its associated update rules, and that the resulting BPS models more closely replicate the process execution control flow relative to data-unaware BPS models.

Adriano Augusto, Raffaele Conforti, Marlon Dumas et al.

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 twelve publicly-available real-life event logs, twelve 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.

Diego Calvanese, Marlon Dumas, Ülari Laurson et al.

The Decision Model and Notation (DMN) is a standard notation to capture decision logic in business applications in general and business processes in particular. A central construct in DMN is that of a decision table. The increasing use of DMN decision tables to capture critical business knowledge raises the need to support analysis tasks on these tables such as correctness and completeness checking. This paper provides a formal semantics for DMN tables, a formal definition of key analysis tasks and scalable algorithms to tackle two such tasks, i.e., detection of overlapping rules and of missing rules. The algorithms are based on a geometric interpretation of decision tables that can be used to support other analysis tasks by tapping into geometric algorithms. The algorithms have been implemented in an open-source DMN editor and tested on large decision tables derived from a credit lending dataset.

Viara Popova, Dirk Fahland, Marlon Dumas

Artifact-centric modeling is a promising approach for modeling business processes based on the so-called business artifacts - key entities driving the company's operations and whose lifecycles define the overall business process. While artifact-centric modeling shows significant advantages, the overwhelming majority of existing process mining methods cannot be applied (directly) as they are tailored to discover monolithic process models. This paper addresses the problem by proposing a chain of methods that can be applied to discover artifact lifecycle models in Guard-Stage-Milestone notation. We decompose the problem in such a way that a wide range of existing (non-artifact-centric) process discovery and analysis methods can be reused in a flexible manner. The methods presented in this paper are implemented as software plug-ins for ProM, a generic open-source framework and architecture for implementing process mining tools.

Muhammad Awais Ali, Marlon Dumas, Fredrik Milani

Predictive process monitoring is an evolving research field that studies how to train and use predictive models for operational decision-making. One of the problems studied in this field is that of predicting the sequence of upcoming activities in a case up to its completion, a.k.a. the case suffix. The prediction of case suffixes provides input to estimate short-term workloads and execution times under different resource schedules. Existing methods to address this problem often generate suffixes wherein some activities are repeated many times, whereas this pattern is not observed in the data. Closer examination shows that this shortcoming stems from the approach used to sample the successive activity instances to generate a case suffix. Accordingly, the paper introduces a sampling approach aimed at reducing repetitions of activities in the predicted case suffixes. The approach, namely Daemon action, strikes a balance between exploration and exploitation when generating the successive activity instances. We enhance a deep learning approach for case suffix predictions using this sampling approach, and experimentally show that the enhanced approach outperforms the unenhanced ones with respect to control-flow accuracy measures.

Orlenys López-Pintado, Marlon Dumas

In business process simulation, resource availability is typically modeled by assigning a calendar to each resource, e.g., Monday-Friday, 9:00-18:00. Resources are assumed to be always available during each time slot in their availability calendar. This assumption often becomes invalid due to interruptions, breaks, or time-sharing across processes. In other words, existing approaches fail to capture intermittent availability. Another limitation of existing approaches is that they either do not consider multitasking behavior, or if they do, they assume that resources always multitask (up to a maximum capacity) whenever available. However, studies have shown that the multitasking patterns vary across days. This paper introduces a probabilistic approach to model resource availability and multitasking behavior for business process simulation. In this approach, each time slot in a resource calendar has an associated availability probability and a multitasking probability per multitasking level. For example, a resource may be available on Fridays between 14:00-15:00 with 90\% probability, and given that they are performing one task during this slot, they may take on a second concurrent task with 60\% probability. We propose algorithms to discover probabilistic calendars and probabilistic multitasking capacities from event logs. An evaluation shows that, with these enhancements, simulation models discovered from event logs better replicate the distribution of activities and cycle times, relative to approaches with crisp calendars and monotasking assumptions.

Muhammad Awais Ali, Marlon Dumas, Fredrik Milani

Predictive process monitoring techniques support the operational decision making by predicting future states of ongoing cases of a business process. A subset of these techniques predict the remaining sequence of activities of an ongoing case (case suffix prediction). Existing approaches for case suffix prediction generate sequences of activities with a single timestamp (e.g. the end timestamp). This output is insufficient for resource capacity planning, where we need to reason about the periods of time when resources will be busy performing work. This paper introduces a technique for predicting case suffixes consisting of activities with start and end timestamps. In other words, the proposed technique predicts both the waiting time and the processing time of each activity. Since the waiting time of an activity in a case depends on how busy resources are in other cases, the technique adopts a sweep-line approach, wherein the suffixes of all ongoing cases in the process are predicted in lockstep, rather than predictions being made for each case in isolation. An evaluation on real-life and synthetic datasets compares the accuracy of different instantiations of this approach, demonstrating the advantages of a multi-model approach to case suffix prediction.

