Textile recovering system and method

Abstract

There is described a handling system for textile subjects, which system comprises receiving, sorting, recovering and recycling textile subjects for reusing the textile of collected textile subjects either in form of second-hand textile subjects e.g. garments or in form of reusable textile fibers, characterized in that said handling system comprises: - a database system, - registration units - a main-sorting station - an infeed unit for a final sorting line, - a lay-up unit - a final sorting line and - recycling stations in which the textile of the textile subjects is recycled to reuseable textile products and / or textile fibers. The system may also comprise a pre-sorting station before the main-sorting station.

Description

[0001] I

[0002] TEXTILE RECOVERING SYSTEM AND METHOD

[0003] Field of the Invention

[0004] The present invention relates to a handling system for textile subjects, which system comprises a main-sorting station for receiving, sorting, recovering and recycling textile subjects.

[0005] Moreover, the invention relates to a method for handling textile subjects, and including receiving, sorting, recovering and recycling the textile subjects.

[0006] The term “textile subjects” as used in the present application will comprise textile sub- jects in form of waste, garments, furniture fabric, linens, blankets, carpets. This list is not exhaustive as any other subject which comprise textile fibers will be meant.

[0007] The term “bulk” as used in the present application will mean an amount of unsorted textile subjects which are not yet sorted into a batch or fraction being well defined.

[0008] The terms “batch” as used in the present application will mean an amount of textile subjects which are sorted into a batch or fraction being well defined and which are or can be registered in the database system.

[0009] The term “chemical” as used in the present application will mean chemical in traditional understanding but will also comprise biological, e.g. chemical recycling also includes biological recycling, and chemical dissolution also includes biological dissolution.

[0010] The term “database system” as used in the present application means a system which comprises storage facilities for data retrieved and for data being pre-entered into the system and a computer for processing the data and for controlling the steps of the method and the units incorporated in the system according to the invention. The com- puter comprises software for registration and prosecution of the data entered in the database system and for controlling. Moreover, a Manufacturing Execution System (MES) is contained in the database system. Moreover, the database system comprises software for transferring data to other external systems e.g. ERP (Enterprise Resource Planning) systems or other database systems.

[0011] The term “clippings” as used in the present application is cut textile subjects suitable to be fed through a fiber-opening line to open the fibers, e.g. cut in guillotine cutters. The term “Identification tag” as used in the present invention is not limited to a physical tag. The identification tag may be a digitally registered location of a batch or a textile subject on a conveyor.

[0012] In a situation where bulk material is received on bulk receiving conveyors (automatic reception) then a physical tag is not placed on the bulk of material. Instead, the material is tracked on the conveyors using sensors for registration of weight and location. There- fore, “identification tag” in this situation can be interpreted as the digital tracking of the material’s weight and physical location on the conveyors.

[0013] Background of the Invention

[0014] The world calls for greener and more sustainable solutions, and a huge area where the sustainability can be optimized and utilized is the textile reutilisation area.

[0015] In many years it has been known to collect textile subjects. The collected textile sub- jects are sorted and textile subjects which are found suitable for reuse have been re- used and sold as recycled textile subjects e.g. garments in second hand shops.

[0016] However, the mechanical recycling and / or reuse of textile subjects is an area which can be optimized. Especially there is a need for a system to recycle fibers from textile subjects which are not suited for as reuseable textile subjects in second hand shops.

[0017] In Europe the 2030 target of the European Green Deal is to increase the lifespan of textile products, their recycling rate, and the proportion of recycled material in new items. The EU strategy also underlines the responsibility towards the entire product value chain, comprising design, sale, usage and reusage.

[0018] The waste production in textile industry is today a serious problem in the textile indus- try. Here waste can be different things.

[0019] Textile production requires use of a lot of water.

[0020] Textile production and washing of synthetic garments releases a high number of mi- crofibers into the seas every year. Furthermore, washing synthetic clothing accounts for a high amount of the release of primary microplastics into the environment. J

[0021] Fast fashion, which constantly offers new garments at very low prices, has led to a sharp increase in the number of garments produced, used, and then discarded. These garments may often end up dumped into landfill. Moreover, companies and shops may often have unsold garments and the textile industries themselves produce samples and offcuts that must be disposed of. Also, this may often end up dumped into landfill.

[0022] Recycling of textile subjects as recycled textile fibers has been considered to be diffi- cult.

[0023] A first reason why recycling textile subjects is so difficult is the structure of the fibers. Often fibers are too short to be reused. In general, it has been considered that the fibers must be at least 15-30 mm long for recycling, but most of the time they are much shorter than that.

[0024] Moreover, many garments mix different materials. Each of them has its own recycling process. For example, recycling polyester is different from recycling cotton. But often the different materials cannot be separated.

[0025] Moreover, recycling itself is often expensive and not economically viable compared to the use of virgin raw materials.

[0026] The current methods of recycling of textile subjects may be effected in several ways.

[0027] Mechanical recycling.

[0028] This is the most common process and consists in fraying of the material and then spin- ning the fibers. It can include chemical treatments. It is mainly suitable for wool, cotton and cotton blends, but it ruins the fibers, so it can be repeated a limited number of times. The final yam is often a mix of recycled and virgin materials to ensure strength and quality.

[0029] Thermomechanical recycling.

[0030] The synthetic fibers are melted and spun again. Unfortunately, it is not suitable for all synthetic fibers.

[0031] Chemical recycling.

[0032] It includes monomer recycling and polymer recycling. The first one is suitable for most textile fibers, however, is only used for synthetic ones. In the second one, the fiber goes under mechanical processes and then to chemical dissolution. However, solvents are often dangerous for health and the environment. Moreover, within the context of the circular economy, bio-based processes could ex- tend mechanical and chemical textile recycling mechanisms in the technical cycle, en- abling greater circularity of textile subjects in the biological cycle before composting takes place.

[0033] Therefore, a system and method for optimizing the possibilities of receiving, sorting, recovering, and recycling collected textile objects are highly requested for reusing the textile subjects either in form of garments or in form of reusable fibers.

[0034] Especially there is a need for a system and a method which makes it possible to track the textile subjects and which also makes it possible to monitor and control the individ- ual process step of the method and the different apparatuses used in the system in order to optimize the sorting and recycling for having an efficient system for handling of the textile object and which makes it possible to optimize the outcome of the treat- ment in form of reuseable second-hand textile subjects e.g. garments or in form of reusable textile fibers.

[0035] Within the area of collecting and reutilisation of textile subjects, and especially within the area of reusing textile fibers, there is a need for a main-sorting of subjects which may be collected in large waste containers in which a broad variety of textile subjects may be collected e.g. from public recycling sites. It is well known that users of recycling sites may throw out very different types of textile subjects in a waste container.

[0036] Within the area of collecting and reutilisation of textile, and especially within the area of reusing textile fibers, it might seem easy and straight forward to provide and perform such recycling and / or reuse processes. However, the need for ideal preliminary pro- cesses before textile subjects of all kinds can be mechanically processed into reusable fibers is both essential but also extremely challenging.

[0037] Before textile can be recycled and / or reused properly, proper identification of one or more characteristics of ail the diverse and various textile subjects is necessary, such that the textile selected for recycling and / or reuse can be identified, classified and / or sorted properly before the subsequent recycling processes.

[0038] Today, solutions for preliminary processes of identifying, classifying and / or sorting tex- tile for recycling and / or reuse purposes are available, however, none of the known sys- tems and methods provide an optimal solution. It is a known problem that even though a huge amount of textile subjects is collected for the purpose of reusing the textile fibers, the majority of the collected textile subjects never ends in a fiber reuse process, simply because the characteristic of the textile comprises in the textile objects cannot be identified properly. This means that the textile subjects are not sorted correctly, and in the end, the final process of making a textile of a collected textile subject into reusable fibers cannot be performed, since the fibers will be destroyed, if they are not opened correctly in the reuse process.

[0039] The sad consequence of this is that the textile meant for recycling and / or reuse is either incinerated or landfilled, which is a waste of resources and damages the environment instead of providing reusable and more sustainable textile fibers.

[0040] In US 2022 / 0161298 A1 a system and a method are described which is used for clas- sifying and sorting of waste materials, especially classification and sorting of plastic materials are described in detail. The sorting is effected to obtain a classification based on different chemical signatures for the material pieces.

[0041] Other materials including textile are only mentioned briefly. There is no specification of which emitters and detectors should be used for other materials in order to establish a classification based on other parameters than the different chemical signatures.

[0042] US 2022 / 0161298 A1 discloses a system and a method which make use of emitting sources and detectors to sense / detect one or more characteristics from a textile piece for classification of different chemical signatures.

[0043] Therefore, this document neither teach use of a database system for retrieving, storing and processing data from radiation sensors or colour image sensors for colour meas- urements.

[0044] WO 2022 / 028746 A1 discloses a textile identification system in which X-ray identifica- tion is used.

[0045] US 7.564.942 B2 discloses an X-ray CT apparatus for CT scanning of a person lying on a moveable table.

[0046] KONSTANTINIDIS FOTIOS K ET AL: ’’Multi-sensor cyber -physical sorting system (CPSS) based on Industri 4.0 principles: A multifunctional approach”, PROCEDIA COMPUTER SCIENCE, ELSEVIER, AMSTERDAM, NL, vol. 217, January 2023 (2023-01), pages 227-237, XP087246859, ISSN: 1877-0509, DOI: 10.1016 / J. PROCS.2022.12.218 discloses a sorting system making use of robots re- ceiving a signal of the position of a subject on a conveyor from a camera.

[0047] US 2019 / 0226125 A1 describes a system where fibers are to be recycled. To recycle the fibers, a two-stage disinfection and sorting process is carried out.

[0048] Between the two disinfection processes, a pre-treatment of the incoming material is carried out. The material is fed in batches via containers from the first disinfection unit to a pre-treatment unit. From the pre-treatment unit, the textile subjects are fed individ- ually via manual processing stations into a detection device.

[0049] A treatment takes place in the detection device, and the individual textile subjects are fed individually to a color detection device.

[0050] From the color detection device, the textile subjects are fed individually to the sorting device, where they are sorted into category containers. The category containers will then be fed via a second disinfection device to a fiber processing device, where the textile subjects are taken out of the containers and cut into smaller pieces for fiber processing.

[0051] The main object is to disinfect textile subjects and then tear these textile subjects to provide “recycling fibers” with physical properties so that they can be reused, for ex- ample for the production of yarn.

[0052] D1 does not mention:

[0053] -a database system,

[0054] -registration units for entering user input,

[0055] -weight sensors,

[0056] -batches except mentioning the containers and the first disinfection device, which leads the textile subjects to the pre-treatment device which may be considered to be a “main- sorting station”,

[0057] -identification tag to each batch,

[0058] -sensors for detecting properties except a color detecting device and a device for loca- tion of metal objects or

[0059] -chemical recycling units. Hence, an improved system and method, preferably an efficient and reliable system and method, suitable for use in the recycling process for identifying, classifying and / or sorting textiles would be advantageous, and in particular a more optimal, reliable, effi- cient, precise, economically, and sustainable system and method would be advanta- geous.

[0060] Object of the Invention

[0061] It is an object of the present invention is to provide a system and a method which makes it possible in an effective way to handle and track the textile subjects and which also makes it possible to monitor and control the individual process step of the method and the different apparatuses used in the system in order to control the sorting and recy- cling procedure for optimizing the outcome of the treatment in form of reuseable sec- ond-hand textile subjects e.g. garments or in form of reusable textile fibers.

[0062] Description of the Invention

[0063] This object of the invention is achieved with a system mentioned in the introduction and which is peculiar in that the reusing is either in form of second-hand textile subjects e.g. garments or in form of reusable textile fibers, or other types of disintegrated textile materials or recycled materials or downcycled materials, wherein said handling system comprises:

[0064] - a database system adapted to at least retrieve, store, and process data measured by sensors being a part of the handling system, and adapted to control an operation of individual system units,

[0065] - registration units for entering user input for delivery data relating to the bulk of textile subjects entering the handling system, and which registration units are connected with the database system in which the delivery data is stored,

[0066] - a main-sorting station comprising

[0067] - a main bulk receiving conveyor having an infeed unit,

[0068] - sensor for registration at least the weight of the textile objects, which sensors are connected with the database system in which the weight registration is transferred to an identification tag,

[0069] - a main-sorting unit for main-sorting the received textile objects and estab- lishing batches of main-sorted textile subjects after the main-sorting of the re- ceived textile object,

[0070] - an application unit for application of the identification tag to each batch,

[0071] - a registration sensor for registration of the identification tag, wherein the identification tag is associated with the batch in the database system, - conveyors for transporting the textile subjects batchwise to a buffer from which the batches of main-sorted textile subjects are transferred to an infeed unit for a final sorting line, wherein the infeed unit comprises

[0072] - a buffer space for containing the textile units,

[0073] - a registration sensor for registration of the identification tag, wherein the identification tag is associated with the batch of main-sorted textile subjects in the database system,

[0074] - a lay-up unit which transfers the textile subjects individually to the final sort- ing line, wherein the lay-up unit comprises

[0075] - at least one gripper for gripping the textile subjects individually from the buffer space

[0076] - a registration sensor for registration that only one textile subject is gripped by the at least one gripper and transferred to a lay-up conveyor for transferring the textile subjects individually to the final sorting line with a mutual distance between consecutive textile subjects, wherein the final sorting line comprises

[0077] - conveyors for transporting the textile subjects individually past several con- secutively arranged property sensors, preferably three or more, for registration of different properties of each textile subject and which property sensors are connected with the database system for transferring the registered properties to the database system in which the registered properties are associated to the registered batch,

[0078] - at least one sorting conveyor for transferring the textile subjects past a num- ber of final sorting units which are connected with the database system and are arranged for an automatically transferring the textile subjects from the at least one sorting conveyor to fraction containers where each fraction container is associated with one of the final sorting units in which automatically transfer- ring is based on the registered properties of the individual textile subjects whereby a fraction of all textile subjects which has same properties will be collected in one of the fraction containers,

[0079] - recycling stations which comprise mechanical recycling units, chemical recycling units and downcycling units in which units the textile of the textile subjects is recycled to reuseable textile products and / or textile fibers, or other types of disintegrated textile materials.

