The Internet of Things (IoT) has long been more than just a trend. By the year 2020 alone, more than 20 billion objects worldwide will be networked with the Internet. According to estimates by the corporate consultants Deloitte about 750 million of these will be in Germany. These networked objects form the basis for supplying millions of items of data to “digital twins”. A digital twin is a virtual image of a physical asset, process or system. Assets at Deutsche Bahn can be locomotives and freight wagons, but also rails, points, station clocks or smart waste bins. “By presenting them in virtual form, you obtain information about the physical condition, for example, whether a waste bin is full or whether a locomotive and its individual components are defective or fully operational”, says Jörn Petereit, Vice President IoT/M2M at DB Systel. In short, the condition in the physical, or real world is transferred to a digital world.
The fundamental concept for this is certainly not new, as digital images have been used in product development for more than 30 years, for example in 3D renderings, asset models or process simulations. The reason: with a digital twin the feasibility of projects is validated and assured. NASA, for example, has been using digital twins for several decades for the complex simulation of spacecraft. But you do not have to reach for the stars: Deutsche Bahn has for many years been using building information modelling (BIM) software for the digital planning, implementation and management of complex construction projects over the entire life cycle.
More information, greater benefits
In the context of IoT, furthermore, data from the digital world is combined with the actual data in real time, which guarantees a quite different return flow of information. If, for example, the physical condition of a freight wagon is recorded with the aid of sensors, this can be transmitted to the digital world almost in real time. The transfer to the digital world creates transparency by which, for example, the maintenance can be optimized and conclusions can be drawn regarding the use of the asset. At Deutsche Bahn there are a great many possibilities for using potential IoT technologies, including about 3,000 signal boxes, 70,000 sets of points, or 5,400 stations with countless platform displays, clocks, lifts and escalators. More than 2,500 locomotives and 80,000 freight wagons in the Cargo department, the passenger transport trains and numerous maintenance depots offer a large number of potential IoT terminal points.
With a digital twin I can now not only ask: ‘Where are you?’ and ‘How are you?’, but also ‘When will you arrive?’.
One example: In the physical world the product benefit in the case of freight wagons is limited to the transport from A to B. The digital image also provides information about condition, mileage or loading. Moreover, the digital image can be expanded to include the context-related information from different data contributors.
The clever thing about this is that the required data already exists in most cases: sensors already transmit data from trains and other assets. With the digital twin, relevant information is linked, and by combining individual values the informative value is increased in real time. In connection with data from existing systems, such as timetable data, geodata or data from sensors on bridges that register the passing of a train, the precise arrival time of the train, for example, can be determined automatically. This predicted arrival time is attached to the virtual image of the train. “This increases the product benefit of the physical asset” explains Petereit. “I am now not only able to ask the freight wagon, or its digital twin: ‘Where are you?’ and ‘How are you?’, but also ‘When will you arrive?’. Digital twins permit the generation of an incremental product benefit, as this will increasingly be generated digitally or software-driven in future”.
A twin is seldom alone
Thanks to the digital twin, the digital value of an asset is greater than the physical value. The reason for this is that with the digital twin, on the one hand the transparency and possibility for diagnosis is increased, while on the other hand the comparison between the real asset and the expanded digital image becomes better and better. This allows conclusions regarding optimum operating parameters to be drawn for each individual twin and also offers added value in the area of efficiency. It will be exciting when all digital twins are considered as part of an ecosystem, as this will enable the interaction of the digital twins with one another. For example, the train in connection with the rail and the maintenance depot. The basis for this ecosystem is the DB IoT Cloud.
There are numerous advantages: digital twins offer greater efficiency, since a permanent monitoring of the physical asset enables the determination of optimum operating parameters and thus an efficient operation with low operating costs. In addition, the quality is significantly increased, as additional data from the entire process chain, in maintenance for example, guarantees comprehensive and continuous quality assurance. Risks are also minimised: permanent monitoring detects anomalies and potential faults at an early stage. “Compared with existing conventional solutions, the comprehensive network creates significantly more flexible control options”, according to Petereit.
In connection with analytical technologies, artificial intelligence and machine learning, new knowledge about the digital twin will in future be fed back into the physical asset and digital process on the basis of the DB IoT Cloud. This permanent circuit permits the generation of an increasing digital benefit across the entire product lifecycle.