Large companies in the IOT space frequently use the term Digital Twin. A Google search will show up multiple results and companies such as GE, IBM, Deloitte, Siemens showing up prominently. The term has a Wikipedia entry as well:
Digital twin refers to a digital replica of physical assets, processes and systems that can be used for various purposes. The digital representation provides both the elements and the dynamics of how an Internet of Things device operates and lives throughout its life cycle.
Here is a simplified way of looking at it and how the concept is supported by the iMsight framework. Conceptually you are modeling the ‘physical thing’ – Aircraft, HVAC system, Wind Turbines, etc. – in software. What you model is can be referred as an ‘Asset’ – that is – it has a business value associated with it. You can have large assets like Aircraft or a Ship or a Wind Turbine or something more simpler like a Freezer at a QSR.
The software model is essentially made up of the following aspects:
- Metadata associated with the Asset – Make, Model, Date of Manufacture, In-service Date, Serial Number or Unique Identifier, any physical location or site associated with it etc.
- Data Sources – this is the interesting aspect – you are getting information on the operational data of the asset – such as engine temperature, location (e.g. moving assets such as trucks or rental equipment), usage (e.g. tracking counts on press or injection molding machine per die or mold). This data is generated by a sensor device.
- Control – these are elements that can be controlled such as – starting/stopping an inverter or an engine, cooling . a freezer, increasing/decreasing speed of process or machine. Control elements work the other way – they are like Data Sinks. That is the software sends data (technically control command) to the asset or the machine to modify its behavior.
Of course the software model can be more sophisticated – the attributes listed above are core to the model.
The Data Source or Control – is made available because there is either a sensor that measures and sends values (data) to the software system, and the reverse is done by sending commands to something called as an actuator. These two are the essential elements of interacting with the physical world using software. Let’s combine them to call it a “Device”.
So Asset is the thing that we want to monitor, manage, control, observe to tap into the business value such as reliability, predicting failure or even monetizing (from the earlier example – instead of selling the die or mold one time – may be selling it for certain number of counts to be used – a subscription service?).
And now let’s bring it together – based on this Asset and Device separation, we have adopted classifying software models of the physical world:
- Integrated Asset – think Aircraft or Wind Turbines – the asset and the device (things that measure or control) are all together. The sensors are all part of the asset itself.
- Pluggable – the best example of this GPS tracking used in Fleet Management with hardware based device plugged into the OBD port of the vehicle. The asset is the vehicle being tracked and managed – and the GPS tracker is the device. The reason this model is called Pluggable is because the device is assigned/associated to the Asset. The data or the control flows through the device (in this case the GPS tracker) but it is the Asset we are interested in tracking. So our software model needs to ensure that even though the data is emanating from the device it pertains to the asset.
- Virtual – the best example is Environmental Monitoring or Soil Sensing. What is the asset in both these scenarios? It is abstract or virtual – in case of Environmental Monitoring it would be the air in a room or in certain part of the city – the air at a certain location is the ‘asset’. The device is the sensor making the environmental measurements.
Implementing the right model of things in software as described above unlocks the business value of IoT. Movinture’s IoT application framework comes with these software models of physical things – aka – Digital Twins – comes baked in.