Collaborative System Failure Prognostics via Federated Longitudinal-Survival Modeling
2026-07-28 • Other Computer Science
Other Computer Science
AI summaryⓘ
The authors developed a method that helps different organizations work together to predict when machines might fail without sharing their private data. They combined ways to learn patterns from sensor data over time with a special hazard model that can be trained separately by each group. Their approach was tested on engine data and did better than when groups worked alone, and almost as well as if all data was combined in one place. This shows their method can help multiple sites do better failure predictions while keeping their data private.
time-to-event modelingfederated learningCox proportional hazards modellongitudinal datadiscrete-time hazardremaining useful life (RUL)sensor datacondition monitoringprognosticsC-MAPSS dataset
Authors
Fan Yang, Madelyn Weller, Dimuthu Fernando, Hila Livneh, Yuxin Wen
Abstract
Time-to-event modeling provides a systematic framework for estimating time-dependent failure risk, reliability, and remaining useful life (RUL) from longitudinal condition monitoring data. However, applying these models to distributed prognostics remains challenging because sensor trajectories and failure-time records are often stored across organizations or operational sites and cannot be centrally pooled due to privacy or proprietary constraints. Moreover, the classical Cox proportional hazards model relies on a nonseparable partial likelihood involving global risk sets, making direct optimization difficult under standard federated learning protocols. This paper presents a federated longitudinal-survival modeling framework for collaborative system failure prognostics. The proposed framework combines longitudinal sensor representation learning with a client-separable discrete-time hazard objective, enabling multiple clients to collaboratively train a prognostic model without sharing raw sensor measurements or individual failure records. Time-dependent representations extracted from multivariate sensor histories are used to estimate interval-specific failure hazards, reliability curves, and system RUL. Experiments on the four C-MAPSS turbofan engine degradation subsets under simulated decentralized settings demonstrate that the proposed framework consistently improves prognostic performance over isolated local training while maintaining performance comparable to centralized training across heterogeneous operating conditions and failure modes. These results demonstrate the potential of federated longitudinal-survival modeling for collaborative, data-aware condition monitoring and system failure prognostics.