Achieving Predictive Precision: Leveraging LSTM and Pseudo Labeling for Volvo's Discovery Challenge at ECML-PKDD 2024

TL;DR

This paper describes a method combining LSTM networks and pseudo-labeling to improve predictive maintenance accuracy for Volvo trucks, achieving a high F1-score and demonstrating effective machine learning application in industrial contexts.

Contribution

The paper introduces a novel combination of LSTM and pseudo-labeling techniques tailored for predictive maintenance in industrial settings, with a focus on data processing and iterative labeling.

Findings

Achieved a macro-average F1-score of 0.879.

Demonstrated the effectiveness of LSTM and pseudo-labeling in predictive maintenance.

Provided insights into applying machine learning in industrial environments.

Abstract

This paper presents the second-place methodology in the Volvo Discovery Challenge at ECML-PKDD 2024, where we used Long Short-Term Memory networks and pseudo-labeling to predict maintenance needs for a component of Volvo trucks. We processed the training data to mirror the test set structure and applied a base LSTM model to label the test data iteratively. This approach refined our model's predictive capabilities and culminated in a macro-average F1-score of 0.879, demonstrating robust performance in predictive maintenance. This work provides valuable insights for applying machine learning techniques effectively in industrial settings.