2026

Schmucker

At Siemens PLM Connection 2026, Katharina Schmucker (Siemens AG, doctoral candidate at IMI under Prof. Ovtcharova) presented an approach for AI-based recursive clustering of technical solutions under the motto "Class trees were yesterday, clustering is today". Unlike static classifications, it uses global and local similarities to uncover relationships and variant structures from fragmented data in a data-driven way. Embedded in an evolving engineering knowledge base with agentic AI methods, it is intended to support configure-to-order and variant management in the future.

IMI

In July, the IMI team joined the 24th Conference of the International Federation of Operational Research Societies (IFORS 2026) in Vienna with three presentations and one session. It was a great opportunity to exchange ideas and connect with researchers from around the world.

IMI

As part of the “FlexTools” research project, a new approach for efficiently managing dense floor storage has been developed. The study, published in the International Journal of Production Research, addresses the Buffer Storage, Retrieval, and Reshuffling Problem. A mathematical model and hierarchical heuristic demonstrate how fleets of autonomous mobile robots can coordinate dynamic storage, retrieval, and reshuffling operations.

Gecco Photographers

The paper ‘CoupleEvo’, presented at GECCO 2026, introduces an LLM-based framework for the automated development of multiple coordinated heuristics for coupled optimisation problems. The paper examines sequential, iterative and integrated evolution of the heuristics.

IMI

As part of the interdisciplinary project “Data- and target-driven sequential decision-making for time-dynamic logistics systems” (4D4L), IMI, IOR and IFL have jointly published first results as a technical report on arXiv. The paper „Context-Aware Synthesis of Optimization Pipelines for Warehouse Optimization“ was authored by Janik Bischoff, Anne Meyer, Uta Mohring, Fabian Dunke, Maximilian Barlang, Özge Nur Subas, Hadi Kutabi, Stefan Nickel and Kai Furmans.

IMI

The research paper "Learning-based 3D CAD model generation in mechanical engineering: A survey" was published in the journal "Advanced Engineering Informatics." The work provides a comprehensive overview of recent developments in learning-based methods for generating CAD geometries and discusses their application in a mechanical engineering context.

IMI

On May 19, 2026, the IMI presented the paper “Foundation Models for Process Parameters Optimization: Evaluating TabPFN for Bayesian Optimization in HMLV Manufacturing” at the CIRP Conference on Manufacturing Systems. The study shows that the tabular foundation model TabPFN is a data-efficient surrogate model for Bayesian optimization of process parameters in high-mix, low-volume manufacturing.

IMI

Das Forschungspapier „Beyond CO2: Integrating Water, Social Indicators, and Uncertainty into Sustainability Knowledge Graphs“ wurde in der Fachzeitschrift Sustainability veröffentlicht. Herzlichen Glückwunsch an die Autorinnen und Autoren!

IMI

Under the title “EcoVity – The digital ecosystem for medium-sized companies for the collaborative development and production of customized commercial vehicles,” we are working with our partners to show how automotive suppliers and SMEs can develop customized commercial vehicles in small quantities in an economical, digital, and cooperative manner in the last mile.