During the past two decades, fields including those of manufacturing, urban planning and healthcare have been using digital twins to manage product life cycles, smart cities and health informatics. But what are digital twins, where do they come from, and how do they work?
What is a digital twin?
A digital twin is a virtual replica of an object or system. Digital twins allow users to run simulations of real situations to see the results of their decisions. Spanning the life cycle of the systems they replicate; the twins use machine learning and reasoning to help decision making. Digital twins operate continuously, updating and adapting in real time, bringing together large amounts of data and complex models, and presenting these in a way that is simple to understand.
Users can enter their variables and see the effects of their decisions on a three-dimensional model. Built on near real-time, historical as well as forecasted changes, all modern digital twins use Artificial Intelligence so users can manipulate and extract data as they need.
Where do they come from?
Digital twins, also known as virtual twins or avatars, have a background in both engineering and the space program with the concept of ‘twinning’ used at a very basic level by NASA during the 1960s.
The first physical twin was crucial to the famous Apollo 13 rescue mission in April 1970. NASA’s digital twin model of Apollo 13 back on earth was a key element of the rescue after oxygen tanks exploded on the space craft threatening the lives of the three astronauts on board. Engineers were able to use the twin to test possible solutions for the drama unfolding over 200,000 miles away.
The concept of the digital twin was suggested for product lifecycle management in the manufacturing industry in the early 2000s. They have now moved from the physical to the virtual with the introduction of The Internet of Things and Cloud computing. Although the idea has been around for several decades; it was only with the advent of the Cloud that the digital twin became more popular and prevalent.
How do they work and where does the data come from?
Digital twins are designed around a two-way flow of information. Information flows from the object or system being mirrored to the system processor that generates the twin. Users run their simulations in the digital twin and the insights and learnings gained from this process can be used to make corrective and preventive actions or decisions on the physical system.
Digital twins use data from a wide variety of sources depending on what systems they mirror. This can include Earth observation data collected from satellites, data from existing in situ information systems, historical data, predictive analysis such as weather forecasting, and even ground level observations.
The information gathered can be real-time updates, historical data, or predictions such as those generated by weather forecasting.
Who uses it and how?
Today digital twins are found almost everywhere. They are prevalent in manufacturing, urban planning, architecture, construction, healthcare, the automotive industry, and to map hydrological systems.
Designers, decision makers, policy planners, economists, scientists, managers, and students are among the many users.
Users can run ‘what if’ simulations, usually via a dashboard, of what could occur in reality if certain decisions were made and acted on, or if variables suddenly changed and were experienced. Users can test scenarios in multiple ways before implementing a new development in the real world allowing them to make better decisions.
What’s the difference between a digital twin and virtual reality?
Digital twins and virtual reality have different purposes and distinct characteristics. The key differences between these technologies lie in their visualization, interaction, data source, industry applications, and hardware.
Virtual reality provides a simulation where the user is immersed in a particular environment and is able to interact with a virtual world. Â Virtual reality provides an immersive experience that can be used for gaming, education, healthcare, manufacturing, and military training.
A digital twin is a virtual replica of a physical object or system. It is dynamic, receiving real-time updates from the object or system that it mirrors. It is this dynamic nature that allows users to perform in-depth analysis, monitoring and predictions.
Digital twin and virtual reality technologies can, however, be used together. One company used a digital twin to replicate farmlands and virtual reality to visualize those farms. Farmers are able to virtually ‘walk’ through their fields, visualize potential problem areas, and make informed decisions about irrigation, fertilization and other operations.
What is the role of AI in digital twins?
Artificial Intelligence (AI) refers to a computer system capable of performing complex tasks that historically only a human could do, such as reasoning, making decisions, or solving problems. Â All digital twins now incorporate some form of AI.Â
AI has the capability of providing additional insights beyond the data the sensors feed into the twin. For example, they can make future predictions of the state of the physical twin, as well dynamically self-adjust to independently determine optimal pathways towards a set of outcomes.
AI occurs when the digital twin has the ability to understand and respond to what it is modelling. A digital twin reaches its full potential when it can issue instructions based on its findings.
Tesla, for instance, creates a digital simulation of every one of its cars. These allow the company’s AI algorithms to determine where faults and breakdowns are most likely to occur and minimize the need for repairs.
How are digital twins used in water management?
Using digital twins in water management is a very new concept. Similar more basic systems have been used in waterworks, waste waterworks, dams, and distribution systems, but digital twins are just starting to find their way in water resources management.
They could provide innovative solutions to tackle the complex challenges of integrated, cross-sectoral, and ecosystem-based water management. Their ability to link real-time data and information with action could be very useful when creating water management policies.
Digital twins could be used to monitor and predict floods and drought and support early warning systems. They could be used to create hydrological models, map a wide variety of scenarios, and optimize water usage. Where several countries share scarce water resources, transboundary water security challenges persist. The use of digital twins can facilitate decisions that promote equitable sharing and maintain balance.
Examples of digital twins that simulate terrestrial water cycles are Digital Twin Earth’s models for the Mediterranean Basin and the Po River valley in Italy, as well as the work by NASA, the European Space Agency, and the International Water Management Institute.