The right resources, right away. That’s the promise of the digital twin environment. Sounds too good to be true? Find out for yourself and see how you can make the most of this cloud mashup available to you with a few clicks via your humble web browser.
Jim Quancy works with Autodesk's developer advocacy and support team. With his global team of software developers, integrators they integrate desktop support for Revit, AutoCAD and Civil 3D to name a few. The Forge cloud platform has been dominating Autodesk’s work for the past five years, and they’ve collaborated with thousands of companies that build web cloud apps that use CAD data in some way, shape, or form. They’re currently working with over a hundred organizations to create digital twin solutions.
Digital twins have been around for a long time, as part of a simulation or analytical view of the real world. They’ve always been known as models of the real world. They provide simulation and analysis capabilities that can be a fancy 3D hologram that you can touch and feel and move things around or a numeric model that runs on Excel.
In the last few years, with the development of data feeds, IoT, and the ability to build more sophisticated models on computers in the cloud, digital twins dramatically expanded from being just drawings. They’ve become more intuitive, and are delivered to the user via graphical representations, which could be schematic, like a water network where someone needs to observe what’s happening in the real world and compare it to analytics to identify opportunities or problems. This is now all possible from a web browser on any device, and as a result, preventative maintenance or spotting something amiss is now much easier.
The context for the model is given in the original design in CAD along with objects and placements of things, like sensors.
The data itself gathered in an IoT middleware database, whether that’s coming from a handful or thousands of sensors. AWS, Azure Digital Twin, Honeywell, Johnson Control or Siemens all have their digital twin middleware.
The model itself also acts as a user interface to present the information to the user in an intuitive way.
Where Is The Data Coming From?
The data is not held on a local machine, and it doesn’t come from a single source.
You may have a 3D model in Autodesk. IoT sensors feed data into your centralized database from various sources such as a security system, machine control system, or traffic control. You can easily have four or five different systems, including geospatial architectures, feeding data, and presenting it in a digital twin environment – a 2D graphic, schematic, or a 3D model of a piece of infrastructure. Federating data is the reality – the world is too big and too complex to be able to represent it with one data source. Once your data is gathered, it’s presented on a web page in an intuitive way, such as a thematic color coding or actual data, depending on what you’re managing.
A relevant example of how data is fed and is displayed in easy to understand ways via digital twins is what we’ve been seeing during the recent COVID-19 outbreak – data, census tracking, and the virus in action across the world.
It depends on the problem you’re trying to solve and how you want to present the information. Think of it as a web cloud mashup.
There are lots of tools and choices to do the analytics, and the most important takeaway is that a digital twin environment can be relatively quick to set up and easy to implement. You can build a website in a couple of weeks and have everything up and running in a couple of months. Remember this can be as simple or as complicated as you want. You can create a basic digital twin by grabbing the pieces and focusing on the experience you want to deliver.
You can now make better decisions quicker. You open a webpage to see what’s going on with temperature, moisture, or vibration, and you can see right away if you need to take action because something is going on there. With standard cloud technologies, it’s easy to tie communication mechanisms together and you, as the user, can be notified if needed by a text message or in Slack if that’s your communication method.
Previously, building something like this, linking several pieces together was risky, costly, and time-consuming.
Anyone and everyone. Companies are doing digital twins for factories, machines, single buildings, large infrastructure projects such as a railway network or a city.
Digital twins can be created at any scale, Machine, city, country or at a global scale
Railway systems and highways in Europe are being recreated as digital twins. They start off with the stations first and then extend them to the rest of the rail infrastructure, the same thing is being done for utilities such as water, electricity and gas systems. They’ve all been thinking about doing it for a long time, and with what’s been happening in the last two months, they’re all ready to do it now. They’ve done their research, asked enough questions. They need to get a digital twin to operate their businesses, minimize the need to travel, and put people in the field.
Digital twins are simulations and use analytics based on real-time data, or just in time data.
There is your model, and you know how things should run. And there is real-time data telling you how things are running in real life. You can see how much easier spotting something amiss is, your machinery is supposed to be running at a temperature range between 120 °C and 140 °C, but it’s been running at 150 °C.
What’s that about?
The machine is not down, no one has made a discovery yet, but something’s amiss. You check the vibration, and you’ve got a bearing problem.
At a rail station, you might be tracking the number of people and their movements. You notice that there are too many people going in and not enough people leaving.
What’s that about?
You look at your CCTV and see what’s amiss there? You can have a sensor go off, notifying you that there’s a temperature spike. There could be a fire in a machine room at the train station, and you’ve just spotted that from halfway across the country before anyone at the train station knew what was going on, and you can take appropriate actions right away. If your IoT system is connected to machinery, which is not unusual at airports, you can turn the ventilation off from your desk, so you don’t feed fresh air to the fire. You can unlock all egress doors, turn on all emergency lighting, and monitor the situation from several different closed-circuit cameras from the same webpage. You’ve notified security, and help is on the way.
