Process for classifying products with a view to their recycling.

Abstract

The present invention relates to a method for classifying a product, with a view to its recycling, the method being characterized in that it comprises the implementation by data processing means (21a) of a first server (2a) of steps of: Obtaining, for each material constituting said product, the weight of said material in the product; Determining subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique;For each sub-assembly, calculation of a recyclability rate of the materials of the sub-assembly and a physicochemical compatibility rate of the materials of the sub-assembly, based on a database of materials defining for each material individually whether it is recyclable and for each pair of materials whether they are compatible. Calculation of a score representative of the overall recyclability of said product based on said recyclability and physicochemical compatibility rates of the materials of each sub-assembly; Assignment to said product of said score representative of the overall recyclability of said product. Fig 1.;

Description

Title of the invention: Method for classifying products with a view to their recycling. [0001 ] GENERAL TECHNICAL FIELD

[0002] The present invention relates to the field of recycling. More specifically, it relates to a method for classifying products, in particular shoes, with a view to their possible sorting and then recycling.

[0003] STATE OF THE ART

[0004] Recycling products requires knowing their composition and assembly, in order to be able to send them to the correct processing line.

[0005] Indeed, certain products will be easy to recycle from the outset, while others will require complex dismantling and then specific operations, particularly chemical or mechanical.

[0006] For example, a sports shoe model may combine a rubber sole and a polyurethane-coated polyester mesh upper, laminated with EVA (Ethylene-vinyl acetate) foam, with the sole being glued to the upper.

[0007] To recycle it today we will simply crush it and sort the materials by density, but this process is not optimal because the usual materials used in shoes often have similar densities, which means that the purity of the material at the end of recycling is not sufficient to allow it to be reused other than by injecting it with a low-rate charge into virgin material, or by using it to create playgrounds, sports fields or cycle paths. One way to increase the purity of the material at the end of recycling would be to first mechanically tear off the sole and recycle the two parts separately, but this technique will be unsuitable for another shoe which would have a sewn sole, or which would be designed in such a way that its materials could be recycled together without prior separation.

[0008] Conversely, for a shoe which would be substantially mono-material, grinding followed by separation by density is useless.

[0009] It is therefore understood that it is necessary to move towards a personalization of the recycling process to optimize the quantity of materials recovered at the output while avoiding unnecessary steps and the associated costs.

[0010] A first technique is to simply look at the product label, on which we generally have the overall composition of the product (list of materials). However, this information does not reflect the distribution of these materials, the assembly techniques.

[0011] It is not possible to place a detailed composition on the label, because this information is a trade secret and many manufacturers would refuse.

[0012] Alternatively, one could have, for any product, a database defining an optimal recycling strategy, without revealing the detailed composition. However, this strategy will not always be able to be implemented so that one will regularly have to implement a "default" strategy, with the same drawbacks as before.

[0013] The invention improves the situation PRESENTATION OF THE INVENTION

[0014] The present invention therefore relates, according to a first aspect, to a method for classifying a product, with a view to its recycling, the method being characterized in that it comprises the implementation by data processing means of a first server of steps of: a. Obtaining, for each material constituting said product, the weight of said material in the product; b. Determination of subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique; c. For each sub-assembly, calculation of a recyclability rate for the materials of the sub-assembly and a physicochemical compatibility rate for the materials of the sub-assembly, based on a materials database defining for each material individually whether it is recyclable and for each pair of materials whether they are compatible; d. Calculation of a score representative of the overall recyclability of said product based on said recyclability rates and physicochemical compatibility of the materials of each sub-assembly; e. Assignment to said product of said score representative of the overall recyclability of said product.

[0015] According to advantageous and non-limiting characteristics:

[0016] Step (e) comprises recording the score representative of the overall recyclability of said product on a label of the product, recording the score representative of the overall recyclability of said product on a radio-frequency chip integrated into the product, and / or associating the score representative of the overall recyclability of said product in a database stored on data storage means of a second server.

[0017] The product comprises at least two parts capable of being obtained by dismantling the product, steps (a) to (c) being implemented for each part of said product, and step (d) comprising the calculation, for each part of said product, of the score representative of the overall recyclability of said part as a function of said recyclability rates and physicochemical compatibility of the materials of each subset of all of said materials constituting said part, and the calculation of the score representative of the overall recyclability of said product as a function of the scores representative of the overall recyclability of each part of the product.

[0018] Steps (b) and (c) are implemented for a plurality of given material separation techniques, said score representative of the overall recyclability of said product being a vector comprising, for each of said separation techniques, at least one value calculated as a function of said recyclability and physicochemical compatibility rates of the materials calculated for said separation technique.

[0019] Step (c) comprises, for each subset, the calculation of a plurality of physicochemical compatibility rates each corresponding to a material regeneration technique of a plurality of given material regeneration techniques, said materials database defining for each pair of materials whether they are compatible in the case of each of said material regeneration techniques, said score representative of the overall recyclability of said product being a vector comprising, for each of said regeneration techniques, a plurality of values calculated as a function of said physicochemical compatibility rates of the materials calculated for said regeneration technique.

[0020] Step (c) is also implemented for the complete set of said materials constituting the product, said score representative of the overall recyclability of said product being a vector further comprising a value calculated as a function of said recyclability and physicochemical compatibility rates of the materials of said set.

[0021] Said given separation technique is selected from density separation, magnetic separation, triboelectric separation, pneumatic separation, and optical separation.

