We encounter large data sets more and more often in everyday life. The important thing here is to put the large amounts of data into a format in which we can read out patterns in order to better understand behavior and make predictions. Digital twins give us exactly this capability. They enable us to improve business performance and make well-informed decisions.
You may have heard the term “digital twin” several times in recent years. By making data available and analyzing it, we have more and more possibilities to discover statistical patterns. Digital twins are a digital replica of an asset that combines both digital and physical data to make accurate predictions about, for example, consumer behavior or the customer journey. This is a concept with which products, as well as machines and their components, are modeled with the help of digital tools.
This modeling includes all geometry, kinematics and logic data. A digital twin is the image of the physical ‘asset’ in real production and allows its simulation, control, evaluation and optimization. Via sensors, the twin is coupled with real objects, connecting the virtual and physical worlds through a wealth of information and data, enabling continuous real-time information and data exchange.
The technological maturity of this model has now reached far beyond basic applications. By 2022, market research firm Gartner expects the mass market to have adopted the concept of a digital twin and made it work for them. The concept holds great potential because it is not limited to a specific niche, but is applicable across the broad industry. Digital twins will be further developed in research and development in the coming years. And it is already becoming clear that this is not a monolithic data model, but rather different aspects of digital representations, functionalities, models and interfaces.
Benefits of Digital Twins
1.) Make more accurate predictions
Various models allow predictions about future use cases based on predefined scenarios. This allows strategic decisions to be made based on an accurate database.
2.) Performance tracking and optimization
Implementing a digital twin enables permanent monitoring and comparison of performance with previously collected data of the respective assets. This enables a continuous improvement process.
3.) A central data lake
Creating a digital twin requires a common data source, which forces companies to reorganize their methods of data collection and analysis. This in turn opens up additional digitization as well as optimization potential.
4.) Enabling automation
Decisions can be labeled and automated based on the insights of the digital twins. This way, the algorithm knows which scenarios were good and which were useless and can improve in the ongoing process. This reduces the number of iterations among decision makers.
5.) Visibility and transparency across all domains
Since digital twins combine multiple data sources, intra-asset effects and dependencies also become visible. This then results in further recommendations for action and potential savings.
What do specific use cases look like?
1.) Faster time to market
The first possible case is a faster time to market scenario. In the case of an automotive OEM, project staff were repeatedly blocked in the development process by sequential decisions. The development of a seat belt system sometimes took over three days, which ensured that the design time was massively drawn out. Here, a model of a digital twin was used with the limiting parameters and goals of the development team. In this case, the time to select the optimal design was reduced to less than one minute.
2.) Lower operational costs
Residents of Carson City, Nevada, USA, suffer from occasional water shortages during dry periods. The city is implementing digital twins to simulate future water supply. (based on peak consumer levels achieved to date) to ensure sustained water availability and reduce operational costs.
Digital twins provide insight into the real water balance of three different countries. Using this data and the resulting recommended actions, Carson City’s water supply was transformed so that it is now automatically controlled. As a result, 15% of the workload of the operational forces was saved and the reliable water supply for 50,000 residents was secured.
3.) Improved product or service development
Another use case can be found in the field of medicine. Doctors have difficulty reconstructing and interpreting the anatomy of patients’ hearts from a set of 2D images. A digital twin of a patient enables the creation of 3-D heart models, which provide in-depth analysis of the specifications and dynamics of individual hearts. A heart model reduces time to surgery by 82% compared to using 2-D medical images. So in this case, a digital twin can be a real lifesaver.
So what can we expect over the next five years?
Initially, there will be a democratization of simulation-based digital twins across all industries. The growing availability of digital twins as a service from major technology companies will increase the number of potential applications. Data is available in large quantities and the next logical step is to generate Smart Data from Big Data and understand the data. Lighthouse projects with a clear business advantage motivate other companies to follow suit.
In the next step, we will see improved end-to-end decision-making capabilities and improvement potentials.This is about connecting what we have learned from the digital twins across all application areas. This process spans from design and development to aftersales. The lessons learned are packaged and visualized to support management in their strategic decisions incorporating new user technologies. End-to-end coverage of digital twins enables optimized automation of processes, which in turn brings a second wave of efficiency gains.
The final development step within the next five years will be autonomous asset optimization in production.
Conclusion on Digital Twins
The boundaries between discrete and durable production, due to AI-driven process automation and machine control, will become increasingly blurred. Digital twins will act as smart software drivers for physical assets through integration into the Internet of Things (IoT). This will allow real-time analytics, feedback and implementation.
Digital twins are essential for Industrie 4.0 and the digitization of manufacturing. Their content is created in the different lifecycle phases of a product or factory, with different tools on diverse platforms. From the first examples in practice, it is already apparent that digital twins must be defined very application-specifically and tailored to each company. Properly applied, they have great potential for increasing efficiency and saving costs.