Toward Digital Twin for Cyber Physical Production Systems Maintenance: Observation Framework Based on Artificial Intelligence Techniques

Abdoune, Farah; Nouiri, Maroua; Castagna, Pierre; Cardin, Olivier

Manufacturing Systems are considered complex engineering systems given the large number of integrated entities and their interactions. Unplanned events and disruptions that can happen at any time in real-word industrial environments increase the complexity of manufacturing production systems. In the fourth industrial revolution (so called Industry 4.0), the industrial sector is rapidly changing with emerging technologies like Cyber-Physical Production System (CPPS), Internet of Thing (IoT), Artificial Intelligence (AI), etc. However, the efficiency and reliability of these systems are still questionable in many circumstances. To address this challenge, an observation framework based on AI techniques aimed at elaborating predictive and reactive planning of the maintenance operations of CPPS is proposed in this paper. The proposed tool aims to improve the system’s reliability and helps the maintenance supervisors to adjust maintenance decisions. In order to assess the performance of the proposed tool, a case study on an industry-type learning factory is considered. A proof of concept shows the efficiency of the framework.