[0022] Step (c) further comprises calculating, for each subset, the proportion of the materials of the subset relative to the total weight of the product, the score; said score representative of the overall recyclability of said product being calculated in step (d) as the average weighted by said proportions, of the products of said recyclability rates and physicochemical compatibility of the materials of each subset.

[0023] Said materials database is stored by storage means of the first server and further defines for each material a behavior associated with said given separation technique.

[0024] Step (a) also comprises obtaining each contaminant from at least one of said materials constituting the product.

[0025] Each contaminant of a material either prohibits the possibility of dismantling, or prohibits at least one given separation technique, or prohibits the recycling of the material.

[0026] According to a second aspect, the invention relates to a method for recycling a product in a system, characterized in that it comprises the implementation of steps of: A. Obtaining by data processing means of a second server of the system a score representative of the overall recyclability of said product, assigned to said product in accordance with the method according to the first aspect; B. Determination by the data processing means of the second server of an optimal recycling strategy based on said score representative of the overall recyclability of said product; C. Recycling of the product by the system in accordance with said determined optimal strategy.

[0027] According to advantageous and non-limiting characteristics, said recycling strategy comprises: - Whether or not the product is dismantled into parts; - The implementation or not of a given material separation technique; - The implementation of a given material regeneration technique.

[0028] According to a third aspect, the invention relates to a server for classifying a product, with a view to its recycling, the server being characterized in that it comprises data processing means configured to: - Obtain, for each material constituting said product, the weight of said material in the product; - Determine subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique; - For each subassembly, calculate a recyclability rate for the materials in the subassembly and a physicochemical compatibility rate for the materials in the subassembly, based on a materials database defining for each material individually whether it is recyclable and for each pair of materials whether they are chemically compatible. - Calculate a score representative of the overall recyclability of said product based on said recyclability rates and physicochemical compatibility of the materials of each sub-assembly; - Assign said product said score representative of the overall recyclability of said product

[0029] According to a fourth aspect, the invention relates to an assembly comprising the server according to the third aspect and the system for implementing the method according to the second aspect, connected.

[0030] According to a fifth and a sixth aspect, the invention relates to a computer program product comprising code instructions for executing a method according to the first aspect of classifying a product, with a view to its recycling; and a storage means readable by computer equipment on which is recorded a computer program product comprising code instructions for executing a method according to the first aspect of classifying a product, with a view to its recycling. PRESENTATION OF THE FIGURES

[0031] Other characteristics and advantages of the present invention will appear on reading the following description of a preferred embodiment. This description will be given with reference to the appended drawings in which:

[0032] [Fig.l] [Fig.l] is a diagram of a system for implementing the method according to the invention;

[0033] [Fig.2] [Fig.2] is a flowchart illustrating the steps of an embodiment of the classification method according to one aspect of the invention;

[0034] [Fig.3] [Fig.3] is a flowchart illustrating the steps of an embodiment of the recycling method according to another aspect of the invention. DETAILED DESCRIPTION

[0035] Architecture

[0036] With reference to [Fig.l], the present invention relates to a method of classifying a product, with a view to recycling said product. The invention also relates to recycling the product in a system 1 as shown in [Fig.l].

[0037] Said product may be any product whose recycling is desired, in particular an item of clothing such as a garment or a shoe, and in the remainder of the description a shoe will be taken, but it may be any type of product, and in particular a spare part, bulk merchandise, etc. It is understood that the present method is typically implemented before the marketing of the product, in particular during its design, so that it is already classified during its life and the result of this classification is used during its final recycling.

[0038] By classification, we mean, as we will see, the allocation to each product of a score representative of the overall recyclability of said product (called overall score) which will surprisingly be sufficient to define the recycling strategy to be applied, in addition to the information already available (label).

[0039] By score we mean any set (i.e. vector) of one or more dimensionless quantities each of which can have several ordered values, typically a number between a minimum and a maximum (for example a percentage between 0 and 100%, or a score between 0 and 10), but we could for example have alternating increasing letters, for example from A to F, the various values of the vector advantageously corresponding to various alternative techniques for separating materials and / or regenerating materials. In the remainder of this description we will take a vector of 2(n+l)m values (in practice for absence / presence of dismantling, n given separation techniques + the case without separation technique, and m given material regeneration techniques) between 0 and 10 but the present invention will not be limited to any particular type of score.

[0040] This overall score expresses the ease of recycling the product: generally the higher the values that make up the score, the less complex techniques will be necessary to recycle the product, and the more material can be expected to be recovered (note that an inverse scale can be provided). And this score remains a code that does not disclose how the product is constructed, so it can be published freely without the risk of revealing confidential information.

[0041] Preferably, the score comprises, as mentioned, several values associated with various recycling strategies, in particular defining material separation techniques and / or material regeneration techniques. It is therefore sufficient to take the highest value and / or to define ranges of score values for each recycling strategy in order to be able to choose effectively. Examples will be seen later.

[0042] The present method is implemented by a first server 2. Advantageously, there is a second server 2b (which is a piece of equipment controlling the recycling as we will see), which will be considered as part of the system 1, even if it can be remote and connected by a network 20 such as the internet network. Note that the same second server 2b can be used for several systems 1.

[0043] Each server 2a, 2b has data processing means 21a, 21b (typically a processor) and data storage means 22a, 22b (a memory, for example a hard disk). As will be seen, the data storage means 22a of the first server 2a can store a database of materials to which we will return later, and / or the data storage means 22b of the second server 2b can store a database of products each associated with their overall score calculated by the present method.

