This repository contains the code of the paper "Comparative Analysis and Evaluation of Aging Forecasting Methods for Semiconductor Devices in Online Health Monitoring" by Adrian Villalobos (Mondragon University, Spain), Iban Barrutia (Mondragon University, Spain), Rafael Peña-Alzola (University of Strathclyde, UK), Tomislav Dragicevic (DTU - Technical University of Denmark), and Jose I. Aizpurua (University of the Basque Country and Ikerbasque Basque Foundation for Science, Spain), published in Engineering Applications of Artificial Intelligence.
The analysis is based on the NASA's MOSFET dataset [1]. The Code folder includes Jupyter Notebooks with (i) short-term Temporal Fusion Transformer (TFT) experiments (ii) long-term Temporal Fusion Transformer (TFT) experiments (leave one out) and (iii) extraction of explainability features and attention points from the TFT.
[1] J. Celaya et al. "MOSFET Thermal Overstress Aging Data Set". In: NASA AMES Prognostics Data Repository (2007). URL: https://www.nasa.gov/intelligent-systems-division/discovery-and-systemshealth/pcoe/pcoe-data-set-repository/
Villalobos, A., Barrutia, I., Peña-Alzola, R., Dragicevic, T., & Aizpurua, J. I. (2025). Comparative analysis and evaluation of ageing forecasting methods for semiconductor devices in online health monitoring. Engineering Applications of Artificial Intelligence, v. 150, n. 110545. https://doi.org/10.1016/j.engappai.2025.110545
Please click on the "Cite this repository" button on the right side to generate the bibtex file.