The wide-scale adoption of Industrial IoT (IIoT) and Industry 4.0 methodologies is spurring interest in digital twins for smarter manufacturing.
Digital twins complemented by IoT sensors on physical systems provide a rich environment for innovative continuous intelligence applications based on real-time analytics. The potential benefit of this combination of technologies is driving the adoption of digital twins to new levels.
Specifically, the wide-scale adoption of Industrial IoT (IIoT) and Industry 4.0 methodologies is spurring interest in digital twins. IIoT brings together connected intelligent devices and analytics in a way that allows organizations to monitor, collect, exchange, analyze, and deliver valuable new insights about their systems and processes. These insights can help drive smarter, faster business decisions.
See also: Germany Turns Manufacturing into an IoT Art Form
Making use of the large volumes of IIoT data, which are continuously generated, requires new thinking. Many companies are complementing the traditional analysis with real-time, artificial intelligence (AI) and machine learning (ML) algorithms to get decision-making information out of the data in a timeframe that will allow a company to take pro-active actions.
When a physical system is paired with a digital representation, such analysis can be a game-changer. IoT-extended AI brings benefits to manufacturers in different aspects of their operations, including smart manufacturing, supply chain optimization, and product or service innovation.
Manufacturing digital twins used for product design and development are anticipated to experience a compound annual growth rate (CAGR) of over 31% through 2025 owing to the implementation of Industry 4.0 by the automotive and manufacturing sectors.
The manufacturing process support and service application segment is expected to witness similar growth at a CAGR of over 30% through 2025. Here, the use of digital twins enables efficient inventory optimization by reducing inventory carrying costs and facilitating improved customer services.
The automotive sector has also been undertaking efforts to deploy digital twin technology to take advantage of the substantial benefits being offered by the technology. For instance, Tesla Motors has been investing in digital twin technology to provide better service and reliability to car owners. The company creates a digital twin of every car it sells and then updates software based on the individual vehicles’ sensor data. This data-driven software development process enables more efficient resource allocation and a better user experience for the vehicle owner.
A Wing and a Pray
Digital twins and real-time analytics are being used in an innovative way in a new project in the UK. VADIS, a project being carried out by the University of Nottingham and aerospace company Electroimpact, aims to improve production in an aircraft assembly line. The project uses intelligence derived in real-time from sensors to adjust the construction process for each part.
In the project, researchers measure the holes that need to be drilled in a wing skin and use those measures to update a digital model. This way, they can adjust the construction process to a specific component.
The process helps get over the deficiencies in past product processes. Specifically, some production processes are still done manually because of the lack of digitalization in the industry, which, in turn, can leave room for error. In wing production, for example, holes may not align exactly as they should, which then requires last-minute adjustments on the assembly line, leading to loss of time.
To address this, VADIS is constructing a frame in which the aircraft skins can be scanned by sensors. This will then be used to create a digital twin of the wing, with all its surfaces and holes registered in the smallest detail. This digital twin would then be used to build the corresponding parts so that they align seamlessly for that specific wing.
The new system will mean that components can be manufactured very precisely off-site, and the operators only need to worry about assembly. They don’t have to worry about re-drilling and re-working at the last minute. In this way, the work of the operators becomes a lot easier.
VADIS aims to be able to replicate a digital wing skin of up to 10 meters long, with an accuracy of 0.06 millimeters.
Architectural Requirements for Success
Such efforts are the tip of the iceberg when it comes to combining continuous intelligence generated from real-time analytics and AI with digital twin technology.
Given the range of possible benefits when these technologies are integrated, it is obvious that companies will need to support a broad set of analytic requirements for IoT applications, including real-time insights, ML, streaming analytics, and high-performance transactional processing.
By selecting the right architecture—one that plugs into an existing infrastructure without disruption—such benefits can be realized. The right operational analytics architecture makes all the needed data available in one place for line-of-business analytics, executive insights, reporting, compliance, governance, and data science that uses ML and AI. (See: Infrastructure Architecture Requirements for Continuous Intelligence.)
