A Self-adaptative Architecture to Support Maintenance Decisions in Industry 4.0

Abstract

The Industry 4.0 era is primarily based on Internet of Things (IoT) devices that generate large amounts of data that can be analyzed to support decisions. However, the system that supports this approach must be highly adaptative and requires knowledge-based and reactive techniques to provide results, characterizing a self-adaptive software solution. In Industry 4.0, preventive maintenance planning is necessary to ensure equipment operation associated with IoT devices. This work presents a self-adaptative architecture to support preventive maintenance. We correlate failure and sensor data and use machine learning and ontologies to analyze these data. We also evaluated the feasibility of the solution in the textile Industry. As a result, we enriched decisionsupport information related to industrial equipment maintenance.