Abstract: Episode mining is an active subfield of data mining in which the aim is to retrieve important knowledge from temporal data and can be used to analyze fault reports and web navigation logs. However, existing methods generally do not consider time gap constraints, and overestimate the frequency of episodes, which may lead to mining a large number of episodes that users are not interested in. To tackle this problem, this paper investigates one-off episode rule (OER) mining with time gap constraints for process event logs and proposes a one-off episode rule mining algorithm called OER-Miner that can mine frequent one-off episodes and the implicit relationship among them. To generate fewer and prune unpromising candidate episodes, OER-Miner utilizes episode join and pruning strategies, respectively. To efficiently calculate the candidate episode support, position indexes, and depth-first search and backtracking strategies are applied to calculate the number of occurrences. Experimental results verify that OER-Miner yields a better performance than seven other competitive algorithms on nine publicly available event logs. More importantly, OER-Miner can be applied to a real-industrial log to identify rework phenomena in the production process by mining strong one-off episode rules, to discover the optimal processes and deficiencies of the system, and provide recommendations for further improvement.

Abstract: Abstract The assessment of process mining techniques using real-life data is often compromised by the lack of ground truth knowledge, the presence of non-essential outliers in system behavior and recording errors in event logs. Using synthetically generated data could leverage ground truth for better evaluation. Existing log generation tools inject noise directly into the logs, which does not capture many typical behavioral deviations. Furthermore, the link between the model and the log, which is needed for later assessment, becomes lost. We propose a ground-truth approach for generating process data from existing or synthetic initial process models, whether automatically generated or hand-made. This approach incorporates patterns of behavioral deviations and recording errors to produce a synthetic yet realistic deviating model and imperfect event log. These, together with the initial model, are required to assess process mining techniques based on ground truth knowledge. We demonstrate this approach to create datasets of synthetic process data for three processes, one of which we used in a conformance checking use case, focusing on the assessment of (relaxed) systemic alignments to expose and explain deviations in modeled and recorded behavior. Our results show that this approach, unlike traditional methods, provides detailed insights into the strengths and weaknesses of process mining techniques, both quantitatively and qualitatively.

Abstract: UiPath Task mining is an AI-powered feature that captures the user data performed on the desktop, records the granular level actions, including each mouse click and keystroke, and provides a visualization of the analyzed data captured with the help of Artificial Intelligence. Task Mining also helps users or business analysts identify the bottlenecks in the process and discrepancies and may even contribute to improving the process. This paper explores the usage and features of UiPath Task Mining. This research study also mentionsthe architecture overview of process mining, integrations, anddashboards, among other features. Different types of mining available in Task Mining are also discussedin this paper. Task capture usage, capabilities, and featuresare discussed. The relevance of Process mining related to streamlining business operations, usage of different available templates that are specific and tailored based on the use case, process apps, different fields in the analysis of process graph, and integration to automation hub are discussed in this research paper.

Abstract: Process mining encompasses a suite of techniques aimed at analyzing event data to gain insights and improve operational processes. One way of achieving this is to discover the driving process of the activities that occurred in a system. Technically, process discovery algorithms are used to transform an event log into a process model which is representative of the activities registered in the given system. This study explores the application of process discovery methods to better understand the learning processes in an introductory programming course for first-year Computer Science BSc students. A total of 52 practical problems were assigned as out-of-class activities via GitHub Classroom, resulting in 2789 commits from 59 students. These commits, along with the students’ exam grades, were recorded in an object-centric event log, subsequently converted into a casebased log for analysis using the PM4Py program library. The study had two primary goals: first, to identify the characteristics of successful learning strategies by comparing process models of students who passed versus those who failed the programming exam; and second, to identify bottlenecks that hindered student progress. By employing the Heuristic Miner and Inductive Miner algorithms, we developed and contrasted learning process models, revealing significant patterns and obstacles within the educational process. The findings provide valuable insights into the factors that contribute to effective learning and suggest areas for enhancing our teaching methodologies.

Abstract: Process mining is revolutionising how organisations optimise operational processes by applying sophisticated algorithms to historical data from contemporary information systems. Despite recent efforts to address the organisational aspects of process mining, a crucial gap in comprehensive understanding of competency requirements for successful implementation remains. This study integrates theoretical insights with empirical job vacancy data to develop a typology of roles, including a competency framework and ideal role types. Using topic modelling, cluster analysis, rigorous evaluation metrics, and consensus-building, this study ensures validity and reliability. It offers practical and theoretical insights to support effective organisational adoption and success in process mining initiatives.

