Spatial Twins vs. Digital Twins

Apple made waves earlier this year with its long-awaited announcement of a spatial computer headset: the Apple Vision Pro. While the company championed the term “spatial computing,” as well as half a dozen other “spatial” concepts, Apple is just the latest company to iterate on the idea of  spatial computing.  

Spatial computing – as a concept - has existed for roughly 20 years and often refers to interaction between the physical and the digital, in which the digital can understand, conceptualize, and even manipulate objects in the physical world. It’s an incredibly complex process and part of the later stage of digital transformation, where the link between physical and digital is both cemented and blurred.  

Perhaps there is no better way to understand what separates spatial computing from traditional computing than examining the differences between a spatial twin and a digital twin. However, before we focus on these two concepts, let’s lay some groundwork on what separates a twin from another technology that it is often confused with – a simulation.

Simulations: Different from spatial and digital twins

PTC defines a digital twin as a virtual representation of a physical product, process, person, or place that can mirror and measure its physical counterpart in real time. The “measure” part is important. Simulations mirror static real-world environments. A simulation of traffic patterns in London, for instance, can be used to predict actual transportation habits – but there’s a catch. There is no conversation in simulation. The simulation simply uploads the data and begins processing possible outcomes.  

Continuing with the London example: imagine a sinkhole opened on the road and closed a 3-block stretch for repairs. The simulation has no idea this has happened. As such, its predictions will not reflect this impactful development. Twins (digital and spatial) are in constant communication with their real-world equivalents. The twin would update its information to reflect the change, while the simulation would not know the change had taken place.  

For manufacturers, it’s an important distinction. Simulations should only be used as predictions – nuanced, data-driven assessments for how a machine or an improvement could work in the real world. The twin, however, is the platform that will actually provide real information once everything is in motion. Any executive who confuses the two may find themselves in trouble from only reading a simulation while the actual factory floor experiences changes in operation that lead to downtime.  

For any manufacturers with further questions on the digital twin concept, I invite you to watch this webinar with PTC’s Chief Strategy Officer Catherine Kniker and ABI Research’s Ryan Martin:  

Spatial twins vs. digital twins: It’s all about context  

A spatial twin is a kind of digital twin, sometimes known as a digital twin of a place. The two concepts are fundamentally linked, and to understand one is to understand both. So then, why should organizations care? Well, while spatial twins are a kind of digital twin, they can provide comprehensive insights that many other kinds of digital twin might miss.   Let’s use an example: A manufacturing company keeps having issues with one of its primary machines in a remote factory location. The device keeps shorting out. The digital twin of the machine detects the issue, but it cannot immediately determine what exactly is causing the issue to happen. There is no expert technician on site and the local staff doesn’t understand what’s happening.  

A spatial twin, however, especially one accessed with an industrial metaverse platform, will allow any employee to augment what they see with new information. In this case, the knowledge that there is a leaky water pipe behind a nearby wall that is interfering with the machine’s power cord. This pipe iss out of sight so the ordinary operators were unaware of its existence, but the spatial twin knows it is there, and knows when its pressure reading isn’t what it should be.  

So, while the digital twin did provide an accurate and up-to-date readout of machine performance, only the spatial twin could provide the needed context of the machine within its environment to deliver the correct solution.  

Making sense of a variable world  

Virtual environments tend to be very precise. They are, by nature, programmed to function within specific parameters (not to say that bugs can’t happen because they certainly can). The physical world, by contrast, is chaos. Variable upon variable upon variable all compounding into changing ecosystems that can be insanely difficult to fully predict. Spatial computing, and spatial twins, can finally capture this chaos – if only in specific physical environments. 

By doing so, they provide us with our first truly accurate and data-driven look into how certain environments function. How long does this average workflow process take? What really happens on the factory floor during the span of a normal day? Where are processes under-performing? Where are they over-performing? Even a dedicated expert human observer would have difficulty capturing all this information – and their very presence would alter the environment in question. Spatial twins provide less biased insights into the actual work environment.  

Spatial twins are a unique, nuanced digital twin type that offers a look into what’s possible with advanced stages of digital transformation, where onsite knowledge is more accessible, and the chaos of the real world makes just a little bit more sense.