Speaker
Sotiris Bersimis
(University of Piraeus, Greece)
Description
Convex hulls represent a powerful geometric tool with broad applications in multivariate data analysis and statistical process monitoring. In this work, we introduce two methodologies that leverage convex hull properties for anomaly detection, change tracking in complex systems, and enhancement of the sensitivity of classical process control charts. Experimental results indicate that the proposed approach achieves excellent performance in the early detection of deviations, positioning it as an innovative tool for advanced statistical monitoring in domains such as industrial production, healthcare, and security systems.
Classification | Mainly methodology |
---|---|
Keywords | Process Monitoring; Poisson Distribution; Price Optimization |
Primary authors
Sotiris Bersimis
(University of Piraeus, Greece)
Mr
Nikos Rokakis
(University of Piraeus, Greece)
Ms
Ekaterini Skamnia
(University of Patras, Greece)
Polychronis Economou
(Department of Civil Engineering, University of Patras)