Orlenys López-Pintado, Jannis Rosenbaum, Marlon Dumas

In business processes, activity batching refers to packing multiple activity instances for joint execution. Batching allows managers to trade off cost and processing effort against waiting time. Larger and less frequent batches may lower costs by reducing processing effort and amortizing fixed costs, but they create longer waiting times. In contrast, smaller and more frequent batches reduce waiting times but increase fixed costs and processing effort. A batching policy defines how activity instances are grouped into batches and when each batch is activated. This paper addresses the problem of discovering batching policies that strike optimal trade-offs between waiting time, processing effort, and cost. The paper proposes a Pareto optimization approach that starts from a given set (possibly empty) of activity batching policies and generates alternative policies for each batched activity via intervention heuristics. Each heuristic identifies an opportunity to improve an activity's batching policy with respect to a metric (waiting time, processing time, cost, or resource utilization) and an associated adjustment to the activity's batching policy (the intervention). The impact of each intervention is evaluated via simulation. The intervention heuristics are embedded in an optimization meta-heuristic that triggers interventions to iteratively update the Pareto front of the interventions identified so far. The paper considers three meta-heuristics: hill-climbing, simulated annealing, and reinforcement learning. An experimental evaluation compares the proposed approach based on intervention heuristics against the same (non-heuristic guided) meta-heuristics baseline regarding convergence, diversity, and cycle time gain of Pareto-optimal policies.

Marlon Dumas, Fabiana Fournier, Lior Limonad et al.

AI-Augmented Business Process Management Systems (ABPMSs) are an emerging class of process-aware information systems, empowered by trustworthy AI technology. An ABPMS enhances the execution of business processes with the aim of making these processes more adaptable, proactive, explainable, and context-sensitive. This manifesto presents a vision for ABPMSs and discusses research challenges that need to be surmounted to realize this vision. To this end, we define the concept of ABPMS, we outline the lifecycle of processes within an ABPMS, we discuss core characteristics of an ABPMS, and we derive a set of challenges to realize systems with these characteristics.

Gamal Elkoumy, Alisa Pankova, Marlon Dumas

The applicability of process mining techniques hinges on the availability of event logs capturing the execution of a business process. In some use cases, particularly those involving customer-facing processes, these event logs may contain private information. Data protection regulations restrict the use of such event logs for analysis purposes. One way of circumventing these restrictions is to anonymize the event log to the extent that no individual can be singled out using the anonymized log. This article addresses the problem of anonymizing an event log in order to guarantee that, upon release of the anonymized log, the probability that an attacker may single out any individual represented in the original log does not increase by more than a threshold. The article proposes a differentially private release mechanism, which samples the cases in the log and adds noise to the timestamps to the extent required to achieve the above privacy guarantee. The article reports on an empirical comparison of the proposed approach against the state-of-the-art approaches using 14 real-life event logs in terms of data utility loss and computational efficiency.

Adriano Augusto, Ahmed Awad, Marlon Dumas

Verifying temporal compliance rules, such as a rule stating that an inquiry must be answered within a time limit, is a recurrent operation in the realm of business process compliance. In this setting, a typical use case is one where a manager seeks to retrieve all cases where a temporal rule is violated, given an event log recording the execution of a process over a time period. Existing approaches for checking temporal rules require a full scan of the log. Such approaches are unsuitable for interactive use when the log is large and the set of compliance rules is evolving. This paper proposes an approach to evaluate temporal compliance rules in sublinear time by pre-computing a data structure that summarizes the temporal relations between activities in a log. The approach caters for a wide range of temporal compliance patterns and supports incremental updates. Our evaluation on twenty real-life logs shows that our data structure allows for real-time checking of a large set of compliance rules.

Kateryna Kubrak, Fredrik Milani, Alexander Nolte et al.

Prescriptive process monitoring methods seek to optimize a business process by recommending interventions at runtime to prevent negative outcomes or poorly performing cases. In recent years, various prescriptive process monitoring methods have been proposed. This paper studies existing methods in this field via a Systematic Literature Review (SLR). In order to structure the field, the paper proposes a framework for characterizing prescriptive process monitoring methods according to their performance objective, performance metrics, intervention types, modeling techniques, data inputs, and intervention policies. The SLR provides insights into challenges and areas for future research that could enhance the usefulness and applicability of prescriptive process monitoring methods. The paper highlights the need to validate existing and new methods in real-world settings, to extend the types of interventions beyond those related to the temporal and cost perspectives, and to design policies that take into account causality and second-order effects.