[0080] The registration sensor for registration that only one textile subject is gripped may be combined with a camera which is also connected with the database system. Hereby a higher security is obtained that only one textile subject is transferred.

[0081] The method according to the present invention is peculiar in that the reusing is either in form of second-hand textile subjects e.g. garments or in form of reusable textile fi- bers, or other types of disintegrated textile materials or recycled materials or downcycled materials, wherein said handling method comprises the steps of:

[0082] - controlling the method steps with a database system adapted to at least retrieve, store, and process data measured by sensors used in steps of the method and adapted to use the processed data to control operation of individual system units used for the method,

[0083] - entering and registration of user input for delivery data relating to the bulk of textile subjects entering the handling system, and forwarding the registrations to the database system in which the delivery data is stored,

[0084] - effecting a main-sorting in the main-sorting station, which main-sorting comprises

[0085] - receiving a bulk of textile subjects on a main bulk receiving conveyor,

[0086] - registration of at least the weight of the textile objects with at least one sensor and transferring the weight registration to the database system in which the weight registration stored and combined with the delivery data, hereby provid- ing a batch of received textile subjects

[0087] - transferring the weight registration and the delivery data to an identification tag,

[0088] - establishing batches of main-sorted textile subjects after the main-sorting of the received textile objects in a main-sorting unit,

[0089] - application of the identification tag to each batch in an application unit,

[0090] - registration of the identification tag with a registration sensor,

[0091] - associating the identification tag with the batch in the database system,

[0092] - transporting the textile subjects batchwise to a buffer,

[0093] - transferred the batches of main-sorted textile subjects to an infeed unit for a final sorting line,

[0094] - buffering the batches of main-sorted textile subjects in a buffer space in the infeed unit,

[0095] - registration of the identification tag with a registration sensor, - associating the identification tag with the batch of main-sorted textile subjects in the database system,

[0096] - transferring the textile subjects individually to a final sorting line with a lay- up unit, by gripping the textile subjects individually from the buffer space with at least one gripper

[0097] - registration that only one textile subject is gripped by the at least one gripper and transferred to individually to the final sorting line with a mutual distance between consecutive textile subjects,

[0098] - conveying the textile subjects individually past several consecutively ar- ranged property sensors preferably three or more,

[0099] - registration of different properties of each textile subject,

[0100] - transferring the registered properties to the database system,

[0101] -saving data for the registered properties in the database system,

[0102] - associating in the database system the saved data of the registered proper- ties of each textile subject to a specific fraction,

[0103] - associating the registered properties to the registered batch,

[0104] - transferring the textile subjects past a number of final sorting units with at least one sorting conveyor

[0105] - sorting the textile subject as the sorting unit is controlled based on the regis- tered properties of the textile subjects

[0106] - automatically transferring the textile subjects from the at least one final sort- ing conveyor to fraction containers based on the registered properties of the individual textile subjects, where each fraction container is associated with one of the final sorting units,

[0107] - collecting the fraction of all textile subjects which has same properties in one of the fraction containers,

[0108] -recycling the textile of the textile subjects to reuseable textile products and / or textile fibers in recycling stations comprising mechanical recycling units, chemical recycling units and downcycling units.

[0109] In one aspect of the invention, the handling system further comprises that a pre-sorting station is provided before the main-sorting station, which pre-sorting station comprises

[0110] - a bulk receiving station comprising a receiving unit for receiving waste con- tainers containing a bulk of the textile subject to be handled and which receiv- ing unit comprises - bulk receiving conveyors,

[0111] - weighing sensors for registration of the weight of the bulk material, which weighing sensors are connected with the database system,

[0112] - transfer conveyors for transporting the batches of pre-sorted and weighed textile subjects to the main-sorting station.

[0113] I the same way the method in one aspect of the invention involves a step comprising pre-sorting before the main-sorting step, which pre-sorting step comprises

[0114] - receiving waste containers containing a bulk of the textile subject to be han- dled in a bulk receiving station which bulk receiving station comprises a re- ceiving unit

[0115] - entering and registration of user input for delivery data relating to the textile subjects received in the bulk receiving station and forwarding the registrations to the database system in which the delivery data is stored,

[0116] - conveying textile subjects in bulk receiving conveyors,

[0117] - registration of the weight of the bulk material with weighing sensors con- nected with the bulk receiving conveyors,

[0118] -transferring the registration of the weight to the database system where it is associated with the registration of user input for delivery data relating to the bulk of textile subjects, hereby providing a batch of received textile subjects,

[0119] - transporting the batch of pre-sorted and weighed textile subjects to the main- sorting station for effecting the main-sorting. in case the material is received in a way that requires the pre-sorting then batches may be established in the pre-sorting step, and it is possible in the main-sorting to skip the steps of

[0120] - receiving a bulk of textile subjects on a main bulk receiving conveyor,

[0121] - registration of at least the weight of the textile objects with at least one sensor and transferring the weight registration to the database system in which the weight registration stored and combined with the delivery data, hereby provid- ing a batch of received textile subjects.

[0122] Accordingly, the main-sorting is effected on a batch and the first step is to provide the batch with the identification tag.

[0123] In one aspect of the invention, the main-sorting station comprises - an unpacking and main-sorting unit for sorting out items which are not textile subjects to be handled in the following sorting and recycling.

[0124] I the same way the method in one aspect of the invention involves that the main-sorting step comprises a step of

[0125] - sorting out items, which are not textile subjects to be handled in the following sorting and recycling, in an unpacking and main-sorting sorting unit.

[0126] Material to be treated in a system and a method according to the present invention may originate from different sources and will therefore be received in different ways. The material may come from waste from households without any sorting, from public recy- cling sites having pre-sorting systems, from industrial facilities or from retail industry.

[0127] The material received for being treated in the system and according to the method of the present invention may be divided into four main groups which each are handled in slightly different ways.

[0128] -Bulk material for pre-sorting

[0129] This is the largest fraction of received material and mainly consists of post-consumer textile waste from households as municipal waste.

[0130] This material is unloaded directly onto the bulk receiving conveyors, outfitted with weighing cells for registration, which registration of the weight is transferred to the da- tabase where it is associated with the delivery data.

[0131] Once received, the material is automatically or manually pre-sorted on pre-sorting con- veyors before it is transferred to the main-sorting line by transfer conveyors. The trans- fer conveyor may be identical with the pre-sorting conveyor.

[0132] The pre-sorting will typically be an evaluation of the humidity. In case the bulk material received is too wet to be handled the pr-sorting conveyor is reversed and the material is discarded as waste.

[0133] Typically, the conveyors for the transfer are roller conveyors also called roller cages.

[0134] -Bulk material for main-sorting or final sorting Pre-sorted material delivered in bulk or batch can come from industrial laundries or municipalities that have pre-sorting capabilities.

[0135] The material is unloaded in the main-sorting station. It may be delivered directly on the main bulk receiving conveyor or on the floor in the main-sorting station to be picked up and fed directly to the main-sorting conveyor, e.g. using a truck. The weight is regis- tered in this step and being registered in the database.

[0136] -Packaged material for main-sorting

[0137] Some municipalities prefer to collect and deliver the waste textile subjects in conveyors in the form of big bags or roller cages.

[0138] This material may be unloaded directly onto the main-sorting conveyors or on the floor in the main-sorting station. The transfer to the main-sorting conveyors is done either manually from cages or by truck from big bags.

[0139] Once the textile subjects are main-sorted, the textile subjects are transported to the final-sorting station. This may be effected on conveyor belts or alternatively the main- sorted textile objects are put into roller cages to be fed into the final-sorting station.

[0140] -Packaged material for final sorting

[0141] Smaller customers, typically from the textile retail industry or industrial laundries can deliver the textile subjects in cardboard boxes, roller cages or big bags.

[0142] The material is either manually unpacked and put into conveyors, e.g. in form of roller cages to be fed into the final sorting line, or they are directly emptied into the infeed for the final- sorting conveyor of the final-sorting station.

[0143] The material is weighed, and the data is registered in the database. This is in one example done manually using trucks or similar.

[0144] For example, the technical units for receiving the material are the large bulk receiving conveyors. They are in principle “just” broad belt conveyors or at least belt conveyors wider than a container. The conveyors are fitted with weighing cells. The weighing cells are coupled with control software which registers the weight on the large bulk receiving conveyors and couples it to user-input delivery data. The main technology in main-sorting are the main-sorting conveyors which are coupled to the bulk receiving conveyors through automation software.

[0145] The pace of the main-sorting conveyors is controlled automatically or manually through buttons at the main-sorting conveyors. This will then control the pace of the upstream conveyors.

[0146] All waste that is removed in the main-sorting step is registered either manually or au- tomatically and coupled to the delivery by a Manufacturing Execution System (MES) contained in the database system.

[0147] All the main-sorted textile subjects coming from the main-sorting conveyors are pro- vided in batch-from, which is typically in roller conveyors in form of roller cages. E.g. the pre-sorted textile subjects are lifted by the last conveyor and dropped into roller cages. Each batch / roller cage is registered in the MES to couple the batch / roller cage with main-sorted textile subjects to the delivery data of the specific delivery.

[0148] The above-mentioned main-sorting provides for an effective way to handle the textile subjects compared to prior art where main-sorting was done manually.

[0149] The system is suited for upgrading to handle large amounts of waste. The bulk receiv- ing conveyor may be connected to a distribution conveyor which distributes the bulk of textile subjects to several main-sorting conveyors. E.g. the bulk receiving conveyor in form of a belt conveyor is connected with a transversal belt conveyor in up streams of the main-sorting conveyors to distribute the material from the bulk receiving conveyor to two or more main-sorting conveyors.

[0150] The main-sorting of the textile subjects is always effected in the main-sorting station independent on whether a pre-sorting is done in the system or is done earlier before the waste is delivered to a system according to the invention.

[0151] Therefore, material entering the main-sorting is freed from items which are not textile subjects to be handled in the following sorting and recycling.

[0152] The main-sorting comprises receiving the bulk of textile subjects on the main bulk re- ceiving conveyor or receiving the batch of textile subjects on the main-sorting conveyors. In the main-sorting unit batches of the main-sorted textile objects are es- tablished and each batch is provided with the identification tag.

[0153] The conveyors for transporting the textile subjects batchwise to the buffer upstream of the infeed unit for the final sorting line are typical roller cages. Alternatively, belt con- veyors may be used.

[0154] Roller cages each with a batch of main-sorted textile subjects are emptied into the infeed unit.

[0155] In case that the roller cage is from a different delivery than the previously emptied cage, the delivery new number is given to infeed unit for the final sorting line through the MES.

[0156] The final sorting line then keeps track of the separation between the two deliveries.

[0157] The roller cages are arranged in a cage-tipping machine tilting the cages. The cage- tipping machine works by arranging the cage into the machine and manually pressing a button or automatically controlled by a sensor which register the insertion of a cage, after which the machine will automatically empty the cage into the infeed unit upstream of the final sorting line. Through the database system the cage tipper communicates with the final sorting line, and the cage tipper is controlled to only tilt the roller cage when there is space in the infeed unit.

[0158] In the infeed unit a registration is effected based on information on the identification tag. The infeed unit comprises the buffer space from which the textile subjects is picked up and individually is transferred to the final sorting line by the lay-up unit.

[0159] The gripper for gripping the textile subjects individually from the buffer space will typi- cally be operated by a robot controlled by the signal from the registration sensor. The registration sensor may be combined with a camera. Hereby it is ensured that only one textile subject is gripped by the at gripper and transferred to the lay-up conveyor up- stream of the final sorting line.

[0160] In one example the infeed unit comprises two grippers each operated by a robot. Hereby a high security is obtained that the textile subjects are individually transferred. The robots may also be called lay-up robots. The function of the two lay-up robots is to feed the lay-up conveyor for the final sorting line with as many textile subjects as possible, without any of them overlapping. In other words, they pick textile subjects from a pile in the buffer and place them on the lay-up conveyor, one-by-one.

[0161] In an example of the infeed unit a conveyor is arranged in the bottom of the buffer for transporting the textile subjects to at pick-up position where the griper grips the textile subjects.

[0162] The infeed unit works by feeding the textile subjects on the conveyor in the bottom of the buffer to the pick-up position. When a textile subject is in view of a camera arranged above the pile of textile subjects in the buffer a first gripper goes down in the and picks up the textile subject using a specialized gripper.

[0163] The first gripper is embodied with gripper fingers which grip the textile subject. Once the textile subject is gripped (a sensor in gripper confirms this), the textile subject is lifted to the top of the buffer. The second gripper takes the textile subject from the first gripper and delivers the textile subject on the lay-up conveyor with a mutual distance between consecutive textile subjects.

