InstandXAI: Industrial Application of Explainable Artificial Intelligence: Predictive Maintenance Planning for Hydraulic Presses Using Learned Pattern Recognition and Contextualized, Personalized Human–Machine Interaction

High-performance predictive maintenance of hydraulic presses aligned with production planning currently requires highly specialized personnel. Commercially available software tools deliver real value only when used with a solid understanding of the underlying calculations and in consideration of complex operational constraints. As a result, the practical suitability and application relevance of such software—especially for small and medium-sized enterprises—remain limited.

The project aims to develop a new technology for AI-based maintenance planning that provides contextualized explanations of computed results and enables interactive assessment of alternative planning options. This will allow maintenance optimization that delivers a major leap in usability and practical benefit through algorithmic generation of individualized rationales and explanations of the results. 

Subproject: Research and software implementation of context-aware, interactive explanation methods for AI-based analysis results