Device and method for predicting mixture hazard reduction for safe chemical product design

Abstract

According to some embodiments, a device for predicting hazard reduction of a mixture in a chemical product comprises: an input unit; a hazard information collection and prediction unit; a mixture hazard prediction unit; a substance of concern selection unit; a hazard reduction substance search unit; a hazard reduction substance selection unit; a mixture hazard reduction prediction unit for predicting a hazard of a mixture including a reduction substance mixed in the chemical product by replacing the substance of concern with the hazard reduction substance; and a hazard reduction evaluation unit for evaluating a degree of hazard reduction of the mixture by comparing a prediction result generated through the mixture hazard prediction unit with a prediction result of the mixture including the reduction substance generated through the mixture hazard reduction prediction unit.

Description

Device and method for predicting mixture hazard reduction for designing safe chemical products

[0001] The present invention relates to a device and method for predicting the hazard reduction of a mixture in a chemical product. More specifically, the present invention relates to a device and method for predicting the hazard reduction of a mixture in a chemical product, which can reduce the hazard of a mixture by selecting a hazardous substance of concern in the mixture and searching for and selecting a hazard reduction substance that can replace the hazardous substance of concern.

[0002]

[0003] When developing chemical products, assessing the hazards of individual components as well as those at the mixture level is crucial. However, to date, hazard assessments are focused solely on individual components. In other words, the hazards of mixtures are not considered.

[0004] Even if a substance is confirmed to be non-hazardous as a component of a product, unexpected hazards may occur when mixed with other substances, so a device and method that can predict and evaluate this is required.

[0005] In addition, although technology exists to predict the toxicity of mixtures, it only predicts toxicity using simple mathematical formulas, so the need for additional technology arises to predict toxicity reduction.

[0006]

[0007] The technical challenge that the present disclosure seeks to solve is to provide a device and method for predicting the hazard reduction of a mixture in a chemical product, which can conduct a review of hazardous substances (substances of high hazard concern, substances of potential hazard concern, etc.) among the components of a product in the early stage of product development for the design of a safe chemical product, search for and select hazardous reduction substances that can replace them, and suggest a hazard reduction strategy at the mixture level when the hazardous reduction substances are mixed.

[0008] Another technical problem to be solved by the present disclosure is to provide a device and method for predicting the hazard reduction of a mixture in a chemical product, which can efficiently select a hazard reduction substance that can replace a substance of concern for hazard among the components of a chemical product by comprehensively utilizing various characteristics such as physicochemical properties, hazards, uses, and / or functions based on the molecular structure of the product components to search for a hazard reduction substance, and to predict the hazard of the mixture and conduct a hazard reduction evaluation.

[0009]

[0010] According to some embodiments, a device for predicting a mixture hazard reduction in a chemical product includes: an input unit for receiving information on a chemical product including a plurality of components; a hazard information collection and prediction unit for collecting hazard information on the plurality of components and collecting and predicting hazards on the plurality of components; a mixture hazard prediction unit for predicting hazards on a mixture composed of the plurality of components and generating a mixture hazard prediction result value; a hazard concern substance selection unit for receiving hazard information on the plurality of components and selecting a hazard concern substance from among the plurality of components; a hazard concern substance search unit for searching a group of hazard reduction substance candidates to replace the hazard concern substances; a hazard concern substance selection unit for selecting at least some of the searched hazard concern substance candidates as hazard reduction substances to replace the hazard concern substances; a mixture hazard concern reduction prediction unit for predicting hazards on a mixture including a reduction substance mixed in the chemical product by replacing the hazard concern substance with the hazard concern substance; and comparing a prediction result generated by the mixture hazard concern prediction unit with a prediction result of a mixture including a reduction substance generated by the mixture hazard concern reduction prediction unit. Includes a hazard reduction assessment department that evaluates the degree of hazard reduction of chemical products.

[0011] In one embodiment, a device for predicting reduction of hazards in a mixture in a chemical product further includes a final result output unit, which, when receiving a determination from the hazard reduction evaluation unit that the hazard of the chemical product has been reduced by replacing the substance of concern with the substance of concern reduction, outputs a final result indicating the degree of hazard reduction in the mixture and that the substance of concern will be replaced with the substance of concern reduction, and, when receiving a determination from the hazard reduction evaluation unit that the hazard of the chemical product has not been reduced by replacing the substance of concern with the substance of concern reduction, outputs a final result indicating that the substance of concern will not be replaced with the substance of concern reduction, and that the substance of concern reduction search unit will search for a group of candidate substances of concern reduction again.