Mahmoud Shoush, Marlon Dumas

Prescriptive process monitoring is a family of techniques to optimize the performance of a business process by triggering interventions at runtime. Existing prescriptive process monitoring techniques assume that the number of interventions that may be triggered is unbounded. In practice, though, specific interventions consume resources with finite capacity. For example, in a loan origination process, an intervention may consist of preparing an alternative loan offer to increase the applicant's chances of taking a loan. This intervention requires a certain amount of time from a credit officer, and thus, it is not possible to trigger this intervention in all cases. This paper proposes a prescriptive process monitoring technique that triggers interventions to optimize a cost function under fixed resource constraints. The proposed technique relies on predictive modeling to identify cases that are likely to lead to a negative outcome, in combination with causal inference to estimate the effect of an intervention on the outcome of the case. These outputs are then used to allocate resources to interventions to maximize a cost function. A preliminary empirical evaluation suggests that the proposed approach produces a higher net gain than a purely predictive (non-causal) baseline.

Volodymyr Leno, Adriano Augusto, Marlon Dumas et al.

Robotic Process Automation (RPA) is a technology to automate routine work such as copying data across applications or filling in document templates using data from multiple applications. RPA tools allow organizations to automate a wide range of routines. However, identifying and scoping routines that can be automated using RPA tools is time consuming. Manual identification of candidate routines via interviews, walk-throughs, or job shadowing allow analysts to identify the most visible routines, but these methods are not suitable when it comes to identifying the long tail of routines in an organization. This article proposes an approach to discover automatable routines from logs of user interactions with IT systems and to synthesize executable specifications for such routines. The approach starts by discovering frequent routines at a control-flow level (candidate routines). It then determines which of these candidate routines are automatable and it synthetizes an executable specification for each such routine. Finally, it identifies semantically equivalent routines so as to produce a set of non-redundant automatable routines. The article reports on an evaluation of the approach using a combination of synthetic and real-life logs. The evaluation results show that the approach can discover automatable routines that are known to be present in a UI log, and that it identifies automatable routines that users recognize as such in real-life logs.

Zahra Dasht Bozorgi, Irene Teinemaa, Marlon Dumas et al.

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 giving a phone call to a customer to obtain missing information rather than waiting passively. 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 the total net gain. The paper proposes a prescriptive process monitoring method that uses orthogonal random forest models to estimate the causal effect of triggering a time-reducing intervention for each ongoing case of a process. Based on this causal effect estimate, the method triggers interventions according to a user-defined policy. The method is evaluated on two real-life logs.

Adriano Augusto, Marlon Dumas, Marcello La Rosa

Enterprise information systems allow companies to maintain detailed records of their business process executions. These records can be extracted in the form of event logs, which capture the execution of activities across multiple instances of a business process. Event logs may be used to analyze business processes at a fine level of detail using process mining techniques. Among other things, process mining techniques allow us to discover a process model from an event log -- an operation known as automated process discovery. Despite a rich body of research in the field, existing automated process discovery techniques do not fully capture the concurrency inherent in a business process. Specifically, the bulk of these techniques treat two activities A and B as concurrent if sometimes A completes before B and other times B completes before A. Typically though, activities in a business process are executed in a true concurrency setting, meaning that two or more activity executions overlap temporally. This paper addresses this gap by presenting a refined version of an automated process discovery technique, namely Split Miner, that discovers true concurrency relations from event logs containing start and end timestamps for each activity. The proposed technique is also able to differentiate between exclusive and inclusive choices. We evaluate the proposed technique relative to existing baselines using 11 real-life logs drawn from different industries.

Manuel Camargo, Marlon Dumas, Oscar González-Rojas

Business process simulation is a well-known approach to estimate the impact of changes to a process with respect to time and cost measures -- a practice known as what-if process analysis. The usefulness of such estimations hinges on the accuracy of the underlying simulation model. Data-Driven Simulation (DDS) methods leverage process mining techniques to learn process simulation models from event logs. Empirical studies have shown that, while DDS models adequately capture the observed sequences of activities and their frequencies, they fail to accurately capture the temporal dynamics of real-life processes. In contrast, generative Deep Learning (DL) models are better able to capture such temporal dynamics. The drawback of DL models is that users cannot alter them for what-if analysis due to their black-box nature. This paper presents a hybrid approach to learn process simulation models from event logs wherein a (stochastic) process model is extracted via DDS techniques, and then combined with a DL model to generate timestamped event sequences. An experimental evaluation shows that the resulting hybrid simulation models match the temporal accuracy of pure DL models, while partially retaining the what-if analysis capability of DDS approaches.