[0044] Said possible system 1 is a recycling system allowing the sorting of the product and then its recycling in accordance with a plurality of predefined alternative recycling strategies. In other words, it comprises a plurality of means for recycling the sorted products (according to the various alternative recycling strategies) which will be returned to later. It is understood that these means may in practice be far apart from each other (i.e. system 1 is a set of installations), and require, for example, the transport of products after sorting (see below). Generally speaking, the recycling of a product includes, if possible, its dismantling (its separation into several parts), a technique for separating the materials in each part (often crushing then separation by density), then the regeneration of the materials (recycled material is obtained that can be incorporated into a new product).

[0045] The system 1 may comprise a sorting unit 10 which will be controlled by said second server 2b, and which is typically of the belt conveyor type (as seen in [Fig.l]), that is to say that it comprises an endless conveyor belt, continuously moving instances of said product, but any other technique may be used (the sorting unit 10 may be roller-based, bucket-based, have arms, or even a cavity crossed by the product in free fall, etc.).

[0046] Sorting means the separation of the product instances according to the classification result. It is assumed that any pre-sorting has already been carried out (in particular by unit 10 or another), allowing for example to take instances of the product which could have a second life (to be reconditioned or repaired), so that only the instances to be recycled are sorted.

[0047] In this respect, the unit 10 preferably comprises means for sorting the product, for example switches, doors, various actuators, or even gripping arms, etc. The sorting means make it possible to place together the instances of the product of the same class (for example all the shoes of the same model but also shoes of different models having similar overall scores), with a view to recycling them.

[0048] The system 1 may further comprise means for obtaining the overall score representative of the recyclability of each product which passes through it. These means may be a simple RFID reader or a tag reader.

[0049] According to one embodiment, the means for obtaining the overall score comprise means for identifying the product, the overall score then being obtained in a product database, in particular stored by the second server 2b, in which each product is associated with its overall score. Said means for identifying the product may be cameras 11 for observing said product so as to identify it (by processing the images by the data processing means 21b of the second server 2b).

[0050] Those skilled in the art may in particular refer to application FR2400826 which describes a high-performance technique for recognizing a product in the system 1 with several cameras 11 from several points of view.

[0051] In any case, we will not be limited to any technique allowing us to obtain said overall score.

[0052] Method

[0053] With reference to [Fig.2], the present method is implemented by the data processing means 21 of the first server 2a, and begins with a step (a) of obtaining, for each material constituting said product, the weight of said material in the product.

[0054] In other words, we obtain a list of materials defining all of said materials constituting the product, each associated with its weight in the product.

[0055] Note that alternatively to the weight, one can have the total weight of the product and the rate of the material in the product, it being understood that the sum of the weights of the materials constituting the product is equal to said total weight.

[0056] Said weights are generally entered on an interface of the first server 2a (and they are generally known during the design of the product), for example via a terminal connected to the first server, but alternatively any physicochemical technique may be used to find these constituent materials from the product, and then step (a) will be a step of acquiring said weights. It will be assumed in the remainder of this description that these weights may be obtained in any way known to those skilled in the art.

[0057] By material is meant any elementary material constituting a component of the product (in other words a basic building block of the product), with the exception of any “contaminants” which will be described later and which are only present in trace amounts.

[0058] Preferably, said materials are chosen from a list of possible materials: it is sufficient to simply enter the weight for each material of the product, and leave 0 for the other materials.

[0059] The materials themselves can be classified by family of materials: metals, plastics, glass, plant materials, etc. In the family of plastics which often constitute products such as clothing, and in particular shoes, the following example list will be taken in the remainder of the description: polyurethane foam (PU), polypropylene (PP), high density polyethylene (HDPE), low density polypropylene (LDPE), polystyrene polyethylene-butylene (SEBS), polycarbonate (PC), rubber, thermoplastic elastomer (TPE), ethylene-vinyl acetate foam (EVA), polyvinyl chloride (PVC), polyester (PES), polyetheretherketone (PEEK), polyethylene terephthalate (PET), polyoxymethylene (POM), polyamide 6.6 (PA6.6), polyamide 11 (PA11), etc. The present method has been tested with a base of approximately 400 materials.

[0060] For example, we obtain the following list for the shoe mentioned in the introduction: Material Weight(g) Proportion PU 160 16.65% PES 235 24.45% EVA 36 3.75% Rubber 530 55.15%

[0061] Preferably, step (a) (as well as steps (b) and (c) as will be seen) is implemented for each part of said product that can be obtained by dismantling the product. We will refer to the term “dismantlable parts” of the product.

[0062] In other words, for each part of the product obtained by dismantling, for each material constituting said part of the product, the weight of said material in the part is obtained. We then obtain as many sets of materials as there are parts, called first set, second set, etc. By "dismantling", we mean the disassembly of the product into units (said dismantable parts) by a given technique, for example mechanical tearing, but also unscrewing, hot detachment, cutting, etc. It is understood that dismantling allows the product to be directly separated into said parts, and that these are well-identified functional units of the product and not just fragments. Note that dismantling can be slightly destructive (the parts may be damaged) but is not a complete destruction such as crushing (where only indistinguishable fragments are obtained in large numbers).Thus the said parts obtained are few in number and very distinct. Note that the number of parts that can be obtained by dismantling is often less than the total number of units initially present if it is not possible to separate some of them without destroying them completely.

[0063] For example, our shoe has a sole and an upper that can be dismantled (by tearing off the sole). On the other hand, the upper and the foam it contains cannot be dismantled without destroying them completely. So in the end we only have two parts that can be obtained by dismantling the shoe: - The sole, - The rest (upper + inner lining).

