Flexibility as key
How Mercedes-Benz is revamping its assembly concept
More than 40 new models by 2027 increase complexity in assembly. Mercedes-Benz relies less on maximum robotics and more on data, digital twins, and AI-supported quality assurance in the ongoing process.
In the assembly hall, it is not the size of a robot cell that determines productivity, but the smooth running of the line. Every model change, every additional drive, every new variant directly affects the cycle. At Mercedes-Benz, this pressure is increasing - significantly. By the end of 2027, the manufacturer plans to produce more than 40 new models in the global production network. Electric vehicles, hybrids, and other derivatives are to run in parallel on existing lines. For assembly, this means above all one thing: increasing variance with constant cycle times. Automation is intended to help integrate new models more flexibly, efficiently, and stably into existing production, according to Mercedes-Benz in response to an AMS inquiry.
Automation aims at stable model changes
The Stuttgart-based company aligns assembly automation not with individual stations, but with the overall system. Under the "Next Level Production" programme, new models are to be integrated into existing lines more quickly and with fewer start-up losses. Following the production launches of the electric CLA at the Rastatt plant and the electric GLC in Bremen, further models from the core and top-end segments will follow. At the same time, locations such as Kecskemét and Sindelfingen are preparing for new electric series. Bremen and Sindelfingen continue to function as lead plants. Assembly is at the centre of the transformation because it brings together all variants.
A key lever is the complete integration of all plants into the MO360 production ecosystem. Through the MO360 data platform, production and quality data are available across locations. In assembly, this enables a continuous view of cycle deviations, process stability, and quality anomalies. In short, automation here arises less from additional technology and more from transparency and rapid response.
Quality is shifted into the ongoing cycle
A key focus of assembly automation is quality assurance within the production line. With the MO360 application "Quality Live," the production of individual vehicles can be tracked in real-time. Data from assembly, testing steps, and re-inspections are combined. The aim of the Swabians is to detect deviations early and avoid rework. In assembly, every error directly affects the cycle. Each additional correction at the end of the line ties up personnel and time. Automation should therefore secure quality earlier; not repair it later.
This approach reflects a fundamental development that is also visible in the question of how far automation can actually go in final assembly. Despite increasing technological deployment, humans remain a central factor. Especially with work steps rich in variants, rigid automation solutions reach their limits. Mercedes-Benz responds by integrating quality and process control more closely with data and assistance systems.
Digital twins transform assembly planning
A crucial tool for this strategy is the digital twin of production. In the Kecskemét plant, a complete hall was digitally mapped for the first time. Based on simulation platforms, assembly processes can be virtually tested, optimised, and secured before they are implemented in reality. This is particularly relevant for assembly when it comes to new models. Start-ups are considered particularly prone to disruptions. Small deviations can affect the entire line flow.
The digital twin makes it possible to identify bottlenecks early and prepare automation in a targeted manner. Changes can be simulated without interrupting ongoing operations. This makes automation more plannable and less risky. This approach fits the general trend where digital twins, cloud systems, and predictive analytics increasingly become the basis of connected production.
AI supports decisions in assembly
Of course, artificial intelligence also plays an increasingly important role in Mercedes-Benz' assembly, but not as a replacement for manual work, according to the OEM. AI is used to optimise products and processes, automate routine tasks, and support employees in data analysis, according to Mercedes-Benz to AMS. AI assistants are already being used in production, for example, to analyse quality data. They serve as a voice-based interface to the data of the MO360 ecosystem. This allows employees to access production metrics, quality information, or maintenance and best practice tips without having to operate complex systems.
An additional important component is the internally developed digital factory chatbot ecosystem, which connects production data, expert knowledge, and best-practice information across locations. Automation here is shown as a shortened decision-making process in the cycle. The role of AI thus shifts from physical automation to cognitive support, as is also evident in the broader use of intelligent systems in manufacturing.
Humanoid robotics still a fringe topic in assembly
In parallel with these developments, Mercedes-Benz is testing humanoid robotics together with Apptronik. The robot "Apollo" is currently being tested in realistic production environments, although the focus is initially on intralogistics, material transport, and simple inspection tasks. In the future, assembly tasks could also be added, such as handling sensitive components.
For current vehicle assembly, however, humanoid robotics does not yet play an operational role. The level of autonomous maturity is considered crucial. Only when systems operate stably, safely, and economically should they move closer to the line. In the short term, assembly remains a hybrid space of human experience and digital assistance.
Assembly automation is measured by stable cycle times
In the end, Mercedes-Benz defines assembly automation not by the proportion of robots, but by the ability to manage complexity. The benchmark is the stable measurement of cycle time. Every automation must be measured by whether it secures ramp-ups, stabilises quality early, and reliably integrates model diversity. Data, digital twins, and AI-supported assistance form the basis for this. The assembly may not necessarily become more spectacular, but it will become more robust. In a production world where lines are no longer thought of linearly, this robustness could gain strategic importance.