[0164] The two grippers are identical or can be of different types. Each gripper may be mounted on and operated by robots, which robots may be 6-axis robots.

[0165] Alternatively, only a first gripper is mounted on and operated by a robot and the second gripper may be mounted in guide rails, e.g. combined with a length adjustable wire to establish a two- dimensional operation of the gripper. The second gripper can have a simpler construction than the first gripper as it “only” has the task to place the textile subject on the lay-up conveyor.

[0166] By gripping the textile subject two times, the risk of double-picks is reduced signifi- cantly.

[0167] In the infeed unit lifting conveyors may be arranged at the side of the buffer for lifting the textile subject off the conveyor in the bottom of the buffer. The pick-up position will be in the area where the textile subjects are lifted up from the bottom conveyor. After this the textile subjects are handled by the two lay-up robots. Once the textile subjects have been put onto the final sorting line, they are arranged on the conveyors for transporting them individually past the property sensors.

[0168] In an example of the invention three sensors are used: an industrial RGB-camera, a NIR sensor and an X-ray sensor. Sensors are described in further detail below. Even though the example describes three sensors it is possible to use less sensors and it is also possible to use more than three sensors according to the present invention.

[0169] Once the textile subject has passed all the sensors, it is transferred to a sorting section comprising the sorting conveyors. This may be effected by squeezing the individually arranged textile subjects between vertical conveyors.

[0170] The sensor setup may in one example be as follows. The RGB and NIR sensors work by measuring reflected light, the X-ray sensor works by measuring penetrating X-rays. This X-ray method is similar to X-ray scanners used in the medical industry for diagno- sis or in airport security and is not to be confused with XRF (X-ray fluorescence) which works by measuring X-rays emitted by a subject (similar to reflection).

[0171] More sensor units could be added in the future including:

[0172] Teraherz waves: Very high frequency electromagnetic waves are passed through the textile subjects and the waves are measured on the other side. It can see “inside ob- jects” and be used to detect hard parts (buttons, zippers and so on). Can be an alter- native / supplement to X-ray. Such a system is e.g. of the type being developed for tex- tile applications by French start-up: “OPTIKAN”. However, other systems using this technology may also be used

[0173] Radar sensor: Similarly to the Teraherz sensor, a radar sensor can also be used to see inside things, but the resolution of the current system developed by Balluff is low.

[0174] High-resolution laser scanner: Are often used in industry for high accuracy QC but could potentially be used to detect the structure of the material (knit, weave, frotte...).

[0175] Makro camera,' (very)high resolution camera: Alternatively, to the high-resolution laser scanner, a secondary camera setup could also be used to detect the structure of the material, if a sufficiently detailed image of the fabric can be obtained. MIR: Works similarly to NIR but looks at higher wavelengths of the electromagnetic spectrum, to provide much higher detailed information.

[0176] XRF: Can be used to classify materials similarly to the NIR-technology.

[0177] Other NIR sensors: The NIR unit currently utilized used spectrography to identify the material. This means that the algorithm looks at the light intensity as a function of wave- lengths, to give a “fingerprint” of the chemical composition.

[0178] E.g. a high-resolution spectrum is measured at 15 points along the conveyor, giving a resolution of 15 pixels over 600mm conveyor. In other words, the NIR has a low spatial resolution, but a high spectral resolution, as we get info about the light intensity at many different wavelengths.

[0179] Alternatively, hyperspectral cameras also look in the NIR-region can have a much higher spatial resolution (more pixels). As a result, the output of the hyperspectral cam- era is a series of greyscale images, at different wavelengths. Such a system can detect smaller areas / objects which might be interesting but can have less accuracy in terms of the chemical detection. It can be interesting for detecting for example mold, which has a large intensity at a specific wavelength, which is why this type of sensor is often used in QC of food processing lines.

[0180] The combination of sensors that can be used gives a unique opportunity to see every- thing of high relevance to the mechanical and chemical recycling.

[0181] A higher number of sensor technologies may be used for optimizing the sorting of the textile subjects.

[0182] The sensor technologies are presented in no particular order:

[0183] 1. Near-Infrared (NIR) Spectroscopy / Hyperspectral imaging

[0184] 2. Visible Light Sorting (Optical Sorting)

[0185] 3. X-ray Technology (XRT and XRF)

[0186] 4. Electromagnetic Sensors

[0187] 5. 3D Laser Scanning

[0188] 6. Hyperspectral Imaging

[0189] 7. Terahertz Wave Sensors

[0190] 8. RFID Tagging 9. Mid- Infrared spectroscopy

[0191] 10. Raman spectroscopy

[0192] 11. UV spectroscopy

[0193] 12. Tactile sensors

[0194] Comments to the relevance of the abovementioned sensor technologies are given be- low.

[0195] 1. Near-infrared (NIR) Spectroscopy is also referred to as SWIR spectroscopy.

[0196] Shining near-infrared ‘light’ on a subject and measuring the reflected light in the NIR and SWIR spectra. Specialized software (Al) is used to compare the spectrum of the reflected light to a database of pre-taught spectra. The spectrum will be specific to the chemical composition of the subject, so therefore the method can be used to recognize any chemically unique material, assuming that it has been pre-taught, and the sensor is sufficiently precise.

[0197] The system uses a multiplexor setup, which means that the reflected NIR-light is meas- ured by a line of separate lenses along the conveyor, each lens corresponding to one pixel. The lenses are each connected to multiplexor units. Inside the multiplexor, the signal from each cable is reflected upwards into a single sensor by a rotating mirror. By rotating at around 30 -100 Hz, the entire line of pixels is all measured one-by-one 30- 100 times pr. second. Each measurement gives the intensity at more than 100 different wavelengths between 1000 nm and 2500 nm.

[0198] The multiplexor setup allows for a high-precision measurement with a high spectral resolution, making the system capable of distinguishing different elements with similar NIR-reflections. Examples include differentiating between different polymers or distin- guishing specific blends of materials from each other.

[0199] Primary application

[0200] Determining the fiber composition of the textile subjects. This is needed to realize the subsequent recycling processes.

[0201] Secondary application

[0202] See some contaminants such as buttons, zippers and prints. Can also be taught to detect moisture contents, dirt, mold, chemical treatments or special dyes in the fabrics. 2. Visible Light Sorting (Optical Sorting)

[0203] Industrial cameras, typically configured in a line-scan setup with a constant and con- trolled lighting environment (vision cell). Various algorithms can be applied, based on the need, ranging from traditional machine vision methods to machine learning meth- ods (Al).

[0204] Primary application

[0205] Color-sorting to avoid having to re-dye the recycled fibers / yarns. This will greatly re- duce the resources needed to recycle the textile subjects. Furthermore, in some chem- ical recycling or downcycling cases, color sorting is needed or can at least increase the value of the sorted textile fractions.

[0206] Secondary applications

[0207] Patterns and textile subjects of different object types can be trained on an Al, for recog- nition of object types in a more abstract context. For example, detecting denim jeans, shirts, shoes, underwear or other specific items which can be valuable to either isolate in its own fraction or keep out of other fractions. This requires sophisticated machine learning algorithms (Al) trained on large datasets.

[0208] Furthermore, having the right camera setup (high-resolution camera with macro lens), one can achieve highly detailed images of the surfaces of the textile subjects. Using machine learning (Al), an algorithm can be taught to recognize different weave / knit patterns to separate the textile subjects by fabric type. This can add value in the context of mechanical recycling (more efficient recycling process).

[0209] Tertiary applications

[0210] Optical sorting can help in identifying impurities such as Lurex thread (metaling thread used for glitter effects), buttons, zippers, care-labels, stitches or polymer prints. This can be useful in relation to sorting of whole garments, but also in relation to sorting of cut / disintegrated textile subjects.

[0211] Lastly, the camera can be used to infer the silhouette of the textile subjects to indicate their size. Very small or very large items might not be fit for recycling. 3. X-ray Technology (XRT and XRF)

[0212] X-ray sensor systems consist of an X-ray emitter (specialized light bulb which emits X- rays) and a sensor to measure the radiation. Two well established applications of this concept is XRT and XRF.

[0213] XRT (X-ray transmission)

[0214] Uses X-ray emitter on one side of the textile subject and measuring the transmitted radiation on the other side of the textile subject. Materials with different densities (and thicknesses) will absorb different amounts of radiation, giving an indication of what is within the textile subjects.

[0215] XRF (X-ray fluorescence)

[0216] Emits X-rays unto the textile subject and measures the radiation emitted back (similar to reflection). The radiation emitted back from the test subject will be characteristic of the chemical composition of the textile subjects (or contaminants on it).

[0217] Applications of XRT:

[0218] Detecting the presence of hard parts on / within the textile subjects like zippers and but- tons based on density differences. Machine vision algorithms (traditional or Al-based) can be used to recognize different geometries / objects on these subjects. This allows for separation of “clean textile subjects” without accessories, which can then be directly recycled (no removal of accessories or secondary sorting needed). Due to the radiation absorbing properties and high densities (compared to textile subjects) of metals, XRT is highly efficient at detecting even small metal pieces within the textile subjects.

[0219] Furthermore, XRT can be used to indirectly measure the mass of the textile object with an accuracy below + / - 10%.

[0220] Applications of XRF

[0221] The technology is highly utilized in scrap metal sorting and mining applications, as it is highly efficient at differentiating between different metal components. This makes the technology suitable for identifying metal impurities on the surface of the textile subjects.

[0222] 4. Electromagnetic Sensors

[0223] Electromagnetic sensors take advantage of metallic components interaction with elec- tromagnetic fields. There are (at least) three different types: -Metal detectors (inductive sensors) generate a magnetic field using a coil. When a metallic object (such as buttons or zippers on textile subjects) enters this magnetic field, it alters the field's shape and strength. The detector senses this change and sig- nals the presence of metal.

[0224] -Eddy current separators are specific to non-ferrous metal detection. They use a high- frequency electromagnetic field to induce eddy currents in non-ferrous metals.

[0225] -Magnetic sensors use magnets or electromagnets to attract and remove ferrous met- als from a mixture of materials. Permanent magnets or electromagnets create a mag- netic field that attracts ferrous metal objects, effectively separating them from non-me- tallic materials.

[0226] Primary application

[0227] Used for detecting metal components, such as zippers and buttons, from textile sub- jects. This is most useful in a secondary sorting of cut / disintegrated textile subjects (for accessories removal) and can also be used as a QC before the fiber opening process.

[0228] 5. 3D Laser Scanning

[0229] Laser scanners emit a laser-line on the conveyor, and as the textile subjects pass un- der the laser line, the shape of the laser line is altered. The sensor has a camera to detect the alteration in the shape of the laser line, to determine a 3D image of the textile. High resolution laser scanners have tolerances small enough, that they can capture the surface texture of the textile.

[0230] Application of low-resolution laser scanners

[0231] Measuring the volume, shape and size of the textile objects.

[0232] Application of high-resolution laser scanners

[0233] High-resolution laser scanners can capture the fine details of a textile's texture and weave pattern. This detailed information can be used to classify textile subjects based on their fabric type, such as distinguishing between knits and woven fabrics, or identi- fying specific weave patterns. This can provide the ability to separate knits from weaves, and potentially even different types of knits / weaves from each other. Having this capability can optimize the output of the mechanical recycling, reducing waste and increasing quality of the process. 6. Hyperspectral Imaging

[0234] Hyperspectral cameras take (relatively) high-resolution images at different wave- lengths, often in the NIR-region (and SWIR-region). So, for every time the camera is triggered, many greyscale images are obtained, each corresponding to a specific wavelength. Compared to NIR spectroscopy, the spatial resolution is much higher (600-1840 pixels), sometimes at a cost of lower spectral resolution (they look at fewer wavelengths). The concept is also referred to as push-broom hyperspectral imaging.

[0235] Primary application

[0236] The primary application of hyperspectral imaging is similar to NIR spectroscopy, i.e. identification of material composition. High-end hyperspectral cameras can provide a spectral resolution similar to the multi-plexor systems, but with a much higher spatial resolution.

[0237] Secondary application

[0238] Apart from being able to match a multi-plexor in terms of precision and robustness in detection of material composition of the textile subjects, a hyperspectral camera can also detect smaller areas / objects more robustly, such as buttons and zippers on the surface of the textile.

[0239] Lastly, hyperspectral systems might be efficient in detecting polymer prints, dirt, mold or other contaminants which can be in small areas or spread in small particles over textile. Examples of this are seen in the food processing industry.

[0240] Tertiary application

[0241] The large spatial resolution makes the system well suited for sorting cut / disintegrated textile clippings to remove accessories,

[0242] 7. Terahertz Wave Sensors

[0243] Electromagnetic waves with wavelengths higher than infrared waves and lower than microwaves (0,1-10 THz). The waves are either passed through the textile subjects and the waves are measured on the other side, or they are reflected by the textile subjects to measure the reflected waves. Different materials will absorb the waves dif- ferently, making it possible to detect different materials.

[0244] Primary application

[0245] In textile sorting, the penetrating waves can be used as an alternative to X-ray. For example, they will penetrate textile subjects more easily than hard plastics and espe- cially metal. When measuring the penetrating waves on the other side, an image can be obtained (using a lot of data processing) similarly to an X-ray image. This allows the sensor to see “inside objects” and detect hard parts (buttons, zippers and so on) within / on the objects.