Abstract: Process mining is increasingly adopted in modern organizations, producing numerous process models that, while valuable, can lead to model overload and decision-making complexity. This paper explores a multi-criteria decision-making (MCDM) approach to evaluate and prioritize process models by incorporating both quantitative metrics (e.g., fitness, precision) and qualitative factors (e.g., cultural fit). An illustrative logistics example demonstrates how MCDM, specifically the Analytic Hierarchy Process (AHP), facilitates trade-off analysis and promotes alignment with managerial objectives. Initial insights suggest that the MCDM approach enhances context-sensitive decision-making, as selected models address both operational metrics and broader managerial needs. While this study is an early-stage exploration, it provides an initial foundation for deeper exploration of MCDM-driven strategies to enhance the role of process mining in complex organizational settings.

Abstract: Processes are fundamental to enterprises, serving as significant engines of optimization and analysis. Understanding business processes is critical, yet integrating formal process ontologies with enterprise data and workflows remains difficult. We refer to this specific kind of ontology application, intended for practitioners working with enterprise data, as operational realization. The varied ontological commitments and highly expressive representation languages of process ontologies create barriers for operational realization, including issues of decidability, a lack of tooling, and operational constraints. This paper presents an architecture for the operational realization of process ontologies, driven by process mining needs. A key aspect of the architecture is the formalization of ontological commitments in tasks like data cleaning and analysis, which rely on implicit assumptions embedded in the interpretation of process data. Our approach builds on existing methodologies, notably ontology-based data access (OBDA), while going further by encoding domain knowledge required to interpret and reason with process data. This structure moves beyond an A-Box and T-Box distinction, explicitly capturing how the process ontology, domain data, and supporting data interpretation theories enable complex process reasoning. By structuring the dynamics of a process ontology, domain data, process data theories, and by characterizing reasoning scenarios, our approach provides a pragmatic foundation for integrating process ontologies into data-driven process workflows. We demonstrate this architecture with real enterprise data, challenge problems, and scenarios already widely used for benchmarking in process mining.

Abstract: Process mining has proven effective in explaining the underlying processes of systems, thereby improving systems’ understanding, analysis, and operational efficiency. Process mining, however, often falls short in addressing multiple dimensions of systems’ behaviors, limiting its ability to provide comprehensive insights for systems’ performance and optimization opportunities. In this paper, we introduce an enhancement to conventional process mining that we term Multi-flow Process Mining (MFPM), which effectively extracts process flows across different system dimensions, such as time, energy, waste, and carbon footprint. MFPM enables a more comprehensive view of a system's dynamics, enabling holistic decision-making for enhanced system efficiency. We detail the framework of MFPM, outline corresponding data requirements, and introduce an expanded version of Petri nets—used here as a modeling formalism to describe and analyze multi-flow system processes. Through a detailed case study, we demonstrate the practical application of MFPM in capturing and analyzing multifaceted aspects of systems.

Abstract: Process mining holds substantial potential to discover and optimize processes utilizing event log data. However, current applications primarily rely on (semi-)structured data from process-aware information systems, limiting their capacity to incorporate multimodal data from diverse sources, particularly in domains like IT service management (ITSM). While existing stand-alone approaches can extract event log data from unstructured sources such as videos, documents, or bot logs, they fall short of leveraging the full range of real-world data available in ITSM. To address this gap, our research focuses on developing a reference architecture for constructing event logs from multimodal data. This architecture integrates diverse data types, construction functions, and process mining use cases. Following a design science research methodology, we aim to evaluate the architecture through a software artifact leveraging real-world ITSM data and incorporating state-of-the-art generative AI. In this study, we present the preliminary reference architecture and share early insights from expert evaluations.

Abstract: Background: Adherence to clinical guidelines supports high quality patient care. Conformance checking, a feature of process mining, can potentially automate the assessment of adherence to clinical guidelines in practice. Objectives: This paper investigates appropriate conformance checking in practice. Methods: Conformance checking in practice was simulated with generated test data, a FHIR server and process mining tools. A corresponding literature review was conducted in parallel. Results: Activities of clinical guidelines or in healthcare processes should be coded using clinical nomenclature to support conformance checking. Conclusion: SNOMED CT should be used as a nomenclature and activities should be coded with SNOMED concepts of the type “procedure”.

