Year: 2026
Type: article
Abstract: Automated textile sorting is increasingly promoted as a key enabler of fibre-to-fibre recycling, yet its effectiveness depends on whether garments are designed in ways that current near-infrared (NIR)-based systems can identify and route. This study uses retailer product data from H&M and Uniqlo in the United Kingdom and Australia to assess garment-level design compatibility with NIR-based automated textile sorting. We develop a rule-based framework that translates documented sorting constraints into measurable indicators of recognition scope, blend complexity, low-share components, black-colour proxy for optical risk, and surface–hidden material mismatch. Across 47,522 colour-specific garment variants, 50.3% violated at least one baseline sorting-relevant rule, with a sensitivity range of 45.0–64.0%. The largest barriers were unsupported recognition scope (36.4%), more-than-two-fibre blends (20.0%), and minor-component detectability (17.3%), followed by the black-colour proxy for optical risk (13.7%) and surface–hidden mismatch (8.2%). These findings highlight design-for-sorting, reliable component-level product information, and continued development of sorting capabilities as complementary prerequisites for scalable textile circularity.
Authors: Kai Li, Grit Walther
Institutions: Leiden University, RWTH Aachen University
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $3,770
Funder: Werner Siemens-Stiftung
OpenAlex ID: W7212360934