Gamal Elkoumy, Alisa Pankova, Marlon Dumas

The applicability of process mining techniques hinges on the availability of event logs capturing the execution of a business process. In some use cases, particularly those involving customer-facing processes, these event logs may contain private information. Data protection regulations restrict the use of such event logs for analysis purposes. One way of circumventing these restrictions is to anonymize the event log to the extent that no individual can be singled out using the anonymized log. This paper addresses the problem of anonymizing an event log in order to guarantee that, upon disclosure of the anonymized log, the probability that an attacker may single out any individual represented in the original log, does not increase by more than a threshold. The paper proposes a differentially private disclosure mechanism, which oversamples the cases in the log and adds noise to the timestamps to the extent required to achieve the above privacy guarantee. The paper reports on an empirical evaluation of the proposed approach using 14 real-life event logs in terms of data utility loss and computational efficiency.

Gamal Elkoumy, Alisa Pankova, Marlon Dumas

Process mining techniques enable organizations to analyze business process execution traces in order to identify opportunities for improving their operational performance. Oftentimes, such execution traces contain private information. For example, the execution traces of a healthcare process are likely to be privacy-sensitive. In such cases, organizations need to deploy Privacy-Enhancing Technologies (PETs) to strike a balance between the benefits they get from analyzing these data and the requirements imposed onto them by privacy regulations, particularly that of minimizing re-identification risks when data are disclosed to a process analyst. Among many available PETs, differential privacy stands out for its ability to prevent predicate singling out attacks and its composable privacy guarantees. A drawback of differential privacy is the lack of interpretability of the main privacy parameter it relies upon, namely epsilon. This leads to the recurrent question of how much epsilon is enough? This article proposes a method to determine the epsilon value to be used when disclosing the output of a process mining technique in terms of two business-relevant metrics, namely absolute percentage error metrics capturing the loss of accuracy (a.k.a. utility loss) resulting from adding noise to the disclosed data, and guessing advantage, which captures the increase in the probability that an adversary may guess information about an individual as a result of a disclosure. The article specifically studies the problem of protecting the disclosure of the so-called Directly-Follows Graph (DFGs), which is a process mining artifact produced by most process mining tools. The article reports on an empirical evaluation of the utility-risk trade-offs that the proposed approach achieves on a collection of 13 real-life event logs.

Manuel Camargo, Marlon Dumas, Oscar Gonzalez-Rojas

A generative model is a statistical model that is able to generate new data instances from previously observed ones. In the context of business processes, a generative model creates new execution traces from a set of historical traces, also known as an event log. Two families of generative process simulation models have been developed in previous work: data-driven simulation models and deep learning models. Until now, these two approaches have evolved independently and their relative performance has not been studied. This paper fills this gap by empirically comparing a data-driven simulation technique with multiple deep learning techniques, which construct models are capable of generating execution traces with timestamped events. The study sheds light into the relative strengths of both approaches and raises the prospect of developing hybrid approaches that combine these strengths.

Zahra Dasht Bozorgi, Irene Teinemaa, Marlon Dumas et al.

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.

Abderrahmane Maaradji, Marlon Dumas, Marcello La Rosa et al.

Business processes are prone to unexpected changes, as process workers may suddenly or gradually start executing a process differently in order to adjust to changes in workload, season, or other external factors. Early detection of business process changes enables managers to identify and act upon changes that may otherwise affect process performance. Business process drift detection refers to a family of methods to detect changes in a business process by analyzing event logs extracted from the systems that support the execution of the process. Existing methods for business process drift detection are based on an explorative analysis of a potentially large feature space and in some cases they require users to manually identify specific features that characterize the drift. Depending on the explored feature space, these methods miss various types of changes. Moreover, they are either designed to detect sudden drifts or gradual drifts but not both. This paper proposes an automated and statistically grounded method for detecting sudden and gradual business process drifts under a unified framework. An empirical evaluation shows that the method detects typical change patterns with significantly higher accuracy and lower detection delay than existing methods, while accurately distinguishing between sudden and gradual drifts.

Bedilia Estrada-Torres, Manuel Camargo, Marlon Dumas et al.

