Predictive Maintenance in Industry 4.0 - Adopting Machine Learning to Overcome Challenges and Open Issues

Abstract

Predictive Maintenance is one of emerging concepts arisen with industry 4.0, playing a vital role in production systems and sustainable manufacturing by introducing maintenance with machine learning. Data gathered from production have increased significantly because of sensing technologies. Even though there are challenges related to organization, finances, and even repair and data source, Maintenance 4.0 has emerged as a strong point for organizations using it. Indeed, it minimizes costs and downtime associated with machine, increasing the life span of machine, and improves quality of production. Industry 4.0 has brought a significant change in manufacturing, forcing industries for adopting innovative approaches to streamline decision-making. Predictive Maintenance is an important component of this revolution, playing a central role by leveraging machine learning among other approaches for this transformation to optimize maintenance schedules, predict equipment failures, and improve operational efficiency. This study is based on the process of “Systematic Literature Review (SLR)” of 216 peer-reviewed papers published from 2019 to 2024, analyzing machine learning used for predictive maintenance in several industries, such as, machinery, manufacturing, smart systems and energy. Unlike previous literatures examining ML methods, this review offers structured taxonomy of approaches for predictive maintenance, highlighting their prevalence and domain-centric uses in industries based on safety.