This chapter presents a design for a Situation-Based Maintenance Model (SBMM) that explains different statistical approaches to predict maintenance. It also gives some example applications to help grasp predictive maintenance before exploring the possible big data models that can predict when maintenance work is most needed. The high-level architecture that reflects the big data predictive maintenance model is presented for the proven potential of future industrial predictive maintenance systems. The growing interest in Industry 4.0 has driven the creation of systems that are capable of real-time data generation. Many different industrial areas can benefit from this grand concept, and analytics is an important area of Industry 4.0. Whether it is structured data from Enterprise Resource Planning (ERP) or Customer Relationship Management (CRM) systems, unstructured data from sensors and machines, or new types of data generated from Radio Frequency Identification (RFID) devices or the Internet of Things (IoT), processing and analyzing extremely large datasets is a challenge that needs to be mastered. This transformation can be achieved through Big Data Analytics. These analytics combine statistical data analysis techniques, models, and algorithms with human ingenuity to yield new insights and optimized decisions.
Keywords: Big data analytics, Industry 4.0, Internet of Things (IoT), Predictive maintenance, Predictive models.
