The Circular Plastic Plant of Amiat: Sorting bioplastics at scale
Amiat manages soil hygiene, waste collection, and disposal services in an integrated manner, with a circular perspective of recovery and valorisation. In the daily management of services, as in future planning, Amiat adopts the principles of circular economy and ecology, responding proactively and effectively to the challenges of protecting the territory and public health.
Amiat collects 450.000 tons per year of waste, serving 1.1 million inhabitants through 19 collection sites and five plants.
The company operates a separate urban waste collection throughout the city of Turin in two different modes:
- Street Separate Collection and
- Home Separate Collection, this is divided into Porta A Porta and Eco-island modes).
In 2024, the data for the waste collection in Turin were as shown in the table on the right.
In all cases, the residual bioplastic content in the separated collected waste streams was limited and manageable. In the case of “plastic” and “organic” streams, bioplastic materials were present at a concentration below 2%, while they were present in trace in the case of paper, glass and metals separately collected, and not present in the case of wood and other separately collected waste (bulky, aggregates, WEEE, textile…).
Innovating the collection: improving the quality of collected waste
Over the past year, Amiat has implemented various strategies to enhance the quality of selected waste and is currently conducting trials with image recognition technology to collect data on the types of waste in bins, their fill levels, and the quality of sorted materials. Additionally, Amiat has identified instances of improper waste disposal near dumpsters. The next step for Amiat was to use the information to refine the best collection patch by processing data collected for each user to reschedule collection routes, and to assess the evolution over time of the quality of Separate Waste Collection.
Amiat, I.BLU sorting plant at Borgaro Torinese: The Circular Plastic Plant
The plant occupies circa 76.000 square meters and has an area of circa 20.000 square meters. It is in the north of Turin, in Borgaro Torinese. The plant uses the following technologies: 1 bag opener machine, 2 rotary screens, 4 ballistic screens, 22 optical readers, 124 conveyor belts, and 2 baling presses. It can process 100.000 tons/year (authorised processing capacity), with a maximum selection rate of 17 tons/hour, 2.430 tons of input plastic storage, and 11.210 tons of output plastic storage. There are still employees at the plant working to support sorting the waste.
Currently, Amiat, in collaboration with the Iren’s Innovation Directorate, is testing advanced technologies and software designed to support operators responsible for qualitative assessment of materials downstream of the automatic cleaning and sorting systems. These solutions aim to enhance the accuracy and efficiency of quality checks following the automated processing stage.
Within ReBioCycle, the input waste from the Circular Plastic Plant has been assessed to verify polymer compositions and quantifications, and to give special attention to the biodegradable plastic content through scheduled analyses of plastic waste. 2.000 tons of input waste have been processed during these tests, which resulted in the quantification of the content of bioplastics into 0,51% ± 0,19%.
During the past six months (April to September 2025), the ReBioCycle partners Iren and Novamont collaborated at the plant to develop a Near Infrared (NIR) database to identify rigid and flexible bioplastics in real-world plastic waste. The activity has been using virgin bioplastic items by leveraging a pre-loaded bioplastic recognition software and integrating it with different types of spectra obtained from virgin bioplastic specimens and real bioplastic items recovered manually from the plastic waste stream. The first full-scale NIR sorting activity was successfully performed.
The following tests were performed with actual plant runs, testing its capacity to select bioplastics and creating a dataset on input quantity and bioplastics output. The bioplastic sample was efficiently selected at the full-scale level, while materials different from bioplastics were removed and quantified.