[0012] According to one embodiment, a device for predicting reduction of hazards of a mixture in a chemical product comprises information on the chemical product received by the input unit, including mixture composition information.

[0013] According to one embodiment, a device for predicting the reduction of hazards of a mixture in a chemical product comprises a hazard information collection and prediction unit that collects and predicts hazards of the plurality of components by utilizing a molecular-based characteristic prediction model and / or database.

[0014] According to one embodiment, the device for predicting the reduction of hazards in a mixture of chemical products includes a hazard reduction material search unit that searches for a group of candidate hazard reduction materials by utilizing a molecular-based characteristic prediction model and / or a database.

[0015] According to some embodiments, a method for predicting reduction of hazards of a mixture in a chemical product includes the steps of: receiving information on a chemical product including a plurality of components through an input unit; collecting hazard information on the plurality of components through a hazard information collection and prediction unit, and predicting hazards of the plurality of components; predicting hazards of a mixture of the plurality of components through a mixture hazard prediction unit, and generating a mixture hazard prediction result value; receiving hazards of the plurality of components through a hazard concern substance selection unit, and selecting a hazard concern substance from among the plurality of components; searching for a hazard concern substance candidate group to replace the hazard concern substance through a hazard reduction substance search unit; selecting at least some of the searched hazard reduction substance candidate group as a hazard reduction substance to replace the hazard concern substance through a hazard reduction substance selection unit; predicting hazards of a mixture including a mixed reduction substance in the chemical product by replacing the hazard concern substance with the hazard reduction substance through a mixture hazard reduction prediction unit; and evaluating the mixture hazard prediction unit through a hazard reduction evaluation unit. It includes a step of evaluating the degree of hazard reduction of the chemical product by comparing the prediction result generated through the mixture hazard reduction prediction unit with the prediction result of the mixture including the reduced substance generated through the mixture hazard reduction prediction unit.

[0016] In one embodiment, a method for predicting reduction of hazards in a mixture in a chemical product further includes the step of, when a determination is received from the hazard reduction evaluation unit that the hazard of the chemical product has been reduced by replacing the hazardous substance of concern with the hazardous reduction substance, outputting a final result indicating the degree of hazard reduction in the mixture and that the hazardous substance of concern will be replaced with the hazardous reduction substance, and when a determination is received from the hazard reduction evaluation unit that the hazard of the chemical product has not been reduced by replacing the hazardous substance of concern with the hazardous reduction substance, outputting a final result indicating that the hazardous reduction substance search unit will search for a group of hazardous reduction substance candidates again without replacing the hazardous substance of concern with the hazardous reduction substance.

[0017] According to one embodiment, a method for predicting reduction of hazard of a mixture in a chemical product is provided, wherein information about the chemical product received by the input unit includes mixture composition information.

[0018] According to one embodiment, a method for predicting reduction of hazard of a mixture in a chemical product comprises: the hazard information collection and prediction unit predicts hazard of the plurality of components by utilizing a molecular-based characteristic prediction model and / or database.

[0019] According to one embodiment, a method for predicting reduction of hazards of a mixture in a chemical product comprises: the hazard reduction material search unit searches for a group of candidate hazard reduction materials by utilizing a molecular-based characteristic prediction model and / or a database.

[0020]

[0021] Through a device and method for predicting the reduction of hazards in a mixture of chemical products according to some embodiments, it is possible to efficiently select hazard reduction substances that can replace hazardous substances among the components of a chemical product, and to utilize the hazard prediction and reduction evaluation of a mixture of hazard reduction substances in a chemical product to design safe chemical products in the early stages of chemical product development.

[0022]

[0023] FIG. 1 is an exemplary flowchart illustrating a method for predicting reduction of hazards of a mixture in a chemical product according to some embodiments.

[0024] Figure 2 is a product composition table showing the composition of some of the household chemical products sold on the market.

[0025] Figure 3 is a table of components of a virtual mixture that constitutes a virtual mixture by selecting and mixing some of the individual chemical substances for toxicity verification.

[0026] Figure 4a is a drawing illustrating an exemplary product that is a target for prediction of hazard reduction.

[0027] Figure 4b is a component table for exemplarily explaining the components of the product selected through Figure 4a.

[0028] Figures 4c and 4d are mathematical formulas for explaining a mixed toxicity prediction model for performing a hazard reduction prediction for the component table of Figure 4b.

[0029] Figures 4e and 4f are exemplary graphs and tables showing the results derived through the mathematical formulas of Figures 4c and 4d.

[0030] Figure 5 is a table illustrating exemplary molecular structure-based properties considered for searching for hazardous reduction substances that can replace hazardous substances among the components in chemical products.

