1,721,276 research outputs found
Data Quality in Process Mining
No full textTo cope with challenges such as tightening budgets and increased care needs, healthcare organizations are becoming increasingly aware of the need to understand their processes in order to improve them. In this respect, process mining has the unique potential to retrieve process-related insights from process execution data. Despite the wide range of algorithms that have been developed over the past decade, the reliability of process mining outcomes ultimately depends on the quality of the input data. Consistent with the notion of “Garbage In, Garbage Out”, applying process mining algorithms to low quality data can lead to counter-intuitive or even misleading decisions. Real-life healthcare event logs typically suffer from a multitude of data quality issues such as missing events, incorrect timestamps and incorrect resource information. Against this background, this chapter provides an introduction to data quality in the process mining field. Three key topics are discussed: (1) data quality taxonomies, i.e. frameworks outlining potential data quality issues, (2) data quality assessment, i.e. the identification of data quality issues, and (3) data cleaning, i.e. efforts towards alleviating data quality issues which are present in an event log
BRINGING STUDENTS TO PRACTICE: PERFORMING A REAL-LIFE SIMULATION STUDY IN AN INTRODUCTORY SIMULATION COURSE
No full textBusiness process simulation can support the analysis and improvement of business processes and, hence, is a valuable technique to teach to students. Consequently, introductory simulation courses are included in a multitude of study programs. Besides providing students with theoretical knowledge and getting them acquainted with simulation software, it is also important to let them experience the complexity involved in conducting a simulation study in practice. In this respect, this paper outlines the inclusion of a real-life simulation study in an introductory simulation course. Besides the content of the simulation study, the instructional design, student feedback, company feedback and challenges perceived by the instructor are outlined. Student feedback shows that, even though the project was perceived as highly challenging, conducting the real-life simulation study provided extensive learning opportunities
Using Indoor Location System Data to Enhance the Quality of Healthcare Event Logs: Opportunities and Challenges
No full textHospitals are becoming more and more aware of the need to manage their business processes. In this respect, process mining is increasingly used to gain insight in healthcare processes, requiring the analysis of event logs originating from the hospital information system. Process mining research mainly focuses on the development of new techniques or the application of existing methods, but the quality of all analyses ultimately depends on the quality of the event log. However, limited research has been done on the improvement of data quality in the process mining field, which is the topic of this paper. In particular, this paper discusses, from a conceptual angle, the opportunities that indoor location system data provides to tackle event log data quality issues. Moreover, the paper reflects upon the associated challenges. In this way, it provides the conceptualization for a new area of research, focusing on the systematic integration of an event log with indoor location system data
Data-Driven Process Simulation
No full textSynonyms-Definition Data-driven process simulation is a technique which constructs a computer model that imitates the internal details of a business process and extensively uses real-life data-recorded by information systems supporting the actual process-to do so. The model is used to evaluate what-if scenarios in order to better understand the actual process behaviour and to predict the impact of potential changes to the process. This is a preprint version of an article published in the Encyclopedia of Big Data Technologies. The final authenticated version is available online at: https://doi.org/10
Unveiling Use Cases for Human Resource Mining: A Framework of Past and Future Research
No full textHuman resources are considered a strategic asset for organizations and play a key role in the execution of business processes. Hence, organizations should provide an environment that enables them to operate in an effective and efficient manner. To shape such an environment, an improved understanding and monitoring of the real-life involvement of human resources in processes and the teams in which they operate would be beneficial. Using event data from information systems, process mining can play a role in this respect. Over the years, several human resource mining methods have been developed, i.e., process mining methods that convey insights related to the human resources in a process using an event log. However, there is a lack of a holis-tic understanding of the breadth of these methods. Against this backdrop, the paper uses a systematic literature review to develop a framework providing an overview of human resource mining use cases. These use cases are classified according to two dimensions: the level of analysis (individual versus multiple human resources) and the focus of analysis (organization-focused versus human-focused). The authors illustrate the versatility of process mining for providing insights into human resources and highlight opportunities for further enriching and extending this area of research to analyze, among other things, how teams of resources can perform better.The authors would like to acknowledge Emma