[0246] Especially for the case of plastics, which can be hard to robustly detect using X-ray

[0247] 8. RFID Tagging

[0248] Embedding RFID tags in textile subjects at the point of manufacture can greatly en- hance sorting capabilities by providing exact material composition, making recycling more efficient. Currently, RFID chips are used by many industrial laundries to ID the textile subjects. The RFID chip contains a code which refers to a database, telling which category the textile belongs to.

[0249] Primary use

[0250] In a waste sorting perspective, RFID chips can contain information on what the textile is made of which accessories are present, which fabric type it is made of, when / where it was made at if recycled fibers are within the object.

[0251] Secondary use

[0252] As many of the waste textile subjects received from industrial laundries already have an RFID-chip, these could be re-coded when leaving the laundry to code the relevant information for sorting for recycling. In cases where an industrial laundry company is interested in investing in sorting capacity to ensure input of recycled fibers for future products, this might be relevant.

[0253] 9. Mid-Infrared (Mid-IR) Spectroscopy

[0254] Mid-IR spectroscopy is a well-established technique in analytical chemistry and mate- rial science, known for its ability to provide detailed information about molecular struc- tures through the absorption of mid-infrared light at specific wavelengths. The method works similarly to NIR-spectroscopy and can also be utilized in hyperspectral solutions. However, Mid-IR is significantly more sensitive towards the chemical structures within a material, making it able to give much more detailed information about the textile sub- jects. Primary use

[0255] Mid-IR can be useful for identifying and sorting textile subjects based on their specific chemical compositions at a higher level of detail than what is possible with NIR. Exam- ples include:

[0256] • Differentiating between polyester types

[0257] ® Detecting chemical treatments or coatings applied to textile subjects.

[0258] The above-mentioned advantages are valuable for recycling processes that require very high purity material streams.

[0259] 10. Raman spectroscopy

[0260] A mature technology used in various fields for molecular and material analysis, using high-power lasers to induce “Raman signals” in the subject. It measures the scattering of the laser light, providing a unique spectral fingerprint for different materials. This is also comparable to NIR spectroscopy. Its direct application in waste textile sorting on an industrial scale is emerging.

[0261] Primary use of Raman spectroscopy

[0262] Raman spectroscopy can complement NIR and Mid-IR techniques by providing addi- tional information that can help distinguish between materials with similar NIR or Mid- IR spectra. The method requires minimal sample preparation, making it suitable for on- line sorting applications where speed and efficiency are critical.

[0263] 11. Ultraviolet (UV) Spectroscopy

[0264] UV spectroscopy measures the absorption of ultraviolet light by substances. Its appli- cation for textile sorting is more limited due to the specific range of materials and con- ditions under which UV absorption provides useful discriminatory information.

[0265] Primary use of UV-spectroscopy

[0266] The method can be used to identify specific dyes, finishers, or other types of surface treatments in the textile subjects.

[0267] 7. Tactile sensors

[0268] While not referring to a specific sensor technology, this last category covers advanced sensor(s) used in combination with robots to manipulate the textile subjects. An example is tactile sensors, which are sensors that can “feel” properties of objects by physically touching / manipulating them.

[0269] Primary use

[0270] The properties to be measured include:

[0271] • Texture, stiffness and surface roughness / smoothness can give information on the type of fabric (knit / weave) or material composition (natural / synthetic)

[0272] • Pressure distribution can give information about hard parts present in the textile subjects.

[0273] The final sorting units may comprise reversible conveyors. In such example the sorting works by many short sections of inclined, reversible conveyors. Once a textile subject reaches the right sorting position, the conveyor reverses to sort the textile subject into the correct fraction container for the textile subject. The fraction container may be in the form of an underlying roller cage. This sorting requires the textile subjects to be separated and works as an alternative to sorting by air-blowing. Alternatively, the final sorting units may be air nozzles which are placed along one side of a long horizontal conveyors to blow the textile subjects of the conveyor once they reach the desired sorting position. Air blowing lines require more space, create significantly more noise and dust pollution, and they require more energy. Conveyor sorting, on the other hand, is more expensive.

[0274] In the database system a recognition software takes all the sensor data and combines it to decide how to sort the textile subject. The recognition software package can be divided into three overall layers:

[0275] A. The data is measured by the three or more different sensors.

[0276] B. Features are extracted from the sensor data, meaning that different properties of the textile are being decided. For example, a NIR sensor software uses a multi-stage machine learning algorithm to decide the material composition that best matches each pixel of the textile subject. And a camera uses traditional machine vision to detect the color category or color distribution of the textile subject.

[0277] C. Having all the outputs from the sensors, a decision logic is applied to decide the sorting position for the textile subject. The decision logic may use different rules to define what types of textile subject that can go into each fraction con- tainer. z /

[0278] The term “color distribution” is a percentage distribution of colors over the surface of the textile subject, while the term “color category” is a single overall color assigned to the textile subject.

[0279] The sorting may be based on different features decided by the operator of the system. In the following there are examples on features to be extracted from the sensor data by extraction algorithms.

[0280] Material composition - for each piece of textile, the distribution material is given in pixels and percentage. Example:

[0281] 60% of the textile is recognized as cotton with less than 5% elastane

[0282] 25% of the textile is recognized as pure cotton

[0283] 5% other (noise).

[0284] Color category or color distribution - a number of categories of color or color distribu- tions are defined using software and predefined data contained in the database system by taking images of the textile subjects and assigning them to a category. The RGB- camera sensor is used for this. For example, 25 images of different blue textile subjects may be used to define a blue category. Database images can be used for this training.

[0285] Textile class - similarly to the color category, the color images of the textile subjects are also used in a machine learning classifier (Al) to recognize specific types of items. This could be denim jeans, workwear (with reflective tape), socks, underwear and oth- ers.

[0286] Significantly more data is needed to robustly train these models (in the order of 500- 1000 images). The model may have an underlying neural network developed for indus- trial laundries and retraining this model for the present invention.

[0287] Textile size - the silhouette of the textile subjects can be extracted to indirectly meas- ure the size of the textile. The output is given as number representing the surface area of the textile.

[0288] Accessories - the X-ray images will be one color in areas where there is metal, and another color in areas where there is hard plastic. However, the hard plastic is harder to detect robustly compared to metal, especially in cases of thick textile subjects, where the plastic accessories are “drowned” in noise from the fabric. The X-ray images are analyzed by a machine learning algorithm (Al) to detect the metal and plastic objects. The machine learning algorithm to be used in the present invention may be developed use in industrial laundries to detect foreign objects that could be in the pockets of the textile subjects. Such algorithm shall also be trained to recognize for example buttons and zippers or similar being part of the textile subjects.

[0289] Mass - the X-ray images are used to estimate the mass of the textile subjects, which is done with an average accuracy of +-5%. This information is mainly used for estimat- ing how full a cage is but can also be used in the sorting rules.

[0290] Fabric types - based on a sensor implemented to measure the structure of the fabric. The extraction algorithms distinguish these patterns to see whether the textile is a weave, knit or others. This information can be used to optimize the mechanical recy- cling process, which works best if the fabric types are separated.

[0291] Acrylic print detections - some textile subjects have thick rubber-like prints on them which contaminates both the mechanical and chemical recycling. The X-ray data can potentially be used for this, by feature extraction algorithms.

[0292] Material categories - presently approximately 30 different material composition can be detected today but more niche mixtures of fiber material may be detected.

[0293] Surface coatings and dangerous chemicals - being able to detect for example water repellant treatments with PFAS or dyes with heavy metals in them can increase the quality and useability of the sorted fractions. This is done using the NIR sensors or other similar sensors, e.g. MIR, UV or hyper spectral sensors.

[0294] Dirty and moldy textile subjects - wet textile subjects are detected using the NIR sen- sor. To see that a textile subject is dirty or moldy when it is dry is done using either the camera, the NIR sensor or a combination of the two or other similar sensors, e.g. MIR, UV, or hyper spectral sensors.

[0295] The decision logic works by setting up sorting recipes in the database system. Different textile types are defined based on the abovementioned features.

[0296] It is possible to set rules for: “more than”, “less than”, “highest” and “lowest” to include or exclude specific features for each sorting. The specific setup of these sorting parameters will optimize the sorting. Sorting may comprise up to 31 fractions of textile subjects, each of which textile subjects has the same properties, and which will be collected in one of the fraction containers.

[0297] The fractions may be sold to costumers as reusable or recyclable textile objects or may be further treated in recycling stations which comprise mechanical recycling units, chemical recycling units and downcycling units before being recycled as reuseable tex- tile products and / or textile fibers.

[0298] The sorting recipes can be adjusted to comply with the demand of chemical recyclers on the market. The system according to the present invention makes it possible to optimize the sorting recipes to make sorted fractions which fulfill the input specifications of these recyclers.

[0299] Furthermore, color categories can be matched with material compositions to match what fibers have the best value on the market for mechanical recycling.

[0300] The traceability starts when a batch of waste textile material is received and weighed to create delivery data for a delivery. Using the unique delivery ID, data is measured throughout the process.

[0301] -In the pre-sorting all the removed material is weighed and registered by type.

[0302] -In the main-sorting every piece of sorted textile subject is tracked to provide all the data for how a delivery has been distributed among the different fractions placed in the fraction containers at the final sorting line.

[0303] -When collecting sorted textile subjects in the fraction containers, the data from each fraction container is combined to provide data on the origins of the textile subjects in each fraction container.

[0304] -When shipping the sorted textile subjects to sub-suppliers, the specific fraction con- tainer numbers must be tracked to keep the traceability through the supply chain.

[0305] The MES is crucial for the traceability. In one embodiment of the invention, the handling system further comprises that the conveyors are chosen among conveyor cages in form of so-called roller cages con- veyor belts and conveyor bags in form of laundry bags and wherein a combination of such conveyors are used in the handling system.

[0306] Material may be moved around in a building in roller cages, belt conveyors and con- veyor bags. This may be effected with a combination of manual handling of the con- veyors or in a transportation system with no manual handling from receiving the waste to finished fractions ready for recycling as reuseable textile products and / or textile fi- bers.

[0307] The fractions may be gathered in bales, which will require automatic baling equipment, which is on the market. Such highly automized systems make sense in countries with high salaries.

[0308] In one embodiment of the invention, the handling system further comprises that the main-sorting station comprises

[0309] - at least one quality registration sensor e.g. a humidity sensor, for registration the quality of the textile objects on the main bulk receiving conveyor which at least one quality sensor is connected with the database system and wherein the bulk receiving conveyor is reversible will be reversed in case the quality registration does not fulfil predetermined quality limits, e.g. humidity level whereby the bulk of textile subjects are discharged.

[0310] As mentioned above wet material will be unsorted when handling the waste in the bulk receiving conveyor to secure the quality of the textile subject handled in the system.

[0311] In one embodiment of the invention, the handling system further comprises that the main-sorting station in front of the main-sorting unit comprises a waste-sorting unit where waste is sorted out before the establishment of the batches.

[0312] Also, this will contribute to secure the quality of the textile subject handled in the sys- tem.

[0313] In one embodiment of the invention, the handling system further comprises that the fraction containers are used as conveyors for transferring the fraction of textile subjects to the recycling stations. Hereby the fractions of the sorted textile subject can effectively be transported either in situ or to another location for the recycling stations. Alternatively, belt conveyors may transfer the fractions of textile subjects to the recycling stations.

[0314] In one embodiment of the invention, the handling system further comprises that the lay-up unit comprise two consecutive grippers arranged so that a first gripper grips the textile subjects individually from the buffer space and a second gripper grips the textile subjects from the first gripper and transfers the individual textile subject to the lay-up conveyor.

[0315] This further secures that only on textile subject is transferred by the grippers.

[0316] In one embodiment of the invention, the handling system further comprises that property sensors of the final sorting line can comprise

[0317] - a vision or optical recognition system,

[0318] - a X-ray recognition system, and

[0319] - an infrared light recognition system for data measurement of each textile subject and wherein the final sorting line further comprises

[0320] -feature extractors for feature classification based on input form the property sensors,

[0321] -classification calculators for classification based on outputs from the feature extractors and a predetermined set of rules and

[0322] -prioritizing units to determine which fraction of textile subject the individual tex- tile subject belongs to.

[0323] The property sensors may be of the type disclosed in the applicants pending interna- tional patent application PCT / DK2024 / 050118 and be operated as disclosed in said patent application.

[0324] The classification data structure, utilized in the final sorting line, can be considered as having the following four layers:

[0325] Level 0: Data measurement is done by the physical sensors. Each sensor may provide a single type of output, which can be used as an input in one or more feature extractors. The sensor outputs can be partially or fully processed results rather than raw data. Examples of this include the NIR sensor, which outputs the recognized material composition of the textile rather than the raw reflectance spectrum, or a color correction applied to the camera image. The data format of the sensor outputs depends on the sensor type: o Camera: RGB image of dimensions: [x,y,3], where x is the resolution across the conveyor width, and y is the length of the image along the conveyor length depending on textile size. o X-ray sensor: Greyscale image of dimensions: [x,y, 1] o NIR sensor: A list containing the number of pixels which are classified into each material category.

