Method for determining application performance indicators of polymers

Abstract

The present invention relates to an apparatus for determining a performance index. A providing unit (112) provides a digital representation of a target polymer, indicative of characterization parameters of the polymer. A providing unit (113) provides a digital representation of application conditions, indicative of application parameters defining the application conditions. A providing unit (114) provides a model based on the provided digital representation of the application conditions. The model is adapted to determine a performance index of the polymer with respect to the application conditions. The model is a data-driven model parameterized with respect to the application conditions, such that a performance index of the polymer can be determined based on the digital representation of the polymer. A determining unit (115) for determining a performance index of the target polymer based on the provided determined model and the digital representation of the target polymer.

Description

[Technical field]

[0001] FIELD OF THEINVENTION The present invention relates to a method, an apparatus and a computer program product for determining indicators for application performance of a target polymer.Furthermore, the present invention relates to a method for determining a target synthesis specification indicative of a target polymer comprising a target indicator for application performance.Furthermore, the present invention relates to a training method, a training apparatus and a training computer program for training a data-driven decision model that can be used by the method, apparatus and computer program product for determining the biodegradability of a polymer and / or for determining a target synthesis specification.Furthermore, the present invention relates to a method and an apparatus for providing an interface for the determination of target indicators of a polymer. [Background technology]

[0002] 2. Background of the Invention In general, polymers are widely used in industrial and / or daily use products due to their wide range of application properties. The use of polymers encompasses coatings, personal care products, cleaning detergents, lubricants, packaging and foams, among others. However, this widespread application and the almost infinite possibilities of polymers often make it difficult to find a polymer for a specific application in a sensible time without a highly manpower-intensive, resource-intensive and cost-intensive search, for example, by synthesizing multiple candidate polymers, measuring their application performance, and then selecting a suitable polymer, if any. Furthermore, for most applications, not only one performance parameter but several different performance parameters must be satisfied. For example, a polymer must not only be suitable for use as a cleaning detergent, but also biodegradable when in contact with wastewater. This makes it even more difficult to find a suitable polymer at an early stage of the product design process. It would therefore be advantageous to provide the possibility to predict the application performance indicators of polymers accurately and in a computationally inexpensive manner. Summary of the Invention [Problem to be solved by the invention]

[0003] Summary of the Invention It is an object of the present invention to provide a method, an apparatus and a computer program product that allows for an accurate determination of application performance indicators of polymers, which are computationally inexpensive and can be reliably applied to new polymers. It is further an object of the present invention to provide a method, an apparatus and a computer program product that allows for an accurate and resource-free determination of the synthesis specifications of polymers that meet predefined targets. It is further an object of the present invention to provide a training method, a training apparatus and a computer program product that allows for providing decision models that can be used in methods, apparatus and computer programs that utilize fewer computational resources and can be trained to provide good decision accuracy. [Means for solving the problem]

[0004] In a first aspect of the present invention, there is provided a computer implemented method for determining an application performance indicator of a target polymer comprising: a) providing a digital representation of the target polymer indicative of or associated with characterization parameters of the polymer; b) providing a digital representation of application conditions for the target polymer, the digital representation of the application conditions indicative of or associated with application parameters defining the application conditions; c) providing a decision model based on the provided digital representation of the application conditions, the decision model adapted to determine the application performance indicator of the polymer with respect to the application conditions, the decision model being a data driven model parameterized with respect to the application conditions such that the application performance indicator of the polymer can be determined based on the digital representation of the polymer; and d) determining the application performance indicator of the target polymer based on the provided decision model and the digital representation of the target polymer.

[0005] Since the decision model is specifically adapted to determine the application performance indicators of the polymer under the specifically provided application conditions defined by the respective application parameters, the application performance indicators of the polymer for the respective application can be determined very accurately. Moreover, since the decision model is specifically trained for one or more specific application conditions, less training data is required for training. Moreover, the decision model becomes more flexible with respect to determining the application performance indicators of new polymers that are not part of the training data set. Thus, the method allows for accurate determination of application performance indicators that are computationally cheap and can be flexibly applied to new polymers as well. Moreover, while currently utilized methods for measuring the application performance indicators of polymers can often be very time-consuming and resource-intensive, the above method allows for essentially immediate results without consuming polymer material. Thus, not only can the technical requirements for finding a suitable polymer be reduced, but also the time required to design a new product for an application can be significantly reduced. Moreover, by providing an easy possibility for accurate determination of the application performance of a product that can already be taken into account during the design process, it becomes possible to design the product in such a way that the actual conditions in which the product will be used later are also taken into account, leading for example to reduced waste generation or increased security.

[0006] Furthermore, the training of the respective decision model can be improved since the characterization parameters, including, for example, chemical and / or physicochemical information of the polymer, such as polymer structure and composition, are utilized. In particular, utilizing the characterization parameters allows the training of such models with less training data, since some of the correlation information that needs to be learned is already presented to the model by using the characterization parameters. This further allows saving the tests and experiments required to provide the training data set.

[0007] The development of new chemical products tailored to application requirements is a major issue in the modern chemical industry. Recently, further requirements are often also raised, for example, related to the environmental impact or safety of chemical products along their life cycle. For example, one important aspect of the environmental impact can be the prevention of microplastics. Microplastics are a growing problem that can be avoided if the polymeric material is biodegradable. A series of standardized tests are currently used to evaluate biodegradation. For biodegradability, various tests exist with specific conditions (e.g., ISO13432, ISO14852, ISO14855, ISO17556 and OECD301). Such standardized tests often strike a balance between time-efficient tests (as short as 14 days, as long as 24 months) and real-life conditions. In fact, to reduce the test time, temperatures higher than the real conditions are often used. Companies developing new polymers need to invest significant resources in self-assessment and certification of the sustainability or safety of their products. A holistic assessment of such additional conditions, including laboratory space and equipment, becomes expensive and time-consuming.

[0008] Therefore, there is a need to identify application performance indicators of new materials early in the development process. The proposed method of determining application performance indicators as disclosed herein allows for a faster and more efficient way of developing new materials. At an early stage, the application performance can be determined even before the synthesis of the polymer. This allows to determine whether the polymer is suitable for market entry. This reduces the time to market. This also allows to reduce waste generation since it is not necessary to synthesize the polymer to determine the application performance. The proposed method provides a digital twin that measures the application performance indicators of the polymer.

[0009] Furthermore, standard measurements and tests for application performance are often time-consuming, for example involving waiting times of months or even years, and are resource-intensive when performing large test series. Especially when developing new polymers for the respective application, these time-consuming tests can strongly limit the development process. In this context, the present invention makes it possible to provide results for new polymers immediately, greatly reducing the time after which the results are available.

[0010] Furthermore, due to the incredibly large number of possible, often not fully explored, polymers potentially suitable for a particular application, today technical product engineers have to synthesize and test a huge amount of possible polymers or consult huge data sets and libraries in which potential polymers are stored in order to find each polymer that can fit the application, given the technical challenge of finding a polymer that is not only suitable for a particular application but also meets the respective target properties, especially the target application performance indicators. Even when using an elaborate design of experimental methods, still a very large number of possible polymers must be synthesized and experimentally tested. In this context, the above-mentioned method allows a user, for example a technical product engineer, to assist in automatically and much faster finding potentially suitable polymers. In particular, by utilizing the above-mentioned method, the user only has to synthesize and test potentially suitable polymers that the method determines to be highly likely to meet the respective target properties. Thus, unnecessary synthesis and testing of polymers can be avoided. Thus, the present method allows a user to perform the technical task of finding suitable polymers for a technical application more quickly and more efficiently.

[0011] The method refers to a computer-implemented method and may therefore be implemented by a general-purpose or dedicated computer adapted to implement the method, for example by executing a respective computer program. The method is adapted to determine, in particular predict, an application performance index of a given target polymer. The application performance index may be any index that defines the performance of the polymer in a particular application. In general, such an application performance index is an intrinsic characteristic of the polymer determined by the physical and chemical properties of the polymer itself. In a preferred embodiment, the performance index is the biodegradability of the polymer. In this case, biodegradability is an index of the polymer's ability to biodegrade in the respective waste environment of the intended application. For example, in the case of a cleaning detergent, the intended application waste environment is wastewater. In general, a biodegradable polymer refers to a polymer that can be degraded by biological processes, and in particular, a biodegradable polymer can refer to a polymer that can be assimilated by bacteria and / or fungi to give environmentally friendly products, i.e., that can be decomposed into non-polluting residues. For example, the biodegradability of a polymer preferably refers to any one of mineralization properties, biotransformation properties and / or degradation of the polymer after a certain time frame. Biodegradability can be used to verify the biodegradation, i.e., degradation characteristics, of a polymer, particularly for a particular biodegradation habitat, i.e., biodegradation habitat. For example, biodegradability can be compared to the biodegradation required for a particular application, and thus it can be determined whether the respective polymer is suitable for the respective application. In another preferred embodiment, the application performance indicator is the adhesive properties of the polymer in a given environment. The adhesive properties can generally determine the adhesive properties of the polymer with respect to adhering at least two surfaces to each other. For example, the adhesive properties can be adhesive strength, removability, timing of the onset of adhesion or loss of adhesion, etc. In a further preferred embodiment, the application performance indicator can be the solubility of the polymer in a given fluid environment.For example, solubility can be defined by the amount of soluble polymer in a particular fluid environment, the concentration in solution, the time to reach a given concentration, etc. In a further preferred embodiment, the application performance index is a washing performance index of the polymer in a given environment. Washing performance can generally determine the ability of the polymer to remove a substance from an object, for example to remove a particular stain color from a fabric. For example, the washing performance index can be any of the amount of substance removed from the object in the washing process, the solubility of a given substance such as fat, and a foaming index. In a further preferred embodiment, the application performance index is the burning behavior of the polymer. In general, burning behavior can be defined by one or more characteristics, which can refer to any property that allows the quantification of the burning behavior of a polymer, for example determining when a polymer will burn in a particular situation. In particular, the burning characteristics reflect the properties of the polymer when subjected to a heat source. Preferably, the one or more flammability properties indicative of the flammability behavior of the polymer included at least one of heat release rate, peak heat release, total heat release, average heat release rate, maximum heat release rate, effective heat of combustion, flame height, burn time, afterburn time, mass loss, mass loss rate, flame growth rate, total smoke production, smoke growth rate, dripping behavior, time to fire and time to extinguish.

[0012] Generally, an application performance index can refer to only one value, such as the ignition temperature or percentage of biodegradation of a polymer after a given time, but can also refer to more than one value, such as the temperature profile or biodegradation profile within a sample made from the polymer when subjected to a given application condition. Generally, an application performance index is an intrinsic characteristic of a polymer. In this context, an intrinsic characteristic of a polymer refers to a property of a polymer that is caused by and thus reflects the nature of the polymer, i.e., structure, composition, etc., for a particular context.

[0013] In general, a polymer can be any material made by polymerization. Preferably, the target polymer is a synthetic polymer. In one embodiment, a synthetic polymer can be a chemical compound produced by chemical production from one or more starting material(s), e.g., monomers, and comprising at least two monomer units. A monomer unit can be considered as a subunit of a polymer. A polymer can be prepared from a monomer by a polymerization reaction. A polymer can be made from a single type of monomer, or from different monomers or different monomers and other polymers. The monomer units can be randomly distributed, have composition drift, or exist as blocks within a polymer. A polymer can be a linear polymer. A polymer can be a branched polymer. A polymer can be a crosslinked polymer. A polymer can be soluble. A polymer can be a solid crystalline, semi-crystalline, or amorphous material. A polymer can have a tertiary structure, e.g., dispersed particles, micelles, or foams.

[0014] In a first step, the method comprises providing a digital representation of the target polymer indicative of or associated with a characterization parameter of the polymer. In particular, providing can refer to receiving the digital representation from a user's input, for example using a respective input unit. Furthermore, providing can also refer to accessing a storage unit in which the digital representation is already stored. Furthermore, providing can also include receiving the characterization parameters, for example via a network connection, from another source and providing the received characterization parameters as a digital representation. Furthermore, indicative of or associated with a characterization parameter of a polymer is defined as allowing access to information of the characterization parameters.

[0015] For example, the digital representation may directly include the characterizing parameters, e.g., in the form of values ​​for the respective quantities. However, the digital representation may also be a link to the respective characterizing parameters through which the characterizing parameters may be accessed, or the digital representation may reference an identifier associated with the characterizing parameters and allowing the respective look-up storage to be utilized to access the characterizing parameters. Furthermore, the digital representation may also point to information that allows the characterizing parameters to be derived using one or more known relationships. For example, the synthetic specifications or structural formula of a polymer may be utilized as a digital representation that allows the respective characterizing parameters to be derived using known chemical and physical laws and relationships.

[0016] In general, throughout the following description, referring to a parameter includes referring to both the respective quantity and the specific value of that quantity, unless explicitly defined otherwise. For example, a parameter that is temperature always refers to the quantity that is temperature and the specific value of temperature that is set for that quantity. In most cases, the explicit value of a parameter may be different for different embodiments and application cases, so the value is not generally mentioned. However, providing a parameter generally means providing information about a quantity, e.g., that the value is a temperature, and also the value of the quantity itself.

[0017] Preferably, the polymer characterization parameters consist of physicochemical parameters of the polymer. Preferably, the digital representation indicates and / or includes polymer physicochemical parameters, preferably referring to polymer descriptors, the polymer physicochemical parameters indicating the physicochemical properties of the polymer. In particular, the polymer physicochemical parameters indicate parameters that quantify the physicochemical properties of the polymer. In this context, the term "physicochemical properties" refers to the physical and / or chemical properties of the polymer. However, the digital representation can also be provided to allow for example to derive the physicochemical properties in the form of polymer descriptors, for example by providing a representation of a polymer in which the respective physicochemical properties are already stored or can be determined, for example, by a respective polymer descriptor calculation. Preferably, the digital representation refers to at least one of the following: recipe, structural formula, brand name, IUPAC name, chemical identifier and CAS number, SMILES name, InCHI name of the polymer.

[0018] In a preferred embodiment, the polymer physicochemical parameter is a parameter that quantifies the physicochemical properties of a subgroup of the polymer. In this embodiment, a digital representation can also be provided to allow deriving the polymer physicochemical parameter by determining a subgroup of the polymer and determining the polymer physicochemical parameter based on the physicochemical properties of the determined subgroup. Generally, a subgroup refers to a part of a polymer, where all subgroups of the polymer together form the polymer. For example, a subgroup can refer to a part of a polymer, where the subgroups are linked together in a chain or network in a continuous manner to form the polymer. Preferably, a subgroup of a polymer refers to a repeating unit that describes a part of the polymer that, when repeated, produces a complete polymer chain. However, in some cases, a subgroup can also refer to a single part of a polymer that is not repeated. Furthermore, it is preferred that the subgroup comprises a repeating part, for example, a subgroup of a polymer can comprise a repeating core that is also present in other subgroups and further additional parts that are not present in other subgroups. Preferably, the subgroup refers to at least one of the polymerized monomers or oligomer fragments. More preferably, the subgroup refers to the polymerized monomers. In this context, polymerized monomers refer to the monomers after their polymerization and are sometimes referred to as "mer units" or "mers". In particular, polymerized monomers do not refer to the monomers present in the reaction mixture before polymerization, i.e. raw materials, but rather to repeating units derived from monomers that have been altered during or after polymerization. The subgroup descriptors determined for polymerized monomers are therefore different from the subgroup descriptors determined for unreacted monomers before polymerization. The inventors have found that polymerized monomers in particular allow for the determination of polymer descriptors from the subgroup descriptors of polymerized monomers, which allows for an accurate determination of biodegradability. In a preferred embodiment, the digital representation of the polymer includes subgroups provided as molecular models showing the chemical structure of the subgroups after polymerization.Even more preferably, the molecular model of the subgroup is determined in a manner suitable for chemical informatics with regard to the number and type of atoms and their connectivity. Furthermore, in addition to or instead of the molecular model of the subgroup treating the subgroup as a monomeric structure, a molecular model can also be utilized which refers to an oligomeric model taking into account the effect of the neighboring molecular structures of the subgroup in the polymer.

