INDUSTRIAL DATA PLATFORMS
FOR THE MANUFACTURING DOMAIN
Two chapters of the DEI WG2 final report (https://ec.europa.eu/futurium/en/dei-implementation) are devoted to "Digital Platforms for Manufacturing Industry" and "Industrial data Platforms". The workshop elaborates on the concept of these platforms in different contexts, from the industrial perspective case to results from pilots in research and innovation funded projects. Each one of these speeches is intended to provide actual cases to the audience in order to demonstrate the concrete achievements obtained.
Smart Manufacturing Industry
Mr. Sergio Gusmeroli (Engineering, Italy)
Mr. Davide Dalle Carbonare (Engineering, Italy)
Since 2016, the Big Data Value Association is running a dedicated working group on Smart Manufacturing Industry that produced a position paper that report an analysis of the challenges identified in three main manufacturing scenarios (Smart Factory, Smart Supply Chain, Smart Product Lifecycle) in relation to the BDV SRIA technical priorities. (presentation)
Digitalization needs Interoperability – Interoperability requests standardization.
The way to do it.
Mr. Thomas Hahn (SIEMENS, Germany)
In the next five to ten years, the real and virtual world will continue to merge. The entire value-added chains will be digitalized, integrated and connected from product design up to on-site customer service – across locations, companies, national boundaries and time zones. Data, Data Economy or generally spoken the importance getting value out of the data is increasing and has a very high relevance for all industries. All this needs interoperability and interoperability needs standards and harmonized policies. Notwithstanding, the creation of and compliance with binding international standards and policies is of central importance to the sustainability of solutions and thus is a competitive strength. What are the circumstances, how can this can be addressed and what organization/teams/tools are needed will be content of the presentation. (presentation)
Securing IoT and Big Data. Exploring a huge iceberg by some illustrative projects.
Dr. Aizea Lojo (IK4-Ikerlan, Spain, Researcher)
Dr. Iñaki Garitano (Mondragon Unibertsitatea, Spain, Researcher/Lecturer)
As in many other domains, cybersecurity is essential in the manufacturing domain. Big Data analysis requires tons of data and multiple processes take part in the generation, acquisition, ingestion and analysis of this data. Even if the analysis is an essential part in order to extract knowledge and value from data, it is just the visible part of a huge iceberg, and cybersecurity is traversal to all of them. IoT devices and sensors are essential to generate data and their orchestration becomes necessary, specially when their number increases and their management requires many resources. Three different and manufacturing related European projects, MATIS, ARROWHEAD and CREMA, and their main outcome will be described within this presentation, with a special focus on cybersecurity. (presentation)
Predictive maintenance for Aerospace Products Manufacturing
Prof. Ernesto Damiani (CINI, Italy, TOREADOR Principal Investigator)
Dr. Mariangela Lazoi (Università del Salento, Italy, TOREADOR Aerospace Pilot coordinator)
DTA use-case of Toreador project aims to build a model-based big data solution for predictive maintenance. The goal is to leverage the huge amount of data, associated with production and inspection phases and coming from different sources, in order to reduce time to market and improve general plant performance. Nowadays, a number of faults is still discovered during the man-carried inspection phase, meaning the scraping of a piece with the resulting cost loss. Operating predictive maintenance would mean discovering the cause of a fault and correcting the system. (presentation)
Steel Production Optimization.
Mr. Marcos Sacristán (Treelogic, Spain, PROTEUS project coordinator)
The steel industry is strategic in the EU economy. Steelmaking is a complex industrial process: the life-cycle of steel is long, from raw material extraction to coating of final products. Defects introduced in early stages have an economic impact in posteriors transformations, thus the sooner defects are detected, the sooner the process can be modified, redefining industrial routes. In this scenario, the Hot Strip Mill process involves multiple and diverse parameters (e.g. temperature, vibration intensity, tension in the rollers, speed) that affect the final dimensional properties of the coil. Data is obtained in real time from a sensor network. Predicting dimensional defects using the massive streaming real-time ‘big’ data is the main target. The use of Big Data technologies supports this objective, in particular using scalable online machine learning for predictive analytics, together with advanced real-time interactive visualization techniques. (presentation)
Manufacturing Industry, Digital Platform, Smart Factory, Smart Supply Chain, Smart Product Lifecycle, Interoperability, Standardization, IoT, Cybersecurity, Predictive Maintenance, Production Optimization
SESSION SPEAKERS / CONTRIBUTORS