A Deep Learning Model for Predictive Maintenance in Cyber-Physical Production Systems Using LSTM Autoencoders

TLDR

Condition monitoring combined with machine learning can improve maintenance in cyber‑physical production systems, but high‑quality, extensive data covering normal and abnormal operations are hard to obtain non‑destructively. This study aims to shift maintenance from scheduled preventive to predictive by enabling planning based on actual machine status. The authors employ LSTM autoencoders that classify real‑world sensor data into condition labels and estimate remaining useful life, trained and tested on manufacturing operation data. The method was evaluated on a steel‑industry production process, demonstrating its applicability.

Abstract

Condition monitoring of industrial equipment, combined with machine learning algorithms, may significantly improve maintenance activities on modern cyber-physical production systems. However, data of proper quality and of adequate quantity, modeling both good operational conditions as well as abnormal situations throughout the operational lifecycle, are required. Nevertheless, this is difficult to acquire in a non-destructive approach. In this context, this study investigates an approach to enable a transition from preventive maintenance activities, that are scheduled at predetermined time intervals, into predictive ones. In order to enable such approaches in a cyber-physical production system, a deep learning algorithm is used, allowing for maintenance activities to be planned according to the actual operational status of the machine and not in advance. An autoencoder-based methodology is employed for classifying real-world machine and sensor data, into a set of condition-related labels. Real-world data collected from manufacturing operations are used for training and testing a prototype implementation of Long Short-Term Memory autoencoders for estimating the remaining useful life of the monitored equipment. Finally, the proposed approach is evaluated in a use case related to a steel industry production process.