The KOMVAR project, funded under the German federal code 03ETE023X, ran from 1 September 2019 to 31 December 2021. Its primary aim was to create a competitive variant‑production system for lithium‑ion batteries that combines a service‑oriented approach to cell development with the capability to manufacture small‑ and medium‑sized series for niche markets. The funding recipient was P3 Automotive GmbH, and the effort was carried out by a consortium of industry and research partners that collaborated on the design, implementation, and evaluation of the production concept.
Technically, the project delivered a comprehensive manufacturing execution system (MES) that integrates real‑time production data with a cloud‑based data‑warehousing architecture hosted on Amazon Web Services. The data model was defined to capture key parameters across the entire production chain, and an ETL pipeline was built to extract, transform, and load data from plant sensors, quality control systems, and enterprise resource planning tools into a central repository. This repository supports both operational analytics and long‑term trend analysis.
Machine‑learning capabilities were added through Amazon SageMaker, enabling predictive maintenance, yield optimisation, and quality forecasting. Live data extraction modules were developed to feed real‑time dashboards that visualise material costs, production throughput, and defect rates. A data‑simulation and augmentation framework was also implemented to generate synthetic datasets for training models when real data were scarce, thereby improving model robustness.
A key outcome was the cost‑simulation tool that estimates cell production costs and profit margins for different cell formats. The tool was demonstrated with two example formats, showing how variations in chemistry, capacity, and connector design influence the overall cost structure. The accompanying dashboard allows stakeholders to interactively explore material cost drivers and to assess the financial impact of design changes.
The MES was prioritised through a structured requirement analysis that identified critical features such as real‑time monitoring, traceability, and integration with existing enterprise systems. The resulting system architecture aligns with industry standards for data management and supports modular expansion for future cell chemistries.
From a collaboration perspective, the consortium maintained a coordinated schedule despite the COVID‑19 pandemic. The project duration was extended by four months to 31 December 2021, a change that was agreed with the funding agency and fully reflected in the revised project plan. P3 Automotive managed the overall project execution, ensuring that the total budget remained within the allocated limits. The consortium also coordinated the filing of intellectual‑property rights, securing patents for the novel data‑integration framework and the machine‑learning algorithms developed.
In summary, KOMVAR produced a scalable, data‑driven production platform that enables rapid deployment of lithium‑ion battery variants for niche applications. The technical achievements—cloud‑based data warehousing, real‑time MES integration, predictive analytics, and cost‑simulation—provide a foundation for future expansion into larger production volumes. The project’s collaborative structure, adherence to an extended timeline, and successful budget management demonstrate the feasibility of integrating advanced manufacturing technologies within a small‑to‑medium‑series production context.