Data-driven Management of Interconnected Business Processes : Contributions to Predictive and Prescriptive Process Mining

Abstract

Business process management (BPM) is an accepted paradigm of organizational design to orchestrate distributed work involving various activities, resources, and actors, connecting the physical and digital world. While traditional research in BPM focused on process models and model-based information systems (e.g., workflow management systems), the focus has recently shifted toward data-driven methods such as process mining. Process mining strives to discover, monitor, and improve business processes by extracting knowledge from process (or event) logs. As process mining has evolved into one of the most active streams in BPM, numerous approaches have been proposed in the last decade, and various commercial vendors transferred these methods into practice, substantially facilitating event data analysis. However, there are still manifold unsolved challenges that hinder the adoption and usage of process mining at the enterprise level. First, finding, extracting, and preprocessing relevant event data remains challenging. Second, most process mining approaches operate on a single-process level, making it hard to apply process mining multiple interconnected processes. Third, process managers strongly require forward-directed operational support, but most process mining approaches provide only descriptive ex-post insights. Driven by these challenges, this thesis contributes to the existing body of knowledge related to data-driven management of interconnected business processes. By proposing methods that enhance and automate the extraction of event logs from typical sources (research paper1) and exploiting novel sources containing process-relevant information (research papers #2 and #3), this thesis contributes to the first challenge of finding, extracting, and preprocessing relevant event data. Regarding the second challenge to apply process mining to a multi-process perspective, this thesis proposes approaches for log-driven prioritization of interconnected business processes (research papers #4 and #5). As the proposed process prioritization methods build on predicting processes’ future performance, they also contribute to the third challenge of providing forward-directed operational support for process managers. Providing accurate predictions leveraging the increasing volume of available data is key to develop predictive and prescriptive process mining approaches. Consequently, the thesis also elaborates on how predictive process monitoring can benefit from the promising trend of deep learning (research paper #6).%%%%Geschaftsprozessmanagement (BPM) ist ein akzeptiertes Paradigma der Organisationsgestaltung zur Orchestrierung verteilter Arbeit, die verschiedene Aktivitaten, Ressourcen und Akteure umfasst und somit die physische mit der digitalen Welt verbindet. Wahrend das wissenschaftliche Hauptinteresse im Bereich BPM fur eine lange Zeit auf Prozessmodellen und modellbasierten Informationssystemen (z.B. Workflow-Management-Systeme) lag, hat sich der Fokus in letzter Zeit zunehmend auf datengestutzte Methoden, wie z.B. das…