[0019] In general, when the digital representation of a polymer does not directly include the polymer physicochemical parameters, the polymer physicochemical parameters are preferably determined by determining subgroups of the polymer. For example, the respective subgroups of the polymer can be determined using known methods. However, the determination of the subgroups of the polymer is preferably performed according to the following embodiment of the invention. In particular, the subgroups are preferably determined such that the bonds between the atoms of the different subgroups in the polymer are as unpolarized as possible, preferably with the smallest possible bond order (e.g., single C-C bonds). Furthermore, the subgroup representing the polymer preferably contains the same number of active non-hydrogen atoms as the polymer. The subgroup may also contain, besides the active atoms, further atoms that can be ignored when calculating the physicochemical parameters of the subgroup. Furthermore, the subgroups are preferably determined such that polymers containing moieties built using different polymerization techniques are sufficiently covered and satisfy the aforementioned conditions. An example is polyethers used as components of polyurethanes. In general, a database or archive can be created with multiple reactions between polymer moieties, and the subgroups can be obtained from the respective structures of the reactions. For example, the subgroups of the polymer can be easily derived using specific chemical languages ​​such as SMILES and SMARTS. For example, a database of reaction SMARTS can be generated and then the corresponding reaction SMARTS can be selected based on the polymerization of each polymer. From the selected reaction SMARTS, the SMILES of the subgroups of the polymer can be directly derived, e.g., using RDkit, the SMILES of the subgroups, i.e., the number of atoms and connectivity, can be determined from the SMILES of the monomers.

[0020] The determined subgroups of polymers are associated with subgroup physicochemical parameters that quantify the parameters characterizing the subgroups in the polymer, preferably the subgroup physicochemical parameters also refer to subgroup descriptors. In particular, if the polymer physicochemical parameters are not directly provided by the digital representation, the polymer physicochemical parameters are preferably determined by determining the respective subgroup physicochemical parameters for each of the subgroups, and the polymer physicochemical parameters are determined based on the subgroup physicochemical parameters of the subgroups, for example by averaging or summing. Thus, the method preferably comprises first providing or determining the subgroups of the polymer from the digital representation of the polymer, then determining or providing the physicochemical parameters of the subgroups, i.e. the values ​​of the parameters that quantify the physicochemical properties, and then determining the physicochemical parameters of the polymer based on the physicochemical parameters of the subgroups of each polymer. Alternatively, a molecular graph representation of the polymer can be used, where the atomically and bond-resolved physicochemical parameters are used as additional information at the vertices and edges of the molecular graph representation.

[0021] Preferably, the polymer physicochemical parameters refer to polymer descriptors that refer to at least one of the following: compositional, counting, structural fragment lists, fingerprints, graph invariants, 3D and / or higher dimensional descriptors that indicate parameters that quantify the physicochemical properties of the polymer. In a preferred embodiment, the polymer descriptor refers to a 3D descriptor, in particular a quantum chemical descriptor. Furthermore, the inventors have found that in particular the molar mass and / or the amount or type of chemical moieties are important for a very accurate description of the biodegradation of a polymer. It is therefore particularly preferred that the physicochemical parameters include the molar mass of the polymer. In general, the physicochemical parameters of the polymer can be derived from the physicochemical parameters of the subgroups, which can therefore also refer to the same descriptors as above. However, the physicochemical parameters can also be derived without the use of subgroups, for example by molecular simulation of the entire polymer. Below, the possible physicochemical parameters are defined in more detail. Also, in these cases, the defined physicochemical parameters can refer directly to the physicochemical parameters of the polymer, or, optionally, to the physicochemical parameters of the subgroups.

[0022] The compositional descriptors may refer to any of the potential, average molecular weight, polydispersity, charge, spin, boiling point, melting point, enthalpy of fusion, dissociation constant, Hansen parameters, protic, polar and dispersive contributions, Abraham parameters, retention index, total polar surface area, receptor binding constant, Michaelis-Menten constant, inhibitor constant, mutagenicity, LD50, bioconcentration, toxicity, biodegradation profile and viscosity.

[0023] The count descriptor can refer to any of the following: sum of atomic electronegativities, sum of atomic polarizabilities, amount of a component, weight fraction of a component, ratio of amount or weight fraction of a component, amount of a subgroup, weight fraction of a subgroup, ratio of amount or weight fraction of a subgroup, number of atoms and non-H atoms, number of specific atoms, particularly H, B, C, N, O, P, S, Si, Hal and heavy atoms, number of H donor and H acceptor atoms, number of bonds, non-H or multiple bonds, number of double, triple and aromatic bonds, number of functional groups or chemical motifs, weight fraction of functional groups or chemical motifs, ratio of functional groups or chemical motifs, ratio of functional groups or chemical motifs and components, sum or count of bond orders, aromatic ratio, number of rings or cycles, number of unpaired electrons, number of rotatable bonds, fraction of rotatable bonds, and number of conformers.

[0024] The polymer descriptor, which refers to a list of structural fragment descriptors, can refer to at least one of a list of molecular fractions, a list of functional groups, a list of bonds, and a list of atoms. The fingerprint descriptor preferably includes at least one of a MACCS key, preferably in bit format or count format, Morgan and other circular fingerprints, preferably in bit format or count format, topological twist, atom pair, infrared and related spectra, PubChem fingerprints, substructure fingerprints, and Klekota-Roth fingerprints. The graph invariant / topological indicators descriptor preferably includes at least one of a topostructural indicator and a topochemical indicator.

[0025] In a preferred embodiment, the polymer physicochemical parameters are 3D descriptors including at least one of the following: volume as a sum over atoms, average volume per atom, area as a sum over atoms, area as an average per atom, area over all atoms, area as an average per atom, solvent accessible surface, dispersion energy, dielectric energy, H donor, H acceptor, polar and non-polar surface area, atomically resolved H donor, H acceptor, polar and non-polar surface area, shape, sphericity, dipole and higher electric moments, polarizability, dielectric energy, proticity, polar and non-polar surface area, orbital energy and orbital gap, ionization energy, electron affinity, hardness, electronegativity, electrophilicity, excitation energy and intensity, infrared and ultraviolet absorption bands, reactivity measurements, redox potential, bond reference point, partial charge, charge surface area, atomic orbital contribution, bond order, atomic radius. In particular, the polymer physicochemical parameters preferably refer to 3D descriptors including at least one of the following: sum of volume over all atoms, average volume per atom, sum of area over all atoms, average area per atom, solvent accessible surface, dispersion energy, dielectric energy, H donor, H acceptor, polar and / or non-polar surface area, atomically resolved H donor, H acceptor, polar and / or non-polar surface area, shape, sphericity, cone angle, polarizability, dielectric energy, proticity, polar and / or non-polar surface area, excitation energy and intensity, infrared and / or UV absorption bands, reactivity measurements, particle charge and / or charge surface area. The preferred high-dimensional descriptors may include at least one of the following: conformational partition function, solubility, vapor pressure, activity coefficient, diffusion coefficient, partition coefficient, surface activity, rotational constant, moment of inertia, radius of gyration, composition drift of the polymer, density, viscosity, conformer-weighted volume and area, conformer-weighted H donor, H acceptor, protonity, polar and / or non-polar surface area, charge distribution, conformational dipole moment and molecular refraction. Preferably, higher-dimensional descriptors are used that include at least one of solubility, vapor pressure and activity coefficient, surface activity, conformer-weighted H donor, H acceptor, protonity, polar and non-polar surface area and charge distribution.

[0026] The method further includes providing a digital representation of application conditions for the target polymer, the digital representation of the application conditions indicating or associated with application parameters that define the application conditions. In particular, providing can refer to receiving the application conditions from a user's input, for example, using a respective input unit. Furthermore, providing can also refer to accessing a storage unit in which the application conditions are already stored. Furthermore, providing can also refer to pre-setting of the application conditions. For example, if the method is utilized in a very specific situation where only one specific application condition is sensible, the respective application condition can be pre-set and thus does not need to be provided as a specific input. Furthermore, providing can also include receiving application parameter values ​​of the application parameters directly from another source, for example, via a network connection, and providing the received application parameter values ​​of the application parameters as the application conditions.

[0027] In general, application conditions may refer to conditions that may exist during the life of a polymer, for example during the manufacture of a product comprising the polymer, during the intended application of a product comprising the polymer, during waste disposal of a product comprising the polymer, etc. Application conditions may be defined by respective application parameters that characterize the application conditions. For example, the application parameters may refer to the salt concentration of an aqueous solution, if the application conditions indicate that such an environment may exist during the life of the polymer, or may refer to the humidity of the air, for example for the application of an adhesive comprising the polymer. Again, indicating or being associated with an application parameter of an application condition is defined as making it possible to access information of the application parameter. For example, the digital representation may directly include the application parameters, for example in the form of values ​​for the respective quantities. However, the digital representation may also be a link to the respective application parameters, via which the application parameters can be accessed, or the digital representation may refer to an identifier that is associated with the application parameters and makes it possible to utilize a respective look-up storage to access the application parameters. Furthermore, the digital representation may also refer to information that makes it possible to derive the application parameters using one or more known relationships. For example, the geolocation of the environment may be utilized as a digital representation that makes it possible to derive the respective application parameters using knowledge about the respective geolocation. In a preferred embodiment, the digital representation is provided by presenting the user with input possibilities for each of the anticipated application environments, for example using a drop-down menu or other presentation that allows the user to select or otherwise indicate the anticipated application environment.Each provided application environment can be considered as a digital representation of the application conditions, from which the respective application conditions can be derived, for example via a lookup on storage where the associated application parameters are already stored, or can be derived based on respective known relationships.

[0028] In a preferred embodiment, if the application performance indicator refers to biodegradation, the application conditions consist of a biodegradation habitat which indicates as application parameters the habitat descriptor values ​​of the habitat descriptors, the habitat descriptors influencing the biodegradation of the polymer in the respective habitat. For example, the biodegradation habitat can indicate a general habitat, for example a wastewater habitat, in which case the respective habitat descriptor values ​​of the habitat descriptors of this habitat are already stored in the respective accessible storage. However, the provided biodegradation habitat can also directly include the respective habitat descriptor values ​​of the biodegradation habitat to provide further specification of the biodegradation habitat, for example a marine benthic organism. In this case, the habitat descriptors as application parameters indicate the environmental characteristics of the habitat. In particular, the environmental characteristics of the biodegradation habitat can influence the biological activity in the respective habitat, for example the presence, growth or absence of certain microorganisms. The environmental characteristics defined by the habitat descriptors therefore also indirectly influence the biodegradation of the polymer in the respective habitat. For example, if a polymer is biodegradable by a particular microorganism that requires a particular salt concentration, the polymer will biodegrade quickly in a habitat that provides such a salt concentration, such as a marine habitat, but much slower in a habitat that does not have the appropriate salt concentration, such as wastewater. Preferably, the biodegradation habitat is any one of a marine habitat, a wastewater habitat, a lake habitat, an anaerobic habitat, a compost habitat, or a soil habitat. In a preferred embodiment, the biodegradation habitat refers to a marine habitat, and the habitat descriptor refers to at least one of the following: salt concentration, sedimentation type, oxygen level, location, sample depth, water temperature, nutrient concentration, pH value, environment type, and microbial community. In a further preferred embodiment, the biodegradation habitat refers to a lake habitat, and the habitat descriptor refers to at least one of the following: salt concentration, sedimentation type, oxygen level, location, sample depth, water temperature, nutrient concentration, pH value, environment type, and microbial community.In a further preferred embodiment, the biodegradation habitat refers to wastewater, and the habitat descriptor refers to at least one of water temperature, microbial community, sludge concentration, nutrient concentration, pH value, test period, and enzyme environment. In a further preferred embodiment, the biodegradation habitat refers to soil, and the habitat descriptor refers to at least one of temperature, sand content, pH value, moisture content, nutrient concentration, microbial community, and enzyme environment. In a further preferred embodiment, the biodegradation habitat refers to compost, and the habitat descriptor refers to at least one of temperature, compost activity, pH value, moisture content, humidity, compost maturity, compost composition, compost origin, nutrient concentration, microbial community, and enzyme environment. In general, the habitat can also refer to the habitat of the standard test used to determine the biodegradability of the polymer. For example, standard tests such as those defined by ISO13432, ISO14852, ISO14855, ISO17556, and OECD301 also define the specific habitat in which biodegradation occurs. Thus, providing a biodegradation habitat can also include providing one of the standard tests, e.g., selecting via user input, where the habitat descriptor refers to the test, i.e., the test environment, and thus to a particular characteristic of the test habitat. Furthermore, the habitat can also be defined by the biodegradation of a reference polymer or other reference chemical. In this case, the habitat can be provided by providing the reference and its biodegradation. In this case, the reference and its biodegradation refer to the habitat descriptor.

[0029] In another preferred embodiment, the application performance indicator is the combustion behavior. In this case, the application conditions preferably refer to a standardized test procedure for the combustion behavior. For example, the combustion test conditions can be provided as application conditions, so as to allow the combustion test method and the combustion test parameters to be derived as application parameters. For example, the combustion test conditions can be provided in the form of an identifier of the combustion test method, and based on the identifier, a storage unit containing the combustion test method and the combustion test parameters can be accessed. The identifier can be, for example, the name or an identification number of the combustion test provided by the standard. However, the combustion test conditions can also be provided directly by the user, for example by means of an input unit, so that the combustion test method and the combustion test parameters are provided. For example, the user can be provided with a selection of known combustion test methods and can then select the respective combustion test method. In general, the combustion test method refers to a standardized test method for experimentally determining one or more combustion properties of a polymer. In general, a plurality of such standardized test methods for one or more combustion properties are known and can be referenced by their respective identifiers. Preferably, the flammability test method refers to one of the following: single flammability item, single flame source test, UL94, UL94 HB, ASTM E84, and cone calorimeter test. The test parameters further refer to a specific test procedure of the flammability test method, e.g., one or more process parameters that specify the flammability test method. For example, the test parameters can indicate a value of the distance of the heat source specified by the flammability test method from the polymer. Further examples refer to, for example, the time that the product formed by the polymer is subjected to the heat source specified by the flammability test method, the temperature profile of the heat source utilized in the flammability test method, the presence or absence and specification of materials between the heat source specified by the flammability test method and the product formed from the polymer, details of the cooling source present during the flammability test method, etc. In particular, the test parameters are specific to each flammability test method and thus depend on the definition of the flammability test method.However, even in standardized test procedures, the test parameters may be varied in some cases such that for the same fire test method, different test parameters may be specified within the range of possible test parameters defined by the fire test method.

