The system of PRECognition combines novel and advanced prediction and fault detection models and constitutes a complete production equipment health monitoring solution for the factory of the future. The project’s purposes are to create a predictive maintenance system that will:
- Detect/predict the risk of failures and effectively manage their possible occurrence
- Reduce fault recovery time and maintenance work, to the appropriate time/scheduling according to the needs of production, while ensuring the smooth operation of the equipment
- Extend the lifetime and efficiency of the production equipment
The PRECognition system consists of the following three modules:
- Incident Prediction and Detection (IPD):
The Incident Prediction and Detection module analyzes the equipment deterioration, which may lead or has already led to failures. More specifically, it focuses on identifying correlations between the normal and abnormal operation of a machine or component, in order to predict the likelihood of a fault, the time horizon of a fault, and the diagnosis of current or ongoing fault events. For this purpose, it interacts with a big data analytics module, while IIoT (Industrial Internet of Things) and big data analytics techniques are applied. The key sub-components of the IPD module are: Fault Detection, Fault Prediction, Results Fusion. - Online Maintenance Manager (OMM):
The Online Maintenance Manager module is the core component of the PRECognition system, as it orchestrates and manages the entities involved and the events failures, damages, maintenance work, etc.) that a Maintenance department of a factory is called upon to deal with during the production procedure. It essentially performs the role of an “Electronic Maintenance Assistant”, which supports, and if necessary, replaces the Maintenance Manager of a factory’s Maintenance department. The main modules of the OMM sub-component are: Ontology, Management of faults and events, Alarm Management and Consulting for operation management and optimization. - Notification and Feedback:
The Alert and Feedback module has the role to create alerts for the appropriate production actors when fault incidents are in progress or be predicted by the IPD module. Additionally, it identifies dynamically who should be notified for each event and what volume of information data is required for each actor. The main sub-components of the IPD module are the following: Recipients Feedback and Data Package creation for Notification.
The expected key results of the project include:
- 15% increase in Total Equipment Yield and Return on Investment (ROI)
- Extension of equipment lifetime by 20%, through reliability, MTBF and MTTR improvement
- Reduction of downtime by 55%
- Contribution to the Digital Transformation of Industry in the context of the 4th Industrial Revolution (Industry4.0)
- Reduction of accidents due to equipment and improvement of working conditions in the production area