Business process simulation is a versatile technique for analyzing business processes from a quantitative perspective. A well-known limitation of process simulation is that the accuracy of the simulation results is limited by the faithfulness of the process model and simulation parameters given as input to the simulator. To tackle this limitation, several authors have proposed to discover simulation models from process execution logs so that the resulting simulation models more closely match reality. Existing techniques in this field assume that each resource in the process performs one task at a time. In reality, however, resources may engage in multitasking behavior. Traditional simulation approaches do not handle multitasking. Instead, they rely on a resource allocation approach wherein a task instance is only assigned to a resource when the resource is free. This inability to handle multitasking leads to an overestimation of execution times. This paper proposes an approach to discover multitasking in business process execution logs and to generate a simulation model that takes into account the discovered multitasking behavior. The key idea is to adjust the processing times of tasks in such a way that executing the multitasked tasks sequentially with the adjusted times is equivalent to executing them concurrently with the original processing times. The proposed approach is evaluated using a real-life dataset and synthetic datasets with different levels of multitasking. The results show that, in the presence of multitasking, the approach improves the accuracy of simulation models discovered from execution logs.

Volodymyr Leno, Marlon Dumas, Marcello La Rosa et al.

Robotic Process Automation (RPA) is a technology for automating repetitive routines consisting of sequences of user interactions with one or more applications. In order to fully exploit the opportunities opened by RPA, companies need to discover which specific routines may be automated, and how. In this setting, this paper addresses the problem of analyzing User Interaction (UI) logs in order to discover routines where a user transfers data from one spreadsheet or (Web) form to another. The paper maps this problem to that of discovering data transformations by example - a problem for which several techniques are available. The paper shows that a naive application of a state-of-the-art technique for data transformation discovery is computationally inefficient. Accordingly, the paper proposes two optimizations that take advantage of the information in the UI log and the fact that data transfers across applications typically involve copying alphabetic and numeric tokens separately. The proposed approach and its optimizations are evaluated using UI logs that replicate a real-life repetitive data transfer routine.

Gamal Elkoumy, Stephan A. Fahrenkrog-Petersen, Marlon Dumas et al.

Process mining is a family of techniques for analysing business processes based on event logs extracted from information systems. Mainstream process mining tools are designed for intra-organizational settings, insofar as they assume that an event log is available for processing as a whole. The use of such tools for inter-organizational process analysis is hampered by the fact that such processes involve independent parties who are unwilling to, or sometimes legally prevented from, sharing detailed event logs with each other. In this setting, this paper proposes an approach for constructing and querying a common type of artifact used for process mining, namely the frequency and time-annotated Directly-Follows Graph (DFG), over multiple event logs belonging to different parties, in such a way that the parties do not share the event logs with each other. The proposal leverages an existing platform for secure multi-party computation, namely Sharemind. Since a direct implementation of DFG construction in Sharemind suffers from scalability issues, the paper proposes to rely on vectorization of event logs and to employ a divide-and-conquer scheme for parallel processing of sub-logs. The paper reports on an experimental evaluation that tests the scalability of the approach on real-life logs.

Daniel Reißner, Abel Armas-Cervantes, Raffaele Conforti et al.

Given a model of the expected behavior of a business process and an event log recording its observed behavior, the problem of business process conformance checking is that of identifying and describing the differences between the model and the log. A desirable feature of a conformance checking technique is to identify a minimal yet complete set of differences. Existing conformance checking techniques that fulfil this property exhibit limited scalability when confronted to large and complex models and logs. This paper presents two complementary techniques to address these shortcomings. The first technique transforms the model and log into two automata. These automata are compared using an error-correcting synchronized product, computed via an A* that guarantees the resulting automaton captures all differences with a minimal amount of error corrections. The synchronized product is used to extract minimal-length alignments between each trace of the log and the closest corresponding trace of the model. A limitation of the first technique is that as the level of concurrency in the model increases, the size of the automaton of the model grows exponentially, thus hampering scalability. To address this limitation, the paper proposes a second technique wherein the process model is first decomposed into a set of automata, known as S-components, such that the product of these automata is equal to the automaton of the whole process model. An error-correcting product is computed for each S-component separately and the resulting automata are recomposed into a single product automaton capturing all differences without minimality guarantees. An empirical evaluation shows that the proposed techniques outperform state-of-the-art baselines in terms of computational efficiency. Moreover, the decomposition-based technique is optimal for the vast majority of datasets and quasi-optimal for the remaining ones.

Manuel Camargo, Marlon Dumas, Oscar González-Rojas

Business process simulation is a versatile technique to estimate the performance of a process under multiple scenarios. This, in turn, allows analysts to compare alternative options to improve a business process. A common roadblock for business process simulation is that constructing accurate simulation models is cumbersome and error-prone. Modern information systems store detailed execution logs of the business processes they support. Previous work has shown that these logs can be used to discover simulation models. However, existing methods for log-based discovery of simulation models do not seek to optimize the accuracy of the resulting models. Instead they leave it to the user to manually tune the simulation model to achieve the desired level of accuracy. This article presents an accuracy-optimized method to discover business process simulation models from execution logs. The method decomposes the problem into a series of steps with associated configuration parameters. A hyper-parameter optimization method is used to search through the space of possible configurations so as to maximize the similarity between the behavior of the simulation model and the behavior observed in the log. The method has been implemented as a tool and evaluated using logs from different domains.

