Picvisa customers discuss diverse sorting needs
Picvisa hosted a discussion among three of its customers – each with different priorities – at the conference during the Textiles Recycling Expo in Brussels in June. The session, called “One technology, three business models: How optical sorting connects the textile circular chain”, was moderated by Silvia Gregorini, Sales Manager, Picvisa, and brought together: Helio Moreira, Manager Head of Sorting, Textile House; Olivier Civil, CEO, Plaxtil Essaimons; and Andrea Sala, Project Evaluation and Development Manager, A2A. Moreira said Textile House’s sorting opera- tions have worked in the reuse market for three decades but the company made the decision to install sorting technology for recycling to maximise the value of those particular streams. Roughly 20% of its throughput is sorted for recycling and 74% for reuse.
Plaxtil integrates back into sorting and preparation Olivier Civil explained that when Plaxtil was established as a textile-to-plastic recycler in Châtellerault, France in 2020 it faced a problem because nobody could prepare textile waste to its specification. So it decided to create Essai- mons, a social company, to prepare and qualify the textile waste before recycling. Using Picvisa technology, it sorts 5,000 tonnes/year of textiles. In the Ecotex Loop project, it is extending sorting services to recyclers, and its objective is to “have many responses for many recyclers”, achieved with Picvisa’s technology, he said. Andrea Sala outlined A2A’s plans to expand into textile recovery with Picvisa
in Biella, Italy. The pilot project has the capacity to process 1,500 tonnes/year, sorting 15 different textile types, in a locally integrated chain. “Once the ecosystem is established and the value chain completed, we are planning to scale the solution up to 20,000 tonnes of treated waste per year,” said Sala.
wool/synthetic blends and technical fibres. The needlepunch nonwoven line enables the production of high- performance materials for applications such as thermal insulation, acoustic barriers, and protective solutions. At Andritz’s Textiles Recycling Expo
stand, Daniel Storms, said the profile of visitors to the event included a growing number of start-up recyclers. On display for visitors to see in action was the Andritz TeXscan predictive sorting unit, an AI-based technology to evaluate fabric quality prior to recy- cling, enabling efficient material routing for chemical recycling, nonwo- ven or spinning. A scoring system is paired with a recommended process- ing option: ringspinning, open-end spinning, nonwoven production and
chemical recycling. In parallel, auto- mated sorting systems classify gar- ments by composition and colour, remove buttons and zippers, and prepare materials for further fibre processing. In a collaboration with optical sorting technology group Pellenc ST, Andritz supplies the MultiTex system, designed to comple- ment manual sorting operations, and the Mistral+ Connect system for high-throughput recycling operations.
Start-ups using AI Technology start-ups in textile sorting also exhibited at the Textiles Recycling Expo, such as Germany-based Reverse Fashion. Founded in 2024 as a spin-off from the Technical University of Berlin, Reverse Fashion completed a
Plaxtil Essaimons’ Ecotex Loop automated sorting operation. Image: Plaxtil
seven-figure extension of its pre-seed financing round in July. It will use the funds in the commercial rollout of its AI-powered sorting technology called
Line.sort which detects damage, wear, and quality in real time. The company is also running a pilot for its Co-sort technology for enhancing manual sorting. Rebecca Guppert, Director of
Partnerships at Refiberd, gave a presentation at the conference on using AI to improve accuracy in textile composition detection. More than 40% of textile composition labels are inaccurate, she said, which impacts value and efficiency in reuse and recycling operations. Refiberd has developed two complementary technologies to address this issue: one uses hyperspectral imaging, along with a proprietary machine learning model, that can accurately detect the fibre composition of textiles within 2% of actual composition; the second is a resale grading tool that uses video cameras to identify brand, garment type, item condition and quality to speed up and standardize the process for recouping value from post-consum- er textiles.
•
www.picvisa.com •
www.valvan.com •
www.trosort.com • DataBeyond LinkedIn Page •
www.andritz.com •
www.pellencst.com •
https://reverse.fashion •
https://refiberd.com
Textiles Loop • Autumn 2026 15
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34