[0030] In a further preferred embodiment, the application performance indicator is an adhesive property. In this case, the application conditions preferably refer to the application conditions. In particular, application parameters can be defined that define the application conditions or procedure for applying the adhesive. Preferably, the application parameters quantify at least one of humidity, one or more materials connected via the adhesive, temperature, contact pressure, application duration, and surface structure and / or texture.

[0031] In a further preferred embodiment, the application performance indicator is solubility and the application conditions refer to a fluid environment. In particular, the application parameters may define characteristics of the fluid environment. For example, the application parameters may quantify at least one of the following: type of fluid, concentration of a component such as salt, temperature, and pressure.

[0032] In a further preferred embodiment, the application performance indicators are formulation properties and the application conditions refer to the formulation. In particular, the application parameters can define characteristics of the chemical composition of the formulation. For example, the application parameters can quantify at least one of the following: type of solvent, concentration of components such as salts or co-solvents, temperature and pressure.

[0033] In a further preferred embodiment, the application performance index is a washing performance index and the application conditions refer to the washing environment. In particular, the application parameters may define the water hardness, the material to be removed, the material and / or structure of the objects to be washed, the washing additives, the washing process applied, the washing temperature and the washing pressure.

[0034] The method further includes providing a decision model based on the provided application conditions. In particular, providing a decision model preferably refers to selecting a decision model based on the provided application conditions. For example, a plurality of decision models may be stored in the storage, each decision model being trained for one application condition, in particular for different values ​​or value ranges of application parameters. Based on the provided application conditions indicating the application parameters, a respective appropriate decision model may be selected from the plurality of decision models. For example, if the indicated application parameters are within the range of the application parameters for which the decision model was trained, the decision model is appropriate. For example, a respective look-up table may be provided that allows an easy comparison between the indicated application parameter values ​​and / or value ranges for which the decision models stored in the storage were trained, so that the appropriate decision model can be directly selected. However, in another embodiment, providing a decision model based on the provided application conditions may also refer to a user selection of a decision model. For example, a user may be provided with a pre-selection of a decision model that refers to the provided application conditions, and then be able to select the respective decision model to be utilized. In general, a possible stored decision model refers to a decision model that is already parameterized based on a respective training data set for one or more application conditions. Since the training data set utilized to parameterize the decision model is historical data, as described in more detail below, the decision model may be trained, and thus generated, at any time prior to the determination of a particular application performance index for a particular polymer, and may be stored in the respective database after training. However, the training, and thus the generation of the decision model, may of course also be performed when it is determined that, for example, a particular decision model for a particular application is required.

[0035] The provided decision model is then adapted to determine the application performance index of the polymer with respect to the application conditions. In particular, the decision model is a data-driven model parameterized with respect to the application conditions so that the application performance index of the polymer can be determined based on the digital representation of the polymer, preferably based on the polymer characterization parameters that quantify the physicochemical properties of the polymer represented by the digital representation. The term "to" should be interpreted herein as the parameterization adapts the decision model to provide the application performance index with respect to the application conditions, e.g. when the polymer characterization parameters are provided as inputs, thus enabling the decision model to provide the application performance index. For example, the decision model associates the polymer characterization parameters of the historical digital representation of the synthetic specification and the historical digital representation of the application conditions with the respective application performance index. This allows the digital representation of the synthetic specification to be determined based on the target application performance index. The term "data-driven" is used herein to emphasize that the model is primarily based on the respective data inputs, and not based on, e.g., intuition, personal experience, or knowledge. Preferably, the decision model refers to a machine learning-based model based on known classification algorithms, such as neural networks, regression models, machine learning algorithms, etc. For most applications in this context, regression models based on, inter alia, linear regression, random forest, boosted tree, Lasso, ridge regression and MARS algorithms have proven to be suitable, while for classification models, inter alia, random forest, logistic regression and SVM algorithms have proven to be suitable. In general, the decision model is parameterized during a training process in which, for example, characterization parameters, preferably polymer descriptors derived from parameters quantifying the physicochemical properties of the polymer, are utilized together with the corresponding application performance indicators for the specific application conditions.Based on such a training data set that is specific to the application conditions, e.g., to a particular application parameter value range and / or value, each parameter of the data-driven model can be determined using known training methods such that the decision model can also determine application performance indicators for polymers that are not part of the training data set.

[0036] Furthermore, in a preferred embodiment, the decision model may also be adapted to determine the application performance index for the polymer further based on the application parameters as input. In particular, the decision model may be trained, for example, by utilizing a training data set comprising the polymer characterization parameters of the polymer and the associated application performance index for the specific application conditions, as described above, resulting in a decision model that indirectly takes the specific application conditions into account. However, the training data set may also optionally comprise specific application parameter values ​​for the respective application conditions. In this case, the decision model may be trained such that, in addition to the polymer characterization parameters, the application parameter values ​​may also be provided as input, and then the decision model determines the application performance index further based on the application parameter values. This has the effect that the application performance index may be determined more accurately, especially in cases where the application performance index strongly depends on the application parameter values ​​of the application conditions. For example, biodegradability in marine habitats may deviate significantly for different regions of the world having different temperatures or nutrient concentrations. Therefore, in such cases, it may be advantageous to directly provide the application parameter values ​​as input to the decision model. However, instead of providing the application parameter values ​​as input to the decision model, it is also possible to train two different decision models and indirectly treat the different regions as different application conditions.

[0037] The method further includes determining an application performance index for the polymer based on the provided decision model and the digital representation of the polymer. In particular, if the digital representation of the polymer includes polymer characterizing parameters, the polymer characterizing parameters are provided as input to the decision model, and the decision model then provides the determined application performance index as output. If the digital representation does not directly include the polymer characterizing parameters, determining the application performance index may also include first determining the polymer characterizing parameters, e.g., as described above. The polymer characterizing parameters so determined may then be provided as input to the decision model. However, the decision model may also be trained, e.g., to directly utilize the digital representation as input, in which case the decision model may determine the characterizing parameters as part of the process for determining the application performance index.

[0038] The determined application performance indicators may then be provided for further processing, for example to an output unit or a computing unit. Preferably, providing the application performance indicators leads to further processing utilizing the determined application performance indicators. In such a case, providing as a separate step may be omitted and replaced by processing of the determined application performance indicators.

[0039] The determination of the application performance indicators using the decision model can be considered as a virtual measurement of the application performance indicators. In particular, the decision model is based on measurement data, e.g., the measured application performance indicators of the polymer used to train the decision model. Thus, the decision model includes information provided by these previous measurements. Furthermore, the characterization parameters may also refer to the measured properties of the polymer in some cases. Thus, the determined application performance indicators of the new polymer determined using the decision model can also be considered as being at least partially based on the measurement results.

[0040] Preferably, processing the application performance indicators includes determining a control signal for controlling a production process based on the determined estimated application performance indicators. The production process may refer to a manufacturing process of the polymer, or may refer to any other process, for example a waste treatment process in which the polymer is utilized or treated. In a preferred embodiment, the control signal indicates a machine executable synthetic specification of the polymer, especially when the comparison indicates that the determined application performance indicator of the polymer is within a predefined range around the provided target application performance indicator.

[0041] Furthermore, the process of processing the application performance indicators may also refer to selecting one or more polymers based on the respective determined application performance indicators. For example, if the respective application performance indicators have been determined for a plurality of potential polymers, the selecting may include comparing the application performance indicators of the different polymers to predefined selection criteria, and selecting the polymer whose determined application performance indicators satisfy these criteria. In particular, in one embodiment, the method includes receiving target application performance indicators for the polymers, comparing the received target application performance indicators to the determined application performance indicators, and providing a control signal in response to the comparison.

[0042] The control signal may refer to any signal that allows further control of the technical system. For example, the control signal may be adapted to control an interface to provide the result of the comparison on the interface. In a preferred embodiment, the comparison refers to a validation of the target application performance indicator, the validation being positive if the determined application performance indicator falls within a predefined range around the target application performance indicator. In this case, the control signal may be adapted to simply control a user interface to provide an indication of a positive or negative validation result. However, preferably, the control signal refers to a recipe, i.e. a synthesis specification, of one or more polymers that meets the specified target application performance indicator, i.e. is positively validated. A recipe, i.e. a synthesis specification, is generally defined as an instruction on how a polymer can be synthesized. In particular, a recipe comprises starting materials and respective parameters for polymerization from the starting materials. Preferably, the control signal comprises a recipe in a form that directly allows an automatic control of a respective industrial system or work facility for producing the polymer. In particular, if the result of the comparison refers to the determined application performance indicator being within a predefined range around the target application performance indicator, the control signal preferably indicates a machine executable synthesis specification of the polymer.

[0043] In a preferred embodiment, the method further comprises providing a synthesis specification as a digital representation of the polymer and determining characterization parameters from the synthesis specification, e.g. in the form of polymer physicochemical parameters, preferably as polymer descriptors. In particular, the synthesis specification, i.e. recipe, comprises information on the polymer synthesis of the polymer, e.g. starting materials and the process by which the respective starting materials are covalently linked to form polymer chains or polymer networks. The method then comprises determining characterization parameters, e.g. polymer descriptors, from the synthesis specification. Optionally, subgroups can be determined from the synthesis specification and then polymer characterization parameters can be determined based on the subgroup characterization parameters of the subgroups, e.g. from a database or by utilizing known characterization parameter determination algorithms. In a preferred embodiment, further from the synthesis specification, the catalyst and / or non-reactive process components to be utilized are determined. In this case, this information is preferably also utilized by the decision model for determining the application performance index together with the characterization parameters, in particular the polymer physicochemical parameters. Preferably, characterization parameters for the catalyst and / or non-reactive process components are also determined, the respective characterization parameters also being used to determine the characterization parameters of the polymer. Preferably, the characterization parameters of the catalyst and / or non-reactive process components, referring to the amount of each component, e.g., molar mass, molar percentage, etc., are considered to determine the polymer characterization parameters of the polymer.

[0044] In a preferred embodiment, determining the polymer characterizing parameters from the synthesis specification includes identifying the type and amount of subgroups based on the synthesis specification, for example as characterizing parameters of the subgroups, and determining the polymer characterizing parameters based on the identified type and amount of the subgroups. In general, the type of subgroups can refer to a given type or class associated with a specific physicochemical property, i.e. physicochemical parameter, of the subgroups and thus of the polymer comprising these subgroups. However, the general physicochemical properties of the polymer, and thus the polymer characterizing parameters, can also depend on the amount of the subgroups present in the polymer, so that amount can also be taken into account. In a preferred embodiment, the determination of the type and amount of the subgroups takes into account information provided by the synthesis specification indicating the type of polymerization. Information on the type of polymerization that can be utilized can refer, for example, to whether the polymerization refers to polycondensation, polyaddition, radical polymerization, cationic polymerization, anionic polymerization, or coordination chain polymerization. Preferably, for each type of polymerization, rules are predefined that can be applied to determine the subgroups of the polymer. For example, rules can be predefined that determine which functional groups of the monomers in the synthesis specification react with which functional groups of the synthesized polymer in what priority order. The rules can be based, for example, on kinetic considerations. Based on the number and type of polymerized functional groups, subgroups can be determined and the number and type of subgroups can be calculated.

[0045] In one embodiment, determining the amount of the subgroups includes determining the amount of at least one of amide, ester, thioester, carbonate, ether, amine, urea, urethane, thiourethane, isocyanurate, biuret, allophanate, acetal, mica adduct, radical polymerized double bond, siloxane, silane, silazane, phosphazene groups, and residual amine, aldehyde, ketone, epoxide, aziridine, isocyanate, alcohol, thiol, carboxylic acid, acyl halide, α,β-unsaturated carbonyl group, α,β-unsaturated carboxyl, and double bond group in the polymer based on a synthesis specification.

[0046] In one embodiment, the digital representation of the application conditions refers to a representation of a predetermined criterion and / or test procedure for the application performance index, and the application parameters define the predetermined criterion and / or test procedure. Generally, there are several standardized test methods and / or criteria for testing the respective application performance index of a chemical. For example, such test methods can be found in DIN or ISO standards. Furthermore, by providing an application performance index test method and providing a decision model trained on the basis of the provided application performance index test method, it is possible to determine, for example, an application performance index that is easily comparable with the application performance index measured respectively using the respective test method. In particular, with respect to this embodiment, it is preferred that the decision model is trained on the basis of a data set in which the test method and / or norm for which the application performance index was determined is clearly specified, whereby the decision model can be specifically trained for one or more test methods and / or norms. For example, the predetermined criterion and / or test procedure can be a biodegradation test method, and the provided biodegradation test method indicates a standardized biodegradation test method for experimentally determining the biodegradation of a chemical. In another embodiment, the predetermined criterion and / or test procedure can be a combustion test method. Preferably, the flammability test method refers to at least one of, but is not limited to, single flammability item (EN 13823:2015), single flame source test (EN ISO 11925-2:2020), UL94 and UL94 HB (EN 60695-11-10:2014 and EN 60695-11-20:2016, respectively), ASTM E84 test, and cone calorimeter (ISO 5660-1:2015).

[0047] In one embodiment, a target application of the polymer is provided with further reference to the intended application of the polymer, and application conditions are provided based on the target application. The target application of the polymer can refer to the intended application situation of the polymer, for example, if it is intended to utilize the polymer as a coating, in personal care products, in cleaning detergents, in lubricants, in agriculture, or in product packaging. Such a target application indicates a particular application condition. For example, if the polymer biodegrades in compost, which provides the respective habitat as application condition for product packaging, it may be interesting to determine biodegradation as application performance indicator. In this regard, a predefined list can be provided on a storage, on which are stored the respective target applications, the corresponding application conditions, and the associated application performance indicators. Then, a target application for the polymer can be provided, for example, by providing a list of target applications to a user and allowing the user to select the respective target application, each target application being connected to one or more application conditions. Then, for each of the application conditions to which the target application is connected, an application performance indicator can be determined, or again, the user can select the respective application condition to which the target application is connected. Additionally or alternatively, information indicative of the intended processing of the polymer can be provided. For example, the end-of-life treatment may indicate whether the polymer is intended to biodegrade in a particular environment or should be subjected to a particular treatment, for example in a bioreactor. Thus, the intended treatment information may also be utilized to determine application conditions for the polymer, as described above.

[0048] In one embodiment, further information is provided indicative of the accessible surface area of ​​the polymer in its intended form, and the decision model is further trained to determine the application performance index based on the accessible surface area, and the method further comprises determining the application performance index further based on the accessible surface area. For example, the information can indicate whether the intended product is provided in a solid, crushed, foamed, pelletized, or any other form. Preferably, the information indicates the surface area of ​​the product per mass, or the geometric shape of the smallest independent part of the product. Generally, the application performance index of a polymer is an intrinsic characteristic of the polymer, but the exact characteristics of the application performance index of a product comprising the polymer may also depend on the surface area that can be accessed by, for example, microbial components for biodegradation or for dissolving the polymer. Thus, further determining the application performance index based on the surface area of ​​the product comprising the polymer allows for increased accuracy in predicting the application performance index of the final product.