[0064] In this case we will have a first set with only PU, PES and EVA (and their weights: 160, 235 and 36 g) and a second set with only the rubber (530 g). We note that it is quite possible that the same material is found in two parts.

[0065] Note that preferably, the method is implemented both for each part and for the entire product, so as to compare the scenarios with and without de- mantling.

[0066] A "recycling strategy" will be defined as a scenario including or not dismantling of the product, the implementation or not of a given material separation technique, and the implementation of a given material regeneration technique (see below). In summary, a recycling strategy includes at least the implementation of a material regeneration technique, and preferably the prior implementation of dismantling and / or a material separation technique. There may be a plurality (n) of given alternative material separation techniques, a plurality (m) of given alternative material regeneration techniques, and therefore, considering the case of the absence of a separation technique, there may be at most 2(n+1)m possible recycling strategies.

[0067] Preferably, it is also possible to obtain in step (a) each contaminant of at least one of said materials constituting the product.

[0068] By contaminant, or “pollutant”, we mean a material generally present in trace amounts, and constituting only an additive (and not a brick), typically an assembly material such as a glue, or even a coating, an insert, etc.

[0069] Indeed, contaminants can have several impacts: - The impossibility of separating parts and therefore of implementing recycling strategies including dismantling (for example certain means of assembling said parts such as glues, or certain materials incompatible with the available dismantling processes); - The impossibility of implementing recycling strategies including certain separation techniques (for example in the case of lamination or coating of a base material with a second material) - The complete impossibility of sorting or even recycling the product / part / material (for example, the presence of carbon black prevents the detection of a material by near infrared, and the presence of a flame retardant prevents any recycling).

[0070] Thus, with regard to the first point, the method advantageously comprises the implementation of step (a) (and following) for each part of the product only in the absence of contaminant incompatible with dismantling, and otherwise the implementation without considering dismantling (for the entire product).

[0071] Then, in a step (b), the method comprises the determination of subsets of said set of said materials constituting the product, such that two materials of two different subsets are ultimately separable by a given separation technique. Said subsets forming a partition of the initial set, i.e. no intersection.

[0072] We understand here that "separable" goes beyond dismantling (we can quite easily having separable materials in parts that cannot be separated by dismantling), and separates fragments of the product (or a dismantled part) based not on the overall architecture of the product but on a physical property. To rephrase further, dismantling divides the product into parts that are identifiable units of the product, whereas separation divides a part / product into sets of fragments with common properties.

[0073] These separation techniques therefore necessarily involve the complete destruction of the product: said separation technique is generally implemented after grinding the product (shredding), i.e. putting it into small, potentially separable fragments. We will therefore preferentially speak of a separation technique after grinding.

[0074] Said given separation technique is advantageously separation by density, for example by floating the fragments of the materials, even if other techniques such as magnetic separation, triboelectric separation, pneumatic separation, optical separation (near infrared), etc. could be used.

[0075] In contrast, the materials of the same subassembly cannot be separated by said technique, even if they are different.

[0076] Please note that it is possible that at the end of step (b) only one subgroup is ultimately defined (which is in fact the original set), if the only materials making up the product are not in practice separable with said given technique (which may mean that this technique is unsuitable for this set of materials, unless the materials are compatible and can be recycled together).

[0077] Preferably, step (b) is implemented for several given separation techniques (i.e., for each given separation technique, subsets of said set of said materials constituting the product are determined such that two materials of two different subsets are separable by said given separation technique). Note that said plurality of given separation techniques may be limited to those compatible with the possible contaminants obtained in step (a).

[0078] To rephrase, the implementation of said technique results in the formation of as many “packets” of material as there are subsets.

[0079] For example, to return to our shoe example, assuming separation by density we have two subsets of the set {PU, EVA, PES, rubber]: on one side {EVA and PU] which float in water, and on the other side {PES and rubber] which sink. Note that if we had taken the magnetic separation technique, we would have a single subset equal to the original set since none of these materials is magnetic.

[0080] If there are several parts that can be obtained by dismantling and therefore several sets of materials, step (b) is implemented for each set, and so we obtain subsets for each set. Thus, if we ever wanted to detach the sole and the upper part, we would have a first set {PU, EVA, PES} (corresponding to the upper part) whose subgroups are {PU, EVA} and {PES}, and a second set {rubber} (corresponding to the sole) having only {rubber} as a subset because it is all alone.

[0081] To define the subsets, any known solution can be used, but in a simple way it is possible to use the said material base mentioned above (or directly define a separation technique base). More precisely, for at least one separation technique, it is possible to define for each material a behavior associated with the said given separation technique, for example via an index (having a value among a plurality of possible index values). For example, for the density technique, the index "a" can be used if the material floats and "b" otherwise.

[0082] The number of index values defines in practice the maximum number of subsets of a set. It is therefore sufficient to read, for each of the materials making up the product, the index corresponding to said separation technique, and to assign the material to a subset resembling all the materials having the same value of this index. In the case of several separation techniques, it is possible to have several indices (one for each technique).

[0083] Then, in a main step (c), the processing means 21a calculate, for each subset (where appropriate for each set of materials corresponding to a part of the product likely to be obtained by dismantling), two components of the overall score which are the recyclability rate and the physicochemical compatibility rate of the materials of the subset (in other words a recyclability rate and a physicochemical compatibility rate of the subset, by each subset).

[0084] The recyclability rate of the materials is preferably a value between 0 and 1, and illustrates which part of the subset materials is recyclable. A rate of 0 means that none of the materials in the subset are recyclable, and a rate of 1 means that all of the materials in the subset are recyclable.