The real-time (or almost real-time) component of the system is unparalleled to anything out there.
Before affordable sensors, armed only with a large amount of data, you’d have to have an analyst spend months or even years trying to make sense of the data, looking at trends, doing analysis, and making recommendations for better decision making.
Now, it’s real-time or close to real-time data you’re armed with. For a traffic-related issue, you can make a decision in minutes. At the train station, you can rejigger the schedule, so you don’t end up with five trains pulling in at the station at the same time, with thousands of people disembarking simultaneously. You can slow down some trains from a browser to avoid a social distancing nightmare situation. You can control egress mechanisms, change lighting, and signage all at the same time on the same page.
Imagine scaling this. In fact, this is precisely what we saw during the COVID-19 updates. Apply it to a building, a neighborhood, and a city. Machines, switch gears, railroad cars, sensors, and data come together. You may immediately think this is intimidating, complex, and overwhelming. It’s not. It’s a web cloud, using web technologies. You just put together views specific to your problems, like single panes of glass. You can see and control data and make decisions smoothly, correctly, and safely.
The first question you should ask before attempting to channel anything into a middleware or simulating anything is, “What’s my biggest problem?”
Is it safety-related? Operations? Maintenance? Efficiency? Energy usage and sustainability? Where is it?
It may help to go back in time and think about decisions that you may regret making or had a second thought about. Where was the pain?
A digital twin can help you by making your data visible and presenting it intuitively instead of you looking at massive data flows that are complicated and time-consuming to understand. You need something that’s in an easy to understand the context, whether it’s related to a machine, a building, or a city, and you can react to it quickly.
Simplicity is the best. Always start from looking at ways that are simple, before moving on to more complex solutions. Perhaps all you need in order to make a decision is data on a page with a flashing red light, and you don’t need anything more than that. Don’t let things digital and complicated get in the way of your problems.
Sometimes, you just need a P&ID diagram of an electrical or sewer network. Sometimes a 3D model because that’s the easiest way to understand what’s happening and what decision you need to make.
Maybe you just need to go and turn a valve on or off, or plan preventative maintenance for the day after tomorrow, or keep an eye on vibration on a bridge and shut it down before something terrible is going to happen.
Start small and build up instead of not doing anything because you’re overwhelmed by the effort and money it takes to build a 3D model.
Agility over Waterfall. Always. If this is all new to you, think of something easy; a bit of a pain, more like a nuisance, and build a digital twin for that to start with. You don’t enter kindergarten and talk about calculus - there is simply not enough foundation there. Build something simple and see the benefits. Spend a few months solving a few problems this way. Then, you can put together your big vision, call it your two or three-year vision, and go after it agile, one piece at a time. As you’ll agree, the vision shifts because we get smarter, or our environment changes. COVID-19 is a perfect reminder of that. Be prepared for what a shift like that would do to your strategy and vision.
Managing the after-effects of earthquakes is where deploying digital twins can make an enormous difference.
After quakes, buildings need to get safety inspections and a tag – red, yellow, or green. Off-limits, grab your wallet, or safe to go back. Classifying thousands of buildings can take several weeks. Now there is a quicker way to do it.
Accelerometers, like the motion sensors in our mobile phones, can be installed in buildings and connected to the internet. During a quake, the sensor output is actively measured. You can see how a building shakes and spot if it’s been damaged from the way it shakes and vibrates. You can have hundreds of thousands of sensors across the entire city and lean on machine learning to identify buildings that are probably fine, suspect, or outright dangerous as there is clearly something wrong with them. Within minutes, the emergency services can go to the buildings where the problem is most likely to exist instead of taking calls from distressed residents about their damaged properties and lost dogs. That’s if they can get through the clogged networks in the first place.
The cost? $100 devices in each building connected to the internet with the data going through to a centralized database where the information is presented near real-time for analysis.
The right resources to the right buildings right away – the analytical way of saving lives. That’s pretty amazing, there is no denying it.
Could you immediately use some of this real-time magic? Would you consider setting up a digital twin, but it sounds all too complicated? Let me know your thoughts.
Geospatial standards, like any standard, are what we agree on as a community. It’s a way to describe how we model geospatial data, exchange it, subset it, process it, visualize it, or reference it. We need standards because we share and integrate data, and we solve complex problems.
Google Earth Engine is acloud computing platform for scientificanalysis andvisualization of geospatial data sets. It isfree to use for research, education, and nonprofit. Google Earth Engine is essentially streaming data. You don’t need to go online to download the data — you just need a browser, and you can access the entire Google Earth Engine data catalog and a bunch of tools to do the analysis and visualization.