[0049] In one embodiment, the polymer characterizing parameters indicated by the digital representation of the polymer refer to at least one of recipe parameters from the polymer synthesis, compositional descriptors, counting descriptors, lists of structural fragments, fingerprints, graph invariants, 3D descriptors and / or higher dimensional descriptors indicative of the chemical nature of the polymer. The respective connections of the digital representation with, for example, previously calculated polymer descriptors or further information about the polymer may already be stored and connected with the respective digital representation. For example, if the digital representation refers to a brand name, the respective structural formulas, subgroups and / or parameters characterizing the subgroups or polymer characterizing parameters corresponding to the brand name may already be stored, for example, in the storage of the owner of the brand name.

[0050] In one embodiment, the application conditions refer to any one of a biodegradation habitat, a product application environment, and a fluid environment as a standardized test procedure for burn behavior, and the application performance indicators refer to any one of biodegradability, burn behavior, adhesion, and solubility, respectively. Examples of each of this embodiment are provided above in the context of the previously described embodiments.

[0051] In a further aspect, there is provided a computer implemented method for determining a target synthesis specification indicative of a target polymer including a target indicator for application performance, the method comprising: a) providing a target indicator for a target application performance of the polymer; b) providing a digital representation of a potential target synthesis specification indicative of or associated with a potential target polymer; c) providing a digital representation of application conditions, the digital representation indicative of application parameters defining the application conditions; and d) providing a decision model based on the provided digital representation of the application conditions, the decision model adapted to determine application performance indicators of the polymer with respect to the application conditions, the decision model comprising: a data-driven model parameterized with respect to application conditions such that application performance indicators for the polymer can be determined based on the digital representation; e) determining an application performance indicator for the potential target polymer based on the provided determination model and the digital representation of the potential target synthetic specifications; and f) comparing the determined application performance indicator for the potential target polymer with the target indicators and based on the comparison, i) determine the potential target polymer as the target polymer and the potential target synthetic specifications as the target synthetic specifications, or ii) provide a new potential target synthetic specification for a new potential target polymer and repeat the determination of the application performance indicator utilizing the new potential target synthetic specification for the new potential target polymer.

[0052] In general, the method also refers to a computer-implemented method, and therefore can be implemented by a general-purpose or dedicated computer adapted to implement the method, for example, by executing the respective computer program. The method is adapted to determine, in particular predict, a target synthesis specification that indicates a target polymer that includes a target indicator. In general, the synthesis specification includes instructions on how a particular associative polymer can be produced. For example, the synthesis specification can refer to the starting products and manufacturing conditions that, if applicable, lead to the synthesis of the polymer in a manufacturing process. Thus, the synthesis specification is always associated with the polymer that is produced when the synthesis specification is carried out, for example, using suitable laboratory or industrial equipment. In particular, the synthesis specification can also be considered as a recipe for how to produce the associated polymer. Since the target synthesis specification and the target polymer correspond to each other, i.e., the target synthesis specification, when executed accordingly, produces the target polymer, in the following, both terms can be utilized simultaneously, for example, when the target synthesis specification is determined, the target polymer is also determined, and vice versa.

[0053] In general, the method can be seen as utilizing the method for determining an application performance indicator as described above for steps b) to e) and providing an additional step for providing and comparing a target indicator. In particular, the same decision model can be utilized for determining the application performance indicator. Thus, all definitions provided for the method for determining the application performance indicator can also be applied to the method.

[0054] In a first step, the method includes providing a target indicator indicative of an application performance indicator of the polymer. In particular, providing can refer to, for example, receiving the target indicator from a user's input applying the respective input unit. Furthermore, providing can also refer to accessing a storage unit in which the target indicator is already stored. Furthermore, providing can also include, for example, receiving the target indicator via a network connection from another source and providing the received application performance indicator. In general, the target indicator can refer to one target value, for example, a target half-life of the polymer in a particular biodegradation habitat, or can refer to a range of values ​​to be met by the polymer. Furthermore, the target indicator can also refer to any kind of target function, for example, a timed sequence of application performance indicators. For example, the target indicator can indicate that the target polymer has a first application performance indicator value range during a first time range, and then a second target application performance indicator value range during a subsequent time range.

[0055] The potential target synthesis specifications can be provided as described above with respect to the method of determining the application performance index using the digital representation. In general, the potential target synthesis specifications can be considered as starting synthesis specifications that indicate for which polymers or in which regions of the potential polymer space the application performance index should be determined first in the search for synthesis specifications that will result in a polymer containing the target index. In general, the potential target synthesis specifications can be provided by a user or automatically, for example according to a predefined rule, or arbitrarily. For example, the user can select a promising potential target synthesis specification as a starting point. However, any target synthesis specification may be utilized, or a set of rules may be utilized to provide the potential target synthesis specifications without user intervention. Preferably, the potential target synthesis specifications are provided based on rules that take into account constraints on the potential target synthesis specification space, i.e. the target polymer space. The target polymer space can be defined by a range of amounts of components or a combination or ratio of components. Furthermore, the target polymer space can be defined by a range of process conditions or a combination or ratio of process conditions. Furthermore, the target polymer space can be defined by a synthesis feasibility index that can be a rule or experience based on or provided by a data-driven model.

[0056] After determining the application performance indicators, the determined application performance indicators of the potential target polymer are compared with the target indicators. Based on this comparison, it is determined whether the potential target polymer is determined as a target polymer and whether the potential target synthesis specification is determined as a target synthesis specification, in which case the iterations may be stopped at this point. Furthermore, based on the comparison, it may be determined to provide a new potential target synthesis specification for a new potential target polymer and to repeat the determination of the application performance indicators utilizing the new potential target synthesis specification for the new potential target polymer. Thus, at this point, iterations are performed in which the determination of the application performance indicators using the determination model and the polymer characterization parameters of the potential target polymers is repeated until one of the potential target polymers is determined as a target polymer. In particular, the comparison may include determining whether the determined application performance indicators of the potential target polymer are within a predetermined range around the target indicator, in which case the target may be considered to be met and the potential target polymer is determined as a target polymer. If the determined application performance indicator is outside the predetermined range around the target indicator, the target is determined to be not met and a new potential target synthesis specification for a new potential target polymer that may meet the target indicator is provided.

[0057] In general, the iteration carried out can refer to any search or directed search of the potential target polymer space. For example, new potential target synthesis specifications or new potential target polymers can be simply selected arbitrarily from a large number of potential target polymers generated in silico. However, specific rules for generating new potential target polymers, and therefore new potential target synthesis specifications, can also be applied based on a comparison between the determined application performance indicators and target indicators, with or without consideration of simultaneous optimization of further target properties of the polymer. In general, known methods for generating new target polymers can be utilized, for example, evolutionary algorithms or Bayesian optimization programs can be used.

[0058] Then, iterations can be performed by utilizing the application conditions and polymer characterization parameters of the new potential target polymer as described above over the step of determining the application performance index of the new potential target polymer. Optionally, the determination of the polymer characterization parameters from the digital description of the new potential target synthesis specifications can also be part of the iterations if the digital description of the new potential target synthesis specifications is not yet provided for the polymer characterization parameters. Furthermore, it is preferred that the same determination model is used in all iteration steps for determining the application performance index. However, in some cases, different determination models can be used in different iteration steps. For example, if other polymer physicochemical parameters for the new potential target polymer are utilized, another determination model may also be more appropriate.

[0059] After the iterations have stopped, for example after a potential target polymer has been determined as the target polymer, or if a new potential target polymer cannot be selected or generated, the results of the iterations can be provided to the user. For example, if none of the possible potential target polymers satisfied the target indicators, the user can be notified of the failure to determine the target polymer. If the target polymer can be determined, the target polymer can be provided to the user as output. For example, the determined target polymer and the target synthesis specification can then be provided to an output unit or a calculation unit for further processing. Preferably, the provision of the target synthesis specification and the target polymer results in further processing utilizing the target synthesis specification.

[0060] Preferably, processing the target synthesis specification comprises determining control signals for controlling a production process based on the determined target synthesis specification. Preferably, the production process refers to a production process of the target polymer utilizing the target synthesis specification. Furthermore, it is preferred that the target specification refers to a machine executable synthesis specification of the target polymer, such that the control signals can directly reference the control of the respective laboratory or process equipment enabling the execution of the synthesis specification to produce the polymer. In one embodiment, providing the target synthesis specification of the target polymer comprises providing control signals adapted to control an industrial plant to produce the target polymer according to the target synthesis specification.

[0061] In one embodiment, a target technical application property of the target polymer is provided, and a potential target synthesis specification is provided based on the provided target technical application property, such that the potential target polymer meets the provided target technical application property. In particular, the technical application property can refer to any property of a polymer and / or a substance at least partially composed of a polymer that allows evaluating the technical applicability of the respective polymer provided after its synthesis. Preferably, the technical application property includes at least one of mechanical properties, optical properties, physicochemical properties, chemical properties, and biological properties. In general, the mechanical properties can refer to any of adhesion, tensile strength, stiffness, hardness, shrinkage, elongation, tear, tear strength, rebound, compressibility, abrasion, leakage, morphology, tactile properties, stress at break, elongation at break, particle size distribution, and packing degree. The optical properties can generally include any of coloration, turbidity, opacity, gloss, reflection, appearance, absorption, scattering, color intensity, cloud point, matteness, optical density, spectrum, and refractive index. Further, the physicochemical properties can refer to any of density, viscosity, K value, molar weight, dispersity, molar mass distribution, particle size distribution, solubility, partition coefficient, interfacial properties, surface tension, dispersibility, storage stability, odor, segregation, coagulation, electrical conductivity, capacitance, surface area, flow time, vapor pressure, VOC, solids content, hygroscopicity, magnetism, miscibility, thixotropy, phase transition properties, glass transition temperature, corrosion inhibition, solvent separation, aggregation, self-heating ability, impact sensitivity, loss on drying, response angle, electrostatic charge, minimum film formation temperature, and charge density. The chemical properties can refer to any of the following: functional group number, atom type number, functional group concentration, atom type concentration, chemical resistance, reaction timing, demold time, growth, hard / soft segment content, crystallinity, reaction temperature, reaction pressure, decomposition, thermal decomposition, photolysis, acidity, pK a, pH, moisture / water content, flammability, burning rate, autoignition, flash point, flammable gas formation, fire susceptibility, deflagration rate, residual monomer count, by-product formation, degree of polymerization, salt content, temperature resistance, oxidation properties, reduction properties, reactivity, ash content, non-volatile matter content, stability, chelating capacity, calorific value, saponification value. Furthermore, the biological properties can include any of biodegradability, biological resistance, toxicity, biotransformation, ecotoxicology, sensitization, bacterial count, enzyme activity, environmental distribution, bioaccumulation, biological exposure. In a preferred embodiment, the technical application property can further refer to biodegradability, e.g., biodegradability in a particular habitat.

[0062] Then, potential target synthesis specifications are provided so that the relevant potential target polymers meet the technical application properties of the provided targets.For example, a database can be utilized in which polymers and corresponding technical application properties are already stored, and a target polymer and associated synthesis specifications that meet the technical application properties of the provided targets can be selected from the database.In general, the polymers that meet the technical application properties of the targets can be considered to form a potential target polymer space that can be explored during an iterative process to find a target polymer.Then, from the selected target polymers that meet the technical application properties of the targets, a first potential target polymer, and therefore a first potential target synthesis specification, can be selected.

[0063] In one embodiment, providing a new potential target synthesis specification is based on modifying the provided target application properties and providing a new potential target synthesis specification, such that the potential target polymer meets the modified target application properties. In particular, if a new potential target synthesis specification must be provided, a comparison of the determined application performance indicators with the target indicators indicates that the determined application performance indicators of the current potential target polymer do not meet the target indicators. In such a case, a new potential target synthesis specification, and thus a new potential target polymer, can be provided such that the new potential target polymer still meets the target technical application properties, if such a respective polymer exists. However, in many cases, it is not possible to provide such a new potential target polymer, or it may not be technically prudent to provide such a new potential target polymer that still meets the target technical application properties. In these cases, it is advantageous to utilize a less strict target technical application property, such as modifying the target technical application property to refer to a range of values ​​instead of one specific value, or to refer to a wider range of values ​​if it refers to a range of values. A new potential target polymer can then be selected or generated such that it meets the modified target application properties.

[0064] In one embodiment, providing potential target synthesis specifications based on the provided target technical application properties includes utilizing an application property model adapted to determine technical application properties of the polymer based on a digital representation of the polymer, the application property model being a data-driven model parameterized to determine technical application properties associated with the polymer based on a digital representation of the polymer including polymer characterization parameters of the polymer. The application property model can refer to any known data-driven model that allows for determining technical application properties based on a digital representation of the polymer including polymer characterization parameters. In general, the application property model preferably follows the same principles as described above with respect to the decision model. In fact, the application property model can be based on or utilize the same machine learning algorithms and training methods, only utilizing different training data, i.e., training data including the other respective technical application properties of the polymer instead of the application performance indicators. Thus, all the embodiments described above with respect to the decision model can also be realized with respect to the application property model for determining technical application properties. Utilizing such a model has the advantage that iterations can be performed in a fast and computationally inexpensive manner not only on the application performance indicators of the polymer, but also on one or more further technical application properties, resulting in a target polymer that satisfies not only the target indicators but also one or more further target technical application properties.

[0065] In a further aspect, an interface method for providing an interface is presented, the interface method comprising: i) receiving as input a digital representation and application conditions via a user interface and providing the received digital representation and application conditions to a processor executing the above-mentioned method; and ii) providing as result a determined application performance indicator or a determined composition specification of the polymer, respectively, to a user via the user interface, the result being received from the processor executing the above-mentioned method.

[0066] In a further aspect, a computer-implemented training method for training a data-driven decision model for parameterizing a decision model is presented, the training method comprising: i) providing training data associated with predetermined application conditions, the training data including: a) digital representations of the training polymers indicating characterization parameters for each of a plurality of training polymers; and b) application performance indicators for each application condition associated with each training polymer; ii) providing a trainable decision model on a data-driven basis; iii) training the provided data-driven decision model based on the provided training data, the trained decision model being adapted to determine application performance indicators of polymers based on the digital representations of the polymers, in particular based on the characterization parameters, preferably polymer physicochemical parameters, more preferably polymer descriptors, indicated by the digital representations; and iv) providing the trained decision model.

[0067] In a further aspect, an apparatus for determining an application performance indicator of a target polymer is presented, the apparatus comprising: a) a digital representation providing unit for providing a digital representation of the target polymer indicative of or associated with characterization parameters of the polymer; b) an application condition providing unit for providing a digital representation of application conditions for the target polymer, the digital representation of the application conditions indicative of or associated with application parameters defining the application conditions; c) a model providing unit for providing a decision model based on the provided digital representation of the application conditions, the decision model being adapted to determine an application performance indicator of the polymer with respect to the application conditions, the decision model being a data-driven model parameterized with respect to the application conditions such that the application performance indicator of the polymer can be determined based on the digital representation of the polymer; and d) a determination unit for determining the application performance indicator of the target polymer based on the provided decision model and the digital representation of the target polymer.