[0031] FIG. 6 is an exemplary block diagram illustrating a device for predicting reduction of mixture hazards in chemical products according to some embodiments.

[0032] FIG. 7 is an exemplary flowchart illustrating a method for predicting reduction of hazards of a mixture in a chemical product according to some embodiments.

[0033] FIG. 8 is an exemplary flowchart for explaining the operation of an input unit (100) of a device for predicting reduction of mixture hazards in chemical products according to some embodiments.

[0034] FIG. 9 is a flowchart specifically explaining a method for predicting the hazard of a mixture component, selecting a substance of concern for hazard, and predicting the hazard of a mixture by using a prediction model and a database, according to some embodiments of a mixture hazard reduction prediction device in a chemical product.

[0035] FIG. 10 is a flowchart specifically explaining a method for a device for predicting the reduction of hazards in a mixture of chemical products according to some embodiments to search for and select hazard reduction substances using a prediction model and a database.

[0036] FIG. 11 is a flowchart specifically explaining a method for predicting and evaluating mixture hazard reduction by selecting hazard reduction substances for hazardous substances of concern among chemical product components, according to some embodiments of a mixture hazard reduction prediction device in a chemical product.

[0037]

[0038] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the attached drawings. The advantages and features of the present disclosure, and methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the attached drawings. However, the technical spirit of the present disclosure is not limited to the embodiments described below and may be implemented in various different forms. These embodiments are provided only to ensure that the present disclosure is complete and to fully inform those skilled in the art of the present disclosure of the scope of the present disclosure, and the technical spirit of the present disclosure is defined only by the scope of the claims.

[0039] When assigning reference numerals to components in each drawing, it should be noted that identical components are assigned the same numerals whenever possible, even if they appear on different drawings. Furthermore, when describing the present disclosure, if a detailed description of a related known configuration or function is deemed likely to obscure the gist of the present disclosure, such detailed description will be omitted.

[0040] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in the same sense as commonly understood by those of ordinary skill in the art to which this disclosure pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. The terminology used herein is for the purpose of describing embodiments and is not intended to limit the disclosure. In this specification, singular forms also include plural forms, unless specifically stated otherwise.

[0041] Spatially relative terms such as "below," "beneath," "lower," "above," and "upper" can be used to easily describe the relationship between one component and other components as depicted in the drawings. Spatially relative terms should be understood to include different orientations of the components during use or operation in addition to the orientations depicted in the drawings. For example, if a component depicted in the drawings were flipped over, a component described as "below" or "beneath" another component could end up "above" the other component. Thus, the exemplary term "below" can include both the above and below orientations. Components can also be oriented in other directions, and thus spatially relative terms can be interpreted accordingly.

[0042] Additionally, terms such as first, second, A, B, (a), (b), etc. may be used to describe components of the present disclosure. These terms are only intended to distinguish the components from other components, and the nature, order, or sequence of the components are not limited by the terms. When a component is described as being "connected," "coupled," or "connected" to another component, it should be understood that the component may be directly connected or connected to the other component, but another component may also be "connected," "coupled," or "connected" between each component.

[0043] The terms "comprises" and / or "comprising" as used in the specification do not exclude the presence or addition of one or more other components, steps, operations and / or elements.

[0044] Before explaining this specification, let us clarify some terms used in this specification.

[0045] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0046] FIG. 1 is an exemplary flowchart illustrating a method for predicting reduction of hazards of a mixture in a chemical product according to some embodiments.

[0047] Referring to FIG. 1, a method for predicting the reduction of hazards of a mixture in a chemical product according to some embodiments first inputs chemical product composition information (S10). The table representing the chemical product composition information may be a product composition table, such as that shown in FIG. 2.

[0048] Figure 2 is a product composition table showing the composition of some of the household chemical products sold on the market.

[0049] Referring to FIG. 2, a product composition table for each of a plurality of products (e.g., Product 1: K Company base film agent, Product 2: B Company water-based wood preservative, and Product 3: C Company paint) is exemplarily presented.

[0050] Each product composition table for multiple products displays multiple substances contained in each of the multiple products, the CAS (Chemical Abstract Service) registration number for each of the multiple substances, and the content ratio of each of the multiple substances contained in each of the products in percentage units.

[0051] Referring back to Figure 1, based on the input chemical product composition information (e.g., the product composition table in Figure 2), multiple substances constituting the product are organized into a product (mixture) and a hazard assessment is performed (S11). Next, the substance with the highest hazard within the virtual mixture is selected as the substance of concern (S12). This will be explained using Figure 3 as an example.