[0326] Level 1 : feature extraction is done by different algorithms, also referred to as feature extractors. Their input is sensor data, and their output is various features. While differ- ent inputs can go into a single feature extractor, different types of outputs can also be provided by it. The data format of feature extractors depends on the type of extractor and can include both continuous and categorical data. o Color detection: The feature extractor divides the textile area into a num- ber of cells and attributes one of the pre-defined base colors to each cell. Outputs the percentage of cells which fall into each base color as a list. o Size detection: The feature extractor finds the area of the textile by counting pixels and outputs it as an integer. o CNN: Object classes: A convolutional neural network (Al) has been trained to classify how well the textile subject falls into each pre-defined object class. The output is a percentile value for each object class, add- ing up to 100%, provided as a list. o CNN: accessories: Another convolutional neural network (Al) has been trained to recognize different types of accessories. E.g. plastic buttons, metal buttons and zippers. The output is a list of each accessories type and the respective confidence value (how certain the algorithm is that an accessory is present) o Mass estimate: Using information about the material and the X-ray im- age, an algorithm has been trained to estimate the mass of the textile subject. The output is a mass in grams. o Material distribution: Based on the NIR-sensor output, the percentage of each material category present in the textile is calculated and pro- vided as a list. Level 2: Classification is performed by testing the feature extractor outputs against a set of rules. Each class be attributed to a sorting position, but multiple classes can also be assigned to the same sorting position. Resultingly, the outputs of the classification stage are Boolean values (true / false). The rule types, which are defined for each class, include: o X=highest or x=lowest: For the color and material distributions, a rule can be set to have either the most or least common color / material be a specific color / material category. Multiple rules of this type can be set for a given feature for a given category. o <x or >x: For any of the features, a rule can be set for the given feature value to be above or below some defined limit. For the color distribution, material distribution, object class and accessory class the rule can be set for any of base colors / material categories / object types / accessory class. Multiple rules of this type can be set for a given feature for a given category. o x+y-highest or x+y=lowest: For material distribution, categories can be combined to set highest / lowest rules for a combination of different ma- terial categories. Multiple rules of this type can be set for a given feature for a given category. o <x+y or >x+y: For material distribution, categories can be combined to set above / below rules for a combination of different material categories. Multiple rules of this type can be set for a given feature for a given cat- egory.

[0327] Level 3: Sorting recipe is used to determine which classes go into which fraction con- tainer, and how the classes are prioritized against each other. Prioritization is neces- sary because a single textile might fit into several different classes, in which case the highest prioritized class is chosen. Prioritization helps simplify the classification rules, and make sure that the value of the sorting is maximized, while the purity of the valua- ble fractions is kept. Resultingly, classes of unwanted contaminants will be prioritized highest (such as wet textile subjects, workwear or small items) to ensure that they do not fall into valuable classes. Moreover, the remaining classes are prioritized depend- ing on their value, making sure that the highest value objects are isolated. Every sorting position is attributed an article ID, which is communicated to the MES for every textile sorted, together with classification information and which delivery the textile came from. While several classes might be mapped to the same sorting position, a single class cannot be mapped to multiple sorting positions. Each sorting position is associated with a fraction container for containing the final sorted textile subjects belonging to the class according to the sorting recipe.

[0328] The classification data structure, presented above, ensures a modular and thus flexible system in several aspects:

[0329] • Rules can be set up to either include or exclude combinations of specific fea- tures from different categories, minimizing the amount of rules necessary to define a category. The rules relate to the feature outputs which are defined in relatable terms, rather than referring to raw outputs, which can be hard to inter- pret.

[0330] » New categories can easily be added to accommodate new needs from the mar- ket, without having to alter the feature extractors or remaining categories. An example of this is a new chemical recycling customer, which has specific re- quirements for their input material. Then, the rules needed to accommodate the requirements are set up in a new category, which can then be mapped to a container to initiate sorting.

[0331] ® Categories are saved and must not necessarily be attributed to a container. This makes it possible to quickly alter active classes, and thus also switch be- tween which fractions that are being produced in the sorting lines. An example: a customer only needs a smaller amount of sorted textile subjects a couple of times every year. The needed class can be defined and attributed to a container to start production for the customer. Once enough material has been accumu- lated to fulfill the order, the class is unassigned from the container. Later in the year, when the customer needs more sorted textile subjects, the class can be attributed to a container again to start producing more sorted textile subjects to the customers specification.

[0332] ® New feature extractors can be added, or the classes / categories within each feature extractor can be altered, without having to consider the classification rules or sorting recipe.

[0333] ® New sensors can rather easily be added or replace existing ones.

[0334] • Tolerances can be iteratively fine-tuned to meet customer needs, as all feature outputs are quantitative. An example: a new color combination is created to meet a customer’s needs of a red+yellow cotton fraction. The class is defined and assigned to a container to produce one cage of sorted textile subjects for the customer. The customer gives feedback that there is too much yellow in the cage, after which the tolerance towards yellow can be altered specifically to create a new fraction for verification. Every time a batch of main-sorted textile subjects is transferred to the final sorting line, the delivery batch which the batch of main-sorted textile subjects belongs to is com- municated to the final sorting line. Combined with data exports of: classification results, mass and delivery origin of each sorted textile subject, it is possible to trace:

[0335] • The mass of textile subjects within a particular container. This information is used for traceability and process control. Without sensors, we know when a container is full by summing up the weight of all textile subjects within the con- tainer. Once the cage is exchanged, this is registered on the sorting line and communicated to the MES.

[0336] • The origins of the sorted material in a container, down to the distribution. An example for a given cage could be: “10% from a delivery batch 412, 40% from a delivery batch 413 and 50% from a delivery batch 414”

[0337] • The distribution of material within a container (according to the classes). Exam- ple: assume that we choose to map two classes into the same container: blue cotton and red polyester. The percentage of blue / red textile subjects in the con- tainerwill depend on the input material of the sorting line and if the definition of the two classes overlap with any other active classes. Through the data ex- ported to MES, we know the exact percentage of blue cotton / red polyester tex- tile subjects within the cage, because we have exported the data for every sorted textile.

[0338] ® The sorting results of a delivery. Once a delivery has been completely sorted by the final sorting line, we know for each textile subject how it was classified and sorted. Furthermore, we know the mass of each textile subject, making it possible to stage how many kg’s went into each container.

[0339] In one embodiment of the invention, the method further comprises that the registration of different properties of each textile subject, comprises identification and registration of one or more of the following features: colour distribution, such as colour gradients, fabric type, such as weave, knit, terry cloth and / or denim, material composition, density, mass, condition, such as cleanness, moisture content and / or mould content, presence of reflective material, presence of pattern, such as stripes and / or squares, presence of polymer prints, presence of hard accessories such as buttons, zippers, pens and / or tools, presence of contaminant chemicals such as oil, presence of multi-layer textile, such as jackets, pillows and duvets, type of textile subject, such as trousers, jeans, a shirt, a bag or a shoe, or the subject is not a textile subject, and wherein the step of sorting the textile subjects comprises a classification of the subject into predetermined classification groups, and / or predetermined sortation groups.

[0340] Typically, the identification of all features and the registrations are used in the database system for controlling the sorting unit whereby a high security in the sorting process is obtained.

[0341] There are several possible processing methods for recycling the textile of the textile subjects to reuseable textile products and / or textile fibers in recycling stations after the sorting in the final sorting line.

[0342] Description of specific features and applications of the system and the method

[0343] Shipping orders

[0344] Shipping orders are used to register the shipment of finished bales containing a specific classification group or sorted textile subjects, shredded textiles or recycled fibers.

[0345] The shipping order function allows the user to create a shipping order (with shipping order ID) to be stored in the database system and where the following features can be pre-defined: o Article ID of a finished bale o Amount of sorted textile [kg] o Number of bales o Specific bale ID's o Customer of the shipment o Destination location for the shipment

[0346] The above specification creates a set of rules in the database system which rules must be fulfilled for the shipping order. When fulfilling the shipping order, a production worker opens up the shipping order and begins to scan identification tags, e.g. a QR-tags, on the bales that are to be shipped. If an Article ID is defined, only bales with this Article ID can be chosen. The same goes if specific bale ID's are chosen for the shipping order. o Bale ID refers to a single specific bale. o Article ID refers to the composition of the material within a bale, many bales can exist with the same Article ID, but only a single Article ID can be within a bale.

[0347] Once the Number of bales or Amount of sorted textile [kg] are fulfilled, the shipping order can be completed to change Destination location for the shipment for each of the scanned bales in the shipping order, such that the digital storage is ajour.

[0348] Further, shipping documents can be automatically generated based on the scanned bales. Containing general information such as destination of the shipment, waste code of the material, total weight and number of bales. Furthermore, an inventory list for the truck is generated with information about each bale in the shipment: o Bale ID o Article ID o Bale weight

[0349] Automatic sorting report generation

[0350] An automatic sorting report generation is used to generate sorting reports for internal information and to distribute to suppliers, documenting the received amounts and how they were sorted.

[0351] When a received delivery completes the sorting process and all the data has been collected during the pre-, main and final sorting steps, a sorting report can automatically be generated by the database system for the delivery as a PDF file. The report contains all the gathered information about the delivery, and may include all of the following information or a part thereof: o Supplier (who delivered the material) o Amount received o Type of material received: Delivery Category (Household post-consumer waste HH-PO, post-consumer industrial laundry waste IL-PO, pre-con- sumer fashion brand FB-PR, etc...) o Type of receival (manual vs. automatic) o Date + time of initiated / completed: receival, pre-sorting, main-sorting, final sorting, sorting complete. o Location of receival and sorting

[0352] Furthermore, all the sorting data is analyzed in the database system and presented in tables with percentage distributions and visualized in plots such as pie charts. The sorting data may include all of the following information or a part thereof: o All waste that was manually removed from the delivery in the pre-sorting and main-sorting, registered as kg by fractions, e.g.: 10 kg wet textiles, 50 kg reuseable, 100 kg multi-layer textiles... o All the sorting data from the automated final sorting, with the data being registered for each sorting textile:

[0353] * Origin of textile: Delivery! D

[0354] ® Weight of textile: [kg]

[0355] * Categorization result: [Category]

[0356] * Sorting output: [pos] + [cagelD]

[0357] Sorting reports can be generated for a single delivery or for multiple deliveries, based on one or more of the following parameters: o One or more specified customers o A given time period o One or more delivery categories

[0358] Automatic traceability report generation

[0359] Automatic traceability report generation is used for each finished bale of sorted textiles, shredded textiles or recycled fibers.

[0360] The traceability report states the exact contents of one or more given bale(s): o 10,1 kg (43 picks) from Delivery ID 1041 (delivery type HH-PO) o 15,4 kg (72 picks) from Delivery ID 2951 (delivery type IL-PO) o etc

[0361] Furthermore, the traceability report contains general information about each bale of sorted textiles, shredded textiles or fiber-to-fiber mechanically recycled textiles: o Time and location of baling o Time and location of shredding (if shredded) o Time and location of fiber-to-fiber mechanical recycling (if recycled) o Time, location and customer of shipping (if shipped) o Weight of the bale [kg] o Fiber quality (if tested): mean fiber length, percentage of short fibers, etc... o Material composition (if tested): 92% cotton, 5% polyester, 2% elastane, 1 % unknown... o Resource consumption o Current storage location of the bale.

[0362] Automatic resource consumption measurement

[0363] Automatic resource consumption measurement is effected in order to evaluate costs associated with the operation of the system.

[0364] Throughout the system, meters may be equipped at: electricity supply, natural gas sup- ply and water supply.

[0365] The meters communicate with the Manufacturing Execution System through the net- work to constantly register the consumption of the three resource types. In this way, the consumption can be monitored in real time, and history data can be easily accessed and visualized.

[0366] Electricity can be measured per. machine / process step such as: receival / pre-sorting, main-sorting, final sorting, shredding, mechanical fiber-to-fiber recycling and baling. The average electricity consumption of each process over time is automatically calcu- lated and compared with average process throughputs in the same period to calculate process efficiencies. These efficiencies can be distributed pr. kg and accumulated for each process step to calculate the power consumption of each sorted bale. Average values for gas and water consumption can be added to have a complete picture of the resource consumption related to each produced bale of sorted textiles.

[0367] Integration with ERP systems

[0368] Integration with ERP (Enterprise Resource Planning) systems is used to automatically adjust the storage and generate sales orders. The integration may also be with other external data systems such as DPP (digital product passport) platforms.

[0369] The MES is integrated with the ERP system through REST API’s, meaning that data is received from the ERP system and delivered to the ERP system.

[0370] Received data includes: o List of customers and related information such as:

[0371] * Customer type

[0372] ® Customer ID

[0373] * Customer name o List of Article ID’s and related information such as:

[0374] * Material type

[0375] * Article ID

[0376] ® Article name

[0377] Delivered data includes:

[0378] When sorting is completed for a delivery, the following information is trans- ferred:

[0379] * Delivery ID

[0380] ® Customer ID

[0381] ® Reci eval date+time

[0382] ® Sorting complete date+time

[0383] ® Recieved kg

[0384] * Registered kg’ s each type in: » Pre-sorting

[0385] « Main-sorting

[0386] « Final sorting o When a bale is made, the following information is transferred:

[0387] * Article ID

[0388] ® Weight [kg]

[0389] * Bale ID

[0390] ® Time of creation

[0391] * Storage location

[0392] If a bale is transported between storage locations, the following information is trans- ferred:

[0393] * Bale ID

[0394] * New storage location

[0395] When a shipping order is completed, the list of scanned bales and the customer infor- mation is transferred to the ERP system to automatically adjust the storage and generate sales order based on article prices. See “Shipping orders” for lists of transferred infor- mation.

