Why digital twins could be the next big thing in IoT

Gartner predicts that by 2021, half of large industrial companies will use digital twins, resulting in those organizations gaining a 10% improvement in effectiveness. The digital twin market is estimated to grow from USD 3.8 billion in 2019 to USD 35.8 billion by 2025, at a compound annual growth rate (CAGR) of 45.4%. 

Reason enough for us as IoT innovators to give a rough overview of the concept, the technologies and applications...

What is a digital twin?

“A digital twin is a digital representation of a physical object. It includes the model of the physical object, data from the object, a unique one-to-one correspondence to the object and the ability to monitor the object.” - Gartner

Digital twins range from people, devices and places to complex environments, such as buildings, factories or cities. It's very likely that every physical object which can be described to the finest detail, will be digitally represented in IoT.

In the industrial sector, for example, digital twins technology allows engineers to first test and simulate before investing time and money in developing actual products. This helps fix bugs and evaluate functionality before even one part of a machine, system or building is created. 

In other words: Digital twins help humans to make better and more educated decisions. But that is not the only advantage of this technology can provide.

What can digital twins do for businesses?

The following three examples show how twin technology can help improve business operations and safety in the very near future.

Improve safety in dynamic environments

Technology is changing many industries so rapidly that it is almost impossible to predict its impact on people and machines with certainty. Operations, which have remained virtually untouched for a long time, are being given new forms and rules by smart systems and AI.

To understand and orchestrate all correlations and interactions in new complex systems, digital representations of individual components or even entire ecosystems are created.

Digital twins enable engineers to improve design and implementation of products by:

1. Observing and controlling operations as they exist to identify optimization potential, failures and risks,

2. Simulating and testing adjustments to identify risks and bugs before implementing,

3. Training employees in handling new systems using Virtual Reality.

Imagine a complex system of people, machines and a physical environment like a large warehouse where thousands of goods are processed daily. If new operational components such as autonomous vehicles such as forklifts are added to such a system, significant implications can be expected. 

To find out how strong possible impacts are and how the new forklift’s artificial intelligence must be designed to optimally fit into the system, engineers have two ways to test new scenarios. First, they test in live operation which is very time-consuming and risky. Second, they simulate interaction of new and existing components in the system before the real forklifts are implemented. In this way, the behavior of the autonomous vehicles can be optimized before they go into service.

Example: Brief overview of how digital twins can improve operations in warehousing

Source: DHL

Improve quality in production and operations

In manufacturing, digital twins will support many industry 4.0 solutions, from automated root cause analysis, to predictive quality, predictive maintenance, inventory intelligence, and supply chain optimization. By equipping machines with sensors, it is possible to record any condition and performance of a machine in real-time and send data to analysis platforms and applications in IoT for processing. 

Twin Technology as part of industrial quality management helps detect error-prone processes at an early stage and to simulate adjustments before they are implemented.

Example: How BMW applies deep learning for visual inspection and quality control

Source: Converge

As devices become smarter every day, more manufacturers will take advantage of local data processing and AI capabilities, also known as edge computing. Instead of simple sensors performing one task, security cameras equipped with video analysis software could be used to monitor quality in industrial production. 

Smart cameras are even able to perform more complex analyses directly in the device instead of just providing raw data. This speeds up data processing considerably and allows engineers to react faster to incidents in systems.

Reduce downtime through predictive maintenance

Digital twins helps industry businesses predict failures in advance and reduce downtime of machines, better manage spare part inventories, monitor and manage fleet, do what-if simulations, and optimize operations. Remote computer vision enables operators to permanently monitor machines and use video analysis, for example, to predict when repairs or maintenance are due. 

Example: Digital twin model helps with the remote maintenance of heating, ventilation and air conditioning systems at an airport.

Source: esri.com

Smart surveillance cameras provide the link between the real and digital world. They ensure that any changes or incidents in real systems are recorded and transferred to the digital twin in real-time.

Extend customer service

Just like digital duplication of objects and machines, digital twins of humans are still at the beginning. Early applications show where the journey could take us in the near future, for example in remote customer service. 

A digital representation of machines or entire systems allows technical service personnel to support customers even without seeing the actual environment. Instead, service engineers act in a digital copy that reflects the actual environment without being on site. This helps saving logistical effort and time, especially when facilities are difficult to access. 

This kind of remote service requires that each digital twin is a 1:1 copy of the real situation at all times. Furthermore, service staff must be able to interact with the digital image. This is where virtual reality technologies come into play.

Sidestep: Support human customer service using Twin Avatars

Digital twins could support or at least partially replace human service agents in 1:1 communication with customers. This is where people literally get digital twins acting on their behalf. Exciting or threatening, the digital copy will very soon converge more and more to the natural model.

Source: ObEN

Worth reading: Get more insights, use cases and resources on Digital Twins from this great article on networkworld.com

Twin deployment challenges

Where advantages lure, challenges lurk. This is especially true when technologies are new to businesses and lack experience. For digital twins to work, mainly three conditions must be met:

• The digital image must have exactly the same properties as the physical model.
• Failures and bugs in systems must be documented in detail and, if possible, automatically transferred to the digital image in real-time.
• A digital twin must be redesigned each time the device configuration or element state changes.

Digital twin technology requires high quality data and the ability to transfer and process data in real-time. Edge computing, such as the processing of video data directly in smart cameras in IoT, will play an important role here.

At Security and Safety Things, we want our open IoT platform to help devices become smarter.