Factory planning is becoming a bottleneck for the battery industry: high capital requirements, long lead times, and little margin for error. The EDAG Group has developed an AI-powered configurator that delivers robust layout concepts for battery cell factories during the early planning phase—in a matter of minutes.
Digitalization, decarbonization, and skilled-labor shortages are putting manufacturers across industries under growing pressure to transform, even as supply chains remain fragile and energy prices volatile. Companies planning a new factory today must make far-reaching decisions about location, capacity, and technology at an early stage—often without reliable benchmarks and within tight budget constraints.
Battery cell manufacturing is under particular pressure. Projects typically cost hundreds of millions and can reach into the low billions; lead times are long, and demand remains difficult to forecast. After several announced gigafactory projects in Europe were delayed or halted in recent years, the industry is under growing pressure to validate site-selection and capacity decisions earlier and with greater confidence—before capital is committed.
This is precisely where the EDAG Group aims to help. In summer 2025, its EDAG Production Solutions business unit developed the Smart Factory Network Configurator. Based on just a few key inputs—such as target capacity, investment budget, and space requirements—the tool generates initial layout concepts in minutes, complete with preliminary material-flow simulation, technology options, and cost estimates.
“The early planning phase determines a battery cell factory’s costs, schedule, and production targets. That is exactly where the Smart Factory Network Configurator comes in: Instead of waiting weeks for initial benchmarks, our customers receive robust layout concepts in minutes,” says Philipp Hummel, Specialist Consultant for Smart Factory Planning. He presented the project at Batterieforum Deutschland 2026 in January.
Inside the Factory Configurator
The tool combines several AI methods. Generative and agentic AI applications, together with machine learning, turn a small set of input parameters into hundreds of thousands of variants. The configurator analyzes the inputs, checks them against integrated domain databases, and automatically generates layout options with cost estimates.
EDAG reports that, in reference projects, this has cut the time from initial concept to feasibility assessment from an average of two months to two to four weeks.
Two Battery Cell Manufacturing Use Cases
Two battery cell manufacturing use cases described by the EDAG Group in a white paper illustrate how the approach works in practice.
In the first case, a customer is planning a new manufacturing facility for NMC811 lithium-ion pouch cells and defines the following at the start of the project:
- Production targets,
- Capital and operating costs, and
- Target values for the layout footprint.
After reviewing an initial design with an annual capacity of 8 gigawatt-hours, the customer requests a configuration that can scale by a factor of 1.5. The EDAG team simulates multiple scenarios in the configurator and prepares a cost estimate for the 12-gigawatt-hour option.
According to the EDAG Group, the outcome is a set of validated layout concepts featuring scalable production structures and transparent investment scenarios for future capacity expansions.
In the second case, a company with sites in Europe is evaluating the relocation of a plant to the United States and provides data from its German facility:
- Capital and operating costs
- Technologies
- Suppliers
- Current processes
For a like-for-like comparison, the system keeps target capacity, capital investment, and space requirements constant across both locations. The differences emerge primarily in the details, including alternative sourcing structures and different energy costs at the U.S. site. If needed, the EDAG team can further optimize the layout for the new site’s logistics requirements.
At the customer’s request, the results can feed into detailed planning and the development of digital twins in metys, EDAG’s Industrial Metaverse platform—for example, to visualize and assess layout changes virtually.
Conclusion: Data Quality Remains Critical
The Smart Factory Network Configurator is not a substitute for the expertise of factory planners or for subsequent detailed planning. It does, however, move the comparison of design alternatives to an earlier stage, when changes can still be made with relatively little effort.
