This deliverable presents an overview of the role and importance of data repositories in Industry 4.0 contexts, such as this project. Furthermore, this document explains the challenges and requirements arising when developing data repositories and provides some recommendations on how to address them, using the i4QDR as an illustrative example.

This deliverable describes the functionalities of the i4QDA solution, and the progress done so far in the development of the solution. Emphasis is given on the developments implemented on M18, when the solution will be applied to sample data of specific pilots, in order to cover the requirements outlined during WP1.

This task addresses a multi-tier infrastructure to address the management of AI-based models in a hybrid cloud edge manufacturing environment. One of the major challenges in this area is scalability. The aim is to reach a high scale by invoking techniques such as a policy-based distribution mechanism. This task supports AI at the edge operations, enabling developers to concentrate on their domain of expertise while this solution provides an automatic placement and deployment apparatus. The AI models deployment task is highly coordinated with the workload distribution mechanism such that AI workloads make use of the corresponding AI models.

This deliverable presents a technical overview of the i4Q Data Repository solution (i4QDR), including an explanation of its mapping against the i4Q Reference Architecture. Furthermore, it provides specific information regarding its implementation status up to M18, describing developments performed so far. In this regard, this document also explains the remaining implementation work, that will be addressed in the rest of task.

This deliverable describes the functionalities of the i4QBDA solution, and the progress done so far in the development of the solution. Emphasis is given on the developments implemented on M18, when the solution will be applied to sample data of specific pilots, in order to cover the requirements outlined during WP1

i4QEW is a toolkit for deploying and running AI workloads on edge computing environment, prevalent in manufacturing facilities, using a Cloud/Edge architecture. i4QEW provides interfaces and capabilities for running different workloads on different industrial devices, efficiently on the edge, including placement and deployment services. AI workloads at the edge are later used by the analysis components to run their inference close to the data sources. Target deployment environments may be very heterogeneous and dynamic, thus deployment needs to take a variety of criteria into consideration. The environment is dynamic thus re-deployment of the entire workload or the adaptation of the underlying model may be required while the workload is running. Deployment shall be based on well-known orchestrators, such as Kubernetes.

This deliverable describes the functionalities of the i4QDIT solution and the progress done so far in the development of the solution. Emphasis is given on the pipelines implemented on M18, when the solution will be applied to sample data of specific pilots, in order to cover some of the requirements outlined during WP1.

Deliverable "D4.4 - i4Q Analytics Dashboard” is a technical specification document, that covers the technical and development aspects of the i4Q Analytics dashboard solution (I4QAD). It describes in detail the role, the functionalities, and the conceptual architecture of I4QAD .

The deliverable D4.7 Infrastructure Monitoring is a Technical Specification document which provides an in-depth analysis and description of the functionalities, and the current implementation status of the i4Q Infrastructure Monitoring solution, while explaining how specific pilot requirements are addressed. It provides a detailed technical overview of the principal features/functional subcomponents of the solution, along with the description of the technical studies, designing, and experimental procedures conducted to implement a fully functional framework. The requirement engineering along with the Machine Learning algorithms designing and development are presented in detail. The experimental procedures conducted both on public and pilot datasets indicate that Infrastructure Monitoring is a highly efficient solution.