Digital Twins in Intralogistics: Between Theory and Practice

Intralogistics has been under enormous pressure to innovate for several years now. Rising customer expectations, volatile markets, and the growth of e-commerce are forcing companies to make their warehouse and material flow systems increasingly efficient, flexible, and transparent. In this context, one term has stood out in particular: the digital twin. It is considered a key technology for digital transformation—but a closer look reveals that there is often still a gap between theory and practice.

To understand this gap, it’s worth taking a closer look at the various stages of development of digital representations: from a simple digital model, through the digital shadow, to the “true” digital twin with bidirectional data exchange. The latter, in particular, has so far been the exception rather than the rule in intralogistics (and is often more appearance than substance).

It all starts with the digital model. It forms the basis of many planning processes and is widely used in modern intralogistics. It is a virtual representation of a system—such as a warehouse or a conveyor system—that usually takes the form of CAD data or simulation models. However, this model is static. It does not incorporate real-time data and has no connection to the actual facility. Changes must be entered manually.

In practice, a digital model is often used in the early project phase. For example, a company might be planning a new distribution center and first creates a layout. Based on this, material flows are defined and initial simulations are carried out. These simulations help identify bottlenecks or evaluate different scenarios. This is where the strength of the digital model becomes apparent: It enables the simulation of system behavior under controlled conditions. However, it always remains a simplified representation of reality.

The next step in development is the Digital Shadow. This is where true digitalization, in the strict sense, begins. Sensors, control systems, and IT systems continuously deliver data from the actual facility, which is fed into the digital model. This creates an up-to-date, data-driven representation of the system. Inventory levels, throughput rates, or disruptions can be visualized in near real time.

This form of representation is already a reality in modern logistics centers. Dashboards show the current status of the facility, bottlenecks become visible, and decisions can be made based on data. Nevertheless, the Digital Shadow remains limited in its functionality: Data flows in only one direction—from reality into the digital model. There is no active feedback.

Only with the Digital Twin does the vision become complete. The key difference lies in the bidirectional data connection. This means that data flows not only from the real world into the digital world, but also back again. The digital model thus becomes an active control tool.

In theory, this sounds convincing. A digital twin could, for example, simulate various strategies for order processing, select the best option, and transfer it directly to the real-world system. Conveyor speeds, priorities, or routes would be adjusted automatically. Predictive maintenance would also be possible: the twin would detect wear patterns early on and trigger maintenance measures before a failure occurs.

In practice, however, this “true” digital twin with bidirectional data exchange is still relatively rare. While many companies talk about digital twins, they actually tend to use digital shadows or advanced simulations. The reason for this lies in the high level of complexity.

A bidirectional twin requires seamless integration of a wide variety of systems: warehouse management systems (WMS), material flow computers (MFR), control systems (PLC), sensors, and often external data sources as well. Added to this are requirements for IT security, data quality, and system stability. Any incorrect decision fed back from the digital twin to the real-world plant can have a direct impact on operations.

That is why true digital twins are found today primarily in pilot projects, highly automated facilities, or at technologically leading companies. They are more like flagship projects than a widespread standard.

To understand the path to the digital twin, it is crucial to distinguish between simulation and emulation. Both concepts play a central role in virtual engineering.

Simulation means replicating a system based on models and assumptions. It primarily involves running through various scenarios. How does throughput change when additional picking stations are added? What are the effects of a changed warehousing strategy? Simulations are particularly valuable in the planning phase, as they enable quick and cost-effective analyses.

Emulation, on the other hand, goes much further. Here, the goal is to replicate the real system as accurately as possible—including the control logic. In intralogistics, this often means that the actual PLC code is executed in a virtual environment. This is often referred to as virtual commissioning.

This form of emulation is an important building block on the path to the digital twin. It makes it possible to test complex systems even before construction begins and to identify errors early on. At the same time, it lays the foundation for later bidirectional coupling, since the system’s behavior is already known with a high degree of precision.

