Smart data adapter: A hybrid pattern-LLM approach Döring, Lina Trojahn, Sebastian Reusch, Pascal Behrendt, Fabian Zadek, Hartmut Janmontree, Jettarat Trojahn, Sebastian Lang, Sebastian Semantic Data Type Detection DigitalTwin Manufacturing SMEs Zero-shot Learning Hybrid AI LLM Adopting digital twin technologies in small andmedium-sized enterprises (SMEs) is often hinderedby heterogeneous, poorly documented productiondata. Current semantic type detection approachesrequire massive labeled datasets making themimpractical for resource-constrained SMEs. Wepropose a zero-shot hybrid framework combiningpattern-based classification with selective largelanguage model (LLM) reasoning formanufacturing-specific data types. The two-stagearchitecture uses rule-based patterns for high-confidence cases, forwarding ambiguous columnsto a multi-step LLM reasoner. Evaluation on fourmanufacturing datasets shows the hybrid approachachieves weighted F1 within 7 24 percentagepoints of pure LLM classification performancewhile reducing LLM invocations by 39% onaverage. Processing time decreased by up to 2.8×.Our framework addresses a critical gap: automated,computationally efficient data type recognition formanufacturing SMEs without requiring trainingdata, contributing to automated simulation anddigital twin construction. Otto von Guericke University Library, Magdeburg, Germany 2026 info:eu-repo/semantics/conferenceObject doc-type:conferenceObject text http://purl.org/coar/resource_type/c_5794 https://www.hsbi.de/publikationsserver/record/7068 Döring L, Trojahn S, Reusch P. Smart data adapter: A hybrid pattern-LLM approach. In: Behrendt F, Zadek H, Janmontree J, Trojahn S, Lang S, eds. <i>19th International Doctoral Students Workshop on Logistics, Supply Chain and Production Management</i>. Otto von Guericke University Library, Magdeburg, Germany; 2026. doi:<a href="https://doi.org/10.25673/123544">10.25673/123544</a> eng info:eu-repo/semantics/altIdentifier/doi/10.25673/123544 info:eu-repo/semantics/closedAccess