Predictive maintenance using deep reinforcement learning.

Supramaniam, Aravind and Syed Ahmad, Sharifah Sakinah and Mohd Yusoh, Zeratul Izzah (2024) Predictive maintenance using deep reinforcement learning. In: 2024 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET).

Text Predictive Maintenance using Deep Reinforcement Learning.pdf Download (818kB)

Abstract

The Fourth Industrial Revolution has impacted various sectors significantly. In machine and equipment- related industries, ensuring an uninterrupted supply to customers is crucial. Therefore, machines and equipment must always be ready and undergo maintenance at the right time to prevent downtime. However, industries believe that simply predicting failures based on historical data is insufficient for effective predictive maintenance. The prediction model must learn from the actions taken on the machines or equipment to develop an optimal maintenance policy. Deep Reinforcement Learning (DRL), a combination of Deep Learning (DL) and Reinforcement Learning (RL), harnesses the strengths of both techniques to handle complex and high-dimensional data. Despite its potential, DRL's application in Predictive Maintenance (PdM) is still emerging. Hence, Predictive Maintenance using Deep Reinforcement Learning (DRL) is proposed. By leveraging the advantages of deep learning (DL) and reinforcement learning (RL), it is possible to predict appropriate maintenance timing and determine the optimal maintenance policy. This paper will discuss the performance of Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) in prediction and policy optimization.

Item Type:	Conference or Workshop Item (Paper)
Divisions:	Faculty of Information and Communication Technology
Depositing User:	NURHASHIRAH BORHAN
Date Deposited:	31 Jul 2026 06:51
Last Modified:	31 Jul 2026 06:51
URI:	http://eprints.utem.edu.my/id/eprint/30093
Statistic Details:	View Download Statistic

Actions (login required)

View Item