[0068] In a further aspect, an apparatus for determining a target synthetic specification indicative of a target polymer including a target indicator for application performance, comprising: a) a target providing unit for providing a target indicator for a target application performance of the polymer; b) a digital representation providing unit for providing a digital representation of a potential target synthetic specification indicative of or associated with a potential target polymer; c) an application condition providing unit for providing a digital representation of an application condition, the digital representation being indicative of application parameters defining the application condition; and d) a model providing unit for providing a decision model based on the provided digital representation of the application condition, the decision model being adapted to determine an application performance indicator of the polymer with respect to the application condition, An apparatus is presented that includes a model providing unit, where the model is a data-driven model parameterized with respect to application conditions such that application performance indicators for the polymer can be determined based on the digital representation of the polymer; e) a determining unit for determining application performance indicators for the potential target polymer based on the provided determining model and the digital representation of the potential target synthesis specifications; and f) a comparing unit for comparing the determined application performance indicators for the potential target polymer with the target indicators and based on the comparison, either i) determine the potential target polymer as the target polymer and the potential target synthesis specifications as the target synthesis specifications, or ii) provide a new potential target synthesis specification for a new potential target polymer and repeat the determination of the application performance indicators utilizing the new potential target synthesis specification for the new potential target polymer.

[0069] In a further aspect, an interface device for providing an interface is presented, the interface device comprising: i) an input interface unit for receiving a digital representation and application conditions as input via a user interface and providing the received digital representation and application conditions to at least one of the above-mentioned devices; and ii) a result interface for providing the determined application performance indicators of the polymer as a result to a user via the user interface, the result being received from the above-mentioned devices.

[0070] In a further aspect, a training apparatus for training a data-driven decision model for parameterizing a decision model is presented, the training apparatus comprising: i) a training data providing unit for providing training data associated with a predetermined application condition, the training data including: a) a digital representation of a plurality of training polymers indicating characterization parameters for each of the training polymers; and b) an application performance indicator for each application condition associated with each training polymer; ii) a trainable model providing unit for providing a data-driven based trainable decision model; iii) a training unit for training the provided data-driven decision model based on the provided training data, such that the trained decision model is adapted to determine application performance indicators of a polymer based on the digital representation of the polymer, in particular based on the characterization parameters, preferably polymer physicochemical parameters indicated by the digital representation; and iii) a training model providing unit for providing the trained decision model.

[0071] In a further aspect of the present invention, the use of at least one of the above methods and devices is provided, the method and / or the device being used to determine an application performance index for a given polymer or to determine target synthesis specifications for any of the following: i) polymers referring in particular to polyesters used in multi-film and packaging applications, such as aromatic-aliphatic copolyesters, ii) polymers referring in particular to polyalkoxylates used in particular to home care and personal care applications, iii) polymers referring to polyurethane dispersions, iv) polymers used in fragrance applications, v) polymers used in paper coatings for packaging applications based on multi-layer blends, and vi) polymers referring to polyurethanes used in adhesives.

[0072] In a further aspect of the invention, a system is provided comprising: i) a control signal comprising a polymer synthesis specification indicating one or more components for producing a polymer, the control signal being generated according to any of the methods described above; and ii) one or more components indicated by the synthesis specification in the control signal.

[0073] In a further aspect of the invention, there is provided the use of control signals generated according to the above-described method for controlling manufacturing processes, particularly manufacturing processes involving the production of polymers.

[0074] In a further aspect of the invention, a control signal is provided, the control signal being generated according to the method described above. Preferably, the control signal comprises a machine executable synthesis specification for generating a target polymer.

[0075] In a further aspect, a computer program product for determining an application performance index for a given polymer is presented, the computer program product comprising program code means for causing an apparatus as described above to perform the method as described above.

[0076] In a further aspect, a computer program product for determining target synthesis specifications is presented, the computer program product comprising program code means for causing an apparatus as described above to perform the method as described above.

[0077] In a further aspect, a computer program product for training a decision model is presented, the computer program product comprising program code means for causing a training device as described above to execute the training method as described above.

[0078] It is to be understood that the above-mentioned method, the above-mentioned device and the above-mentioned computer program product have similar and / or identical preferred embodiments, in particular as defined in the dependent claims.Furthermore, the above-mentioned training method, the above-mentioned training device and the above-mentioned training computer program product also have similar and / or identical preferred embodiments, in particular as defined in the dependent claims.

[0079] It is to be understood that a preferred embodiment of the invention can also be any combination of the dependent claims or the above embodiments with the respective independent claim.

[0080] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter. [Brief description of the drawings]

[0081] [Figure 1] FIG. 1 illustrates a schematic, exemplary embodiment of a system including an apparatus for determining target synthesis specifications indicative of a target polymer including a target application performance index. [Diagram 2] 1 shows, generally and exemplarily, a flow chart of a method for determining target synthesis specifications indicative of a target polymer including a target application performance index. [Diagram 3] 1 shows, generally and exemplarily, a flow chart of a method for training a decision model for determining application performance indicators of a polymer. [Figure 4]1 shows, by way of example only, a flow chart of a preferred and more detailed embodiment of a method for determining a target synthesis specification indicative of a target polymer comprising a target indicator. [Diagram 5] 1 shows, by way of example only, a flow chart of a preferred and more detailed embodiment of a method for determining a target synthesis specification indicative of a target polymer comprising a target indicator. [Figure 6] 1 illustrates, in a schematic and exemplary manner, an optional extension of the method for determining application performance indicators of a polymer. [Figure 7] 1 shows, in a schematic and exemplary manner, a block diagram of a system architecture of a system and apparatus for determining application performance indicators of a polymer. [Figure 8] 1 shows, in a schematic and exemplary manner, a block diagram of a system architecture of a system and apparatus for determining application performance indicators of a polymer. [Figure 9] 1 shows, in a schematic and exemplary manner, a block diagram of a system architecture of a system and apparatus for determining application performance indicators of a polymer. [Figure 10] 1 shows schematic and exemplary output and input screens of an exemplary user interface; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0082] Detailed Description of the Embodiments 1 shows an embodiment of a system 100 including a device 110 for determining a target indicator, e.g., a target synthesis specification indicative of a target polymer with a target biodegradability. Furthermore, the system 100 includes a training device 130 for training a biodegradation model used in the device 110, a database 140 capable of storing the results of the determination of the target synthesis specification, and a production system 120 for producing a product including the determined target polymer that can be controlled using the determined target synthesis specification. In general, the device 110 for determining a target synthesis specification can be considered to include a device for determining an application performance indicator, which includes some of the logical units and functions of the device 110.

[0083] The apparatus 110 comprises a target indicator providing unit 111, a digital representation providing unit 112, an application condition providing unit 113, a model providing unit 114, a determining unit 115, an iterative control unit 116, and optionally an output and / or control unit 117 that may be adapted to output the determined target synthesis specification and / or to provide a control signal for controlling a production process of the production system 120 based on the determined synthesis specification. In particular, the digital representation providing unit 112, the application condition providing unit 113, the model providing unit 114, and the determining unit 115 can be considered to form an apparatus for determining application performance indicators, which is part of an apparatus for determining a target synthesis specification indicative of a target polymer comprising the target indicators.

[0084] The target indicator providing unit 111 is adapted to provide a target indicator indicative of a desired application performance indicator of the polymer. The target indicator providing unit 111 can refer to, for example, an input unit through which a user can input the respective target indicator. Furthermore, the target indicator providing unit 111 can refer to or be part of a user interface that allows a user to interact with the device 110 for providing the target indicator. However, the target indicator providing unit 111 can also refer to or be communicatively coupled to, for example, a storage unit in which target indicators for a particular application are already stored.

[0085] The digital representation providing unit 112 is adapted to provide a digital representation indicative of a potential target synthesis specification indicative of polymer characterizing parameters of the potential target polymer. The digital representation providing unit 112 may, for example, refer to an input unit through which a user can input the respective digital representation. Furthermore, the digital representation providing unit 112 may refer to or be part of a user interface that allows a user to interact with the device 110 and / or the database 140. However, the digital representation providing unit 112 may also refer to or be communicatively coupled to a storage unit in which the digital representations of the polymers are already stored. In general, the digital representation may directly include characterization parameters, for example polymer physicochemical parameters indicative of parameters quantifying physicochemical properties of the respective polymer. However, instead of directly providing the polymer characterizing parameters, it is also possible to provide only the synthesis specification of the polymer. In this case, the digital representation providing unit 112 is preferably further adapted to determine the polymer characterizing parameters from the synthesis specification. In particular, the digital representation providing unit 112 is preferably adapted to identify, from the potential target synthesis specification, the type and amount of a subgroup of polymers and to determine the polymer characterizing parameters based on the identified type and amount of the subgroup. In particular, the digital representation providing unit 112 may be adapted to determine a respective subgroup characterizing parameter for each identified subgroup, for example by accessing a database in which respective characterizing parameters for a plurality of most relevant subgroups are stored. The characterizing parameters of the polymer may then be determined based on the subgroup characterizing parameters of the subgroups, preferably also based on the determined amount and type of the subgroups, for example by a weighted average of the subgroup characterizing parameters of the subgroups. The digital representation providing unit 112 is then adapted to provide a digital representation comprising the polymer characterizing parameters, for example to the determining unit 115.

[0086] The application condition providing unit 113 is adapted to provide the application conditions. The application condition providing unit 113 can, for example, refer to an input unit through which a user can input the respective application conditions. For example, a user interface can be provided that allows the user to select from several predefined application conditions. In a preferred embodiment, the application condition providing unit 113 can be communicatively coupled to or refer to a user interface that allows the user to indicate an intended application for the polymer, and the application condition providing unit can then be adapted to provide the application conditions based on the intended application. For example, if the application indicates use in a cleaning detergent, the application conditions can refer to a habitat in which the polymer is intended to biodegrade, for example wastewater. The application condition providing unit can then be adapted to determine the respective wastewater biodegradation habitat as the application condition, and the device can determine the biodegradation of the polymer in this wastewater.

[0087] In general, the application conditions indicate and / or are defined by application parameter values ​​of application parameters that affect the application performance indicators of the polymer under the respective application conditions. In particular, the application parameters indicate characteristics of the application conditions, e.g., for a marine habitat, the salt concentration may strongly affect the biodegradation of the polymer in the marine habitat. The specific application parameter values ​​typical for each application condition can be stored in the database. However, the user can also input the respective specific application parameter values, e.g., if the application parameter values ​​for the respective habitat are known to deviate from the typical application parameter values.

[0088] The model providing unit 114 is adapted to provide a decision model based on the provided application conditions. In particular, the model providing unit 114 is preferably adapted to select a decision model from a plurality of decision models already stored in the database. For example, the decision model may be trained on training data corresponding to one or more specific application conditions. These specific application conditions may be defined on specific application parameter values ​​or value ranges that define which application conditions the respective decision model is suitable for. For example, a look-up table may be provided that allows the model providing unit to select which of the decision models is appropriate based on the application conditions, for example based on the application parameter values ​​of the application conditions. However, the model providing unit 114 may also comprise or point to an input unit to which a decision model may be provided, for example by a user selection or user input indicating which decision model should be used.

[0089] The decision model is a data-driven model parameterized to determine the application performance index of the polymer based on the digital representation, preferably based on the polymer characterization parameters that can indicate the physicochemical properties associated with the polymer. Optionally, the decision model can also be trained to further utilize the provided application parameter values ​​as input. In a preferred embodiment, the data-driven model refers to a machine learning model that utilizes, for example, a regression model-based algorithm or a classifier model-based algorithm. The regression model-based algorithm can be based on any of the following algorithms: neural network algorithm, linear regression algorithm, LASSO algorithm, ridge regression algorithm, MARS algorithm, random forest algorithm, and boosted tree algorithm. The classifier model-based algorithm can be based on any of the following algorithms: random forest algorithm, logistic regression algorithm, and SVM algorithm. The inventors have found that for most applications, in particular, linear regression, random forest, and MARS-based algorithms are suitable.

[0090] The decision model can be trained, for example, with the aid of a training device 130. In particular, the training device 130 comprises a training data providing unit 131 for providing training data for training the data-driven decision model. The training data includes: a) digital representations indicative of polymer characterization parameters of a plurality of training polymers; and b) application performance indicators associated with each training polymer for one or more different application conditions. Optionally, the training data set can further include application condition values ​​for the specific application conditions for which the application performance indicators of each of the polymers were determined. Preferably, in the training data, the application performance indicators provided for each training polymer refer to application performance indicators measured according to the same measurement method. However, the application performance indicators can also be provided for different measurement methods, in which case it is preferably clearly indicated which application performance indicators are associated with which measurement method, so that the decision model can be trained to distinguish between the different measurement methods. In general, the training data can be designed to cover a predefined condition space of the decision model to be trained, the condition space being defined by the value ranges of the respective application parameters for which the decision model is trained. For example, the training data can be designed to cover a predefined polymer type for a predefined application condition. Known methods can be used to design and optimize training data for a given condition space, such that the condition space is adequately covered by the training data and random outliers are avoided.

[0091] Further, the training device 130 includes a model providing unit 132 adapted to provide a data-driven trainable decision model, for example a decision model including parameters that can be set during a training process to train the decision model. For example, the trainable decision model may already be stored in a storage unit that the model providing unit 132 can access to provide it. Further, the training device 130 includes a training unit 133 for training the provided data-driven decision model based on the provided training data. In particular, training can refer to varying the parameters of the decision model based on the respective training data until the decision model is adapted to determine the application performance index of the polymer based on the digital representation, in particular based on the polymer characterization parameters. In general, any known training algorithm for training a data-driven, in particular machine learning-based model, can be utilized. Preferably, during the training of the decision model, the characterization parameters of the polymer that most affect the application performance index in the respective application conditions are also determined, and then the model is trained based on these most influential characterization parameters. To determine these most influential characterization parameters, for example, a cluster analysis or PCA analysis tool can be utilized. In particular, the characterization parameters can be utilized to determine the application space of the training data, which is then defined by the characterization parameters of the polymer and the application parameters covered by the data. The determination of the most influential characterization parameters and / or application parameters can be performed as a dimensionality reduction of the application space. An algorithm can then be applied to optimize the training data within the application space, for example, to cover the application space with as little training data as possible.

[0092] The training device 130 then comprises a trained model providing unit 134 adapted to provide the trained decision model, for example to a storage unit in which trained decision models for different application conditions and / or different types of polymers and / or polymer characterization parameters, respectively, are stored. However, the trained model providing unit 134 can also be adapted to provide the trained decision model directly, for example to the decision model providing unit 114 of the device 110.

[0093] In all cases, the decision model providing unit 114 is then adapted to provide a suitable trained decision model to the decision unit 115. The decision unit 115 can then utilize the decision model and the provided digital representation to determine the application performance index. In particular, the decision unit 115 can be adapted to utilize the polymer characterization parameters represented by the digital representation as input to a decision model trained as already described above and then provide a determination of the trained application performance index as output. The application performance index so determined can then be provided for further processing. For example, the application performance index can be processed by being displayed on a display or by generating a control signal based on the application performance index, for example to control a production process associated with the polymer. Preferably, the application performance index is further processed by validating the application performance index, particularly with respect to the intended application. A respective synthesis specification can then be provided based on the results of the validation.