[0085] The physicochemical compatibility rate of the materials is also preferably a value between 0 and 1, and illustrates to what extent the materials of the subassembly can be physically or chemically regenerated together. Indeed, it is recalled that the materials of the same subassembly are not separable, at least by the technique envisaged. And it is quite possible to have a mixture of two materials each perfectly recyclable, but which become impossible to recycle if they are mixed due to a physicochemical incompatibility.

[0086] By "physicochemical compatibility" is meant for a regeneration technique given materials (i.e. compatibility of the materials of the subassembly with each other when said regeneration technique of these materials is implemented), which is itself a physicochemical process. In other words, a regeneration technique adapted to a first material or a second material taken alone (i.e. individually recyclable) may prove inapplicable if these first and second materials are mixed (typically the materials will interfere). We then say that the two materials present a physicochemical incompatibility.

[0087] It is therefore understood that, for a given material regeneration technique, several materials are said to be compatible if they can be regenerated even if they are still mixed in the subset, whereas several materials are said to be incompatible if the fact that they are mixed in the subset will cause the regeneration technique to fail. Note that materials may be incompatible for a first regeneration technique, but compatible for another.

[0088] Today, the most widely used regeneration technique is the mechanical regeneration technique (ultrafine grinding known as micronization), but we can also have thermal regeneration (fusion), or chemical regeneration (by chemical reaction, typically by using solvents) or even the combination (e.g.: thermo-mechanical recycling). Thus, in the case of classic chemical regeneration, the compatibility is chemical compatibility.

[0089] Preferably, there are a plurality of given material regeneration techniques, and step (c) comprises the calculation, for each subset, of a physicochemical compatibility rate corresponding to one of said given material regeneration techniques. For example, if three alternative regeneration techniques are considered, there would be the calculation for each subset of the recyclability rate and three physicochemical compatibility rates. In the remainder of this description, a single (chemical) regeneration technique will be considered, and therefore there will be a single physicochemical compatibility rate.

[0090] Each of these recyclability and physicochemical compatibility rates can be calculated from “elementary” rates advantageously entered in the aforementioned materials database. For each material, we can have: - an individual recycling rate of the material (preferably binary, i.e. the two values 1 and 0 signifying that the material is individually recyclable or not, even if we can envisage intermediate rates signifying partial recyclability of the material). - The list of materials with which it is compatible, and possibly the list of materials with which it is partially compatible. Again, said physicochemical compatibility is understood for a given material regeneration technique, and we can have several lists in case of plurality of regeneration techniques considered.

[0091] Materials not included in these lists are considered incompatible. This list(s) define(s) a physicochemical compatibility rate for each possible pair of materials (depending on the regeneration technique concerned), either 0 if the two materials are not in their reciprocal lists, or 1 if the two materials are in their reciprocal lists of compatible materials, and for example 0.5 if the two materials are not in their reciprocal lists of compatible materials but in their possible reciprocal lists of partially compatible materials. Note that alternatively the physicochemical compatibility rate of a pair can be binary (0 or 1). Alternatively, the physicochemical compatibility rates of each pair can be precalculated in the form of a matrix (and therefore if there are several regeneration techniques there are several matrices).Note that the number of pairs of materials is very high (potentially several hundred thousand), so that in a particularly preferred manner said physicochemical compatibility matrix is completed by artificial intelligence: being known (by real tests) the physicochemical compatibility rates of a certain number of pairs (a few tens to a few hundred), we can use these pairs as learning examples to determine all the other values.

[0092] Then, from these elementary rates, for a subset, we can calculate: - The recyclability rate of the subset as a function of the individual recycling rates of each material of the subset, preferably the average (i.e. sum of the individual rates divided by the number of materials of the subset), each material being optionally weighted by its proportion. Note that in the event of the presence in a material of the subset of a contaminant making its recycling purely and simply impossible, the recyclability rate of the subset can arbitrarily be set to 0. - The physicochemical compatibility rate of the subset as a function of the physicochemical compatibility rates of each pair of materials in the subset, preferably the average (i.e. sum of the rates of the pairs of materials divided by the number of possible pairs of materials in the subset - the latter being equal to n(nl) / 2 if n is the number of materials in the subset), each material being again optionally weighted by its proportion. Note that if the subset only includes one material, the formula cannot be applied because there is no possible pair, but there is no compatibility problem so the rate is arbitrarily set to 1.

[0093] In the case of weighting by proportion, step (c) advantageously comprises- carefully beforehand, for each sub-assembly, the calculation of the proportion of each material of the sub-assembly in relation to the total weight of the materials of the sub-assembly, i.e. weight of the material in the sub-assembly divided by total weight of the sub-assembly (= sum of the weights of the materials of the sub-assembly).

[0094] In our example, PES and rubber are recyclable (individual rate of 1) but not PU and EVA (individual rate of 0), so the recyclability rate of the subset of materials that float is 0 (=(0+0) / 2) and the recyclability rate of the subset of materials that sink is 1 (=(1+1) / 2). The weightings do not change the calculation.

[0095] However, unfortunately PES is not compatible before rubber and PU is not compatible with EVA, so the physicochemical compatibility rates of the two subsets are 0 (=0 / (1 *2 / 2)).