[0052] Figure 3 is a table of components of a virtual mixture that constitutes a virtual mixture by selecting and mixing some of the individual chemical substances for toxicity verification.

[0053] Referring to FIG. 3, for example, if there are 20 or more components shown in the product composition table of the chemical product disclosed through FIG. 2, a virtual mixture is composed by extracting 7 or more and 9 or less components that are mainly utilized among the 20 or more components. A plurality of virtual mixtures mixed in this way can be formed (e.g., virtual mixture 1, virtual mixture 2, virtual mixture 3, and virtual mixture 4). The components of each of the plurality of virtual mixture composition tables (virtual mixture 1 to virtual mixture 4) representing the constituent materials of the plurality of virtual mixtures may partially overlap with each other, but not all of the compositions are identical.

[0054] Each of the plurality of virtual mixture composition tables (virtual mixture 1 to virtual mixture 4) displays the CAS (Chemical Abstract Service) registration number for each of the plurality of substances and the content ratio of each of the plurality of substances contained in each of the products in percentage units.

[0055] A method for predicting reduction of hazard of a mixture in a chemical product according to some embodiments performs a hazard assessment based on a plurality of virtual mixture composition tables (virtual mixture 1 to virtual mixture 4), selects a substance with the highest hazard among the mixture components as a substance of concern for hazard, searches for a hazard reduction substance that can replace the substance of concern for hazard, and selects the most appropriate hazard reduction substance that can replace the substance of concern for hazard.

[0056] In a method for predicting the hazard reduction of a mixture in a chemical product according to some embodiments, a method for predicting the hazard of a mixture and evaluating whether the hazard is reduced is exemplarily explained through FIGS. 4a to 4(f).

[0057] Figures 4a4a are diagrams illustrating exemplary products that are targets of hazard reduction prediction. Figure 4b is a composition table for exemplarily explaining the composition of the product selected through Figures 4a4a. Figures 4c and 4d are mathematical formulas for explaining a mixed toxicity prediction model for performing hazard reduction prediction for the composition table of Figure 4b. Figures 4e and 4(f) are exemplary graphs and tables illustrating the results derived through the mathematical formulas of Figures 4c and 4d.

[0058] Referring to FIG. 4a, an exemplary chemical product is selected to explain a method for predicting reduction in the hazard of a mixture in a chemical product according to some embodiments.

[0059] Referring to Fig. 4b, a product composition table for the product selected through Fig. 4a is exemplarily presented. Among these, a hazard assessment is performed on a virtual mixture containing components 1 to 12 using a method for reducing the hazard of a mixture in a chemical product according to some embodiments. At this time, the assessment can be performed using the CA (Concentration Addition) of Fig. 4c and the IA (Independent Action) of Fig. 4d, respectively, targeting results grouped by the characteristics of the components in the virtual mixture, or utilizing the TSP (Two Stage Prediction) that sequentially integrates them.

[0060] To explain CA in more detail according to Fig. 4c, when a mixture is formed for two substances 1 and 2, it can be established as in the mathematical formula 1 below.

[0061] [Mathematical Formula 1]

[0062] Here x i (i is a natural number) is y in Fig. 4c i can correspond to . In addition, Ci (i is a natural number) is P in Fig. 4c iIt can correspond to . This also applies to the explanation of mathematical formula 2 below. If the CA formula regarding the toxicity of a mixture composed of two substances is expanded to n substances, it can be established as mathematical formula 2 below.

[0063] [Equation 2]

[0064] That is, the reciprocal value of this can be the output value in the CA calculation result.

[0065] P in Fig. 4c i It represents the relative proportion of each component in the mixture, and means the value obtained by dividing the concentration of the component by the concentration of all components in the mixture, and ECy i represents the concentration of individual chemical components that cause effect X%.

[0066] To explain IA in more detail according to Fig. 4d, either mathematical equation 3 or mathematical equation 4 below can be established for two different hazardous substances. More specifically, mathematical equation 3 is the overall calculation equation for IA, and mathematical equation 4 is the IA calculation equation for a mixture composed of two substances.

[0067] [Equation 3]

[0068] [Equation 4]

[0069] Here, E(c i )(i is a natural number) is e in Fig. 4d A , e B , ... can be corresponded to. If E(c1) = 1-(1-E(c1)) is substituted into the first term of the right side of mathematical expression 3, it can be expressed as an expression for E of FIG. 4d for a mixture of multiple constituent substances.

[0070] E(C) in Equations 3 and 4 mix) means the effect of the entire mixture, and E(c1) can mean the effect of the mixture component c1.

