No description available.

This Master's describes knownledge discovery from process logs by using process mining algorithms. Chosen algorithms are described in detail. These aim to create process model based on event log analysis. The goal is to design such components, which would be able to import the process and run the simulations. Results from components can be used for short term planning.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / Process Mining is a discipline that recognizes three types of analysis: Discovery, monitoring, and process improvement. Organizations are focusing on redesigning and automating their major processes, according to a report published in 2018 [1]. In this way, a challenge n process mining is to show the results of the process analysis in a way that is understandable to non-expert users. Therefore, this research paper introduces a matrix of guidelines to guide process mining specialists/tool developers to improve the results of the analysis in process mining projects. This matrix is composed of 2 study fields that throughout the literature have been merging their virtues. First, process mining under 2 of its 3 types of projects: (1) based on objectives and (2) based on questions. The last type is based on data (exploratory analysis). Second, visualization of data with its techniques to represent data graphically. This research proposes a matrix of guidelines that integrates the discipline of process mining and the set of data visualization techniques based on the purpose of each graph (technique), the question / objective to be achieved and the importance that colors take in the analysis results in the process mining projects. / Revisión por pares

El presente trabajo propone un modelo para la evaluación de variables en el sector salud utilizando Process Mining y Data Visualization soportado por la herramienta Celonis. Esto surge ante la problemática orientada a la dificultad en la comprensión de las actividades que están involucradas en los procesos negocios y los resultados de este. El proyecto se centra en la investigación de dos disciplinas emergentes. Una de estas disciplinas es Process Mining y se enfoca principalmente en los procesos, en los datos por cada evento, esto con el fin de descubrir un modelo, ver conformidad de los procesos o mejorarlos (Process Mining: Una técnica innovadora para la mejora de los procesos, 2016). La segunda disciplina es Data Visualization, esta permite presentar los datos en un formato gráfico o pictórico ("Data Visualization: What it is and why it matters", 2016). El proyecto implica principalmente investigación, en primer lugar, se analizan las técnicas de Process Mining y Data Visualization. En segundo lugar, se separan las características y cualidades de las disciplinas, y se diseña un modelo para la evaluación de variables en el Sector Salud utilizando Process Mining y Data Visualization, generando un valor agregado, dado que al tener un formato gráfico o pictórico que representa adecuadamente los resultados de usar una técnica de minería de procesos, la comprensión y el análisis en la toma de decisiones es más precisa. En tercer lugar, se valida el modelo en una institución que brinda servicios en el Sector Salud, analizando uno de los procesos core. Finalmente, se elabora un plan de continuidad para que el modelo propuesto se aplique en técnicas de optimización de procesos en las organizaciones. / The present work proposes a model for the evaluation of variables in the health sector using Process Mining and Data Visualization supported by the Celonis tool. This arises from the problem oriented to the difficulty in understanding the activities that are involved in business processes and their results. The project focuses on the investigation of two emerging disciplines. One of these disciplines is Process Mining and it focuses mainly on the processes, on the data for each event, this in order to discover a model, see conformity of the processes or improve them (Process Mining: An innovative technique for the improvement of the processes, 2016). The second discipline is Data Visualization, this allows data to be presented in a graphic or pictorial format ("Data Visualization: What it is and why it matters", 2016). This project mainly involves research, first, Process Mining and Data Visualization techniques are analyzed. Second, the characteristics and qualities of the disciplines are separated, and a model is designed for the evaluation of variables in the Health Sector using Process Mining and Data Visualization, generating added value, given that by having a graphic or pictorial format that adequately represents the results of using a process mining technique, understanding and analysis in decision making is more accurate. Third, the model is validated in an institution that provides services in the Health Sector, analyzing one of the core processes. Finally, a continuity plan is drawn up so that the proposed model can be applied to process optimization techniques in organizations. / Tesis

