First, let’s talk a little about the terminology. What is a digital twin? According to engineering researchers at the University of Salerno, the definition of a digital twin is: “a virtual representation of a physical product, asset, process, system, or service that allows us to understand, predict, and optimize their performance for better business outcomes“. Simply put, it’s a virtual model of a target object that can be replicated down to the last detail with a possibility of a continuous update from the data. With the right equipment and software, you can make a digital twin of almost anything, starting with any small physical object all the way up to large-scale networks, even entire cities. The virtual twin can be as simple or as complex as you want it to be, depending on the amount and quality of data you use for building it. For a better understanding, let’s take road networks as an example.
Real World Scenarios of Digital Twins: Infrastructure and Road Networks
A digital twin of a road network can show 3D spatial data of that network. This means that you can see every physical asset related to the network – such as the pavement, traffic signs, road furniture, and asset items, as well as the severity of road defects and condition of the road and the objects. Even the slightest cracks and the smallest depressions on your roads can be detected on the digital twin images and models, as well as attributes of road assets, such as the type, size, and condition of traffic signs. Furthermore, not only can you see the defects and attributes, but you can also use the data for a plethora of purposes. One of which is predictive maintenance – a perfect road maintenance method that saves time and money.
But let’s not get ahead of ourselves. Before running any statistical analysis, you need to build the twin first and, of course, the precision of the model depends on the quality and the amount of data that has been captured to build the digital twin.
To build a digital replica of your road network, you first have to collect the road network data that you wish to model. It’s essential to use connected sensors when building the model – this ensures high-quality data and gives the needed input for accuracy, such as geolocation. If you don’t know where the defects are located in your road network, the model isn’t of much use. Capturing the sensor data requires you to drive through the entire road network that you wish to model. But this is much more efficient than the manual labor required to take analog notes on every road.
Why would you want a virtual representation of your road network? The short answer is to get an overview of the state of your roads. We will continue with this topic in our next post where we go into more detail regarding the benefits of building a digital twin of your road network and why capturing data using digital twin technology is far more cost-effective and efficient than the traditional methods.