From Internet of Things Data to Business Processes: Challenges and a Framework

• DOI: 10.48550/arxiv.2405.08528
• Publication Date: 2024-05-14
• Authors: Juergen Mangler; Ronny Seiger; Janik-Vasily Benzin; Joscha Grüger; Yusuf Kirikkayis; Florian Gallik; Lukas Malburg; Matthias Ehrendorfer; Yannis Bertrand; Marco Franceschetti; Barbara Weber; Stefanie Rinderle‐Ma; Ralph Bergmann; Estefanía Serral; Manfred Reichert
• Keywords: Software Engineering (cs.SE); FOS: Computer and information sciences; Computer Science - Software Engineering; Artificial Intelligence (cs.AI); H.3.3; Computer Science - Artificial Intelligence; H.3.3; I.2.1; I.2.1

The IoT and Business Process Management (BPM) communities co-exist in many shared application domains, such as manufacturing healthcare. community has a strong focus on hardware, connectivity data; the BPM focuses mainly finding, controlling, enhancing structured interactions among devices processes. While field of Mining deals with extraction process models analytics from event logs, data produced by sensors often is at lower granularity than these process-level events. fundamental questions about extracting abstracting process-related streams sensor values are: (1) Which can be clustered together part events?, (2) signify start end (3) are related but not essential? This work proposes framework to semi-automatically perform set steps convert low-level into higher-level events that suitable for mining. meant provide generic sequence abstract guide extraction, abstraction, correlation, variation points plugging specific analysis techniques algorithms each step. To assess completeness framework, we present challenges, how they tackled through an example instantiate real-world demonstration smart manufacturing. Based this future research conducted manner refining improving individual steps.