Context: Process mining is an approach that exploits event logs to discover real processes executed in organizations, enabling them to (re)design and improve process models. Goal modelling, on the other hand, is a requirements engineering (RE) approach mainly used to analyze what-if situations and support decision making. Problem: Common problems with process mining include the complexity of discovered “spaghetti” processes and a lack of goal-process alignment. Current process mining practices mainly focus on activities and do not benefit from considering stakeholder goals and requirements to manage complexity and alignment. The critical artifact that process mining practices rely on is the event log. However, using a raw version of real-life event logs will typically result in process models being too complex, unstructured, difficult to understand and, above all, not aligned with stakeholders’ goals. Method: Involving goal-related factors can augment the precision and interpretability of mined models and help discover better opportunities to satisfy stakeholders. This thesis proposes three algorithms for goal-oriented process enhancement and discovery (GoPED) that show synergetic effects achievable by combining process mining and goal-oriented modelling. With GoPED, good historical experiences will be found within the event log to be used as a basis for inferring good process models, and bad experiences will be found to discover models to avoid. The goodness is defined in terms of alignment with regards to three categories of goal-related criteria: • Case perspective: satisfaction of individual cases (e.g., patient, costumer) in terms of some goals; • Goal perspective: overall satisfaction of some goals (e.g., to decrease waiting time) rather than individual cases; and • Organization perspective: a comprehensive satisfaction level for all goals over all cases. GoPED first adds goal-related attributes to conventional event characteristics (case identifier, activities, and timestamps), selects a subset of cases concerning goal-related criteria, and finally discovers a process model from that subset. For each criterion, an algorithm is developed to enable selecting the best subset of cases where the criterion holds. The resulting process models are expected to reproduce the desired level of satisfaction. The three GoPED algorithms were implemented in a Python tool. In addition, three other tools were implemented to complete a line of actions whose input is a raw event log and output is a subset of the event log selected with respect to the goal-related criteria. GoPED was used on real healthcare event logs (an illustrative example and a case study) to discover processes, and the performance of the tools was also assessed. Results: The performance of the GoPED toolset for various sizes and configurations of event logs was assessed through extensive experiments. The results show that the three GoPED algorithms are practical and scalable for application to event logs with realistic sizes and types of configurations. The GoPED method was also applied to the discovery of processes from the raw event log of the trajectories of patients with sepsis in a Dutch hospital, from their registration in the emergency room until their discharge. Although the raw data does not explicitly include goal-related information, some reasonable goals were derived from the data and a related research paper in consultation with a healthcare expert. The method was applied, and the resulting models were i) substantially simpler than the model dis-covered from the whole event log, ii) free from the drawbacks that using the whole event log causes, and iii) aligned with the predefined goals. Conclusion: GoPED demonstrates the benefits of exploiting goal modelling capabilities to enhance event logs and select a subset of events to discover goal-aligned and simplified process models. The resulting process model can also be compared to a model discovered from the original event log to reveal new insights about the ability of different forms of process models to satisfy the stakeholders’ goals. Learning from good behaviours that satisfy goals and detecting bad behaviours that hurt them is an opportunity to redesign models, so they are simpler, better aligned with goals, and free from drawbacks that using the whole event log may cause.

Many software companies today collect command usage data by monitoring and logging user interactions within their applications. This is not always utilised to its full potential, but with the use of state-of-the-art process mining techniques, this command usage log data can be used to gain insights about the users' workflows. These insights can then be used to improve the software application and boost user productivity and efficiency. One area where this might be especially relevant is within the radiology domain, where the radiology labour shortage renders every efficiency-improvement valuable. Connected to this, this thesis aimed to evaluate a number of process mining techniques on real radiology PACS command usage data from Sectra to identify which techniques might be useful for analysing user workflows. Three process discovery algorithms (Alpha, Heuristics, and Inductive Miner - infrequent) were used on two datasets and evaluated based on a number of quantitative metrics (fitness and simplicity) and qualitative aspects (interpretability and usefulness). The qualitative aspects of the resulting process models were assessed through interviews with domain experts at Sectra, and the Heuristics Miner was found to discover the most useful models that could be interpreted and analysed by domain experts, mainly due to its simpler process model notation. To reduce model complexity, three different filtering methods based on sequential pattern mining were evaluated as a pre-processing step before applying the discovery algorithms. This resulted in improvements for the Alpha and Inductive Miner - infrequent, although none of the methods improved the Heuristic Miner models against the baseline. Trace clustering was also explored to address model complexity with the aim of identifying trace execution variants. Several configurations of trace representation techniques and clustering algorithms were used, and the neural-network-based approaches, Word2Vec and Autoencoder, emerged as the alternatives that achieved the best scores in the clustering evaluation. A few clusters with well-separated trace execution variants were identified - although most clusters were still complex and dominated by similar events. Finally, a prototype application with integrated process mining concepts was created based on the findings from the previous interviews. This was then evaluated with domain experts at Sectra, with the aim of investigating what concepts are practically useful for assisting with domain analysis. The findings indicate a clear use case for such an application to analyse sequential relations and command usage patterns in PACS user workflows, providing a data-driven and on-demand approach for hypotheses testing. Simpler concepts like manual filtering and aggregation were found to be practically useful and prioritised by the domain experts, while the opinion was more divided on the automatic pre-processing methods.