[0094] The preferred verification is performed by the iteration control unit 116. The iteration control unit 116 is adapted to control the iteration process for determining the target synthesis specification. In particular, the iteration control unit 116 is adapted to compare the determined application performance indicators of the potential target polymer with the target indicators. Then, based on this comparison, the iteration control unit 116 is adapted to determine whether a further iteration step is necessary to determine the target synthesis specification or whether the iteration has reached an end, in particular whether the potential target polymer can be set as the target polymer and thus the potential target synthesis specification as the target synthesis specification. Preferably, the comparison of the determined application performance indicators of the potential target polymer with the target indicators refers to determining whether the determined application performance indicators are within a predetermined range around the target indicators, for example by determining whether the difference between the determined application performance indicators and the target indicators is below a predetermined threshold. However, the comparison can also refer to more complex mathematical functions, and the condition for the potential target polymer to be determined as the target polymer can refer to any condition based on the comparison of the determined application performance indicators with the target indicators. Generally, if a predetermined condition is met, for example if the determined application performance indicator is within a predetermined range around the target indicator, the iteration control unit 116 determines that the potential target polymer is the target polymer and the potential target synthetic specification is the target synthetic specification, and terminates the iteration.

[0095] If the above condition is not met, for example if the determined application performance index is not within a predefined range around the target index, the iteration control unit 116 is adapted to determine that a further iteration step is necessary. In this case, the iteration control unit 116 is adapted to provide new potential target synthesis specifications for the potential target polymer and repeat the determination of the application performance index utilizing the new potential target synthesis specifications for the new potential target polymer. For example, the new potential target polymer synthesis specifications can be provided on a database in which a number of potential target synthesis specifications are already stored, from which the iteration control unit 116 can select the new potential target synthesis specifications arbitrarily or according to a predefined rule. Such a rule can be, for example, a function of the comparison of the determined application performance index and the target index of the potential target polymer. For example, the function can refer to the magnitude of the difference between the determined application performance index and the target index, the smaller the difference, the more parts of the new potential target polymer are similar in the potential target polymer. In such a case, these rules may result in the iterative control unit 116 being adapted to select new potential target polymers that are more similar to the potential target polymer if the determined application performance indicators for the potential target polymer are already similar to the target indicators, and less similar if the difference between the determined application performance indicators and the target indicators is large. However, completely different rules may also be applied. Furthermore, the iterative control unit 116 may be adapted to generate new potential target synthesis specifications, for example, based on the potential target synthesis specifications and predefined rules, or arbitrarily. Again, the same principles as described above may be applied to the rules.

[0096] Furthermore, the iteration control unit 116 may be adapted to apply a stopping criterion to the iterations that indicates that no suitable target synthesis specification has been found for the respective target index. For example, the iteration control unit 116 may be adapted to apply a stopping criterion that refers to a predetermined number of iteration steps, i.e., to determining a predetermined number of new potential target synthesis specifications. However, other stopping criteria may also be utilized.

[0097] An output unit, for example a display, can then be adapted to output the determined target synthesis specification or the target polymer, for example in the form of a visual representation of the polymer, an identification of the polymer, a chemical formula representing the polymer, etc. Furthermore, the output unit can additionally or alternatively be adapted to provide the determined target synthesis specification to a database 140 for storing the respective determined target synthesis specification in association with the respective target indication for future use. Optionally, the apparatus 110 can comprise a control unit 117 adapted to provide a control signal based on the determined target synthesis specification to control the production process of the production system 120. In particular, the control signal is preferably indicative of a machine executable synthesis specification of the target polymer to be produced based on the determined target synthesis specification for producing a target polymer satisfying the target indication. However, the control unit 117 can also be adapted to provide a control signal indicative of a machine executable synthesis specification for another product utilizing or including the respective target polymer, for example to control the production process of another product based on the determined target synthesis specification.

[0098] FIG. 2 shows a schematic and exemplary flow chart of a method for determining a target synthesis specification indicative of a target polymer comprising a target indicator. In general, the method can also be considered as including the steps of a method for determining an application performance indicator. The method 200 includes a first step 210 of providing a target indicator. Furthermore, in step 220, a digital representation of a potential target synthesis specification is provided, indicative of polymer characterization parameters of the potential target polymer. In particular, the provision of the target indicator and the digital representation can follow the principles described above with respect to the target indicator providing unit 111 and the digital representation providing unit 112, respectively. Furthermore, in step 230, application conditions are provided, indicative of application parameter values ​​of application parameters that affect the application performance indicator of the polymer in the respective application conditions. For this step 230, the principles described above with respect to the habitat providing unit 113 can also be applied. Furthermore, in step 240, a decision model is provided, adapted to determine the application performance indicator of the polymer based on the digital representation. As already discussed in more detail above, providing a decision model can also refer to the selection of a decision model based on the provided application conditions. Furthermore, the decision model is a data-driven model parameterized with respect to the application conditions so that the application performance index of the polymer can be determined based on the polymer characterization parameters. In general, steps 210, 220, 230, and 240 can be performed in any order or simultaneously. In the next step 250, the application performance index is determined based on the provided digital representation of the potential target polymer and the decision model. Steps 220, 230, 240, and 250 can be considered as steps of a method for determining the application performance index, after these steps, the determined application performance index can be provided for further processing, for example, as in step 260. Then, in step 260, the determined application performance index of the potential target polymer is compared with the target index.Based on this comparison, the potential target polymer is determined as the target polymer, the potential target synthesis specification is determined as the target synthesis specification, or a new potential target synthesis specification for a new potential target polymer is provided, and the determination of the application performance index using the new potential target synthesis specification for the potential target polymer is repeated. In an optional step 270 after the target synthesis specification is determined using the above steps, the determined target synthesis specification can be provided to a user via an output unit together with the determined target polymer and the target index. Furthermore, in step 270, the potential target synthesis specification can also be used to generate a control signal that allows the control of the production process of a product, for example, a target polymer or a product containing a target polymer, as already described in detail above.

[0099] FIG. 3 shows a schematic and exemplary flow chart of a method for training a data-driven based decision model, for example used in the method 200 described with reference to FIG. 2. In general, the method 300 can be performed by, for example, each unit of the training device 130 described with reference to FIG. 1. The method 300 comprises a step 310 of providing training data for training the data-driven based decision model. The training data comprises a) digital representations indicative of polymer characterization parameters of a plurality of training polymers, and b) application performance indicators associated with each training polymer in a respective application condition, for example for a particular application parameter value. In particular, the training data can be provided according to the principles described above with reference to the training data providing unit 131 described with reference to FIG. 1. The method further comprises a step 320 of providing a data-driven based trainable decision model, for example a machine learning based decision model such as a neural network. In general, steps 310 and 320 can be performed in any order, or even simultaneously. The method 300 then further includes a step 330 of training the provided data-driven decision model based on the provided training data, for example by varying parameters in the data-driven based trainable decision model such that the trained decision model is adapted to determine the application performance indicators of the polymer based on the digital representation of the polymer. In step 340, the trained decision model can be provided, for example, by storing the trained decision model in a storage or by directly providing the trained decision model to the device 130 as described with respect to FIG.

[0100] In the following, a more detailed preferred embodiment of the above-mentioned method and the corresponding device is described. A schematic and exemplary flow chart of an exemplary and preferred embodiment of the method is provided by FIG. 4. In this exemplary embodiment, the method starts with requesting, for example, via a user interface, target indicators for a target application, for example, target biodegradability or target flammability. Furthermore, in a next step, the optimization is initialized by providing potential target synthesis specifications, i.e. starting recipes. Optionally, constraints on the recipe, i.e. synthesis specifications, can be taken into account in this process, for example, if the user provides such constraints. The constraints can refer, for example, to constraints in the manufacture of the polymer, constraints on the starting materials to be used to synthesize the polymer, etc. Furthermore, application conditions are requested, which indicate, for example, biodegradation habitat or flammability behavior tests. Furthermore, the application conditions can also indicate further information regarding the target polymer to be satisfied. For example, the requested application conditions can refer to the intended application, which indicates the application conditions. Based on the above steps, the optimization for determining the target polymer, i.e. the target synthesis specifications, can be initialized. In a first step of the optimization, polymer physicochemical parameter values ​​can be derived from the provided starting recipe, i.e. from the provided potential target synthesis specifications. A more detailed preferred possibility for deriving the physicochemical parameter values ​​is described with respect to FIG. 6. However, the derivation of the polymer physicochemical parameters can also refer to accessing a storage in which the respective physicochemical parameter values ​​for each potential target polymer are already stored. Furthermore, if the provided digital representation of the potential target synthesis specification already includes the polymer physicochemical parameters, this step can also be omitted. Based on the required application conditions, a respective decision model can be provided. Based on the provided decision model and the digital representation of the potential target synthesis specification, a target application value, i.e., an application performance index, for the potential target polymer can be provided.In the next step, it is determined whether the determined performance value, i.e. the determined application performance index, meets the target value, i.e. the target index, within a given limit. If not, i.e. if this condition is not met, the formulation of the potential target synthesis specification is modified and a new target synthesis specification is determined, optionally taking into account the constraints previously provided. Then, iterations can be started anew for the new potential target synthesis specification. In one respect, if the determined performance value meets the target value within a limit, i.e. if the respective condition is met, the potential target synthesis specification is determined as the target synthesis specification and is provided, for example, to a user or a control unit to produce the respective determined target polymer.

[0101] FIG. 5 shows, in a schematic and exemplary manner, a further preferred embodiment of the above-mentioned method for determining a target synthesis specification with a given target indicator, in which in addition to the target indicator, it is desirable for the target polymer to also satisfy a further target value, i.e. a target technical application property. The additional target technical application property can refer to any technical application property, for example biodegradability in a particular habitat, or any other technical application property. In general, in particular, the method follows the same principles as described above with reference to FIG. 4. However, due to the additional target value, additional conditions have to be met during optimization. Therefore, in the following, only the main differences with the above-mentioned method are pointed out. In particular, in this preferred embodiment, the optimizer module not only optimizes for the first target value, i.e. for the target indicator, but also for the second target value. Preferably, also for the second target value, a decision model is utilized that is adapted to determine the value of the technical application property based on the physicochemical parameters of the polymer. Thus, in addition to the above-mentioned method for the second target application, a second decision model is provided, i.e. an application property model, which allows to determine the application property value based on the polymer physicochemical parameters for the second target application. The second decision model can be based on the same algorithm as the decision model described with reference to FIG. 4, for example, only trained on a different data set to determine another property of the polymer. The comparison then refers not only to determining whether the determined application performance indicators meet the target indicators within limits, but also to determining whether the determined second application property values ​​meet the target second application property values ​​within limits. A predetermined rule can be utilized to determine in which cases the iteration continues, i.e., a new formulation is provided as a new potential target synthetic specification, and for which conditions the potential target synthetic specification is determined as the target synthetic specification. For example, the user can predetermine weights for weighting which conditions must be satisfied to what extent.For example, it may be more important to the user that an application performance index is met, but other target application characteristics are less important. In this case, the limit on which the second target application characteristic can be met can be set broader, or the satisfaction of this condition can be weighted less strongly. In this regard, Pareto optimization methods can also be utilized to find the optimal trade-off between different targets. If at some point in the iterations the conditions are met and determined to satisfy the predefined rules, the respective potential target synthesis specifications are determined as target synthesis specifications and can be provided to the user as output or utilized to generate a control file for producing the respective target polymer.

[0102] Fig. 6 shows, in a schematic and exemplary manner, a preferred method 500 for deriving characterization parameters, in particular physicochemical parameter values, from a digital representation of a new polymer, in this embodiment with reference to polymer descriptors. In a first step 510, a digital representation of the polymer is provided. The digital representation may directly contain the polymer descriptor, in which case the steps shown in Fig. 6 up to step 550 may be omitted. However, in many cases the polymer descriptor must first be determined on the basis of the provided digital representation, which in such cases refers for example to a recipe for the synthesis of the polymer, or to a chemical representation of the polymer showing the chemical components and bonds in the polymer. In this step 510, the digital representation may include any one or more of the following information: amounts of monomeric components, amounts of non-monomeric components such as initiators, fillers, additives, etc., reaction conditions such as temperature, vessel, pressure, stirring speed, condition profiles, e.g. temperature profile, pH value, solvent, feed profile, type of polymerization, e.g. radical, cationic, anionic, polycondensation, polyaddition, polyether formation, post-treatment such as amounts of components, conditions, and temperature and feed profile, type of post-treatment, e.g. radical, cationic, anionic, polycondensation, polyaddition, polyether formation, chemical information on the components such as mixtures, connectivity of non-polymeric pure compounds, composition of polymeric pure compounds based on subgroups, connectivity of monomers related to subgroups in polymeric pure components, for block copolymers, information on which block each monomer and reactive prepolymer is incorporated into, and for structured / layered materials and composites, information on which phase / layer each component is included in. If such information is not directly provided by the digital representation, then in optional step 520, the non-active ingredients and subgroups may also be derived from the digital representation, for example from a recipe.

[0103] If the information provided indicates the presence of a mixture, in a next step the mixture is decomposed into its pure components and each polymer component is treated as an input polymer. Furthermore, the polymer composition can also be converted to mole %, weight %, volume %, or absolute moles, as appropriate.

[0104] In the next step 530, the polymerizable components can be converted into subgroups, e.g. repeating units, where the subgroups are determined as different types. For example, the polymerizable subgroups can be determined based on the connectivity information of the non-polymeric pure compounds by using SMARTS, e.g. via a KNIME workflow. Also, the connectivity information of all possible subgroups can be derived from the connectivity information of the non-polymeric pure compounds by using reaction SMARTS, e.g. via a KNIME workflow.

[0105] After the subgroups and their types are determined, in step 540, the type of the descriptor to be utilized may be provided. However, the descriptor may also be determined without first selecting the type of subgroup. In order to reduce the computational resources for the method, in step 541, it is preferably determined whether the subgroup descriptor associated with each type of subgroup is already stored in the database, for example, whether an entry for the subgroup with the same connectivity information is already present in the database. If so, the respective associated subgroup descriptor may be directly downloaded, for example, in step 544. If the determined type of subgroup is not stored in the database, the subgroup descriptor associated with each type of subgroup may be determined, for example, in step 542. For example, a 3D structure of each type of subgroup may be derived based on the connectivity information, and either an automatic computation of the subgroup descriptor may be initiated, for example, using a computer cluster, or an existing machine learning determination may be utilized as the subgroup descriptor. In general, if computation for a new subgroup is required, in step 543, the result is preferably stored in the database after the computation is completed. Optionally, further subgroup descriptors can be provided from subgroup topological analysis, quantum chemical calculations, molecular dynamics calculations, coarse-grained methods, finite element calculations and dynamics simulations. In particular, polymer reaction engineering methods can be used to derive subgroup descriptors that allow the microstructure of the polymer to be taken into account.

