Special Session 6: Industrial Reliability Analysis and Optimization under Limited Data
Please submit your manuscript via https://easychair.org/conferences/?conf=icsrs2026 and select Special Session 6
Chairs
Brief Introduction
Industrial systems often have limited failure observations, which may be insufficient for reliable modeling and decision making. Recent AI methods, particularly large language models, can introduce engineering knowledge to support model development. Numerical simulation can provide additional information when failure data are scarce. This special session focuses on reliability analysis and optimization by integrating limited operational data with AI-based knowledge and simulation evidence. Particular interest is given to new applications enabled by such integration and to practical challenges, including knowledge reliability, model calibration, information inconsistency, uncertainty quantification, and decision optimization.
Sub-topics
1. Degradation modeling with sparse data
2. RUL prediction under limited degradation trajectories
3. Fault diagnosis with limited labeled failures
4. Process monitoring based on generative AI methods
5. Reliability assessment using simulation and field data
6. LLM-assisted reliability modeling
7. Model calibration with limited validation data
8. Condition-based maintenance under imperfect monitoring
9. Uncertainty quantification for prediction and optimization
10. Data-knowledge inconsistency in reliability modeling