Orlenys López-Pintado, Marlon Dumas, Luciano García-Bañuelos et al.

Blockchain technology provides a tamper-proof mechanism to execute inter-organizational business processes involving mutually untrusted parties. Existing approaches to blockchain-based process execution are based on code generation. In these approaches, a process model is compiled into one or more smart contracts, which are then deployed on a blockchain platform. Given the immutability of the deployed smart contracts, these compiled approaches ensure that all process instances conform to the process model. However, this advantage comes at the price of inflexibility. Any changes to the process model require the redeployment of the smart contracts (a costly operation). In addition, changes cannot be applied to running process instances. To address this lack of flexibility, this paper presents an interpreter of BPMN process models based on dynamic data structures. The proposed interpreter is embedded in a business process execution system with a modular multi-layered architecture, supporting the creation, execution, monitoring and dynamic update of process instances. For efficiency purposes, the interpreter relies on compact bitmap-based encodings of process models. An experimental evaluation shows that the proposed interpreted approach achieves comparable or lower costs relative to existing compiled approaches.

Stephan A. Fahrenkrog-Petersen, Niek Tax, Irene Teinemaa et al.

Predictive process monitoring is a family of techniques to analyze events produced during the execution of a business process in order to predict the future state or the final outcome of running process instances. Existing techniques in this field are able to predict, at each step of a process instance, the likelihood that it will lead to an undesired outcome.These techniques, however, focus on generating predictions and do not prescribe when and how process workers should intervene to decrease the cost of undesired outcomes. This paper proposes a framework for prescriptive process monitoring, which extends predictive monitoring with the ability to generate alarms that trigger interventions to prevent an undesired outcome or mitigate its effect. The framework incorporates a parameterized cost model to assess the cost-benefit trade-off of generating alarms. We show how to optimize the generation of alarms given an event log of past process executions and a set of cost model parameters. The proposed approaches are empirically evaluated using a range of real-life event logs. The experimental results show that the net cost of undesired outcomes can be minimized by changing the threshold for generating alarms, as the process instance progresses. Moreover, introducing delays for triggering alarms, instead of triggering them as soon as the probability of an undesired outcome exceeds a threshold, leads to lower net costs.

Aivo Toots, Reedik Tuuling, Maksym Yerokhin et al.

Pleak is a tool to capture and analyze privacy-enhanced business process models to characterize and quantify to what extent the outputs of a process leak information about its inputs. Pleak incorporates an extensible set of analysis plugins, which enable users to inspect potential leakages at multiple levels of detail.

Orlenys López-Pintado, Marlon Dumas, Luciano García-Bañuelos et al.

Blockchain technology enables the execution of collaborative business processes involving mutually untrusted parties. Existing platforms allow such processes to be modeled using high-level notations and compiled into smart contracts that can be deployed on blockchain platforms. However, these platforms brush aside the question of who is allowed to execute which tasks in the process, either by deferring the question altogether or by adopting a static approach where all actors are bound to roles upon process instantiation. Yet, a key advantage of blockchains is their ability to support dynamic sets of actors. This paper presents a model for dynamic binding of actors to roles in collaborative processes and an associated binding policy specification language. The proposed language is endowed with a Petri net semantics, thus enabling policy consistency verification. The paper also outlines an approach to compile policy specifications into smart contracts for enforcement. An experimental evaluation shows that the cost of policy enforcement increases linearly with the number of roles and constraints.

Diego Calvanese, Marlon Dumas, Fabrizio Maria Maggi et al.

The Decision Model and Notation (DMN) is a recent OMG standard for the elicitation and representation of decision models, and for managing their interconnection with business processes. DMN builds on the notion of decision tables, and their combination into more complex decision requirements graphs (DRGs), which bridge between business process models and decision logic models. DRGs may rely on additional, external business knowledge models, whose functioning is not part of the standard. In this work, we consider one of the most important types of business knowledge, namely background knowledge that conceptually accounts for the structural aspects of the domain of interest, and propose decision knowledge bases (DKBs), which semantically combine DRGs modeled in DMN, and domain knowledge captured by means of first-order logic with datatypes. We provide a logic-based semantics for such an integration, and formalize different DMN reasoning tasks for DKBs. We then consider background knowledge formulated as a description logic ontology with datatypes, and show how the main verification tasks for DMN in this enriched setting can be formalized as standard DL reasoning services, and actually carried out in ExpTime. We discuss the effectiveness of our framework on a case study in maritime security.