[0106] In step 531, the amount of subgroups, i.e. the amount of each type of subgroup, is determined, for example, based on the polymer recipe information provided in step 532. For example, this amount can be determined by counting the amount of polymerizable groups per polymerizable component, optionally including a prepolymer. In this case, information on the polymerizable groups can be obtained from the non-polymeric components, and the amount so determined can be determined based on the composition of the polymeric component, in addition to counting the number of polymerizable groups, optionally not polymerized, of the subgroups of the polymeric component. It is also preferable to exclude from the amount obtained the amount of polymerizable groups resulting from agents used for post-treatment after polymerization.

[0107] However, although it is preferred that the polymer descriptors are derived from the polymer subgroups, in other embodiments of the invention the polymer descriptors can also be derived in other ways, for example by determining the polymer descriptors directly from the complete polymer. Furthermore, the polymer descriptors for each polymer may also already be stored on a storage unit, such that derivation of the polymer descriptor from a digital representation of the polymer can refer to determining information about the polymer from the digital representation that makes it possible to access a database and retrieve the corresponding polymer descriptor.

[0108] Optionally, the derived amounts of subgroups can be used for further interpretation of the polymer composition, for example, to determine the total number of polymerized functional groups, such as double bonds, amine groups, alcohol groups, thiol groups, carboxylic acid groups, isocyanate groups, epoxide groups, and functional groups formed, such as amide groups, ester groups, thioester groups, urea groups, urethane groups, thiourethane groups, ether groups. Also, molar weighted total number of polymerized functional groups, mass weighted total number of polymerized functional groups, total number of residual functional groups, e.g. double bonds, amine groups, alcohol groups, thiol groups, carboxylic acid groups, isocyanate groups, epoxide groups, molar weighted total number of residual functional groups, mass weighted total number of residual functional groups, sum of all residual functional groups, ratio between functional groups after polymerization, number of crosslinks in the polymer, mole fraction of crosslinks in the polymer, optionally also mass weighted, per subgroup, optionally also average number of atoms per weight, per subgroup, optionally also average number of non-H atoms per weight, per subgroup, optionally also average number of bonds between non-H atoms per weight, per subgroup, optionally also average number of rotors per weight, per subgroup, optionally also average number of rotors between non-H atoms per weight, per subgroup, optionally also average number of rings per weight, per subgroup , optionally as well as the average polar surface area per weight, per subgroup, optionally as well as the average refractive index per weight, the total number of blocks, the molar size of the first block, the molar size of the last block, the HLB value of the polymer, optionally the area weighted HLB value, the HLB value of the block having the lowest HLB value, optionally the area weighted HLB value, the HLB value of the block having the highest HLB value, optionally the area weighted HLB value, the HLB value of the first block, optionally the area weighted HLB value, the HLB value of the last block, optionally the area weighted HLB value, the mass of the first block, the mass of the last block, the area of ​​the block having the lowest HLB value, the area of ​​the block having the highest HLB value, the difference in the HLB values ​​of the blocks, optionally the area weighted HLB value, the hydrophilic area of ​​the polymer, the lipophilic area of ​​the polymer, the number of arms for ring opening polymerisation, or the length of the arms for ring opening polymerisation can be determined.

[0109] In step 550, the determined amount and type of subgroups and associated subgroup descriptors can be utilized to calculate a polymeric descriptor. For example, the polymeric descriptor can be determined by one or more of the molar weighted average (e.g., arithmetic mean, harmonic mean or logarithmic mean), mass weighted average (e.g., arithmetic mean, harmonic mean or logarithmic mean), volume weighted average (e.g., arithmetic mean, harmonic mean or logarithmic mean), surface area weighted average (e.g., arithmetic mean, harmonic mean or logarithmic mean) of the associated descriptors of the subgroups. Additionally, the polymeric descriptor can be determined by determining one or more of the molar weighted standard deviation, mass weighted standard deviation, volume weighted standard deviation, surface area weighted standard deviation, molar weighted maximum, mass weighted maximum, volume weighted maximum, surface area weighted maximum, molar weighted minimum, mass weighted minimum, volume weighted minimum, surface area weighted minimum, molar weighted sum, mass weighted sum, volume weighted sum, surface area weighted sum, and maximum difference from the associated subgroup descriptors.

[0110] In step 560, the derived or provided polymer descriptors can then be provided to a trained decision model for determining application performance indicators, for example as described with respect to FIG. 4. The decision model can be trained based on automated statistical pre-processing of the training data, in particular the training polymer descriptors, for example using feature engineering. For example, feature engineering can include first determining a number of different polymer descriptors for the polymer, for example based on subgroup descriptors of subgroups, and pre-selecting from this number of descriptors those that have a predetermined probability of being associated with the application performance indicators of the polymer in specific application conditions. Based on the relevant descriptors, a cluster analysis is preferably performed to identify groups of strongly correlated descriptors. Such groups make it possible to select only one of the members of the group, i.e. only one descriptor of the group, to represent the entire group of descriptors. Thus, based on the cluster analysis, the number of relevant descriptors can be further reduced. The same process can optionally be performed on the application parameters to determine the application parameters that are most relevant for the determination of the application performance indicators of the polymer in specific application conditions. Based on the remaining polymer descriptors, and optionally also on the application parameters, an application space can be determined and optimized. The application of the trained decision model to, for example, a particular application condition or polymer descriptor can then be determined by the space spanned by the training data forming the application space. This space can be optimized, for example, by modifying the training data to cover the application space periodically, by removing strong outliers, by adding training data to parts of the space not yet covered, etc. This also makes it possible to maximize the applicability space. The decision model is then trained on the basis of the optimized training data.The decision model can generally refer to sparse, e.g., spline, LASSO regression, PLS, and non-sparse, e.g., ridge regression, tree methods, kernel-based methods, statistical learning models for relating polymer descriptors to application performance indicators in specific application conditions. Furthermore, the decision model can further provide a reliability estimate of the decision depending on the respective decision model used. In step 570, the determined application performance indicators, i.e., technical application characteristics, can be provided to a user, e.g., via a user interface.

[0111] FIG. 7 shows a block diagram of an exemplary system architecture of an automated laboratory system 1000 for synthesizing polymers, having a laboratory equipment control device 1102, a network 1150 and synthesis specifications, i.e. recipes, modules 1100 / 1110, and a client device 1108. The automated laboratory system includes a laboratory equipment control device layer 1152 as part of the laboratory equipment control device 1102, and a synthesis specification module layer 1154 associated with a synthesis specification module, and a remote control or client layer 1156 associated with the client device 1108. The laboratory equipment control device layer can be divided into several layers, namely hardware, middleware, and interface layers. The hardware layer relates to hardware resources, such as sensors and actuators, especially for controlling the synthesis of polymers. The middleware relates to any of the known middleware for laboratory or plant synthesis operations. One example is LABS / QM, which provides different abstractions, such as low-level device control and message passing, for hardware, network, and operating system. The communication layer relates to a communication protocol, which may be REST or may be implemented over different transport protocols (i.e. UDP, TCP, Telemetry) that allow the exchange of messages between the lab equipment control device and the lab equipment devices. Such a software architecture allows to control and monitor the lab equipment without the need to interact with hardware.

[0112] The synthesis specification module layer 1154 may include a mass storage layer, a computing layer, and an interface layer. The storage layer is configured to provide mass storage for a data-driven decision model for providing a recipe, i.e., a synthesis specification, for a polymer based on application performance indicators, as described in detail above. In particular, the functions performed by the device as described above may be provided as program code means stored in the mass storage. Furthermore, the synthesis specifications for multiple polymers may be stored in the mass storage. Such data may be stored in a structured database, such as a SQL database, or a distributed file system, such as HDFS, or a NoSQL database, such as HBase, MongoDB, etc. The computing layer may include an application layer that allows customizing the functionality provided by standard cloud services to perform computing processes based on target properties. Such functionality may include determining a digital representation of a target polymer based on target indicators and a decision model, generating a synthesis specification from the digital representation of the target polymer, and providing the synthesis specification as control data to a laboratory equipment control device.

[0113] The interface layer may implement a web service, a network interface as UDP or TCP, or a web socket interface. For communication with the lab equipment control device, a REST API is implemented.

[0114] The client layer 1156 provides an interface for end users. For end users, the client layer 1156 can launch a client-side web application that provides an interface to the synthetic specification module layer 1154 or the laboratory equipment control device layer 1152. The user can be provided with a UI to select a target indicator and an application condition for the target indicator, which can also include a range of application performance indicator values. In other examples, the user can be provided with a UI to select two or more target indicators and their respective values. The application can be configured to allow the user to remotely monitor and control the laboratory equipment control device and operation. In other examples, the client device layer and the synthetic specification module layer can be integrated into one device. The alternatives described herein are for illustrative purposes only and should not be considered limiting.

[0115] 7 shows a block diagram of an example system architecture of a system and apparatus for generating a decision model for determining application performance indicators, a network 2150, a model generation module 2100 / 2110 which may be considered or comprise a training model apparatus, a synthesis specification module 1100 / 1110, and a client device 2108. The system for generating the decision model includes a model generation module layer 2154 as part of the model generation module and a client layer 2156 associated with the client device 2108.

[0116] The model generation module layer 2154 may include: a mass storage layer, a computing layer, an interface layer, and a storage layer configured to provide mass storage for the data-driven decision model as described above. Further, the mass storage is configured to store the composition specifications of the polymer and the measured biodegradability of one or more habitats. Such data may be stored in a structured database, such as a SQL database, or a distributed file system, such as HDFS, HBase, NoSQL database, such as MongoDB, etc. The computing layer may include an application layer that allows customizing the functionality provided by standard cloud services to perform computing processes for generating a decision model for determining application performance indicators of the polymer. Such functions include receiving, for at least two previously measured polymers, respective digital representations associated with a synthetic specification, measurement data of at least one application performance indicator in at least one habitat for each of the at least two previously measured polymers, receiving in the model generation module a digital representation of at least one unmeasured polymer, training a model according to the above-mentioned training principle based on the digital representations of the at least two previously measured polymers, measurement data of the application performance indicator in at least one application condition for each of the at least two previously measured polymers, and preferably a similarity indicator between the digital representations associated with the synthetic specification for each of the at least two previously measured polymers and the respective digital representations associated with the synthetic specification for the at least one unmeasured polymer, and providing a decision model of the application performance indicator via an output interface. The model generation module layer may be configured to deploy the generated model and the synthetic specification database to the synthetic specification module layer. This may include storing the generated model and the synthetic specification database in a mass storage device associated with the synthetic specification module.

[0117] The model generation module layer may further be configured to determine from the synthesis specification a digital representation of the polymer associated with the synthesis specification. The digital representation may include a set of polymer physicochemical parameters and polymer physicochemical parameter values ​​associated with the synthesis specification for each measured polymer. One way to derive these polymer physicochemical parameters may be to apply the SMILES algorithm or any other principle already mentioned above. If the model is generated based on the digital representation derived from the recipe, the relationship between the synthesis specification and the physicochemical parameters may be stored in a mass storage device associated with the model generation module. In such a case, developing the model includes providing the relationship.

[0118] The interface layer may implement a web service, a network interface as UDP or TCP, or a web socket interface. For communication with client devices, a REST API is implemented in this embodiment. The client layer 2156 provides access to a mass storage device containing the synthetic specifications of the polymers and at least one application performance indicator for at least two polymers. The client layer further provides an interface for end users. For end users, the client layer 2156 may launch a client-side web application that provides an interface to the model generation module layer 2154 or a mass storage device associated with the client layer. A UI can be provided to the user to select a test method as the application condition for which the application performance indicator is determined. The user may further be provided with a UI to select synthetic specification data. The user interface may also provide an option to upload the selected data to the model generation module layer and, optionally, to start model generation.

[0119] FIG. 9 shows an exemplary system 700 for producing a chemical product based on a synthesis specification generated according to the present invention. In this embodiment, the system comprises a user interface 710 and a processor 720 associated with a control unit 740. The user interface 710 and the processor 720 can be associated with or implemented according to the above-mentioned principles, and in particular can be adapted to perform a computer-implemented method for determining a target polymer and / or synthesis specification based on a determined application performance indicator, as described above. The control unit 740 is configured, for example, to receive control data generated according to the present invention, as described above, in particular to receive control data generated based on a synthesis specification of a polymer including a target indicator. In this embodiment, the control data is provided from a database 730, but in other embodiments, the control data can also be provided from a server or any other computing unit for distributing data. The vessels 750, 752 each contain a component of a chemical product, such as a prepolymer, a catalyst, etc. Typically, there are more than two vessels, however, in this embodiment, for illustrative purposes, only two are shown. Valves 760, 762 are associated with the vessels 750, 752. Valves 750 and 752 can be controlled to charge the appropriate amount of each component to reactor 770 according to the synthesis specification. A motor 800 of mixer 780 can also be controlled by the control unit according to the synthesis specification. An optional heater 790 can also be controlled according to the synthesis specification. Finally, an outlet valve 810 in fluid communication with the reactor can be controlled by the control unit to provide chemical products to a vessel or test system 820.

[0120] FIG. 10 exemplarily and diagrammatically shows a possible user interface for interacting with a processor executing the above-mentioned method for determining a target polymer with a target indicator, which in this example is the target biodegradability. In this example, an input screen is shown on the left. The input screen allows for the definition of the target biodegradability, for example in the form of a minimum and maximum value. In this example, the target biodegradability refers to the determination of the final aerobic biodegradability of the polymer in soil by measuring the oxygen demand in a respirometer or the amount of carbon dioxide evolved within two years according to the ISO 17556 standard. Optionally, the input can also refer to another target application characteristic, in this example the target kinematic viscosity, and a weighting of the two targets. Furthermore, the input screen can allow for providing constraints on the target polymer, for example a constraint on the polymer class, which can also be selected via a drop-down menu, as shown. Here, as an exemplary case, the polymer is constrained to the class of polyalkoxylates. Furthermore, as shown in FIG. 10, additional constraints for the polymer can be provided. However, this input option can also be omitted and, as mentioned above, a general synthesis specification can be provided by a database or generated by known methods. Furthermore, the input screen may also allow for providing further or other constraints on the target polymer or target synthesis specifications. Furthermore, the input screen may allow for the selection of the respective habitat and, optionally, the habitat descriptor value. Here, in this example, the habitat is determined based on the selected metric for biodegradability, which indicates that the habitat refers to soil. In general, the input may also refer to further information, such as defining the habitat descriptor, the intended application, the measurement method, etc. In this example, it may be shown that good prediction accuracy may be achieved when utilizing, as the polymer type, a descriptor that refers to the molecular weight of the polymer, a descriptor that refers to the amount of subgroups and chemical motifs, and a descriptor that refers to the hydrophilicity of the polymer.The values ​​of such polymer descriptors can then be determined according to the above principles for the starting polymer and also for each modified polymer until the target polymer can be found. An example of an output screen is shown on the right side of FIG. 10. In this example, the output screen provides a target polymer that also meets the target biodegradability and target dynamic viscosity. Furthermore, each component of the polymer is provided in the form of component type and associated components, amounts and blocks. Optionally, further information about the determined target polymer (e.g. associated synthesis specifications) can also be provided.