[0396] Live KPI visualizations

[0397] The MES comprises a dedicated page for a Dashboard showing live KPI (Key Perfor- mance Indicators) data and visualizations. The data is extracted directly from the data- base system and shown in tables and pre-defined graphs such as bar charts, pie charts, meters and other visualizations.

[0398] Description of the Drawing

[0399] Various examples are described hereinafter with reference to the figures. Like reference numerals refer to like elements throughout. Like elements will, thus, not be described in detail with respect to the description of each figure. It should also be noted that the figures are only intended to facilitate the description of the examples. They are not intended as an exhaustive description of the claimed invention or as a limitation on the scope of the claimed invention. In addition, an illustrated example need not have all the aspects or advantages shown. An aspect or an advantage described in conjunction with a particular example is not necessarily limited to that example and can be practiced in any other examples even if not so illustrated, or if not so explicitly described.

[0400] Exemplary embodiments of the invention are described in the figures, wherein:

[0401] Fig. 1 illustrates a schematic fragmentary overview of a plant according to the pre- sent invention,

[0402] Fig. 2 illustrates a layout for a system according to the present invention,

[0403] Fig. 3 illustrates an enlarged view of part of the layout illustrated in Fig. 2,

[0404] Fig. 4 illustrates a schematic overview of a lay-up unit which transfers the textile subjects individually to the final sorting line,

[0405] Fig. 5 illustrates a schematic overview of a first part of a final sorting line where sev- eral consecutively arranged property sensors are arranged in connection with a conveyor transporting the textile subjects individually past the sensors,

[0406] Fig. 6 illustrates an example of a spectral graph obtained with a NIR sensor,

[0407] Fig. 7 illustrates an example of a series of greyscale images obtained with a hyper- spectral camera,

[0408] Fig. 8 illustrates a schematic overview of conveyors in a part of the final sorting line where several short sections of reversible belt conveyors are used to sort the textile subjects into fraction containers in form of roller cages,

[0409] Fig. 9 illustrates an example of the data structure of the classification in the sorting process,

[0410] Fig. 10 illustrates a handling chain of sorted textile subjects for recycling in recycling stations,

[0411] Fig. 11 illustrates a schematic overview of a fiber opener used in a recycling station, Fig. 12 illustrates a schematic overview of unit for a secondary sorting of clippings used in a recycling station,

[0412] Fig. 13 illustrates a schematic overview of a first type of a separator unit for a second- ary sorting of shredded textiles used in a recycling station, and

[0413] Fig. 14 illustrates a schematic overview of a second type of a separator unit for a secondary sorting of shredded textiles used in a recycling station. Detailed Description of the Invention

[0414] Exemplary examples will now be described more fully hereinafter with reference to the accompanying drawings. In this regard, the present examples may have different forms and should not be construed as being limited to the descriptions set forth herein. Ac- cordingly, the examples are merely described below, by referring to the figures, to ex- plain aspects.

[0415] Fig. 1 illustrates a plant in which a system 1 according to the invention is contained.

[0416] The system comprises a pre-sorting station 2. The pre-sorting station 2 comprises a bulk receiving station 3 in which a receiving unit 4 receives waste containers 5 in which textile subjects are delivered to the system.

[0417] Moreover, the pre-sorting station 2 comprises a bulk receiving conveyor 6. The bulk receiving conveyor 6 comprises weight sensors which register the weight of the bulk material. The weight sensors (not illustrated) are load cells on which the bulk receiving conveyor 6 are arranged. The weight sensors are connected with a database system 7 (see Fig. 5).

[0418] Moreover, an unpacking and pre-sorting unit 8 is arranged at the end of the bulk re- ceiving conveyor 6. The unpacking and pre-sorting unit 8 is connected with a transfer conveyor 9 for transporting the presorted bulk of textile subjects to a main-sorting sta- tion 10.

[0419] The main-sorting station 10 comprises sensors 16’ for registration the weight of textile objects which have passed the pre-sorting and is received by an infeed unit 13 of a main bulk receiving conveyor 11. The infeed unit 13 establish batches of pre-sorted textile subjects. The weight sensors will normally be load cells on which the main bulk conveyor 11 or the infeed unit 13 is arranged.

[0420] In the main-sorting station 10 main-sorting units 12 are provided for effecting a main- sorting of the received textile objects. The main-sorting units 12 provide batches 20’ of main-sorted textile subjects. These batches 20’ are transferred to an infeed unit 14, e.g. by roller cages 20. The infeed unit transfers the textile subjects to a final sorting line 15 (see Fig. 2). The database system 7 is adapted to retrieve, store and process data measured by sensors which will be explained below. Moreover, the database system 7 is adapted to control the operation of the individual system units.

[0421] In fig. 1 a final sorting line 15 is not included. This is seen in fig. 2.

[0422] Fig. 2 illustrates a layout corresponding to fig. 1 , however, also illustrating the infeed units 14 in final sorting lines 15 comprising several final sorting line conveyors 15’.

[0423] Fig. 3 illustrates an enlarged view of the layout illustrated in fig. 2 and illustrates the layout of the main-sorting station 10 illustrated in fig. 1. When receiving the container 5 in the pre-sorting station 2, the bulk receiving conveyor 6 may also be monitored by photo sensors 16 (only one is illustrated in fig. 3). The photo sensors 16 will monitor whether the bulk receiving conveyor 3 is empty or full. When initiating an unloading of the container 5 the bulk receiving conveyor 3 will start to move and will move as long as the photo sensors 16 register that material are present on the bulk receiving con- veyor 3.

[0424] When the container 5 is emptied the weight on the bulk receiving conveyor 3 will be registered and communicated to the database system 7. In the database system the material received will be associated with an indication for the delivered amount of waste. The bulk receiving conveyor 3 may be divided into a first conveyor 3’ and a second conveyor 3”. The conveyor 3’ may be connected with a weighing sensor 140 and be denoted as a weighing conveyor and the conveyor 3” may be denoted as a transferring conveyor.

[0425] Accordingly, it is possible to have a large amount of waste on the bulk receiving con- veyor 3 and waste received from different customers.

[0426] At the end of the bulk receiving conveyor 3 there is provided a transverse transfer conveyor 9 on which transfer conveyor the textile subject matters are presorted in the unpacking and pre-sorting unit 8. After this pre-sorting the pre-sorted textile subjects are transferred by the transfer conveyor 9 to a main bulk receiving conveyor 11. The sensors 16’ are provided for registration of weight and possibly also other delivery data relating to the bulk of textile subjects entering the handling system on the bulk receiving conveyor 11. The bulk of textile subjects may be received directly on the bulk receiving conveyor 11 without being pre-sorted. In this situation the sensors 16’ register the weight of the received bulk of textile subjects on the bulk receiving conveyor 11 .

[0427] Registration units 116 for entering user input for delivery data relating to the bulk of textile subjects entering the handling system is provided for entering input data either in connection with bulk receiving conveyor 6 or in connection with the main bulk receiv- ing conveyor 11 or the infeed unit 13. The registration units 116 are connected with the database system 7 in which the delivery data are stored.

[0428] The sensor 16’ for registration at least the weight of the textile objects is connected with the database system 7. In the database system the weight registration and user input data is transferred to an identification tag 137 associated with a batch. The sys- tem comprises an application unit 138 for application of the identification tag 137 to each batch.

[0429] As illustrated in figs. 1-3 the main bulk receiving conveyor 11 is connected to several transversely arranged main-sorting conveyors 17. Adjustable guiding plates 18 or sim- ilar solutions or other solutions (e.g. overlapping conveyors) may direct the textile sub- jects of the main bulk receiving conveyor 13 to the transversely arranged main-sorting conveyors 17. In the embodiment illustrated the main-sorting unit 12 is arranged in form of the main-sorting conveyors 17. Fig. 3 illustrates that persons 152 may monitor the textile subjects on the main-sorting conveyors 17 and effect a manual sorting, e.g. if wet material is located. Alternatively, the main-sorting is effected automatically.

[0430] At the end 19 of each main-sorting conveyor 17 there is provided a roller cage 20 for containing the batches 20’ of main-sorted textile subjects which are transferred to the infeed unit 14 for the final sorting line 15. Alternatively, to the roller cages 20 a direct infeed may be provided by main-sorting conveyors 17.

[0431] All above mentioned conveyors will typically be belt conveyors.

[0432] A photo sensor 21 at the end 19 of the main-sorting conveyor 17 ensures that a roller cage 20 is present, otherwise a signal from the photo sensor 21 will be sent to the database system 7 which stops the main-sorting conveyor 17. When the main-sorting conveyor 17 is stopped also the main bulk receiving conveyor 11 is stopped and likewise the transfer conveyor 9 for transporting pre-sorted textile subjects will also be stopped.

[0433] A roller cage 22 for fractions final sorted textile subjects is illustrated in fig. 8. Such roller cage 22 could also be used as roller cage 20 for transferring the main-sorted textile subjects.

[0434] Fig. 4 illustrates an enlarged view of an embodiment for an infeed unit 14. Textile units 23 are provided in a buffer space 24 in the infeed unit 14. At the bottom of the buffer space 24 a first conveyor 25 transports the textile subjects 23 to a pick-up area 26.

[0435] The buffer space 24 comprises a registration sensor 28 for registration of the identifi- cation tag 137 (see Fig. 3) associated with the batch of the main-sorted textile subjects in the database system.

[0436] A lay-up unit 29 is provided for transferring the textile subjects 23 to the final sorting line 15.

[0437] The lay-up unit 29 comprises a first gripper 30 and a second gripper 31. Each of the grippers 30, 31 are operated by robots 27 integrated in each of the grippers 30,31.

[0438] The first gripper 30 will pick-up a textile subject 23 from the pick-up area 26. The textile subject 23 as seen in fig. 4 will be gripped by the second gripper 31 coming in from right in fig. 4 and transfers the textile subject 23 to an infeed conveyor 32 for the final sorting line 15.

[0439] The infeed unit 14 comprises a registration sensor 33 which is used to locate that the first gripper 30 only grips one textile subject at a time. The registration sensor 33 may be used together with a camera or the registration sensor 33 may be provided in form of a camera.

[0440] Each of the grippers are provided with gripper fingers 134 which pick up the textile subject 23. When the textile subject is gripped the registration sensor 33 will confirm that one textile subject 23 is gripped.

[0441] Then the textile subject 23 is lifted to the top of the infeed unit 14 to be gripped by the second gripper 31 . When the second gripper 31 has gripped the textile subject 23, the first gripper 30 will release the grip and the second gripper 31 will transfer the textile subject 23 according to the arrow 135 and place it on the infeed conveyor 32.

[0442] By gripping the textile subject two times, the risk of double-picks is reduced.

[0443] The infeed conveyor 32 transfers the textile subjects individually with a mutual distance between consecutive textile subjects 23 to a first section of the final sorting line 15 where it is arranged on a sorting conveyor 33 as illustrated in fig. 5.

[0444] In fig. 5 an example of a first section 51 of the final sorting line 15 is disclosed. The final sorting line 15 will normally comprise several final sorting line conveyors 15’ de- pending on the capacity of the system. However, a final sorting line 15 may also com- prise one final sorting line conveyor 15’ only.

[0445] The textile subjects 23 are illustrated arranged on the final sorting conveyor 33 for transporting the textile subjects 23 through the sorting steps effected in the final sorting line 15.

[0446] The first section 51 of the final sorting line 15 comprises property sensors for determin- ing the properties of each individual textile subject 23. This result is used in a second section 52 of the final sorting line 15 for establishing fractions which are sorted-out into fraction containers in which the textile subjects each have the same properties.

[0447] In the first section of the final sorting line 15 the textile subjects 23 are transported past several property sensors. In the embodiment a NIR sensor 34, an RGB sensor 35 and an X-ray sensor 36 are illustrated as examples on possible property sensors. The final sorting conveyor 33 may be separated into different conveyors as different properties for the conveyors may be needed. This is especially the case for the X-ray sensor 36 which is used for measuring penetrating X-rays as illustrated in fig. 5.

[0448] The NIR sensor 34 will measure reflected light from light emitters 37 and the RGB sensor 35 will also measure reflected light from light sources 38.

[0449] Signals from the sensors 34,35,36 will be transferred to the database system 7 as illustrated by dotted lines 39, 40 and 41. The database system 7 is illustrated with a computer 42 and a monitor 43 to give the operator a visual indication of the different functions and measurements of the system. Fig. 6 illustrates the result from the NIR sensor.

[0450] The NIR sensor 34 will use spectrography to identify the material. Accordingly, an al- gorithm in the database system will look at the light intensity as a function of wave- lengths. This will give an indication of the chemical composition at several points along the conveyor width and length to provide a 2D image of pixels. To each of the pixels a material is attributed. Such spectrograph is illustrated in fig. 6.

[0451] Alternatively, a hyperspectrai camera could also be used for the NIR. This will have much higher spatial resolution (more pixels) but a lower spectral resolution as they look at fewer wavelengths.

[0452] As a result, the output of a hyperspectrai camera will be a serious of greyscale images at different wavelengths. This is illustrated in fig. 7. This method can be used for de- tecting smaller areas / objects which might be interesting. However, it will have less ac- curacy in terms of chemical detection. Therefore, it could be used in combination with NIR sensor using spectrography as mentioned above.

[0453] However, the hyperspectrai camera will be of use e.g. for detecting mold which has a large intensity at specific wavelengths. It should be mentioned that such type of sensor often is used in quality control of food processing lines.

[0454] Fig. 8 illustrates the principle for the second section 52 of the sorting line 15.

