Unexpected  machine failures can lead to costly and even catastrophic consequences. In order to avoid them, the use of intelligent tools that analyze the machine condition and behavior is necessary. Such condition monitoring systems can then provide automatic health indicators and alarms for faulty behavior and even predict a future failure or the remaining useful life of the equipment. Analyzing machine condition is becoming possible due to the fast evolution of advanced sensors, data collection and storage systems and intelligent data analytical tools.

At the smart maintenance team of IDP we work together with our partners from various industry and public sectors in order to develop AI-based tools for fault detection, diagnosis and prognostics of machine condition, which are tailored to their specific application. We then use these tools for the optimization of maintenance decision making. Our smart maintenance algorithms utilize methods ranging from statistical analysis through machine learning and deep learningalgorithms in combination with physics-based models.

• MSE Module Lifecycle-Management von Infrastrukturen
• Bachelor Module Instandhaltung (Verkehrsysteme)
• Bachelor Module RAMS (Verkehrsysteme)
• CAS Instandhaltungsmanagement
• CAS Industrie 4.0

• Physics-Informed AI for Energy Loss Diagnostics in Solar Power Plants
  We are developing a software module for intelligent detection and diagnosis of energy losses in grid-connected PV systems.
• Decision support system for predictive maintenance of laser cutting machines (Bystronic Laser AG)
  We are developing a new data-driven decision support system for predictive maintenance of laser cutting machines, with various software modules integrated into the Bystronic IoT system.
• Machine Learning Based Fault Detection for wind Turbines (Nispera AG)
  We are developing intelligent machine learning algorithms for fault detection and predictive maintenance of critical components of wind turbines.