Van Coillie for supporting the initial full-text screening of the first
literature search in 2021. Moreover, we would like to sincerely thank
the editor and the reviewers for their insightful and constructive
feedback. This has significantly contributed to the development of the
paper during the review process
Process mining in healthcare – An updated perspective on the state of the art
No full textABSTR A C T Process mining is the research domain focusing on the development of innovative methods to gather insights from event logs. It has been used for various use cases within the healthcare domain with the ambition to instigate evidence-based process improvement. Over the past years, the research interest in process mining in healthcare has been increasing. This paper presents the results of an extensive systematic literature review on process mining in healthcare in which 263 papers have been reviewed. Besides providing the most recent overview of literature and the extensive number of reviewed papers, we complement existing reviews by considering three novel review dimensions: (i) the process mining project stages, (ii) the involvement of domain expertise, and (iii) the Key Performance Indicators (KPI) considered during the process mining analysis. Orthogonal to these three novel dimensions, we also highlight the evolution of the research domain by considering time trends within the review dimensions. The review generates new perspectives on process mining in healthcare as a research domain. For instance, process redesign is rarely part of a process mining project, domain experts are mostly asked for validating insights, and less than half of the published papers considers one or more specific KPIs to direct their analysis
Patient flow data registration: A key barrier to the data-driven and proactive management of an emergency department
No full textEditorial Patient flow data registration: A key barrier to the data-driven and proactive management of an emergency department Emergency department (ED) crowding is a global issue, and one of the most researched operational challenges in healthcare [1-4]. Several adverse outcomes are linked to crowding, including reduced patient satisfaction, increasing patient mortality rates, and rising stress levels among ED staff [2,5]. Besides input-related aspects (e.g., the influx of low-acuity patients), crowding often originates from issues including inadequate staffing or inpatient boarding. Such issues impede smooth patient flows and, hence, contribute to crowding [1,2]. A wide range of studies have been conducted to investigate measures to improve ED operations (e.g., crowding scales, team triage, fast track, escalation protocols, …) [1,2]. It has been suggested that these approaches are predominantly reactive, influenced by a paradigm focused on predicting and controlling crowding-related issues [6]. In the same paper, a paradigm of analysing and managing is proposed. In the view of the authors, this latter paradigm has several advantages as it allows for a proactive approach to management, instead of an approach focused on solving crowding-related issues. Data-driven decision-making A wide range of studies has been conducted to investigate measures to improve ED operations and decision-making [1,2]. While it has been argued that clinical data can be leveraged in this respect [7,8], the same holds for patient flow data which is recorded in the ED's information system. Patient flow data relates to data recordings which enable the reconstruction of the patient's trajectory throughout the ED, ranging from his/her registration, triage and examination, to discharge from the ED or admission into the hospital. As this information is recorded for each patient, it enables an ED to take a bird's eye view of a set of patients flow data instead of focusing on an individual patient. While patient flow data is increasingly being recorded by electronic information systems, paper-based systems are still frequently used in EDs around the world. The potential of electronically recorded patient flow data has been demonstrated in digital tools such as dashboarding and process mining. A dashboard uses patient flow data to provide real-time insights into the current situation of the ED by visualising some key metrics such as the door-to-doctor time or the length-of-stay [9,10]. In this way, a dashboard constitutes a key instrument to proactively manage an ED as it can, for instance, inform ED staff about problems that are likely to occur in the near future. While dashboarding is typically centred around key performance metrics, process mining provides a set of algorithms to discover the end-to-end process of patients from patient flow data. In other words, patient flow data is used to retrieve a process model which visualises the real patient flow at the ED. Process mining algorithms serving a wide range of other goals have been developed, including to automatically check the compliance between the actual patient flow and a normative model (e.g., originating from clinical guidelines) [11,12]. By providing