Abstract: Due to the omnipresence of information systems in today's society, there are data traces of almost all human activities. These data traces allow us to analyze how processes are executed, identify problems, and improve such processes. The area of computer science defining algorithms to analyze processes from data is called process mining. Traditionally, there is an expected data format for the data generated by processes, the traditional event log format. Such an event log consists of a collection of events that describe the execution of a certain activity at a certain point in time. Additionally, there are two central assumptions for each event: It can be uniquely associated with one end-to-end run through the process, the case identifier, and every case identifier is an instantiation of the same process. We call these assumptions identifier singularity and identifier homogeneity.These two assumptions limit the applicability of process mining in practice: One might be interested in analyzing two different but intersecting processes at once and how different, overlapping instantiations of processes behave. Imagine a complex production process: Different parts are manufactured in individual, different processes but assembled together, meaning that there are events that refer to multiple case identifiers of different types. Event data where events refer to multiple identifiers of different types is called object-centric event data. Since process mining algorithms require traditional event logs, we would need to enforce the two assumptions of identifier singularity and homogeneity for object-centric event data, which leads to quality problems and information loss that might impact the accuracy of process mining insights. The process of enforcing these assumptions is called flattening.Flattening moves object-centric event data towards traditional process mining and was the state of the art for applying most process mining algorithms to object-centric event data when we started working on this thesis. However, there is also another way of applying process mining algorithms to object-centric event data: One can move process mining towards object-centric event data, providing an object-centric adaptation of algorithms. So far, only a few algorithms have been translated to the object-centric setting and there has been no comprehensive effort across all process mining tasks. This missing piece in process mining research is the central motivation for this thesis. We define the two main research questions based on this observation:How to translate traditional process mining concepts and algorithms into object-centricity? How are process mining insights and their accuracy advanced by object-centric process mining? To address these research questions, we translate the traditional case and trace concept into object-centricity, lifting sequential traces from traditional process mining into graph-based process executions for object-centric process mining. We provide process execution extraction techniques and show under which circumstances they are guaranteed to be free of certain quality problems that can produce misleading insights. We then translate the concepts of variants, replay fitness and precision, alignments, feature extraction, and feature encoding into object-centricity. We provide visualization techniques for these concepts and evaluate how object-centric variants, discovery, fitness and precision, alignments, features, and encodings advance process mining. We also provide tools and case studies that practically show these improvements and make them reproducible.In our evaluations, we find that object-centric process mining advances traditional process mining across all tasks. We can discover object-centric process models of equal quality using less data. Conformance metrics are more accurate due to the absence of quality problems, and predictive monitoring results are superior due to more precise feature extraction and encoding. This increased accuracy comes, in general, at the expense of higher computation times as we move from sequences to the more general concept of graphs. This also means that object-centric process mining is a generalization of traditional process mining.Our research of this thesis focuses on the core algorithms of process mining. The most important focus of future work deals with integrating this work into the larger overall process mining pipeline, addressing challenges from event data extraction, transformation, and loading such as streaming or distributed computing, all the way to the end user, who requires a responsive and accessible visualization and storage solutions. Our contributions also enable the definition of new problems and algorithms, such as root-cause analysis across objects or object-relationship conformance checking.

Abstract: In the realm of healthcare, optimizing processes is imperative for delivering timely and effective patient care. Process mining, a technique rooted in data and process science, offers a promising avenue for uncovering insights and streamlining operations within healthcare organizations. This chapter explores the applications of process mining in hospitals, focusing on its ability to identify inefficiencies, optimize resource allocation, and enhance overall operational efficiency. Furthermore, it discusses the challenges faced in healthcare operations and proposes solutions to bridge the gap between visible and invisible problems. The chapter concludes with recommendations for leveraging process mining alongside innovative technologies like robotic process automation to revolutionize healthcare delivery and improve patient outcomes. Furthermore, process mining serves as a valuable tool for regulatory compliance and quality assurance in healthcare. By capturing and analyzing process data, organizations can ensure adherence to standards, protocols, and best practices, thereby mitigating risks and improving patient safety. In summary, the utilization of process mining in conjunction with cutting-edge technologies and quality improvement methodologies offers a comprehensive approach to transforming healthcare delivery. By fostering a culture of data-driven decision-making and continuous improvement, healthcare organizations can enhance efficiency, effectiveness, and patient-centricity across the care continuum.