Discrete-event simulation (DES) is an invaluable tool which organizations can use to help better understand, diagnose, and optimize their operational processes. Studies have shown that for the typical DES exercise, the greatest amount of time is spent on developing an accurate model of the process that is to be studied. Process mining, a similar field of study, focuses on using historical data stored in software databases to accurate recreate and analyze business processes. Utilizing process mining techniques to help rapidly develop DES models can drastically reduce the amount of time spent building simulation models, which ultimately will enable organizations to more quickly identify and correct shortcomings in their operations. Although there have been significant advances in process mining research, there are still several issues with current process mining methods which prevent them from seeing widespread industry adoption. One such issue, which this study examines, is the lack of cross-compatibility between process mining tools and other process analysis tools. Specifically, this study develops and characterizes a method through which mined process models can be converted into discrete-event simulation models. The developed method utilizes a plugin written for the ProM Framework, an existing collection of process mining tools, which takes a mined process model as its input and outputs an Excel workbook which provides the process data in a format more easily read by DES packages. Two event logs which mimic real-world processes were used in the development and validation of the plugin. The developed plugin successfully extracted the critical process data from the mined process model and converted it into a format more easily utilized by DES packages. There are several limitations which will limit model accuracy, but the plugin developed by this study shows that the conversion of process models to basic simulation models is possible. Future research can focus on addressing the limitations to improve model accuracy.

This study intended to validate if clustering could be used to simplify models generated with process mining. The intention was also to see if these clusters could suggest anything about user efficiency. To that end a new metric where devised, average mean duration deviation. This metric aimed to show if a trace was more or less efficient than a comparative trace. Since the intent was to find traces with similar characteristics the clustering was done with characteristic features instead of time efficiency features. The aim was to find a correlation between efficiency after the fact. A correlation with efficiency could not be found.

Das stetige Wachstum von Datenmengen, besonders in Unternehmen, setzt die Neu- und Weiterentwicklung geeigneter datengetriebener Analysemethoden voraus, die die gesammelten Informationen in einen Kontext setzen und einen operativen Mehrwert für die Unternehmen erzeugen. Insbesondere die echtzeitnahe Analyse von Geschäftsprozessdaten, die in den Unternehmensinformationssystemen gespeichert werden, lassen sich mit Hilfe von Analysewerkzeugen, wie es das Process Mining zur Verfügung stellt, auswerten und generieren Einblicke in die Prozesse der Unternehmen. Für vertrauenswürdige Ergebnisse wird jedoch eine hohe Qualität der zu analysierenden Daten vorausgesetzt. Die vorliegende Arbeit befasst sich mit der Entwicklung eines Monitoring- und Klassifikationsverfahrens für Business Process Event-Streams zur Verwertung im Kontext des Online Process Mining. Zu den erarbeiteten Artefakten dieser Arbeit zählen ein auf den Bedarfen abzielender Anforderungskatalog, ein Konzept, das eine Streaming-Architektur, ein Klassifikationsmodell, eine rekonstruktionsbasierte Anomalieerkennung, einen Online Learning Workflow und Erklärungskomponenten umfasst sowie eine prototypische Umsetzung der Konzepte. Über technische Experimente auf Basis unterschiedlicher Datengrundlagen und optimierten Umgebungsparametern werden die Funktionsweise und Güte des erarbeiteten Monitoring- und Filterverfahrens überprüft. Durch die Einbettung des Event-Filters in eine Streaming-Architektur, die Kombination verschiedener Strukturen des maschinellen Lernens und der damit einhergehenden Generalisierungsfähigkeit des Modells sowie der Fähigkeit des Modells kontinuierlich zu Lernen gelingt es existierende Ansätze um Aspekte der echtzeitnahen Verarbeitung und dem Monitoring von Event-Streams zu erweitern und die Genauigkeit der Anomalieerkennung auf Event-Ebene zu verbessern. Durch Monitoring, Klassifikation und Filterung der eingehenden Event-Daten kann die Datenqualität für die Anwendung nachgelagerter Process Mining Aktivitäten erhöht werden.

Background: Game analytics is a field that aims to analyze games and help in the enhancement of game development. Data mining is a prominent technique for game analytics. Recent advances in the field of process mining have motivated users to apply process mining to real-world scenarios in order to derive process-oriented insights. In this study, We provide a discussion on how process mining can be used in game analytics. Objective: The goal of this study is to apply process mining to player data from a live game, analyze the results, and determine whether these results can be interpreted, whether we can derive any patterns or insights that can be useful for game designers, and whether process mining can be used in-game analytics and, if so, what kind of versatility it can offer. Also, this study provides approaches on how process mining can be used in player churn prediction and determination of abnormal player activity sequences. Method: Firstly, a literature review is performed to comprehend all of the process mining techniques and metrics used to evaluate the discovered process models. Then experiments are conducted by applying process mining on data from a live game, determine a churn predictor using process mining and determining a technique to identify abnormal player sequences. Results: Process discovery algorithms are applied on data from a live game, the results are analyzed. Several process models are discovered to identify player churn and it is compared with a baseline machine learning churn predictor trained on the same data to that of process mining. Abnormal player activity sequences of the gameare determined using process mining and compared with expected player sequences and analyzed with the help of game designers. Conclusion: Process mining can be utilized in game analytics to discover new process-oriented insights. When compared to typical data mining techniques, the results gained by process mining are more versatile. It also has other capabilities such as detecting unusual sequences in data.