Every week we discuss one of 2017’s technology trends. The last three weeks we discussed the theme intelligent, with blogs about AI, intelligent applications, and intelligent things. The coming three weeks this series of tech trends will cover the theme digital. In this blog you read all you need to know about ‘digital twins’, which seem to be used to measure, for instance, the entire lifecycle of products.
2017’s Tech trend 4: digital twins
Testing physical prototypes, such as a jet engine, could be very time-consuming and expensive. They will be a lot easier and faster with the help of Digital Twins. A digital twin is a complete virtual prototype of an entire system, which means that it is a realistic virtual counterpart of physical objects. These copies are created in a way that they consist all the characteristics of the original object. In this way realistic simulations take place without investing actual time and resources in the production.
Creating digital twins is possible for various products, from a simple table leg to more complex objects, such as the jet engine. But digital twins are not only good for product design, but also for manufacturing process planning and feedback loops. There are several digital twins, such as: parts twins, product twins, process twins, and system twins, which deliver value in different ways. Part twins can deliver an early warning about a problem and tell you about a failure that one will encounter. Second, product twins optimize the remaining life of an asset, which they do against maintenance costs. What process twins do is finding ways to optimize scheduling manufacturing processes. System twins balance things, like revenue against remaining life against maintenance costs.
Gartner expects that in three to five years over 21 billion sensors and endpoints, and digital twins will exist for potentially billions of things. This requires close collaboration between technicians, engineers and operations personnel who understand the operation and maintenance of real-world things, and data scientists and other IT professionals who use digital twins.
See, think & do
According to Colin Parris (Vice President for Software Research, GE Global Research) digital twins operate in three stages: see, think, and do. In the see stage it is actually gathering data of its situation to give a warning or prediction. The digital twin takes the data, such a boil temperature and the rotor speed, in and uses a model. This model creates a cum cumulative damage understanding of what is happening on that rotor. At the time it hits a threshold, it can give a warning that there is a problem and then can predict the problem. “It is a model that is constantly updating itself to reflect the precise conditions of the asset and then learning from others in the fleet.” After it does that is has the ability to give these insights at the right time.
The second stage is the think stage, where the digital twin could, for example, give you various options to pursue. To do that it will run over 50.000 simulations. It is looking at previous data, it is looking at fleet data, it is looking at models, and it is looking at forecast for revenue and cost. After these simulations it will give a number of options. Then it reasons across the option, based upon the risk and the confidence it can deliver.
The last stage is the do stage, which is about informing and executing what needs to be done. A digital twin could, for example, give a manual option and an option for using an application. The latter can run much more precisely and it can precisely monitor, which causes the term of stresses on the rotor to be minimal. “These are the digital twins working with us with the minds of the machines creating these capabilities.”
Can everything have a digital twin?
Using digital twins could spare us billions of dollars. Consider, for instance, unscheduled maintenance events with aircrafts. Those cause not only great inconvenience for passengers because of their flight being delayed or cancelled, but they also cost a lot of many for the airline. With the digital twin these unplanned downtime is routine and it decreases airline costs for this matter. But digital twins are also active in for instance, wind farms and in health care. Does this mean that everything can have a digital twin? And will we in, the future, live in some kind of perfect world where nothing can go wrong because digital twins gets ahead of the problem?
Read and watch this interesting article and video about digital twins by Colin Parris.