Abstract: In the contemporary dynamic business environment, the dependability of process mining algorithms is intricately tied to the quality of event logs, often marred by data challenges stemming from human involvement in business processes. This study introduces a novel approach that amalgamates insights from prior works with unsupervised techniques, specifically Principal Component Analysis (PCA), to elevate the precision and reliability of event log representations. Executed through Python and the pm4py library, the methodology is applied to real event logs. The adoption of Petri nets for process representation aligns with systematic approaches advocated by earlier studies, enhancing transparency and interpretability. Results demonstrate the method's efficacy through enhanced metrics such as Fitness, Precision, and F-Measure, accompanied by visualizations elucidating the optimal number of principal components. This study offers a comprehensive and practical solution, bridging gaps in existing methodologies, and its integration of multiple strategies, particularly PCA, showcases versatility in optimizing process mining analyses. The consistent improvements observed underscore the method's potential across diverse business contexts, making it accessible and pertinent for practitioners engaged in real-world business processes. Overall, this research contributes an innovative approach to improve event log quality, thereby advancing the field of process mining with practical implications for organizational decision-making and process optimization.

Abstract: Business processes are generally time-sensitive, impacting factors such as customer expectations, cost efficiencies, compliance requirements, supply chain constraints, and timely decision-making. Time analysis is therefore crucial for customer understanding and process congestion minimisation. Existing process mining methods mainly employ basic statistics, process discovery and data mining techniques. These approaches often lack a structured model or profile to characterise the data related to the duration of individual process tasks. Consequently, it can be difficult to comprehensively understand critical observations such as trends, peaks, and valleys of task durations. This paper proposes a parsimonious generic representation of task duration data that addresses these limitations. A mixture model comprising gamma, uniform and exponential distributions is proposed that allows for peaked components corresponding to durations terminating near a particular value (the peak) with, in addition, flatter components for durations terminating more randomly between the peaks. The modelling is validated using examples from patient billing and the telecom industry. In each scenario, the corresponding fitted models offer a good representation of the underlying process tasks. The model can therefore be used to improve knowledge of these tasks in terms of the mixture components and what they might represent, such as the root causes of task termination. The paper also considers information criteria more appropriate for large data sets where very small effects can appear “significant” using techniques developed for smaller data set

Abstract: This paper explores the synergies between process mining and Data Envelopment Analysis (DEA) in business process management. Process mining offers insights into organizational workflows by analyzing event logs, while DEA provides a robust framework for efficiency assessment across diverse sectors. In the initial stage, hierarchical data analysis was conducted to construct process information diagrams and categorize process traces based on various factors, including decision points for each case. In the second stage, process analysis concepts and process discovery algorithms were utilized to define and compute key performance indicators (KPIs) from the perspectives of organizational resources and individual cases, followed by constructing process diagrams using Alpha, Inductive, and Heuristic miners. In the third stage, a scenario-based robust DEA model was applied to rank the KPIs of employees derived from the event log data. Furthermore, we examine how the proposed scenario-based Robust DEA revolutionizes efficiency assessment, providing decision-makers with a comprehensive framework for evaluating organizational performance. Ultimately, by integrating ML prediction and behavior analysis, organizations can anticipate future trends, optimize operations, and drive continuous improvement. Ultimately, process mining and DEA represent indispensable tools for organizations seeking to enhance operational efficiency, mitigate risks, and achieve strategic objectives in today's dynamic business environment. The findings demonstrate that this comprehensive integration of techniques provides profound insights into the event logs, identifying strengths, weaknesses, bottlenecks, and overall system efficiency, thereby facilitating organizational improvement and optimization.

Abstract: Amid the sweeping currents of digital transformation reshaping industries and unrelenting competitive pressures, organizations face the ongoing challenge of reinventing their operations. Companies have to constantly explore novel approaches to streamline their operations, eliminate wasteful practices, and sharpen their decision-making capabilities. Process mining, powered by tools like Celonis, has emerged as a revolutionary technology that bridges the gap between data and actionable insights. By visualizing and analyzing real-time processes across departments, process mining enables businesses to identify bottlenecks, streamline workflows, and unlock unprecedented levels of operational efficiency.Focusing on its ability to provide end-to-end metric oriented visibility into organizational processes, drive data-driven decision-making and uncover inefficiencies, this paper explores the transformative potential of process mining.It delves into the core methodologies of process mining and highlights its impact on critical business areas such as cost reduction, compliance, resource optimization, and customer satisfaction. Organizations can make informed adjustments, foster a culture of continuous improvement and gain a deeper understanding of their operations by leveraging the advanced capabilities of Celonis, as per the author.Through case studies and industry benchmarks, this paper demonstrates how process mining has enabled companies across sectors to achieve measurable outcomes, including shorter cycle times, enhanced process conformance, and significant financial savings. As businesses strive for agility and excellence, process mining serves as a vital tool in their journey toward sustained growth and success.