Written by Tim Huckaby
Digital Twins & Device Twins in Enterprise IOT Deployments
If there ever was a popular buzz word in the software industry that is so epically not understood it is the term “Digital Twins”. IK has been one of the many pioneers in implementation of solutions leveraging Digital Twins. So, in this article, I will explain what Digital Twins is and why it is relevant to enterprise software applications; especially those rooted in data produced by IOT devices.
Digital Twins Definition – Simple Use Case
Frequently I find the best way to start describing a technology is with a use case. So, let’s start there:
Imagine a high-tech manufacturing company with 5,000 IOT devices used in the process of manufacturing their products. The IOT devices could be sensors on a machine that makes parts or the machine itself. Or the assembly machine or its parts. Or a myriad of other IOT devices that a manufacturing plant may have. Now picture the seemingly herculean task of:
- Managing thousands of IOT devices.
- Securing thousands of IOT devices. They all have an IP address.
- Receiving telemetry data out of those devices in a near real-time way and persisting that data to cold, warm and hot storage.
- Integrating the telemetry data out of all those devices so it can be visualized in reporting systems.
- Accommodating the scale of the enormous amount of data produced.
- Building meaningful, impactive and immersive Business applications based on that IOT data designed to understand, control, simulate, analyze, and improve real-world business operations.
In this example Digital Twins would be a digital replica of all 5,000 IOT devices; a virtual representation of the entire IOT device environment. In this example, Digital Twins would be the software architecture foundation that accelerates development time. You spend less time on gluing all these core pieces together, securing them, and more time on the functionality of enterprise business applications at production scale. In this example Digital Twins brings the physical assets (IOT devices) to life, to monitor and track.
Digital Twins enables you to model an environment, and then bring this digital twin to life in a scalable and secure manner. Digital Twins is an IOT platform which enables the creation of next-generation IOT solutions that model the real world.
Digital Twins Value Proposition
Digital Twins manages the complexities of:
- Scale – If you are building or using enterprise software you know that you do not get scale “for free”. Scale must be architected and can frequently be a challenge. Digital Twins handles the scale in IOT device solutions.
- Security – Let’s face it, the real scare of the proliferation of IOT devices is the increasingly growing attack surface for hackers. Digital Twins handles the security of the virtual device.
- Complexity of environments
- Siloed data spread across the enterprise ecosystem.
- The combination of several software architectures
- Near real time synchronization of data
- Integration of IOT devices and services
- Relationships between devices
Digital twins incorporate all of these building blocks to make solutions enterprise ready and tailor-fit for IOT; Solutions that are secure and scalable.
Digital Twins – A More Encompassing Definition
Now here is where it gets complicated and where the Digital Twins confusion frequently comes from. Digital Twins is more than just IOT devices. Digital Twins can include people and places and things and processes and other systems. Digital Twins is not only insights about the IOT device itself, but the device in context of the full environment it is in.
Digital Twins can be the platform that allows you to model any environment and connect sensors and business systems to that model. Ideally, all these things work together to provide better insights beyond just the telemetry information from the IOT device. The power of this means you can control the present, track the past and predict the future from a holistic view of your busines.
So, lets restate the definition of Digital Twins to: A digital replica of real-world things, places, business processes, and people.
The use cases for Digital Twins are seemingly endless, but here are some common ones:
- Business apps that capture efficiencies like crowd flow
- Spaces and facilities like a smart building
- Products or manufactured goods – insights from a particular product
- Supply Chain and its processes.
Digital Twins Costs
Costs? Well, there are some great Digital Twins solutions out there: Cisco, IBM, GE, Bosch, Ansys, PTC, Oracle, Intel, and Siemens just to name a few. Some of them are in Open Source. And each of them license differently, of course. We have expertise and the most experience with Microsoft Azure Digital Twins. Microsoft Azure uses a consumptive model; the more you consume the more you pay. It’s cloud pricing so it’s quite cost effective:
Message $1 per million messages
Operation $2.50 per million operations
Query unit $0.50 per million query units
Real World Example – Digital Twins
Designed and built for Microsoft, IK created the SmartHotel360 demo and reference application. SmartHotel360 is a Mixed Reality solution built on top of Azure Spatial Anchors and Digital Twins that is compatible with Android and HoloLens devices through Unity to visualize the information of the hotel room sensors used in IOT solutions. You can see the IK developers talk about it on Microsoft’s IOT show here. You can find the source code and the details of the solution here.
As if Digital Twins was not confusing enough there are also Device Twins. A Device Twin is the virtual image of a device, mostly containing the status and configuration data. It also carries device specific metadata. Device twin data is synchronized with the cloud. Applications consume and reference the Device Twin in the cloud instead of burdening the device directly.
This eases cycles on the device itself, including the burden of the secure connection to the device. In Microsoft’s Azure IOT, Device Twin information includes the general properties of the device, tags and metadata. Actions and events are not part of the Device Twin; that data would be typical of a Digital Twin. In AWS IoT the exact same thing, Device Twins, are called Device Shadow.
Real World Example – Device Twins
Designed & built for a world-wide sporting event, IK used Device twins and Azure IOT Hub to build an enterprise piece of software that is scalable and secure. It is extensible to connect business systems and designed to leverage data from physical assets like IOT devices; specifically, thousands of cameras that produce computer vision telemetry. That computer vision telemetry is used for the real-time analysis and decision making required operational efficiency in a sporting event of this magnitude. Computer Vision telemetry is collected in terms of Occupancy, Queue Times, Flow Rates, Crowd Density, foot traffic flow and even bus occupancy from literally thousands of cameras. Using Device Twins facilitates the easy installation, configuration and management of 10s, if not 100s of thousands of devices. This solution meets the operational efficiency challenges this sporting event faces today and the many sporting events into the future. It is container based and extensible to leverage the telemetry of any computer vision pre-trained machine learning model in a secure and scalable way.
Hopefully by this point, you understand Digital Twins and Device Twins as more than just buzzwords. They are a critical piece in constructing, securing, scaling, and managing complex IoT solutions. Digital Twins also goes beyond IOT devices themselves, including people, places, things, processes and other systems and enabling context and insights beyond the telemetry of IOT devices. It’s the technology that powers enterprise-ready, next generation IOT solutions that are scalable, secure, and cost-effective.