A Cost-Sensitive Transformer Model for Prognostics Under Highly Imbalanced Industrial Data

The rapid influx of data-driven models into the industrial sector has been facilitated by the proliferation of sensor technology, enabling the collection of vast quantities of data. However, leveraging these models for failure detection and prognosis poses significant challenges, including issues like missing values and class imbalances. Moreover, the cost sensitivity associated with industrial operations further complicates the application of conventional models in this context. This paper introduces a novel cost-sensitive transformer model developed as part of a systematic workflow, …

challenges class collection cost cs.ai cs.lg data data-driven detection enabling failure industrial missing values sector sensor technology transformer transformer model values vast