[0096] If we implement a dismantling, we have: - for the upper part (two sub-assemblies) • recyclability rates 0 and 1 • the physicochemical compatibility rates 0 and 1 (because PES alone) - for the sole (a single sub-assembly) • the recyclability rate of 1 • the physicochemical compatibility rate of 1 (because rubber alone)

[0097] If there are several separation techniques, as explained, steps (b) and (c) are implemented for each of said given separation techniques. There are therefore several sets of subsets (for each separation technique), and thus for each of the subsets of each of these sets, the recyclability and physicochemical compatibility rates of the materials are calculated.

[0098] Note that step (c) can also be implemented in the absence of a separation technique, i.e. for the complete set of said materials constituting the product. To reformulate, we have a single subset, which is the complete set of materials of the product. The idea is to consider the absence of a separation technique as a special case of a separation technique.

[0099] In the example of the shoe, without separation technique, we keep the whole {PU, EVA, PES, rubber], with a recyclability rate of 0.5 = (=(0+0+1+1) / 4) and a physicochemical compatibility rate always of 0.

[0100] If we implement a dismantling (tearing off the sole) without separation technique (i.e. with a single assembly corresponding to the complete assembly for each part), we have: - for the upper part • the recyclability rate of 0.33 • the physicochemical compatibility rate 0 - for the sole • the recyclability rate of 1 • the physicochemical compatibility rate of 1 (because TPE alone)

[0101] Thus, with n given separation techniques, step (c) can be implemented up to n+1 times.

[0102] Then, in a step (d), the method comprises the calculation of said score representative of the overall recyclability of said product as a function of said recyclability rates and physicochemical compatibility of the materials of each subassembly.

[0103] Preferably, step (d) comprises the calculation of the average (advantageously weighted by the respective proportions of the detachable parts relative to the product) of the products for each subset of the recyclability rate with the physicochemical compatibility rate, i.e. unfortunately 0 (=0*0+1*0) in our example.

[0104] Indeed, the product of the recyclability and physicochemical compatibility rates of a subset is an estimate of the rate of recoverable materials in the subset. The value obtained after weighted average is therefore an approximation of the overall proportion of recoverable materials in the product, when the corresponding separation technique is used.

[0105] In the event of dismantling, the average can be taken (where appropriate weighted by the respective proportions of the parts likely to be obtained by dismantling in relation to the total weight produced) of the overall scores calculated for each part, or they can be combined in a more precise manner (for example by concatenating the corresponding vectors).

[0106] In this respect, step (c) advantageously also comprises the calculation of the proportion of each subset relative to the total weight of the product, i.e. weight of the materials in the subset (= sum of the weights of each material in the subset) divided by total weight of the product (= sum of the weights of the subsets = sum of the weights of all the materials). In the example of separation by density, the subset {EVA, PU] is in a proportion of 20.4% and the subset {PES, rubber] is in a proportion of 79.6%.

[0107] If there are several separation techniques and / or in the case of the absence of a separation technique, step (d) is implemented as many times.

[0108] Similarly, if there are several regeneration techniques, step (d) is implemented as many times.

[0109] If there are both several separation techniques and several regeneration techniques, step (d) is implemented as many times as there are combinations of a separation technique and a regeneration technique.

[0110] We therefore obtain a vector of values, each corresponding to a combination of techniques (i.e. the overall score is a vector of values each associated with a combination of techniques), the dimension of the vector being preferably equal to the number of possible recycling strategies, which we recall can be equal to 2(n+l)m with n the number of different separation techniques, considering the cases of dismantling or not and the possibility of an absence of separation technique, and m the number of regeneration techniques.

[0111] Thus, by first putting the value without separation then the value with separation by density, and only the chemical regeneration (n=l, m=l, i.e. 4 possible strategies), in our shoe example: - if we do not consider any dismantling implementation (i.e. a single part), the two values obtained are (0;0); - if we dismantle the sole / upper part, the values obtained are (0.55; 0.796). Indeed, tearing off the sole allows the rubber to be put aside and recycled (55.15% in proportion), and if we then implement separation by density on the upper part, we isolate the PES, hence an additional 24.45%, each subgroup having recyclability and physicochemical compatibility rates of 1.

[0112] We can stop there, and define that the vector of said values obtained constitutes the overall score. The values can be converted into a more or less rounded note (for example out of 10) or a letter for greater readability, by providing for example 10 / 10 for more than 95%, 9 / 10 between 85% and 95%, etc.

[0113] Preferably, these values are only one component of the overall score, and we will, for example, take into account a number of separation steps (or even the total number of steps, i.e., including dismantling, sorting, etc.) and / or a number of sub-assemblies comprising recoverable materials (the other “lost” sub-assemblies are ignored). Indeed, a product that would be composed of k materials, each 100% recyclable, but involving the separation of the materials into k groups, would have the same recoverability (100%) as a single-material product, even though the latter is much simpler to recycle and therefore much more economically profitable.

[0114] Thus, by naming respectively, for a given separation technique, first note, the value obtained as a function of said recyclability and physicochemical compatibility rates of the materials of each subset, second note, a value varying inversely (in a decreasing manner) with the number of stages of the given separation technique (for example 0 / 10 for 3 stages or more, 3.33 / 10 for 2 stages, 6.66 / 10 for 1 stage, and 10 / 10 for no separation technique - it is recalled that there may be the additional case of an absence of separation technique), and third note, a value varying inversely with the number of subsets containing recoverable materials (i.e. a subset for which the product recyclability rate and compatibility rate is not zero - we can take a scale similar to the second score, for example 0 / 10 for 4 subsets or more, 3.33 / 10 for 3 subsets, 6.66 / 10 for 2 subsets, and 10 / 10 for a single subset), the overall score can thus be a vector of weighted averages (with coefficients representative of the importance of these three scores, for example 0.7, 0.1, and 0.2, but the person skilled in the art will know how to adapt according to the situations) of the first, second and third scores for each separation technique.