[0455] The second section 52 comprises a number of sorting conveyors 44 for transferring the textile subjects past a number of final sorting units 22’. Each sorting unit 22’ comprises a sensor 139 which is connected with the database system. The sorting conveyors 44 are connected with the database 7 and are arranged for an automatically transferring the textile subjects from the individual sorting conveyors 44 to fraction containers 22 based on signal to the database from the sensors 139. Each fraction container 22 is associated with one of the final sorting units 22’ in which automatically transferring is based on the registered properties of the individual textile subjects 23. Hereby a frac- tion of all textile subjects which has same properties will be collected in one of the fraction containers 22, Each of the successive sorting conveyors 44 of the final sorting line 15 comprises short sections of inclined, reversible conveyors as illustrated in the top of fig. 8.

[0456] When a textile subject reaches a sorting position determined by the sensors 139, the associated sorting conveyor 44 will reverse to sort-out the textile subject 23 into a frac- tion container in form of an underlying roller cage 22. In the roller cage 22 a fraction of the sorted textile subject will be gathered based on the property determination effected in the first section 51 and based on determination in the database system 7.

[0457] Accordingly, all textile subjects in the sorted fraction in one roller cage 22 will have the same properties. The roller cage will be used for transferring the textile subjects to recycling stations in which each of the different fractions will be recycled. The recycling may be effected as mechanical recycling, chemical recycling or in down recycling units.

[0458] Hereby the textile subjects for each of the fractions will be recycled as reuseable textile products and / or textile fibers.

[0459] Fig. 9 illustrates a classification data structure having four layers.

[0460] A first layer will be data measurement. A second layer will be feature extraction. A third layer will be classification of the textile objects and a fourth layer will be sorting recipe.

[0461] In fig. 9 it is illustrated that the first layer may comprise camera, X-ray sensors and NIR- sensors which have already been explained above.

[0462] As indicated in fig. 9 it is possible to have color image, X-ray image or material classi- fication results from the data measurement.

[0463] In the database system the next level will be extraction of features like color detection, size detection, CNN object classification, CNN accessories, mas estimates and mate- rial distribution.

[0464] These results will be used as indicators e.g. color, size, mass or material distribution, for the next level where there is effected a classification. The classification can be as illustrated in the example small objects, white cotton trou- sers, white cotton w / o accessories, off-white cotton, cotton / elastane or other classifi- cations.

[0465] Based on the classification which is registered in the database system a sorting is ef- fected based on predetermined sorting recipes whereby the fraction is placed in the fraction containers.

[0466] In Fig. 10, a flowchart is seen outlining the different possible processing methods for sorted recyclable textile subjects.

[0467] It shall be kept in mind, that the 4 different approaches to disintegration of the textile subjects are applied after the textile waste has been through pre-sorting, main-sorting and final sorting.

[0468] In this way, the following types of impurities have been removed:

[0469] ® Reuseable textile subjects

[0470] • Multi-layer textile subjects

[0471] • Not textile waste

[0472] ® Polymer prints

[0473] ® Wet, dirty and moldy textile subjects

[0474] Furthermore, the textile subjects have been sorted in homogeneous fractions accord- ing to:

[0475] • Material composition

[0476] • Color composition

[0477] ® Object type (no workwear or underwear, denim is in its own fraction)

[0478] » Accessories (fractions w / o acc. have no metal and only few hard plastic parts)

[0479] • Mass and size (no very small or very large objects)

[0480] And potentially, they can also be sorted by:

[0481] ® Fabric structure (knit, weave, terry (towels) and denim in separate fractions)

[0482] ® Chemicals (keep fractions free from hazardous chemicals or isolate specific dye chemicals)

[0483] Mechanical fiber opening for respinning Requiring thorough sorting, careful processing and rather expensive equipment, me- chanical fiber opening is mostly applied to materials with a high fiber value such as cotton, cotton-rich polyester blends, and wool.

[0484] When opening the fibers for re-spinning, it must be done with great caution to ensure the fibers are sufficiently open while still maintaining a decent fiber length. If the ma- chines are run too harshly, too many fibers will be broken. If the machines are run too gently, the capacity will be too low, and the textile subjects will not be sufficiently opened. For this reason, sorting according to fabric structure is important for mechan- ical fiber opening for respinning. Rough fabrics such as denim will behave differently than finer fabrics such as thin jersey t-shirts. Therefore, different parameters are needed for opening the different fabric types efficiently.

[0485] Before fiber opening for respinning, the textile subjects cannot be shredded in an in- dustrial shredder but must be cut instead in guillotine cutters. Furthermore, the textile subjects must not be cut in too small pieces (approx. 100x100mm), as this would de- stroy too many fibers. The cut textile subjects which are called clippings are then fed through a fiber-opening line to open the fibers:

[0486] A. Mechanical fiber opening lines use large blow rooms which are filled with homoge- neous material (often 1-5 t) of clippings. The material is blended and a water or oil emulsion spray can be applied to ease the subsequent fiber-opening.

[0487] B. The water or oil emulsion will cure in the material for 0-24 hours.

[0488] C. The mixed clippings will be fed to a number of fiber openers (2-10), having course openers in the beginning and ending with very fine openers. A fiber opener is based on the principles of a carding machine and a schematic is seen in Figure 11 .

[0489] Having mentioned the different possible processing methods for sorted recyclable tex- tile subjects an explanation of recycling stations will be illustrated.

[0490] Fig. 11 illustrates a carding machine which comprises a main drum with either saw- toothed wire wrapped around it, or it is completely covered in spikes / needles 45. When rotating in combination with other smaller rollers 46 the textile is slowly being feared apart. Using air blowing techniques, hard parts such as wood and plastic will often be filtered out, and unopened textile subjects are recirculated 47. Short fibers and dust are also removed with the air. Around 15-30% of the material will get lost as short fibers and dust.

[0491] D. At the end of the line, open fibers are produced with good fiber lengths. This means that the fibers have sufficient length for respinning. The open fibers can be used in a spinning mill, where they are fed into the spinning lines together with virgin fibers to create a mixture of virgin and recycled material. Instead of virgin fibers, high quality recycled fibers can also be used, such as chemically recycled fibers, biochemically re- cycled fibers, thermo-chemically recycled fibers, thermo-mechanical recycled fibers, or rPET (polyester fibers made from plastic bottles). Sometimes, the fiber opening lines will be adjusted to not fully open the fibers, as the remaining fiber-opening will then occur when carding the recycled fibers with virgin fibers. This can help optimize fiber quality but sets requirements at the spinning step of the supply chain in terms of equip- ment and process knowledge.

[0492] The downside of mechanical fiber opening for respinning is, that the equipment was initially designed for carding clean cotton or opening cut-off textile waste (homogenous fractions without any hard parts or stitches). When buttons and zippers or other hard parts get into the fiber openers, they will create a large degree of wear and tear and can damage the needles on the drums. Plastic will often get stuck between the needles, requiring operators to stop the machine, take out the large drums and manually remove all the plastic stuck between the needles. Meanwhile, metal gives much more wear and tear, and can furthermore introduce a risk of sparking that can cause fires. For this reason, the hard parts must be removed prior to feeding the fiber opening lines.

[0493] Furthermore, some soft impurities must also be removed to ensure a pure and high- quality fiber output. Keep in mind that the soft impurities will not directly interfere with the fiber opening process but can cause issues in spinning. Soft impurities include pol- ymer prints, care labels, rough stitches such as in denim and workwear, fabric logos, other soft trim items, and clippings with foreign fiber types. Some of these will get sep- arated in the fiber opener, but not all.

[0494] As the textile subjects cannot be disintegrated into small pieces before removing the impurities, air density sorting is not robust. Instead, a detection followed by sorting is needed to separate the clippings with hard impurities. A concept for automatic separation of clippings with impurities is seen in Fig. 12. The concept uses a vibrational feeder 48 to distribute the clipping unto the conveyor where they pass under sensor(s) 49. The following sensors can be used (and combined):

[0495] • Camera to visually identify impurities. Can detect prints, colors, hard parts, care labels, hard stitches, and care labels. Unless a special configuration with two sensors is used, only one side of the clipping can be inspected.

[0496] ® NIR / hyperspectral / XRF sensor to detect foreign fibers, chemicals, care labels, hard parts, and some soft trim items. Has the same challenge as the camera.

[0497] ® XRT sensor to detect any hard parts. Due to the transmissive nature of the sensor, it is expected to robustly be able to detect the hard parts (due to density difference).

[0498] Having detected the clippings with impurities they must be separated in a separator unit 50 from the clean clippings. Small air nozzles at the end of the conveyor spanning the conveyor width can be used to blow away the clippings with impurities to fall into the eject bin. Alternatively, a mechanical switch can be used to block off the clean bin when impurities are present. Both concepts are currently applied in different industries. The air blowing concept has been developed by different companies for textile sorting, by adapting existing solutions for plastics, glass and metal waste sorting. However, cost-effective and efficient solutions have not been brought to market yet.

[0499] Alternatively to the automatic sorting, manual removal of the impurities can also be done. The process is extremely labor intensive and is not suitable for large scale recy- cling.

[0500] Downcyclinq to non-woven products

[0501] Some fractions such as polyester rich fractions and polyester-rich cotton blends can be used for non-woven products such as felt products in construction, furniture, pack- aging, automotive industries and other industries.

[0502] Adaptions of the mechanical fiber opening lines have been developed to accept textile subjects with accessories and handle those. The machines work by having rougher “pre-openers” before the finer needle rolls. In combination with air sorting, the clippings are roughly opened up and hard parts are separated. However, this process is harsher on the material, so lower fiber lengths are obtained, which are mostly suited for non- woven products. Alternatively, the textile subjects can be shredded and sorted using industrial shredders and air density sorters such as a “zigzag” or cyclone air density.

[0503] Both concepts work by using air flow to separate denser materials from bulk materials. For shredded textile subjects, the buttons, zippers, and other hard pieces will be denser than the shredded fabrics. For the concept to work, the material must be shredded to a minimum size. If the pieces are too big, the behavior of a piece without any hard parts will be too similar to a piece with a small button on it. Resultingly, the air sorting would net work. Once the material is shredded and hard parts have been separated, it must be fed to a fiber opening machine. Such a fiber opening machine is much simpler and cheaper than the mechanical fiber openers for respinning and the fiber openers which can accept accessories. The open fibers that have been shredded before fiber opening will be shorter than the fibers which have been directly opened by a fiber opener which can accept hard parts.

[0504] Resultingly, there are two ways of obtaining open fibers which can be used for downcycled products in the form of non-wovens:

[0505] ® Medium length fibers which have not been shredded. These will be more ex- pensive to obtain but can be used for higher quality non-wovens such as thinner and finer felt products.

[0506] ® Short length fibers which have been shredded. These will be cheaper to obtain but can only be used for lower quality non-wovens. It is uncertain which prod- ucts can be made from these fibers.

[0507] Downcyclinq to insulation material

[0508] The method of shredding the fibers before opening them in simpler fiber opening ma- chines as described above, can also be used to obtain insulation material. This is cur- rently being done with wood, paper and cardboard waste to create a bulk insulation material which can be blown into buildings to insulate them. Fire retardant minerals are applied to the material, which can then replace traditional mineral-based insulation ma- terial which are energy intensive to produce.

[0509] Testing the open fiber pulp from textile subjects has shown promising results in insula- tion properties, dust content and the applicability of fire retardants.

[0510] Therefore, the stream can be useful for fractions which cannot be otherwise recycled and can provide a cheap value stream as hard parts are accepted in the material. The mechanical fiber opener that removes hard parts can potentially also provide pulp for this purpose, however it seems that costs of operating this equipment are too high compared to the value of the insulation pulp.

[0511] Chemical recycling

[0512] Chemical recycling of textile subjects is a field in rapid development. Within cotton, large scale facilities are in existence, but they struggle with strict input requirements and very low margins. The reason being, that the chemical recycling process is com- plex, requires highly specialized equipment and is energy intensive. The output of chemical cotton recycling is a viscose pulp, which must afterwards be further pro- cessed into a viscose fiber. This adds another link in the value chain before a recycled fiber is obtained. Resultingly, very low prices are offered for feedstock to the current chemical recyclers.

[0513] Meanwhile, polyester recycling is emerging, with large-scale projects underway in Eu- rope. According to the company plans, large scale chemical recycling of polyester will be a reality by 2025 or 2026. Furthermore, projects are underway for chemical recycling of cotton / polyester blends as well, but this technology is less mature.

[0514] Common for chemical recycling plants is that they need the input cleaned from hard parts, especially metal. Furthermore, they offer very low prices for their feedstock, so highly efficient sorting and disintegration is needed.

[0515] It is preferred to use the process of shredding the sorted textile subjects and removing hard parts using zigzag and cyclone separators.

[0516] A big challenge with the chemical recyclers is that the different technologies are intol- erant to different specific chemicals / materials. Examples include elastane, specific syn- thetic fibers, metals, plastics, specific dyes and surface treatments. This sets large re- quirements to the sorting (and hard part separation).

[0517] Fig. 13 illustrates a first example for a separator 53 used for textile objects where the textile material has been shredded. These materials may be sorted by using an air density sorter as illustrated in fig. 13. This is a zigzag separator.

