How digital twins are changing cargo terminal decision-making

In a June 2025 blog, Infosys called digital twins ‘a game-changer for air cargo simulation’. The phrase captures a shift that is quietly taking hold across cargo terminals worldwide, where digital twins are emerging as command centres that influence how operators decide where to place people, pallets and planes in real time.

What is a digital twin?
The concept of the digital twin traces its roots to NASA’s physical spacecraft simulators of the 1960s, which were used to mirror real missions on the ground and played a critical role during the Apollo 13 crisis. The idea was later formalised in 2002 by the scientist and writer Michael Grieves, who articulated the framework of linking a physical asset with its digital representation, while the term “digital twin” itself was coined by economist John Vickers around 2010.

At its core, a digital twin is a continuously updated virtual representation of a physical operation. In air cargo, this means a live digital model of warehouses, aircraft positions, ULD inventories, ground equipment and landside flows, all linked to real operational data. Unlike traditional simulations that are run occasionally and then set aside, a digital twin stays alive, evolving as conditions change on the ground.

This persistence is what gives the concept its practical value. Digital twins matter because they allow cargo terminals to see the consequences of decisions before those decisions are made. By combining live operational data with predictive models, they help operators anticipate disruption, reduce risk and coordinate responses across people, equipment and infrastructure in real time.

By analysing turnaround times, clearance velocity, slot utilisation, and resource constraints, such models help surface structural inefficiencies and key improvement levers.
Vineet Malhotra, Kale Logistics Solutions

From static plans to living models
For decades, cargo terminals have relied on static planning tools. Seasonal forecasts, Excel sheets, rule-of-thumb capacity models and the experience of a handful of supervisors formed the backbone of daily operations. These methods worked when schedules were stable and demand patterns predictable.

That stability has steadily eroded. E-commerce surges, geopolitical shocks and increasingly volatile fuel and labour costs have exposed the limits of static planning. In this environment, a digital twin represents a fundamental shift. Instead of reacting to deviation after deviation, planners can explore scenarios as conditions evolve. The result is not perfect foresight, but better preparation and faster, more confident decisions when disruption inevitably occurs.

How the twin sees a cargo terminal
Making this work requires stitching together systems that have traditionally operated in isolation. Sensors on dollies, forklifts and loading equipment stream location and status data in real time. Warehouse management systems contribute shipment details and handling milestones. Flight operations systems add schedules, delays and aircraft type, while customs and security platforms layer in clearance status and inspection holds.

Machine-learning models then sift through this growing data set to detect patterns, flagging where bottlenecks are likely to form or which shipments may be at risk of delay. As Vineet Malhotra, Co-founder and Director at Kale Logistics Solutions, explains, “The aim is to create near-real-time virtual representations of cargo flow across airports, ports, and logistics corridors, bringing together data from Cargo Community Systems, slot management platforms, and customs interfaces to simulate cargo movement and test scenarios.”

Decisions are no longer one-shot bets
This shift is most visible in how operational decisions are made. Traditionally, cargo duty managers would lock in shift plans hours in advance, then spend the rest of the operation responding as reality diverged from the plan. Each decision was, in effect, a one-shot bet.

Digital twins introduce a more iterative approach. Before freezing a build-up sequence, teams can test delayed flights, inspection holds or reduced ramp capacity. According to Malhotra, these models support “predictive and what-if analysis,” allowing stakeholders to evaluate options such as “optimising resource allocation” and “strengthening operational resilience.”

The same logic applies beyond live operations. Digital twins are increasingly used to simulate cargo hub layouts and process flows before physical changes are made. By testing alternative designs and equipment allocations virtually, operators can base investment decisions on quantified outcomes rather than intuition alone.

At the sidelines of Transport Logistic and Air Cargo Southeast Asia 2025 in Singapore, digital twins was an area of focus for many, reflecting how the industry is beginning to link predictive modelling, shared data and operational resilience more closely than before. In an exclusive interview, Zeta Lou, Head of Commercial Operations Asia-Pacific at CHAMP Cargosystems, shared that digital twins sit firmly in the context of safety and prediction. “We also see the rise of a digital twin for safety predictive management,” she said, adding that such technologies have the “potential to revolutionise how cargo will be monitored and secured end to end.”

From a broader supply chain perspective, Alex Teo, CEO, Managing Director and Vice President Southeast Asia at Siemens Digital Industry Software, framed digital twins as a way to connect decisions across the value chain. “You can build digital twins of your supply chain,” he said. “You basically create a digital representation of your supply chain and then apply business scenarios before you start executing.” His colleague Emre Akbağ, Global Business Leader for Digital Logistics at Siemens, emphasised preparedness over reaction, noting, “This is a scenario that we can build in the digital twin,” adding that teams have “done the rehearsal already, so they don’t have to organise an emergency meeting or a fire drill.”

As the layers of the Digital Twin matures, it will add better dynamic orchestration, allowing each actor to adjust operations proactively based on the realtime state of the cargo flow.
Frédéric Brun, Liege Airport

Predict, don’t just react
Prediction is where digital twins begin to show their full potential. Sensor-driven models of ground assets and terminal infrastructure support predictive maintenance by identifying abnormal patterns before failures occur, allowing work to be scheduled outside peak periods.

On the process side, digital twins help forecast congestion and pressure points. As Frédéric Brun, Head of Commercial Cargo and Logistics at LGG, puts it, “By combining real-time operational data with predictive modelling, the Digital Twin will help identify early signs of congestion and bottlenecks.”

Brun is careful to underline that this is a journey rather than a finished product. “While it is not yet a fully deployed forecasting tool, we are establishing the foundations to anticipate disruptions before cargo arrives,” he says.

As forecasting becomes more reliable, attention naturally turns to how those insights are shared and acted upon across a crowded operational landscape.

A new cockpit for collaboration
Air cargo handling has long been defined by fragmented decision-making. Airlines, handlers, truckers, freight forwarders and authorities often operate with different data sets, priorities and timelines. Digital twins offer a way to bring those perspectives together.

“Improved coordination is rooted in shared visibility,” Brun says. When stakeholders work from the same real-time view, conversations shift from reconciling data to deciding how best to respond.

From the platform perspective, Malhotra highlights the role of “live status updates, automated alerts, and shared operational visibility” in reducing communication gaps and enabling faster, more coordinated responses during disruption.

Once a shared operational picture is in place, optimisation becomes possible. Decisions no longer need to be framed as simple trade-offs but can be evaluated across service levels, cost and sustainability at the same time.

The limits and the learning curve
Digital twins demand high-quality, consistent data from systems that were often designed independently. Many cargo terminals still rely on legacy platforms and manual processes, making integration a significant challenge.

There is also a human dimension. Experienced supervisors who have built their careers on intuition and local knowledge can be sceptical of algorithm-driven recommendations, especially when early models challenge established practise. As Malhotra notes, digital twin modelling helps reveal “where delays originated, how bottlenecks propagated, and which operational dependencies proved most vulnerable.”

In an industry measured in minutes and kilogrammes, the advantage increasingly lies with those who can test decisions before acting on them. For cargo terminals navigating constant disruption, the quality of the virtual model is fast becoming as important as the efficiency of the physical one.