Orlenys López-Pintado, Luciano García-Bañuelos, Marlon Dumas et al.

Blockchain platforms, such as Ethereum, allow a set of actors to maintain a ledger of transactions without relying on a central authority and to deploy scripts, called smart contracts, that are executed whenever certain transactions occur. These features can be used as basic building blocks for executing collaborative business processes between mutually untrusting parties. However, implementing business processes using the low-level primitives provided by blockchain platforms is cumbersome and error-prone. In contrast, established business process management systems, such as those based on the standard Business Process Model and Notation (BPMN), provide convenient abstractions for rapid development of process-oriented applications. This article demonstrates how to combine the advantages of a business process management system with those of a blockchain platform. The article introduces a blockchain-based BPMN execution engine, namely Caterpillar. Like any BPMN execution engine, Caterpillar supports the creation of instances of a process model and allows users to monitor the state of process instances and to execute tasks thereof. The specificity of Caterpillar is that the state of each process instance is maintained on the (Ethereum) blockchain and the workflow routing is performed by smart contracts generated by a BPMN-to-Solidity compiler. The Caterpillar compiler supports a large array of BPMN constructs, including subprocesses, multi-instances activities and event handlers. The paper describes the architecture of Caterpillar, and the interfaces it provides to support the monitoring of process instances, the allocation and execution of work items, and the execution of service tasks.

Ilya Verenich, Marlon Dumas, Marcello La Rosa et al.

Predictive business process monitoring methods exploit historical process execution logs to generate predictions about running instances (called cases) of a business process, such as the prediction of the outcome, next activity or remaining cycle time of a given process case. These insights could be used to support operational managers in taking remedial actions as business processes unfold, e.g. shifting resources from one case onto another to ensure this latter is completed on time. A number of methods to tackle the remaining cycle time prediction problem have been proposed in the literature. However, due to differences in their experimental setup, choice of datasets, evaluation measures and baselines, the relative merits of each method remain unclear. This article presents a systematic literature review and taxonomy of methods for remaining time prediction in the context of business processes, as well as a cross-benchmark comparison of 16 such methods based on 16 real-life datasets originating from different industry domains.

Volodymyr Leno, Abel Armas-Cervantes, Marlon Dumas et al.

Automated process discovery is a class of process mining methods that allow analysts to extract business process models from event logs. Traditional process discovery methods extract process models from a snapshot of an event log stored in its entirety. In some scenarios, however, events keep coming with a high arrival rate to the extent that it is impractical to store the entire event log and to continuously re-discover a process model from scratch. Such scenarios require online process discovery approaches. Given an event stream produced by the execution of a business process, the goal of an online process discovery method is to maintain a continuously updated model of the process with a bounded amount of memory while at the same time achieving similar accuracy as offline methods. However, existing online discovery approaches require relatively large amounts of memory to achieve levels of accuracy comparable to that of offline methods. Therefore, this paper proposes an approach that addresses this limitation by mapping the problem of online process discovery to that of cache memory management, and applying well-known cache replacement policies to the problem of online process discovery. The approach has been implemented in .NET, experimentally integrated with the Minit process mining tool and comparatively evaluated against an existing baseline using real-life datasets.

Irene Teinemaa, Niek Tax, Massimiliano de Leoni et al.

Predictive process monitoring is concerned with the analysis of events produced during the execution of a process in order to predict the future state of ongoing cases thereof. Existing techniques in this field are able to predict, at each step of a case, the likelihood that the case will end up in an undesired outcome. These techniques, however, do not take into account what process workers may do with the generated predictions in order to decrease the likelihood of undesired outcomes. This paper proposes a framework for prescriptive process monitoring, which extends predictive process monitoring approaches with the concepts of alarms, interventions, compensations, and mitigation effects. The framework incorporates a parameterized cost model to assess the cost-benefit tradeoffs of applying prescriptive process monitoring in a given setting. The paper also outlines an approach to optimize the generation of alarms given a dataset and a set of cost model parameters. The proposed approach is empirically evaluated using a range of real-life event logs.

Irene Teinemaa, Marlon Dumas, Anna Leontjeva et al.

