Public Deliverables and Publications

Public Deliverables


D4.4 Data-driven engineering methods and techniques

This deliverable is the final and public version of the WP4 series of deliverables on data-driven engineering methods and techniques in IVVES. It reports on the baseline for existing solutions and techniques provided by technology providers, and initial state of industrial use-cases.


D6.1 IVVES Dissemination plan

This document describes the plan for using and disseminating the knowledge in the context of the IVVES project, through various means including internal and external communication channels, the distribution of dissemination material and participation in dissemination activities.

D5.1 Requirements for the IVVES experimentation framework

By making sure that all the IVVES outcomes come with a well-defined and clear package of training and possibilities to try them out (experiment with them), we hope to give dissemination of the IVVES outcomes a real boost.

D4.1 Data-driven engineering: state of the art & gap analysis

WP4 in IVVES, titled ‘Data-Driven Engineering’, focuses on implementing solutions for identifying data correlations and behavioural patterns throughout the entire product life cycle with respect to component failures, and with the ultimate goal of enabling predictive maintenance and anomaly detection.

D3.2 Validation methods and techniques for evolving systems considering use case requirements

In this deliverable we present validation methods and techniques which we have been developing or anticipating to adapt to the use cases. We selected existing validation methods and techniques we plan to adapt and implement in existing tools developed by partners.

D3.1 – State of the art of validation methods and techniques for complex evolving systems

This report introduces the idea of continuous quality assessment process which spans on entire ES lifecycle and maps methods, techniques and existing tools helping partners to navigate in the domain and apply right approach for the right lifecycle stage.

D2.2 – Training data quality

Data quality is an essential subject considering trustworthiness of ML systems. Data quality issues can be caused by many reasons, like missing data/values, outliers in the data or imbalance of the data. To respond to these data quality issues, we introduce methods for generating simulated data and validating incoming data in this chapter.

D2.1 – State of the Art on Validation Techniques for ML

Due to the popularity of applying machine learning and artificial intelligence to solve an ever-increasing variety of challenges, the need for verifying and validating models is also increasing accordingly. In order to meet this growing need, different approaches focusing on model interpretability, data augmentation and various testing approaches will be discussed in this document.


Solution Development Partners

Service Provider Partners