Conceptualizing an Industry 4.0’s Predictive Maintenance System in a Medical Devices Manufacturing Enterprise

Abstract

AbstractThe research presents a case study from a medical devices manufacturing enterprise planning for preventive maintenance from the perspective of Industry 4.0. The research aims to generate a preliminary study in the enterprise onto maintenance and use the information later to plan for a predictive maintenance system. The preliminary study focused on 12 Computer Numerical Control (CNC) 5-axis milling machines, which run the most critical processes of the enterprise. A total of 82 breakdowns of these machines were detected and investigated for over 1.5 years. They were categorized and clustered on the basis of the suitable dimensions (frequency, duration, and organization financial loss). The findings reveal that 18 types of breakdowns constituted over 85% of the total breakdown. In total, 80% of the downtimes were not over 10 h. December was observed having the highest financial loss attributed to downtime. A causality analysis was performed, and the causes (parameters) were placed in three categories to underline the degree of real-time monitoring difficulty. The management of the enterprise deliberated on the results and conceived action plans, which involve development of a computerized maintenance system and vendor collaborations. On the basis of the concept of Health and Usage Monitoring Systems (HUMS), a conceptual predictive maintenance system is presented to provide a predictive breakdown and system modelling. The case study shows the enterprise’s endeavor for predictive maintenance planning. In terms of research and practical contribution, this research helps reduce the gap in literature and application by demonstrating an industry-based preliminary study onto the most common machine (e.g. CNC) in the case study company from the maintenance perspective.KeywordsBig dataCNC millingHUMSMachine learningMaintenanceIndustry 4.0IoTPredictive