Industrial Cyber-Physical System Evolution Detection and Alert Generation

• Main Authors: Aitziber Iglesias, Goiuria Sagardui, Cristobal Arellano
• Format: Article
• Language: English
• Published: MDPI AG 2019-04-01
• Series: Applied Sciences
• Subjects: Cyber-Physical Systems (CPS); scalability test; Internet of Things (IoT)
• Online Access: https://www.mdpi.com/2076-3417/9/8/1586

Industrial Cyber-Physical System (ICPS) monitoring is increasingly being used to make decisions that impact the operation of the industry. Industrial manufacturing environments such as production lines are dynamic and evolve over time due to new requirements (new customer needs, conformance to standards, maintenance, etc.) or due to the anomalies detected. When an evolution happens (e.g., new devices are introduced), monitoring systems must be aware of it in order to inform the user and to provide updated and reliable information. In this article, CALENDAR is presented, a software module for a monitoring system that addresses ICPS evolutions. The solution is based on a data metamodel that captures the structure of an ICPS in different timestamps. By comparing the data model in two subsequent timestamps, CALENDAR is able to detect and effectively classify the evolution of ICPSs at runtime to finally generate alerts about the detected evolution. In order to evaluate CALENDAR with different ICPS topologies (e.g., different ICPS sizes), a scalability test was performed considering the information captured from the production lines domain.