This paper aims to provide a systematic review of how agent-based modeling (ABM) has been applied in maintenance contexts across various industries. By consolidating findings from a range of peer-reviewed studies, this review identifies current research trends and pinpoints underexplored opportunities for future work.
A systematic literature review was carried out following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework. Relevant papers were gathered from Web of Science and supplementary sources such as Google Scholar, applying predefined search queries and exclusion criteria. A final set of 69 papers was analyzed in detail, focusing on industry applications, methodological combinations, software usage and stated future research directions.
The review reveals that ABM and multi-agent systems (MAS) approaches are widely used, in the context of maintenance, in energy, infrastructure, aviation and manufacturing, often integrating simulation, optimization and machine learning techniques. Less attention has been given to healthcare, oil and gas and non-aviation military maintenance, highlighting areas with significant potential for further study. Researchers emphasize the need for real-world validation, larger-scale modeling and the incorporation of human and environmental factors. Additionally, variations in software choice reflect differing priorities, from visualization and simplicity (e.g. NetLogo) to robust coordination in multi-agent environments (e.g. JADE) and comprehensive, hybrid simulation capabilities (e.g. AnyLogic).
This review uniquely synthesizes current research on ABM in maintenance, providing a comprehensive overview of its applications, benefits and future research directions.
