Digital Product Passports for Circular Supply Chains
The pilots taught us a pragmatic rule: at design time, anchor on the data model with the strongest legal anchor( Catena-X for battery, the future textile delegated act for textile) and treat every other standard as a target for two-way mapping rather than a competitor. Since May 2026 that anchor has strengthened at the horizontal level— meaning at the cross-sector, technologyneutral layer that sits above any single industry ' s schema: the JTC 24 European Standards give every sector schema a common baseline to map onto, narrowing the fragmentation surface even before the sector delegated acts arrive [ 19 ].
A further dimension of this fragmentation, raised by reviewers of this paper, concerns qualitygrade cascading in recycled materials. When a recovered material does not meet the quality or cost thresholds required for reuse in its original application— for example, battery-grade cobalt that cannot be re-refined to cell specification— it is typically redirected to lower-grade applications such as industrial alloys, pigments, or construction materials. This is a normal and economically rational step in circular material flows. It matters for DPP design because none of the schemas examined in our pilots— CX-0143, the GBA Battery Passport, or the EU Battery Regulation— currently provides a standardised field for recording a material ' s intended postrecycling application tier or the quality grade at which it was recovered. The result is that two passports may report identical recycled-content figures even when one refers to a closed-loop, same-application recovery and the other to a cascade into a secondary market. Closing this gap requires the sector schemas to expose a recovery-quality or application-tier attribute alongside the recycled-content percentage, so downstream actors and regulators can distinguish highvalue closed-loop recovery from lower-value cascade use. This is a data-model gap, not a governance gap, and it is tractable within the JTC 24 horizontal standards framework now available.
3.2 SEMANTIC DIVERGENCE ACROSS INITIATIVES
Even where two consortia describe the same data point, they often describe it differently.“ Recycled content,” for instance, is computed under different boundary conditions in the GBA Battery Passport, in CIRPASS-2 textile pilots, and in the EU Battery Regulation’ s own definitions. Adisorn, Tholen, and Götz observed early that DPP data needs a shared semantic layer if it is to inform circular decisions [ 13 ]. Berger et al. showed that DPP system requirements consistently include semantic harmonisation but rarely deliver it [ 14 ]. Zhang and Seuring’ s structured review of 82 DPP supply-chain use cases concluded that semantic divergence is the most consistently underestimated implementation risk [ 10 ].
The textile programme produced the matching example: the field labelled“ recycled content” was reported by one brand as a share of total product weight and by another as a share of the dominant fibre only, so identical labels carried incompatible denominators. The same ambiguity appeared in“ country of origin,” which one supplier resolved to the country of final assembly and another to the country of fibre production.
In both pilots, the operational response was to introduce a thin semantic translation layer between source-of-truth ERPs and PLMs and the DPP publication layer. This is closer to the Asset Administration Shell( AAS) submodel pattern than to a single canonical schema, and it scales better across sectors because it leaves source systems unchanged [ 15 ].
EDM Association – Journal of Innovation 11