What is it about?

The study incorporates machine learning algorithms with recursive feature elimination and exposes latent degradation assuming hydraulic pump degradation, which was not observed by lubricant condition monitoring. The study shows that unobserved degradation led to >20% loss based on the initial state in the experiment.

Featured Image

Why is it important?

The idea is to transition towards easy-to-access and low-cost indicators such as pressure and flow for gaining insight into the state of the system with the help of a machine and deep learning algorithms.


The researcher's idea is to continue investigating sustainable maintenance perspectives using cheap and easy-to-access indicators; however, with the required help of a machine and deep learning algorithms.

Dr Marko Orošnjak
University of Novi Sad, Faculty of Technical Sciences

Read the Original

This page is a summary of: From predictive to energy-based maintenance paradigm: Achieving cleaner production through functional-productiveness, Journal of Cleaner Production, July 2023, Elsevier,
DOI: 10.1016/j.jclepro.2023.137177.
You can read the full text:



The following have contributed to this page