[0115] We can possibly predict that if the first score is 0, then the corresponding score is directly 0 without calculating the second and third scores (since in any case there is nothing to recycle).

[0116] Please note, in the event of dismantling (several parts), 1 must be added to the number of separation steps (dismantling being an additional step).

[0117] In summary, in the case with dismantling (otherwise we have 0 everywhere): Without separation With separation 1st grade 5.5 8 2nd grade 6.66 (1 step) 3.33 (2 steps) 3rd grade 10 (1 subset) 6.66 (2 subsets) Overall grade 6.5 7.3

[0118] Overall score (0; 0; 6.5; 7.3). If we want letters, we can put (F, F, C, B)

[0119] We therefore see that dismantling is obligatory, with a slightly higher note higher in the case of density separation, which shows that the effort of carrying out this separation is worth the quantity of additional materials recovered. Nevertheless, the score without separation remains correct, so that even if the shoe is put into recycling in a system that would not be capable of carrying out density separation, the recycling remains interesting.

[0120] In order to further favor a small number of materials, we can apply to each note a penalty increasing with the number of materials, for example 1-1 / number of materials.

[0121] Then, the method comprises a final step (e) of assigning to said product said score representative of the recyclability of said product. The idea is that the score can subsequently be easily accessible during sorting and then recycling of the product.

[0122] Conventionally, step (e) may comprise the inscription of the score representative of the overall recyclability of said product on a label of the product, close to the composition. In the case of a score with several values, the inscription may specify to which dismantling / separation techniques / regeneration techniques these values refer. Color codes may be used.

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132] For example, the following summary table can be used: Without density separator With density separator Whole shoe 0 0 Dismantling 6.5 7.3 Alternatively or in addition, step (e) may comprise recording the score representative of the overall recyclability of said product on a radiofrequency chip integrated into the product, and / or associating the score representative of the overall recyclability of said product in a database (of products) stored on data storage means 22b of the second server 2b (that of the system 1). Recycling process According to a second aspect, the invention relates to the recycling method of the method in the system 1, which typically comprises a second server 2b connected to the first server 2a. While the classification method is typically implemented at the beginning of the product life, the recycling method is implemented at the end of the product life. With reference to [Fig.3], the recycling process begins with steps (A) and (B) implemented by the data processing means 21b of the second server 2b. In step (A), they obtain the score representative of the overall recyclability of said product (assigned to said product in accordance with the process according to the first aspect). This can be done in several different ways, with means of obtaining the score depending on how the score is assigned: - optical reading of a label (for example with a camera 11 connected to the server 2b); - reader of a radiofrequency chip (for example with an RFID reader connected to the server 2b); - identification of the product, and obtaining the code in a particular product database stored on data storage means 22b of the second server 2b; - etc. For this identification, we can again use cameras 11 and refer to request FR2400826. Then, in a step (B), an optimal recycling strategy is determined based on said score representative of the overall recyclability of said product, in particular from among a plurality of recycling strategies permitted by the system 1. Step (B) is typically accomplished by identifying the highest score value corresponding to a strategy permitted by System 1, i.e. directly by making a action based on the overall score.

[0133] It is recalled that each recycling strategy includes the dismantling or not of the product into parts and the implementation or not of a given material separation technique and the implementation of a material regeneration technique, i.e. 2(n+1)m cases. Thus, in our example, the dismantling + density separation strategy is chosen (chemical regeneration being fixed). Note that other criteria such as exceeding a threshold can be used, in particular with value ranges, taking into account the current size of the system 1 etc. For example, there is little difference between the scores with or without density separation, so that in certain cases it may be more interesting to do without separation.

[0134] Note that the set of recycling strategies permitted by system 1 may be smaller than the set of recycling strategies considered when classifying the product. For example, system 1 may not allow separation by density, and in this case, the chosen strategy is directly dismantling without separation.

[0135] Finally, in a step (C), the product is recycled by the system 1 in accordance with said determined optimal strategy. As explained, depending on the strategy to be applied, the data processing means 21b can control a sorting unit 10 of the system 1 so as to direct the product towards various recycling means corresponding to the various recycling strategies permitted by the system 1.

[0136] Server

[0137] According to a third aspect, the invention relates to the first server 2a for implementing the method according to the first aspect.

[0138] Thus, this first server 2a comprises, as explained, at least data processing means 21a and a memory 22a. This is typically a server for classifying a product with a view to its recycling.

[0139] The data processing means 21a are configured to implement steps consisting of: - Obtain, for each material constituting said product, the weight of said material in the product; - Determine subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique; - For each subassembly, calculate a recyclability rate for the materials in the subassembly and a physicochemical compatibility rate for the materials in the subassembly, based on a materials database defining for each material individually whether it is recyclable and for each pair of materials whether they are compatible. - Calculate a score representative of the overall recyclability of said product based on said recyclability rates and physicochemical compatibility of the materials of each sub-assembly; - Assign said product said score representative of the overall recyclability of said product.

[0140] According to a fourth aspect, the invention proposes an assembly comprising said first server 2a, as well as at least one second server 2a of a system 1 connected (via the network 20), for implementing the recycling method according to the second aspect.

[0141] Again, the system 1 and / or the first server 2a and / or the second server 2b may be confused.

