Tiresias 2027
From Reactive Troubleshooting to Predictive Management of High-Tech Systems
High-tech systems are becoming increasingly complex, while demands for performance, quality, and availability continue to rise. Performance issues often result from interactions between hardware, software, processes, and environmental conditions, making them difficult to diagnose. At the same time, the availability of experienced diagnostic experts is declining. As a result, industry is moving from reactive troubleshooting towards predictive system management, where degradation is detected early and maintenance can be planned proactively. Tiresias develops new methods that combine AI with engineering knowledge to detect, explain, and ultimately predict performance degradation.
AI and Engineering knowledge combined for smarter diagnostics
Tiresias develops a hybrid diagnostic approach that combines data-driven AI with explicit engineering and system knowledge. AI methods are used to automatically identify recurring degradation patterns in operational data, while knowledge-based models help explain these patterns and relate them to their underlying causes. This results in a powerful and explainable diagnostic framework that enables faster root-cause analysis, better-informed maintenance decisions, and a more proactive approach to lifecycle management of complex high-tech systems.