[0121] Below are given some examples of the application of the invention to determine biodegradability. In particular, in these examples, a model is trained to determine the biodegradability of a polymer in a particular habitat. For each example, a biodegradation model was trained for a particular class of polymers using a training data set that includes 50-100 different polymers in the polymer class and their respective biodegradation in the respective habitat. Furthermore, each model was trained to determine the biodegradability as the percentage of the polymeric material converted to CO2 after 28 days, based on the theoretical oxygen demand of the polymeric material calculated assuming total carbon conversion. Furthermore, for better comparability, the model applied to polyalkoxylates in wastewater was trained according to the standard OECD 301-test, and the model applied to polyesters in soil was trained according to ISO 17556.

[0122] In a first example, a biodegradation model was trained for polyalkoxylates in wastewater, based on a two-step approach, where in a first step a random forest model was trained and used to select polymer descriptors, which were then utilized in a second step in a trained linear regression model to determine biodegradability. The trained biodegradation model, in this case a linear regression model, was then applied to Plurafac LF 221 using the respective physicochemical properties. The output of the biodegradation model resulted in 91% biodegradation, which means that 91% of the polymeric material was converted to CO2 based on the calculated theoretical oxygen demand of the polymeric material assuming total carbon conversion, and the measured biodegradation for this polymer is reported to be greater than 60%.

[0123] In the second example, a biodegradation model was trained for polyalkoxylates in wastewater in the same manner as in the first example. In particular, the same biodegradation model can be used for both examples. The biodegradation model was then applied to Pluriol E 200 using its respective physicochemical properties. The output of the biodegradation model resulted in 96% biodegradation, meaning that 96% of the polymeric material was converted to CO2 based on the calculated theoretical oxygen demand of the polymeric material assuming total carbon conversion, and the measured biodegradation for this polymer is reported to be greater than 70%.

[0124] In the third example, a biodegradation model was trained for amine-terminated polyalkoxylates in wastewater in the same manner as in the first example. Notably, the same biodegradation model can be used for both examples. The biodegradation model was then applied to Jeffamine D 230 using its respective physicochemical properties. The output of the biodegradation model yielded 7% biodegradation, meaning that 7% of the polymeric material was converted to CO2 based on the calculated theoretical oxygen demand of the polymeric material assuming total carbon conversion, and the measured biodegradation of this polymer is reported as 7.2%.

[0125] In a fourth example, a biodegradation model is trained for polyester in soil. In this example, the trained biodegradation model is based on a trained partial least squares model. The biodegradation model is then applied to Lupraphen 1619 / 1 using its respective physicochemical properties. The output of the biodegradation model results in 49% biodegradation, which means that 49% of the polymeric material is converted to CO2 based on the calculated theoretical oxygen demand of the polymeric material assuming total carbon conversion, and the measured biodegradation of this polymer is reported to be greater than 60%.

[0126] With expected measurement errors of + / - 10% for wastewater and + / - 10% for soil, the results of the trained biodegradation model are within the appropriate accuracy range for the intended application. Thus, training using a training dataset with as few as 50 polymer data points already provides adequate accuracy. Utilizing a dataset with more polymers may lead to even higher accuracy. Thus, the biodegradation model can be trained accordingly according to the accuracy appropriate for the respective application.

[0127] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

[0128] With respect to the processes and methods disclosed herein, the operations performed in the processes and methods may be implemented in different orders. Furthermore, the outlined operations are provided only as examples, and some of the operations may be optional, combined into fewer steps and operations, supplemented with additional operations, or expanded into additional operations, without detracting from the essence of the disclosed embodiments.

[0129] In the claims, the word "comprising" does not exclude other elements or steps and the indefinite article "a" or "an" does not exclude a plurality.

[0130] A single unit or device may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0131] The procedures performed by one or more units or devices, such as providing a digital representation and a decision model, determining application performance indicators, providing application performance indicators, etc., may be performed by any other number of units or devices. These procedures may be implemented as program code means of a computer program and / or as dedicated hardware.

[0132] The computer program product may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless communication systems.

[0133] Any unit described herein may be a processing unit that is part of a conventional computing system. The processing unit may include a general-purpose processor, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or any other dedicated circuit. Any memory may be a physical system memory that may be volatile, non-volatile, or some combination of the two. The term "memory" may include any computer-readable storage medium, such as non-volatile mass storage. If the computing system is distributed, the processing and / or memory capabilities may be distributed as well. The computing system may include multiple structures as "executable components." The term "executable components" is a structure well understood in the computing arts as being a structure that may be software, hardware, or a combination thereof. For example, if implemented in software, one skilled in the art will understand that the structure of the executable components may include software objects, routines, methods, etc. that may be executed on the computing system. This may include both executable components in the heap of the computing system or executable components on a computer-readable storage medium. The structure of the executable components may reside on a computer-readable medium such that, when interpreted by one or more processors of a computing system, e.g., by processor threads, it causes the computing system to execute a function. Such structures may be directly computer readable by a processor, e.g., where the executable components are binary, or may be structured to be interpretable and / or compiled, e.g., in either a single step or multiple steps, to generate such a binary that is directly interpretable by a processor. In other cases, the structures may be hard-coded or hard-wired logic gates that are implemented exclusively or nearly exclusively in hardware, such as in a field programmable gate array (FPGA), application specific integrated circuit (ASIC), or any other dedicated circuit.Thus, the term "executable components" is a term for structures well understood by those skilled in the art of computing, whether implemented in software, hardware, or a combination. Any embodiment herein is described with reference to operations performed by one or more processing units of a computing system. When such operations are implemented in software, one or more processors direct the operation of the computing system in response to executing the computer-executable instructions that make up the executable components. A computing system may also contain communication channels that enable the computing system to communicate with other computing systems, for example, via a network. A "network" is defined as one or more data links that enable the transport of electronic data between computing systems and / or modules and / or other electronic devices. When information is transferred or provided to a computing system via a network or another communication connection, for example, either wired, wireless, or a combination of wired or wireless, the computing system properly regards the connection as a transmission medium. A transmission medium may include a network and / or data links that may be used to carry desired program code means in the form of computer-executable instructions or data structures and may be accessed by a general-purpose or special-purpose computing system or combination. Although not all computing systems require a user interface, in some embodiments a computing system includes a user interface system for use in interfacing with a user. The user interface serves as an input or output mechanism to a user, for example via a display.

[0134] Those skilled in the art will appreciate that at least a portion of the present invention may be implemented in a networked computing environment having many types of computing system configurations, including personal computers, desktop computers, laptop computers, message processors, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, cell phones, PDAs, pagers, routers, switches, data centers, wearables such as glasses, etc. The present invention may also be implemented in a distributed system environment where tasks are performed by both local and remote computing systems that are linked through a network, for example, by either hardwired data links, wireless data links, or a combination of hardwired and wireless data links. In a distributed system environment, program modules may be located in both local and remote memory storage devices.

[0135] Those skilled in the art will also appreciate that at least a portion of the present invention may be implemented in a cloud computing environment. A cloud computing environment may be distributed, but this is not required. When distributed, a cloud computing environment may be distributed internationally within an organization and / or have components owned across multiple organizations. In this specification and in the claims that follow, "cloud computing" is defined as a model for enabling on-demand network access to a shared pool of configurable computing resources, e.g., networks, servers, storage, applications, and services. The definition of "cloud computing" is not limited to any of the numerous other advantages that may be obtained from such a model when deployed. The computing system of the figure includes various components or functional blocks that may implement various embodiments disclosed herein, as described. The various components or functional blocks may be implemented on a local computing system or on a distributed computing system that includes elements that reside in the cloud or that implement aspects of cloud computing. The various components or functional blocks may be implemented as software, hardware, or a combination of software and hardware. The computing system depicted in the figures may include more or fewer components than those depicted in the figures, and some of the components may be combined where circumstances warrant.

[0136] Any reference signs in the claims should not be construed as limiting the scope.

[0137] The present invention relates to an apparatus for determining a performance index. A providing unit provides a digital representation of a target polymer, indicative of characterization parameters of the polymer. The providing unit provides a digital representation of application conditions, indicative of application parameters defining the application conditions. The providing unit provides a model based on the provided digital representation of the application conditions. The model is adapted to determine a performance index of the polymer with respect to the application conditions. The model is a data-driven model parameterized with respect to the application conditions, such that the performance index of the polymer can be determined based on the digital representation of the polymer. A determining unit for determining a performance index of the target polymer based on the provided determination model and the digital representation of the target polymer.

Claims

1. 1. A computer-implemented method for determining an application performance index of a target polymer, the method (200) comprising: providing 220 a digital representation of the target polymer indicative of or associated with a characterization parameter of the polymer; providing (230) a digital representation of application conditions for the target polymer, the digital representation of the application conditions indicating or associated with application parameters that define the application conditions; providing (240) a decision model based on the provided digital representation of the application conditions, the decision model being adapted to determine an application performance indicator for a polymer with respect to the application conditions, the decision model being a data-driven model parameterized with respect to the application conditions such that the decision model is capable of determining an application performance indicator for a polymer based on the digital representation of the polymer; and determining (250) the application performance indicators for the target polymer based on the provided decision model and the digital representation of the target polymer.

2. The method of claim 1 , wherein the digital representation of the application conditions refers to a representation of predetermined criteria and / or test procedures for the application performance indicators, and the application parameters define the predetermined criteria and / or test procedures.

3. 2. The method of claim 1, wherein the application conditions refer to one of a biodegradation habitat, a product application environment, and a fluid environment as a standardized test procedure for combustion behavior, and the application performance indicators refer to one of biodegradability, combustion behavior, adhesion, mechanical properties, and solubility, respectively.

4. 2. The method of claim 1, wherein the decision model is parameterized to further determine the application performance indicator based on the application parameters, such that the application performance indicator is determined based on the provided decision model, the digital representation of the polymer, and further based on the application parameters.

5. 1. A computer-implemented method for determining a target synthesis specification indicative of a target polymer containing target metrics for application performance, the method (200) comprising: Providing a target indicator of the polymer's target application performance (210); providing (220) a digital representation of a potential target synthesis specification representing or associated with a potential target polymer; providing (230) a digital representation of an application condition, said digital representation indicating application parameters that define said application condition; providing (240) a decision model based on the provided digital representation of the application conditions, the decision model being adapted to determine an application performance indicator for a polymer with respect to the application conditions, the decision model being a data-driven model parameterized with respect to the application conditions such that the decision model is capable of determining an application performance indicator for a polymer based on the digital representation of the polymer; determining (250) the application performance indicators for the potential target polymer based on the provided decision model and the digital representation of the potential target synthetic specifications; comparing (260) the determined application performance indicators for the potential target polymer with the target indicators, and based on the comparison, either i) determine the potential target polymer as the target polymer and the potential target synthesis specifications as the target synthesis specifications, or ii) provide new potential target synthesis specifications for a new potential target polymer and repeat the determination of the application performance indicators using the new potential target synthesis specifications for the new potential target polymer.

6. 6. The method of claim 5, further comprising providing target technology application characteristics for the target polymer, and providing the potential target synthesis specifications based on the provided target technology application characteristics such that the potential target polymer satisfies the provided target technology application characteristics.

7. 7. The method of claim 6, wherein providing new potential target synthesis specifications is based on modifying the provided target application properties and providing the new potential target synthesis specifications such that the potential target polymer meets the modified target application properties.

8. 7. The method of claim 6, wherein providing the potential target synthetic specifications based on the provided target technology application properties comprises utilizing an additional decision model adapted to determine technology application properties of a polymer based on the digital representation of the polymer, the additional decision model being a data-driven model parameterized to determine the technology application properties associated with a polymer based on the digital representation of the polymer.

9. 6. The method of claim 5, wherein providing the target synthesis specification for the target polymer comprises providing a control signal adapted to control an industrial plant to manufacture the target polymer according to the target synthesis specification.

10. 1. An interface method for providing an interface, the interface method comprising: receiving as input via a user interface a digital representation and an application condition, and providing said received digital representation and said application condition to a processor executing a method according to any one of claims 1 to 4 or claims 5 to 9; and providing the determined application performance indicators or the determined synthetic specifications of the polymer, respectively, as results to a user via a user interface, the results being received from the processor executing the method of any one of claims 1 to 4 or claims 5 to 9.

11. 1. A computer-implemented training method for training a data-driven based decision model to parameterize the decision model, the training method (300) comprising: providing (310) training data associated with predetermined application conditions, the training data including: a) a digital representation of a plurality of training polymers indicating characterization parameters for each of the training polymers; and b) application performance metrics for each of the application conditions associated with each training polymer; Providing a data-driven based trainable decision model (320); training (330) the provided data-driven based decision model based on the provided training data, wherein the trained decision model is adapted to determine application performance indicators of the polymer based on a digital representation of the polymer, in particular based on the characterization parameters, preferably polymer physicochemical parameters, more preferably polymer descriptors, indicated by the digital representation; and providing (340) the trained decision model.

12. 1. An apparatus for determining an indicator of application performance of a target polymer, said apparatus (110) comprising: a digital representation providing unit (112) for providing a digital representation of the target polymer indicative of or associated with a characterization parameter of the polymer; an application condition providing unit (113) for providing a digital representation of application conditions for the target polymer, the digital representation of the application conditions indicating or associated with application parameters that define the application conditions; a model providing unit (114) for providing a decision model based on the provided digital representation of the application conditions, the decision model being adapted to determine an application performance index of a polymer with respect to the application conditions, the decision model being a data-driven model parameterized with respect to the application conditions such that the decision model is able to determine an application performance index of a polymer based on the digital representation of the polymer; a determination unit (115) for determining the application performance index of the target polymer based on the provided determination model and the digital representation of the target polymer.

13. 1. An apparatus for determining target synthesis specifications indicative of a target polymer containing target indicators for application performance, the apparatus (110) comprising: a target providing unit (111) for providing a target indication of the target application performance of the polymer; a digital representation providing unit (112) for providing a digital representation of a potential target synthesis specification representing or associated with a potential target polymer; an application condition providing unit (113) for providing a digital representation of an application condition, said digital representation indicating application parameters that define said application condition; a model providing unit (114) for providing a decision model based on the provided digital representation of the application conditions, the decision model being adapted to determine an application performance index of a polymer with respect to the application conditions, the decision model being a data-driven model parameterized with respect to the application conditions such that the decision model is able to determine an application performance index of a polymer based on the digital representation of the polymer; a determination unit (115) for determining the application performance indicators of the potential target polymer based on the provided determination model and the digital representation of the potential target synthetic specifications; an iteration control unit (116) for comparing the determined application performance index of the potential target polymer with the target index, and based on the comparison, either i) determine the potential target polymer as the target polymer and the potential target synthesis specifications as the target synthesis specifications, or ii) provide new potential target synthesis specifications for a new potential target polymer and repeat the determination of the application performance index using the new potential target synthesis specifications for the new potential target polymer.

14. 13. A computer program for determining an application performance index for a given polymer, said computer program comprising program code means for causing an apparatus according to claim 12 to perform the method according to any one of claims 1 to 4.

15. A computer program for determining target synthesis specifications, said computer program comprising program code means for causing an apparatus according to claim 13 to carry out the method according to any one of claims 5 to 9.

Images