CC BY 3.0 UnportedTeriete, TimTimTerieteLangenfeld, LenaLenaLangenfeldErlach, KlausKlausErlachBauernhansl, ThomasThomasBauernhansl2025-11-132025-11-132024https://publica.fraunhofer.de/handle/publica/499269https://doi.org/10.24406/publica-633910.15488/1770910.24406/publica-63392-s2.0-85205965727The first step of the value stream method—widely used for lean production optimization-is the deduction of product families. Aggregating products with production similarities into product families and analysing the value streams of these families decreases the analytical effort and increases clarity and informative value of the results. In the design phase, product families serve as a basis for production segmentation to achieve a laminar order flow. The digital transformation of production-also known as Industry 4.0-allows to use data science in the value stream method. Existing approaches describe data-driven product family deduction using cluster analysis without considering the overall context of a digitalised value stream method. This paper adapts the product family concept to allow integration into the digitalised value stream method, considering three different traditional approaches to product family deduction. While increased clarity and informative value of the mapping results remain primary reasons for using product families and result in specific methodological requirements, automation decreases the need to reduce the analytical effort. It also changes the position of the product family deduction in the procedure of a digitalised value stream method.enfalsedigitalisationIndustry 4.0product familyValue stream mappingvirtual representativeconference paper