Predictive process monitoring is concerned with the analysis of events produced during the execution of a business process in order to predict as early as possible the final outcome of an ongoing case. Traditionally, predictive process monitoring methods are optimized with respect to accuracy. However, in environments where users make decisions and take actions in response to the predictions they receive, it is equally important to optimize the stability of the successive predictions made for each case. To this end, this paper defines a notion of temporal stability for binary classification tasks in predictive process monitoring and evaluates existing methods with respect to both temporal stability and accuracy. We find that methods based on XGBoost and LSTM neural networks exhibit the highest temporal stability. We then show that temporal stability can be enhanced by hyperparameter-optimizing random forests and XGBoost classifiers with respect to inter-run stability. Finally, we show that time series smoothing techniques can further enhance temporal stability at the expense of slightly lower accuracy.

Niek Tax, Marlon Dumas

Sequential pattern mining techniques extract patterns corresponding to frequent subsequences from a sequence database. A practical limitation of these techniques is that they overload the user with too many patterns. Local Process Model (LPM) mining is an alternative approach coming from the field of process mining. While in traditional sequential pattern mining, a pattern describes one subsequence, an LPM captures a set of subsequences. Also, while traditional sequential patterns only match subsequences that are observed in the sequence database, an LPM may capture subsequences that are not explicitly observed, but that are related to observed subsequences. In other words, LPMs generalize the behavior observed in the sequence database. These properties make it possible for a set of LPMs to cover the behavior of a much larger set of sequential patterns. Yet, existing LPM mining techniques still suffer from the pattern explosion problem because they produce sets of redundant LPMs. In this paper, we propose several heuristics to mine a set of non-redundant LPMs either from a set of redundant LPMs or from a set of sequential patterns. We empirically compare the proposed heuristics between them and against existing (local) process mining techniques in terms of coverage, redundancy, and complexity of the produced sets of LPMs.

Irene Teinemaa, Marlon Dumas, Marcello La Rosa et al.

Predictive business process monitoring refers to the act of making predictions about the future state of ongoing cases of a business process, based on their incomplete execution traces and logs of historical (completed) traces. Motivated by the increasingly pervasive availability of fine-grained event data about business process executions, the problem of predictive process monitoring has received substantial attention in the past years. In particular, a considerable number of methods have been put forward to address the problem of outcome-oriented predictive process monitoring, which refers to classifying each ongoing case of a process according to a given set of possible categorical outcomes - e.g., Will the customer complain or not? Will an order be delivered, canceled or withdrawn? Unfortunately, different authors have used different datasets, experimental settings, evaluation measures and baselines to assess their proposals, resulting in poor comparability and an unclear picture of the relative merits and applicability of different methods. To address this gap, this article presents a systematic review and taxonomy of outcome-oriented predictive process monitoring methods, and a comparative experimental evaluation of eleven representative methods using a benchmark covering 24 predictive process monitoring tasks based on nine real-life event logs.

Jan Mendling, Ingo Weber, Wil van der Aalst et al.

Blockchain technology promises a sizable potential for executing inter-organizational business processes without requiring a central party serving as a single point of trust (and failure). This paper analyzes its impact on business process management (BPM). We structure the discussion using two BPM frameworks, namely the six BPM core capabilities and the BPM lifecycle. This paper provides research directions for investigating the application of blockchain technology to BPM.

Luciano García-Bañuelos, Alexander Ponomarev, Marlon Dumas et al.

Blockchain technology enables the execution of collaborative business processes involving untrusted parties without requiring a central authority. Specifically, a process model comprising tasks performed by multiple parties can be coordinated via smart contracts operating on the blockchain. The consensus mechanism governing the blockchain thereby guarantees that the process model is followed by each party. However, the cost required for blockchain use is highly dependent on the volume of data recorded and the frequency of data updates by smart contracts. This paper proposes an optimized method for executing business processes on top of commodity blockchain technology. The paper presents a method for compiling a process model into a smart contract that encodes the preconditions for executing each task in the process using a space-optimized data structure. The method is empirically compared to a previously proposed baseline by replaying execution logs, including one from a real-life business process, and measuring resource consumption.

Niek Tax, Ilya Verenich, Marcello La Rosa et al.

Predictive business process monitoring methods exploit logs of completed cases of a process in order to make predictions about running cases thereof. Existing methods in this space are tailor-made for specific prediction tasks. Moreover, their relative accuracy is highly sensitive to the dataset at hand, thus requiring users to engage in trial-and-error and tuning when applying them in a specific setting. This paper investigates Long Short-Term Memory (LSTM) neural networks as an approach to build consistently accurate models for a wide range of predictive process monitoring tasks. First, we show that LSTMs outperform existing techniques to predict the next event of a running case and its timestamp. Next, we show how to use models for predicting the next task in order to predict the full continuation of a running case. Finally, we apply the same approach to predict the remaining time, and show that this approach outperforms existing tailor-made methods.