Process Mining on Distributed Data Sources
Title: Process Mining on Distributed Data Sources
Abstract: Recent years have seen an immense uptake of the Internet-of-Things and applications that build on sensor data. Such applications have emerged in various domains. In logistics, services use sea ship transponder data for monitoring the movement, loading and unloading of vessels. In healthcare, hospitals install Real-Time Locating Systems to track events relating to staff, patients, and equipment in clinical pathways. Technically, the mentioned scenarios have in common that they support complex processes in a distributed environment with a distributed infrastructure that collects sensor data for managing a complex system. This means that the sensor data in the above scenarios integrates into larger business processes. To efficiently address these scenarios with process mining requires novel process mining techniques for distributed event data. This talk will present applications for distributed process mining as well as new techniques to address the field.
Bio: Agnes Koschmider is a professor of Business Informatics at the University of Bayreuth. Prior to this position Agnes Koschmider was professor of Business Informatics at the Computer Science Institute of the University of Kiel. She completed her PhD and her habilitation in Applied Informatics at KIT. Agnes conducts research on methods for data-driven analysis and explanation of processes using artificial intelligence. Her work centers on process analytics, in particular the development of pipelines that efficiently handle the entire chain from raw data (time series, sensor events, and video data) to process discovery. Such data pipelines have applications across a wide range of disciplines.