As high-tech systems become more capital-intensive, minimizing unplanned downtime is crucial. Diagnosing root causes of failures is increasingly difficult due to system complexity, especially when failures cause multiple effects (‘collateral damage’). These failures create an avalanche of unwanted effects, confusing service technicians.

The current approach of service technicians relies mainly on their individual experience, mental models, and trial-and-error methods. A major reason for this is the lack of available knowledge on the service side regarding system design. Traditionally, this knowledge has been located in the design and engineering departments. However, the current trend of digitalization and the adoption of model-based system engineering are making this knowledge more accessible, offering the possibility of model-based diagnostics. Such a method must be scalable and accurate, where accuracy means indicating the correct action plan in case of a system failure.

This project builds on a series of model-based diagnostics projects that TNO-ESI has carried out in recent years, primarily in the field of system failure diagnostics, and in collaboration with their industrial and academic partners.

The primary goal of Carefree 2025 is to further develop concepts for an iterative diagnostic approach. This approach will guide service engineers to collect actionable insights first and then identify the most likely root causes of machine-down problems. A secondary goal is to explore diagnostics for more complex issues, such as performance degradation, beyond the binary on/off failures of hardware components. Additionally, these diagnostic models can be used to assess the system’s diagnosability and to develop efficient and effective service procedures early in the design phase.