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

TL;DR

This paper presents a novel cost-sensitive transformer model tailored for failure prognosis in industrial settings, effectively handling class imbalance and missing data to improve predictive accuracy and operational reliability.

Contribution

The paper introduces a new cost-sensitive transformer framework with a hybrid resampler and regression imputer, advancing failure prediction in imbalanced industrial datasets.

Findings

Significant performance improvements over state-of-the-art methods.

Effective handling of class imbalance and missing data.

Validated on real industrial datasets from trucks and manufacturing.

Abstract

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, which also integrates a hybrid resampler and a regression-based imputer. After subjecting our approach to rigorous testing using the APS failure dataset from Scania trucks and the SECOM dataset, we observed a substantial enhancement in performance compared to state-of-the-art methods. Moreover, we conduct an ablation study to analyze the contributions of different components in our proposed method. Our findings highlight the potential of our method in addressing the unique challenges of failure prediction in industrial settings, thereby contributing to enhanced reliability and efficiency in industrial operations.