Application of machine learning and artificial intelligence technologies for process data analysis

The objective of this activity is to develop an algorithm for manufacturing process robustness, anomaly detection and failure modes investigation in space applications. 

Over the past years, the tendency to capture huge volumes of historical data describing process operations together with complex experimental datasets has become a reality. In this regard, the use of artificial intelligence (AI) and in particular machine learning (ML) for data mining addresses the question of which is the best way to use this historical data to discover regularities and to facilitate future decisions. Following the significant progress and recent success in many science and engineering domains, the activity is aiming at exercising AI/ML technique to provide substantial benefits to the advanced manufacturing processes domain (e.g. additive manufacturing). Advanced and new manufacturing processes and technique may benefit from the integration and operational usage of AI/ML techniques. The use of these technologies might be beneficial to extract relevant information from big data generated through the different steps of the new advanced manufacturing processes. Furthermore, the falling cost of large data storage devices and the increasing facility of collecting data over networks; the improvement of computational power, enabling the use of computationally intensive methods for data analysis in parallel to the development of robust and efficient machine learning algorithms further highlight the need and actuality of this activity. In the frame of this activity, a software prototype shall be developed, able to extract data from existing sources and categorize them in useful information in order to the use the state of the art AI/ML algorithms for advanced manufacturing processes modelling, process anomalies detection and failure mode investigation. The main steps that shall be completed in the frame of this activity are the following:Identification of case studies (e.g. additive manufacturing process modelling, in situ monitoring, NDI inspections, defect identifications etc.) and related data sources available;Review of the state-of-the art, preliminary use cases selection, definition of the preliminary requirements and identification of the validation process to be applied;Design and definition of the software preliminary architecture;Detailed design and definition of the software architecture;Software Implementation and integration;Software testing and performance assessment;Software validation and risk assessment based on the reliability of the software in the decisional process.