[0518] Material infeed 65 enters through an inlet 55 in the top and air is blown through an inlet 54 in the bottom and creates an upwardly airstream 57. A heavy fraction 56 in dark color fall through the airstream 57 and leaves the separator 53 through a bottom outlet 58 in the bottom, A light fraction 59 in grey color goes upwardly with the airstream 57 and leaves the separator 53 through a top outlet 60 in the top of the separator. In fig. 14 an alternative example for an air density sorter is illustrated. This is a cyclone separator 61. The cyclone separator 61 has an inlet 64 for material infeed 65 of material and air, a bottom outlet 62 for a heavy fraction 63 and a top outlet 66 for a light fraction 67. As it occurs from figs. 13 and 14 the material infeed 65 may be divided into a light fraction and a heavy fraction.

[0519] After this separation in the separators, textile fibers may be recycled or reused e.g. as raw material in a non-woven material.

Claims

CLAIMS1. A handling system (1) for textile subjects, which system comprises a main-sorting station (10) for receiving, sorting, recovering and recycling textile subjects (23) for re- using the textile of collected textile subjects characterized in that the reusing is either in form of second-hand textile subjects e.g. garments or in form of reusable textile fi- bers, or other types of disintegrated textile materials or recycled materials or downcycled materials, wherein said handling system comprises:- a database system (7) adapted to at least retrieve, store, and process data measured by sensors being a part of the handling system (1), and adapted to control an operation of individual system units,- registration units (116) for entering user input for delivery data relating to the bulk of textile subjects entering the handling system, and which registration units (116) are connected with the database system (7) in which the delivery data are stored,- the main-sorting station (10) comprising- a main bulk receiving conveyor (11) having an infeed unit (13),- sensor (16’) for registration at least the weight of the textile objects, which sensors (16’) are connected with the database system (7) in which the weight registration is transferred to an identification tag (137),- a main-sorting unit (12) for main-sorting the received textile objects and es- tablishing batches (20’) of main-sorted textile subjects after the main-sorting of the received textile object,- an application unit (138) for application of the identification tag (137) to each batch,- a registration sensor (21) for registration of the identification tag, wherein the identification tag is associated with the batch in the database system,- conveyors (5) for transporting the textile subjects batchwise to a buffer from which the batches of main-sorted textile subjects are transferred to an infeed unit (14) for a final sorting line (15), wherein the infeed unit (14) comprises- a buffer space (24) for containing the textile units (23),- a registration sensor (28) for registration of the identification tag, wherein the identification tag is associated with the batch (20’) of main-sorted textile sub- jects in the database system (7),- a lay-up unit (29) which transfers the textile subjects (23) individually to the final sorting line (15), wherein the lay-up unit (29) comprises- at least one gripper (30,31) for gripping the textile subjects (23) individually from the buffer space (24)- a registration sensor (33) for registration that only one textile subject (23) is gripped by the at least one gripper (30,31) and transferred to an infeed con- veyor (32) for transferring the textile subjects individually to the final sorting line (15) with a mutual distance between consecutive textile subjects (23), wherein the final sorting line (15) comprises- conveyors (15’, 33, 44) for transporting the textile subjects individually past several consecutively arranged property sensors (34,35,36), preferably three or more, for registration of different properties of each textile subject (23) and which property sensors (34,35,36) are connected with the database system (7) for transferring the registered properties to the database system (7) in which the registered properties are associated to the registered batch (20’),- at least one sorting conveyor (44) for transferring the textile subjects past a number of final sorting units (22’) which are connected with the database sys- tem and are arranged for an automatically transferring the textile subjects from the at least one sorting conveyor (44) to fraction containers (22) where each fraction container (22) is associated with one of the final sorting units (22’) in which automatically transferring is based on the registered properties of the individual textile subjects (23) whereby a fraction of all textile subjects which has same properties will be collected in one of the fraction containers (22),- recycling stations which comprise mechanical recycling units (48,53,61), chemical recycling units and downcycling units (45,46) in which units the textile of the textile subjects is recycled to reuseable textile products and / or textile fibers.

2. The handling system according to claim 1 , wherein a pre-sorting station (2) is pro- vided before the main-sorting station (10), which pre-sorting stating comprises- a bulk receiving station (3) comprising a receiving unit (4) for receiving waste containers (5) containing a bulk of the textile subject to be handled and which receiving unit comprises- bulk receiving conveyors (6),- weighing sensors (140) for registration of the weight of the bulk material, which weighing sensors (140) are connected with the database system (7),- transfer conveyors (9) for transporting the batches of pre-sorted and weighed textile subjects to the main-sorting station (10).

3. The handling system according to claim 1 or 2, wherein the conveyors (3,5,11 , 15’, 33, 44, 22) are chosen among conveyor cages (22), conveyor belts (3,5,11 , 15’, 33, 44) and conveyor bags and wherein a combination of such conveyors are used in the handling system (2).

4. The handling system according to any of the preceding claims, wherein the main- sorting station (10) comprises- at least one quality registration sensor (12) e.g. a humidity sensor, for regis- tration the quality of the textile objects on the main bulk receiving conveyor which at least one quality sensor is connected with the database system (7) and wherein the bulk receiving conveyor (11) is reversible will be reversed in case the qual- ity registration does not fulfil predetermined quality limits, e.g. humidity level whereby the bulk of textile subjects are discharged.

5. The handling system according to any of the preceding claims, wherein the main- sorting station (10) in front of the main-sorting unit (12) comprises a waste-sorting unit (8’) where waste is sorted out before the establishment of the batches (20’).

6. The handling system according to any of the preceding claims, wherein the fraction containers (22) are used as conveyors for transferring the fraction of textile subjects to the recycling stations (48,53,61).

7. The handling system according to any of the preceding claims, wherein the lay-up unit (29) comprise two consecutive grippers (30,31) arranged so that a first gripper (30) grips the textile subjects (23) individually from the buffer space (24) and a second grip- per (31) grips the textile subjects (23) from the first gripper (30) and transfers the indi- vidual textile subject to the infeed conveyor (32).

8. The handling system according to any of the preceding claims, wherein property sensors of the final sorting line (15) at least comprise- a vision or optical recognition system (35,38),- a X-ray recognition system (36), and- an infrared light recognition system (34,37)for data measurement of each textile subject and wherein the final sorting line (15) further comprises-feature extractors (level 1 in Fig.9) for feature classification based on input form the property sensors,-classification calculators (level 2 in Fig. 9) for classification based on outputs from the feature extractors and a predetermined set of rules and-prioritizing units (level 3 in Fig. 9) to determine which fraction of textile subject the individual textile subject belongs to.

9. The handling system according to any of the preceding claims, wherein the main- sorting station (10) comprises- an unpacking and waste-sorting unit (8’) for sorting out items which are not textile subjects to be handled in the following sorting and recycling.

10. The handling system according to any of the preceding claims, wherein the data- base system contains predefined data for o Amount of sorted textile [kg] o Number of bales o Specific bale ID’s o Customer of the shipment o Destination location for the shipment and wherein these data are used in a set of rules to be fulfilled for creating a shipment order in the database.

11. The handling system according to any of the preceding claims, wherein the data- base system contains delivery data for: o Supplier (who delivered the material) o Amount received o Type of material received: Delivery Category (Household post-consumer waste HH-PO, post-consumer industrial laundry waste IL-PO, pre-con- sumer fashion brand FB-PR, etc...) o Type of receival (manual vs. automatic) o Date + time of initiated / completed: receival, pre-sorting, main-sorting, final sorting, sorting complete. o Location of receival and sortingand sorting data generated in the database system and including all of the following information or a part thereof: o All waste that was manually removed from the delivery in the pre-sorting and main-sorting, registered as kg by fractions, e.g.: 10 kg wet textiles, 50 kg reuseable, 100 kg multi-layer textiles... o All the sorting data from the automated final sorting, with the data being registered for each sorting textile:® Origin of textile: DeliverylD* Weight of textile: [kg]* Categorization result: [Category]* Sorting output: [pos] + [cagelD] and wherein these data are used in the database system for generating sorting reports.

12. The handling system according to any of the preceding claims, wherein the data- base system contains data for each bale which data comprise: o 10,1 kg (43 picks) from Delivery ID 1041 (delivery type HH-PO) o 15,4 kg (72 picks) from Delivery ID 2951 (delivery type IL-PO) o Time and location of baling o Time and location of shredding (if shredded) o Time and location of fiber-to-fiber mechanical recycling (if recycled) o Time, location and customer of shipping (if shipped) o Weight of the bale [kg] o Fiber quality (if tested): mean fiber length, percentage of short fibers, etc... o Material composition (if tested): 92% cotton, 5% polyester, 2% elastane, 1 % unknown... o Resource consumption o Current storage location of the bale. and wherein these data are used for creating a traceability report in the database.

13. A method for handling textile subjects (23), and including receiving, sorting, recov- ering and recycling the textile subjects for reusing the textile of collected textile sub- jects, characterized in that the reusing is either in form of second-hand textile subjects e.g. garments or in form of reusable textile fibers, or other types of disintegrated textile materials or recycled materials or downcycled materials, wherein said handling method comprises the steps of:- controlling the method steps with a database system (7) adapted to at least retrieve, store, and process data measured by sensors used in steps of the method and adapted to use the processed data to control operation of individual system units used for the method,- entering and registration of user input for delivery data relating to the bulk of textile subjects (23) entering the handling system, and forwarding the registrations to the da- tabase system (7) in which the delivery data is stored,- effecting a main-sorting in the main-sorting station (10), which main-sorting comprises receiving a bulk of textile subjects on a main bulk receiving conveyor (11),- registration of at least the weight of the textile objects with at least one sensor (3”) and transferring the weight registration to the database system (7) in which the weight registration stored and combined with the delivery data,- transferring the weight registration and the delivery data to an identification tag (137),- establishing batches (20’) of main-sorted textile subjects after the main-sort- ing of the received textile objects in a main-sorting unit (12),- application of the identification tag (137) to each batch (20’) in an application unit (138),- registration of the identification tag with a registration sensor,- associating the identification tag with the batch in the database system (7),- transporting the textile subjects batchwise to a buffer,- transferred the batches of main-sorted textile subjects to an infeed unit (14) for a final sorting line (15),- buffering the batches of main-sorted textile subjects in a buffer space (24) in the infeed unit (14),- registration of the identification tag with a registration sensor,- associating the identification tag with the batch of main-sorted textile subjects in the database system (7),- transferring the textile subjects (23) individually to the final sorting line (15) with a lay-up unit (29), by gripping the textile subjects individually from the buffer space (24) with at least one gripper (30,31),- registration that only one textile subject (23) is gripped by the at least one gripper (30,31) and transferred to individually to the final sorting line (15) with a mutual distance between consecutive textile subjects,- conveying the textile subjects individually past a number of consecutively arranged property sensors (34,35,36),- registration of different properties of each textile subject,- transferring the registered properties to the database system (7), -saving data for the registered properties in the database system,- associating in the database system the saved data of the registered proper- ties of each textile subject (23) to a specific fraction,- associating the registered properties to the registered batch (20’),- transferring the textile subjects past a number of final sorting units (22 ) with at least one sorting conveyor (44),- sorting the textile subject as the sorting unit is controlled based on the regis- tered properties of the textile subjects- automatically transferring the textile subjects from the at least one sorting conveyor (44) to fraction containers (22) based on the registered properties of the individual textile subjects, where each fraction container (22) is associated with one of the final sorting units (22’),- collecting the fraction of all textile subjects which has same properties in one of the fraction containers (22),-recycling the textile of the textile subjects to reuseable textile products and / or textile fibers in recycling stations comprising mechanical recycling units (48,53,61), chemical recycling units and downcycling units (45,46).

14. The method according to claim 13, wherein a step comprising pre-sorting is pro- vided before the main-sorting step, which pre-sorting step comprises- receiving waste containers (5) containing a bulk of the textile subject (23) to be handled in a bulk receiving station (3) which bulk receiving station com- prises a receiving unit (4),- entering and registration of user input for delivery data relating to the bulk of textile subjects received in the bulk receiving station (3) and forwarding the registrations to the database system (7) in which the delivery data is stored,- conveying textile subjects in bulk receiving conveyors (3, 3’, 3”),- registration of the weight of the bulk material with weighing sensors (140) connected with the bulk receiving conveyors (3, 3’, 3”),-transferring the registration of the weight to the database system (7) where it is associated with the registration of user input for delivery data relating to the bulk of textile subjects, hereby providing a batch of received textile subjects,- transporting the batch of pre-sorted and weighed textile subjects to the main- sorting station (10) for effecting the main-sorting.

15. The method according to claim 13 or 14, wherein the registration of different prop- erties of each textile subject, comprises identification and registration of one or more of the following features: colour distribution, such as colour gradients, fabric type, such as weave, knit, terry cloth and / or denim, material composition, density, mass, condition, such as cleanness, moisture content and / or mould content, presence of reflective material, presence of pattern, such as stripes and / or squares, presence of polymer prints, presence of hard accessories such as buttons, zippers, pens and / or tools, presence of contaminant chemicals such as oil, presence of multi-layer textile, such as jackets, pillows and duvets, type of textile subject, such as trousers, jeans, a shirt, a bag or a shoe, or the subject is not a textile subject, and wherein the step of sorting the textile subjects comprises a classification of the subject into predetermined classification groups, and / or predetermined sortation groups.

16. The method according to claim 13, 14 or 15, wherein the main-sorting step com- prises- sorting out items, which are not textile subjects to be handled in the following sorting and recycling, in an unpacking and main-sorting sorting unit (8).

Images