[0142] Computer program product

[0143] According to a fifth and a sixth aspect, the invention relates to a computer program product comprising code instructions for the execution (on the data processing means 21a of the first server 2a) of a method according to the first aspect of classification of a product, with a view to its recycling, as well as storage means readable by computer equipment (for example the data storage means 22a of the first server 2a) on which this computer program product is found.

Claims

Claims

1. Method for classifying a product, with a view to its recycling, the method being characterized in that it comprises the implementation by data processing means (21a) of a first server (2a) of steps of: a. Obtaining, for each material constituting said product, the weight of said material in the product; b. Determination of subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique; c. For each sub-assembly, calculation of a recyclability rate for the materials of the sub-assembly and a physicochemical compatibility rate for the materials of the sub-assembly, based on a materials database defining for each material individually whether it is recyclable and for each pair of materials whether they are compatible; d. Calculation of a score representative of the overall recyclability of said product based on said recyclability rates and physicochemical compatibility of the materials of each sub-assembly; e. Assignment to said product of said score representative of the overall recyclability of said product.

2. Method according to claim 1, in which step (e) comprises writing the score representative of the overall recyclability of said product on a label of the product, recording the score representative of the overall recyclability of said product on a radiofrequency chip integrated into the product, and / or associating the score representative of the overall recyclability of said product in a database stored on data storage means (22b) of a second server (2b).

3. Method according to one of claims 1 and 2, in which the product comprises at least two parts capable of being obtained by dismantling the product, steps (a) to (c) being implemented for each part of said product, and step (d) comprising the calculation, for each part of said product, of the score representative of the overall recyclability of said part based on said recyclability rates and physicochemical compatibility of the materials of each subset of all of said materials constituting said part, and the calculation of the score representative of the overall recyclability of said product based on the scores representative of the overall recyclability of each part of the product.

4. Method according to one of claims 1 to 3, in which steps (b) and (c) are implemented for a plurality of given material separation techniques, said score representative of the overall recyclability of said product being a vector comprising, for each of said separation techniques, at least one value calculated as a function of said recyclability and physicochemical compatibility rates of the materials calculated for said separation technique.

5. Method according to one of claims 1 to 4, in which step (c) comprises, for each subset, the calculation of a plurality of physicochemical compatibility rates each corresponding to a material regeneration technique of a plurality of given material regeneration techniques, said materials database defining for each pair of materials whether they are compatible in the case of each of said material regeneration techniques, said score representative of the overall recyclability of said product being a vector comprising, for each of said regeneration techniques, a plurality of values calculated as a function of said physicochemical compatibility rates of the materials calculated for said regeneration technique.

6. Method according to one of claims 1 to 5, in which step (c) is also implemented for the complete set of said materials constituting the product, said score representative of the overall recyclability of said product being a vector further comprising a value calculated as a function of said recyclability rates and physicochemical compatibility of the materials of said set.

7. A method according to one of claims 1 to 6, wherein said given separation technique is selected from density separation, magnetic separation, triboelectric separation, pneumatic separation, and optical separation.

8. A method according to one of claims 1 to 7, wherein step (c) further comprises calculating, for each subset, the proportion of the materials of the sub-assembly in relation to the total weight of the product, the score; said score representative of the overall recyclability of said product being calculated in step (d) as the average weighted by said proportions, of the products of said recyclability rates and physicochemical compatibility of the materials of each sub-assembly.

9. Method according to one of claims 1 to 8, wherein said database of materials is stored by storage means of the first server (2a) and further defines for each material a behavior associated with said given separation technique.

10. A method according to one of claims 1 to 9, wherein step (a) also comprises obtaining each contaminant from at least one of said materials constituting the product.

11. A method according to claims 3, 4 and 10 in combination, wherein each contaminant of a material either prohibits the possibility of dismantling, prohibits at least one given separation technique, or prohibits recycling of the material.

12. Method for recycling a product in a system (1), characterized in that it comprises the implementation of steps of: A. Obtaining by data processing means (21b) of a second server (2b) of the system (1) a score representative of the overall recyclability of said product, assigned to said product in accordance with the method according to one of claims 1 to 10; B. Determination by the data processing means (21b) of the second server (2b) of an optimal recycling strategy as a function of said score representative of the overall recyclability of said product; C. Recycling of the product by the system (1) in accordance with said determined optimal strategy.

13. Method according to claim 12, wherein said recycling strategy comprises: - Whether or not to dismantle the product into parts; - Whether or not to implement a given material separation technique; - Whether or not to implement a given material regeneration technique.

14. Server (2a) for classifying a product, with a view to its recycling, the server (2a) being characterized in that it comprises data processing means (21a) configured to: - Obtain, for each material constituting said product, the weight of said material in the product; - Determine subsets of a set of said materials constituting the product, such that two materials of two different subsets are separable by a given separation technique; - For each subset, calculate a recyclability rate of the materials of the subset and a physicochemical compatibility rate of the materials of the subset, based on a database of materials defining for each material individually whether it is recyclable and for each pair of materials whether they are chemically compatible.- Calculate a score representative of the overall recyclability of said product based on said recyclability rates and physicochemical compatibility of the materials in each sub-assembly; - Assign to said product said score representative of the overall recyclability of said product.

15. Assembly comprising the server (2a) according to claim 14 and the system (1) for implementing the method according to claim 13 connected.

16. Computer program product comprising code instructions for executing a method according to one of claims 1 to 11 for classifying a product, with a view to its recycling, when said program is executed on a computer.

17. Storage means readable by computer equipment on which is recorded a computer program product comprising code instructions for the execution of a method according to one of claims 1 to 11 for classifying a product, with a view to its recycling.

Images