The concept of Virtual Engineering describes the systematic use of digital models, simulations, and emulations for the planning and optimization of technical systems. It is now widely used in intralogistics. New warehouses are rarely planned without prior simulation.

A typical example is the design of an automated small-parts warehouse. Simulations are used to test different layout options, determine the specifications of conveyor systems, and develop strategies for putaway and retrieval.

This helps reduce planning errors and safeguard investments.

A relatively new approach is Immersive Engineering. Here, technologies such as virtual reality (VR) and augmented reality (AR) are used to bring digital models to life. Planners, operators, and even customers can enter a virtual warehouse, understand processes, and make decisions more intuitively.

In practice, this is used, for example, for training purposes. Employees can practice maintenance tasks in a virtual environment without having to shut down real equipment. Safety aspects can also be communicated more effectively this way.

A realistic picture emerges when looking at specific applications. In many logistics centers, digitalization begins with a digital model created during the planning phase. This model is then used for simulations to optimize material flows.

After commissioning, a Digital Shadow is often implemented. Sensors provide data on inventory levels, movements, and status. This data is displayed in visualization systems and enables transparent monitoring of operations.

One example is a large e-commerce distribution center. Thousands of orders are processed here every day. A Digital Shadow shows in real time how heavily individual areas are utilized. Bottlenecks can be identified, and measures can be taken quickly—such as by redeploying staff.

A true digital twin would go one step further in this scenario. It would not only indicate that a bottleneck is developing but would also independently calculate and implement alternative strategies. However, this exact form of autonomy is still rare.

Another example is the virtual commissioning of a high-bay warehouse. Emulation is used to test the control software before the facility physically exists. This reduces risks and significantly shortens the ramp-up phase. Here, too, it becomes clear that many elements of the digital twin are already in place but have not yet been fully integrated.

The three concepts—digital model, Digital Shadow, and digital twin—each offer specific advantages.

The digital model stands out for its simplicity and low cost. It is ideal for planning but quickly reaches its limits during ongoing operations.

The Digital Shadow brings transparency to operations. Real-time data enables informed decisions and improves responsiveness. However, it lacks the ability for active control.

Finally, the Digital Twin promises maximum efficiency. Through bidirectional coupling, systems can be optimized, processes automated, and decisions accelerated. At the same time, however, it presents significant challenges: high costs, complex integration, and requirements for data quality and IT security.

Bidirectional data exchange is particularly critical. It requires a high degree of trust in digital systems—and appropriate safeguards against errors.

The trend toward a true digital twin will continue to accelerate in the coming years. In particular, the integration of artificial intelligence will play a central role. AI can analyze large amounts of data, recognize patterns, and identify opportunities for optimization.

Industrial IoT will also further strengthen the foundation. The more devices are connected, the more accurate the digital representation of reality becomes. At the same time, control capabilities will improve.

Immersive technologies will also gain in importance. The combination of digital twins and VR could lead to operators monitoring and controlling their facilities in virtual control rooms in the future.

In the long term, even the vision of largely autonomous intralogistics is realistic. Systems could optimize themselves, plan maintenance independently, and react to changes without the need for human intervention.

Digital twins are far more than just a buzzword—they are a central building block of digital transformation in intralogistics. However, it is important to distinguish between the various levels of maturity. While digital models and digital shadows are already widely used, the true digital twin with bidirectional data exchange remains the exception so far.

The technology exists, but its implementation requires significant investment, expertise, and a well-thought-out strategy. Companies that pursue this path, however, can benefit from substantial efficiency gains and competitive advantages in the long term.

The future of intralogistics will be increasingly digital, connected, and intelligent. The digital twin plays a significant role in this—even if it is still more often a vision than a lived reality today.

An interesting panel discussion on this topic also took place at LogiMAT 2026. The video is available here for anyone interested:

See our other articles
Back to the homepage