The production centers of tomorrow will not only produce parts and products, but also will manufacture and benefit from endless sources of information.
It is estimated that in the coming years, the AI based automated systems will operate in a holistic and flexible way, allowing the workers, robot assistants and additive and substractive manufacturing systems, optimize the material and energy flows.
Thanks to the advances in algorithms and simulation technologies, SIEMENS, a leading global technology company, is able to create most of the products in a virtual world, thanks to the so-called “Digital Twins”. But as this process evolves, no only the geometric characteristics of an object can be developed, but also its functional characteristics, such as the coefficients of expansion and contraction or resistance to heat, and also its safety optimization, which is already being tested in the virtual world. SIEMENS estimates that the entire manufacturing processes are on the way of being developed, tested and optimized in this way.
The Digital Twin Data
But this does not end here. Once one object – from a gas turbine to a complete production facility – has been optimized in the virtual world and its physical model has been built, tested and operated in the real world, a new dimension opens up in the virtual world : data from the physical world can flow in order to refine and increase the accuracy of the original Digital Twin throughout the product life cycle. “The Digital Twin concept completes the knowledge cycle from design and testing to production and operation, and from data acquisition and analysis to improved service, and back to the start,” says Dr. Norbert Gaus, Chief of Research in Digitalization and Automation of Corporate Technology at SIEMENS.
For instance, this permits in the near future, once a product has been created in the virtual world, that its data will be transferred without interruption to production facilities where humans, assisted by semi-autonomous robots, will use manufacturing methods additive, subtractives and traditionals to automatically translate that data into physical objects. In addition, as these production steps are being performed, the products will be simulated in real time, allowing to compare performance models with real performance and thus to continuously improve quality and predictive maintenance.
These production plants will be Cyberphysical, which means that all their robots, machines and processes will operate as a self-organized IOT based on Artificial Intelligence (AI), which will constantly optimize material and energy flows within the production facilities.
An example of the extent to which artificial intelligence (AI) and neural networks can optimize a complex system, is its application to a SIEMENS Gas Turbine. “Our AI system is capable of reducing the emissions by an additional of 10% to 15%“, says Gaus. “This new manufacturing world will open the door to the production of affordable and individually produced parts, tailored to the demands and specific customer requirements, as well as the use of materials designed to increase the relationship between performance and weight of parts“, he adds.
For that reason, SIEMENS, through its laboratories around the world, keeps investigating and is already generating prototypes of this technology as a manufacturing solutions. To do this, the company is merging its knowledge of Big Data with the virtual and physical worlds of MindSphere, its cloud-based Internet of Things operating system, to open the way to digital planning (virtual reality) methods, additives manufacturing, software for robotic systems or new technologies for Industry 4.0 environments.
Source: SIEMENS Press