Original article Total Telecom:Read More
Interview
We spoke to Hexagon’s Director of EMIA (Utilities and Communications), Jean-Francois Allard, to discuss taking digital modelling to the next level
Telcos are some of the most data-rich organisations on the planet but using that data effectively remains a major challenge. For Jean-Francois Allard, Director of EMIA (Utilities and Communications) at Hexagon, digital twins – digital replicas of physical assets enabling real-time interaction – are a vital platform for leveraging that data.
“It’s about creating a continuous loop, where changes in the physical world are reflected in the digital world and vice versa,” he explained. “To say it shortly, our main goal is to put data in action.”
Allard argues it is useful to consider digital twins not as a novel system, but rather as a maturation of earlier digital models.
Previously, digital models focussed primarily on digitising physical assets into a central inventory. These models were limited by their relatively static nature, requiring significant manual oversight to generate actionable insights. As such, these models have quickly evolved into what Allard terms ‘digital shadows’, where sensors and field data provide a one-way stream of real-time information into the model.
A digital twin takes this paradigm to its logical conclusion, making the system synchronised and bidirectional, allowing it to autonomously react to incoming data in real-time.
“You can have in real time both real world and digital world that are synchronized so that fully autonomously, this is the ultimate goal,” says Allard.
While industries like nuclear power are already approaching this level of sophistication for security reasons, Allard estimates that telecom is still “5 to 10 years” away from full autonomy.
Managing the complexity of telco networks
What makes a telecoms digital twin more complex than a water or power grid? According to Allard, it is the interplay between physical infrastructure and logical services. In what he terms a “relational digital twin,” the system must track everything from the trench in the ground to the specific IPTV service running through a single strand of fibre.
“In telecom, you manage the physical aspects… but also what’s inside,” Allard explains. “You need to know this hierarchy at all times[…] so that you can manage the services, the fibre links, and fibre channels in near real-time.”
This precision understanding of huge, distributed networks and their various technical elements not only allows for more efficient automation, it also improves customer experience. By using OTDR (Optical Time Domain Reflectometer) technology integrated with a digital twin, for example, Hexagon can locate a network failure within 10 metres on a 100-kilometre backbone network, resulting in faster repair and less downtime.
In a competitive market where customers can switch providers with ease, this
Infusing digital twins with AI
As the industry looks toward the next decade, AI is the clear catalyst for scaling these digital twins. However, Allard warns that the “intelligence” of the system is strictly capped by the quality of the underlying data.
“AI today is only valid if the data you are using to feed your AI is valid. Good data will provide good AI analysis. Bad data can only bring you bad answers,” he cautions.
When the data is sound, the applications can be highly effective. Allard highlights two primary use cases:
- Automated digitisation: Using machine learning and point cloud analysis to turn field photographs into “intelligent” data automatically.
- Plain language interactivity: Combining available data with generative AI allows engineers to perform complex analysis “just by talking with your own words” to the system. This is faster than performing the analysis manually and often yields more effective results.
“What if you are in the green field and you want to find a route but you have no information but pictures?” Allard asks. “AI can help you. AI can identify a route just based on the pictures. So, this is where AI can go beyond just using existing data – it can create a new path that you didn’t even didn’t think about.”
Getting the foundation right
Building an effective digital twin is a major undertaking. For Allard, the key is to build with purpose rather than chasing a “rich” but unusable model. He suggests starting with the basics, like connection points, nodes, and cabinets, rather than trying to map every inch of the network at once.
“Anticipate […] Ask what you want to do with your data,” Allard concludes. “Very often we see customers having a very, very rich model, but it’s so complex that it’s very difficult to take a benefit from it.”
Keep up to date with all the latest telecoms news with the Total Telecom newsletter
Also in the news
World Communication Award Winners 2025
Ofcom clears the way for satellite-to-smartphone services
LG Uplus’s AI voice call app glitch leaks user data
The post Putting data in action: Transforming digital shadows into digital twins appeared first on Total Telecom.