insights into end-to-end processes based on data, process mining results can support evidence-based process improvement and decision-making in EDs. Difficulties in gathering adequate data Despite the enormous potential of data-driven tools and techniques to proactively manage an ED and assist with evidence-based process improvement, many EDs worldwide are unable to make use of this due to issues with patient flow data registration. An ED can, for instance, be confronted with a low information system maturity, in which essential parts of a (or the whole) patient's trajectory are still recorded in paper files. While this makes real-time patient flow analysis almost impossible , it also makes systematic post-hoc analysis (e.g., using a process mining algorithm) complicated due to the efforts required to create a sufficiently large patient flow dataset. This would require the digitisa-tion of the relevant data points from the paper files of a sufficiently large number of patients, which might not be feasible in practice given the high work pressure of an ED. Moreover, paper files tend to focus on clinical data and not on providing information about a patient's tra-jectory with associated timestamps. For instance: files might contain the clinical parameters observed during a clinical examination, but not the exact times at which the examination started and ended. Some steps in the process might not even be recorded at all. While paper-based systems still exist in many EDs, an increasing number of EDs possess state-of-the-art integrated electronic information systems to support all of their operations. However, the fact that data is recorded in an integrated electronic system does not guarantee that patient flow data will be of good quality. This can be attributed to the fact that data registration still depends on a manual act by ED staff. For example: to analyse patient flow processes, it is key that activities are recorded at the moment at which they are executed. In practice, it is often seen that ED staff perform a series of activities on several patients only to record them at a later point in time, potentially even in a random order [13]. Such recording behaviour can be, at least partly, explained by the fact that ED patients often require immediate clinical action and, hence, timely data recording is of secondary importance. Moreover, when the average level of data and process literacy tends to be low in an ED, the potential of the patient flow data might not be recognised. Attention to these topics in medical and nursing education tends to be limited, entailing the risk that ED staff do not appreciate the added value of accurate and timely patient flow data registration.Martin, N (corresponding author), Hasselt Univ, Fac Business Econ, Martelarenlaan 42, B-3500 Hasselt, Belgium.
[email protected]; [email protected]
From Insights to INTEL: Evaluating Process Mining Insights with Healthcare Professionals
No full textAs healthcare organisations are looking for ways to improve their processes, process mining techniques are increasingly being used. Current process mining methods do not offer support for translating process mining insights into actionable improvement ideas. By performing action research at two healthcare organisations, we introduce and illustrate the INTEL funnel, a novel three-staged method consisting of process familiarisation, domain explanation and improvement ideation. Our method complements existing process mining methods and constitutes the first attempt to open the black box regarding the path from process mining insights to actionable process improvement ideas. In this way, it can contribute to a more systematic uptake of process mining in healthcare practice
DaQAPO: Supporting flexible and fine-grained event log quality assessment
No full textProcess mining can provide valuable insights in business processes using an event log containing process execution data. Despite the significant potential of process mining to support the analysis and improvement of processes, the reliability of process mining outcomes depends on the quality of the event log. Real-life logs typically suffer from various data quality issues. Consequently, thorough event log quality assessment is required before applying process mining algorithms. This paper introduces DaQAPO, the first R-package which supports flexible and fine-grained event log quality assessment. It provides a rich set of tests to identify a wide range of event log quality issues, while having sufficient flexibility to allow the detection of context-specific quality issues
Demystifying Data Governance for Process Mining: Insights from a Delphi Study
No full textData governance is recognised as a new capability for organisations to maximize the value of data. Process mining is essential for the resilient growth of businesses, making process data a strategic asset for organisations. Even though the availability of reliable process data is vital for obtaining dependable insights into process mining techniques, there exists no framework that explains how to govern process data holistically. We address this gap by presenting the first data governance framework for process mining that was derived from a Delphi study conducted with a panel of academics and practitioners from around the world. The framework provides multiple avenues for future research
- …