[0071] The reaction curves and reaction table results of FIG. 4e and FIG. 4(f) can be obtained by utilizing CA and IA of FIG. 4c and FIG. 4d, respectively, or by utilizing TSP.

[0072] The x-axis of the response curve according to Fig. 4e is in (nM) units expressed in Log10 scale, and the y-axis shows the toxic effect in fraction scale.

[0073] Looking at the reaction table results according to Figure 4f, we can see the CA value for 50% toxicity.

[0074] A closer look reveals that the x-axis in Figure 4e represents the concentration of the mixture, typically expressed on a log scale. The unit is not fixed and may vary depending on the input value, but in this figure, it is expressed as nM (nano Mole) for illustrative purposes.

[0075] The y-axis in Figure 4e represents the harmful effects according to the mixture concentration. Effects can be expressed in units of fraction (0-1) or percentage (0-100, %).

[0076] The dotted line at 0.5 in Figure 4e indicates the concentration that exhibits a 50% toxicity effect when performing an analysis for toxicity (hazard), and is a dotted line added to emphasize the concentration that exhibits a 50% toxicity effect when performing an analysis for toxicity (hazard).

[0077] Figure 4f is highlighted to focus on the toxicity effect at a 50% toxic concentration, but this is only an example. Accordingly, a toxicity assessment method based on the response curve of the CA value can be used to assess the reduction in toxicity of a mixture.

[0078] Methods for predicting the reduction of toxicity in chemical mixtures according to some embodiments can improve the accuracy of prediction models by utilizing experimental validation data for various mixture combinations. This validation data includes toxicity measurement results based on toxicity concentrations for each actual mixture combination and can be utilized for training and validation of prediction models.

[0079]

[0080] That is, through the explanation according to Fig. 4, it is possible to evaluate the reduction of mixture toxicity through a toxicity evaluation method.

[0081]

[0082] The toxicity assessment method illustrated in Figure 4 is based on an experimentally validated mixture dataset. For example, the toxicity concentrations of multiple different mixtures (e.g., mixtures A, B, C, etc.) can be experimentally measured and compared with the predicted values ​​of the CA, IA, and TSP models to assess the model's predictive accuracy.

[0083] The mixture concentration-response curves and toxicity value tables presented in Figures 4e and 4(f) are the result of synthesizing experimental data on mixtures composed of various combinations of multiple components. Specifically, toxicity tests were performed on mixtures containing multiple components with different material combinations and mixing ratios, and each EC 10 , EC 20 , EC 30 , EC 40 , EC 50 The concentration values ​​of the back can be derived.

[0084] In one embodiment, mixture toxicity data can be collected and utilized in the following manner: Reference toxicity values ​​can be established by measuring concentration-response curves through toxicity testing on reference mixture combinations.

[0085] For example, by adjusting the content of relatively harmful ingredients among the ingredients included in the existing mixture, and increasing the content of less harmful ingredients, new mixtures based on the mixture composition ratio can be formed, and toxicity tests can be performed on the mixtures.

[0086] Additionally, toxicity tests can be conducted on reconstituted mixtures in which highly hazardous substances among existing mixture components are selected as hazardous substances of concern and replaced with substances of reduced hazard.

[0087] Depending on the specific example, the toxicity reduction effect of each improved mixture can be quantified by comparing the toxicity values ​​of each improved mixture with those of the reference mixture. Based on this comparative data, the optimal material combination of the components constituting the mixture can be selected.

[0088] Although Figure 4f presents a table of toxicity values ​​for a single mixture, this is only an example, and in some embodiments of the present invention, comparative toxicity concentration data for multiple different mixtures (e.g., original mixture, reconstituted mixture A, reconstituted mixture B, etc.) can be generated. For example, the EC of the original mixture 50 When this is 342.657 nM, the reconstituted mixtures with different composition ratios of components or application of hazardous reducing substances have different EC 50 Values ​​can be expressed, allowing for comparison of concentration values ​​to select the optimal improvement plan. For example, information on the composition ratio of mixture components or harmful reduction substances with higher overall concentration values ​​can be selected as the optimal improvement plan.

[0089] In some embodiments, the comparative analysis of toxicity concentrations by mixture can be implemented in a matrix format by expanding the existing table format of Figure 4f, where rows represent toxicity effect levels (10%, 20%, ... 90%) and columns represent toxicity concentration prediction results for different mixtures. This allows for direct comparison of the concentrations required by each mixture at the same toxicity effect level. While the aforementioned examples disclose concentration values ​​for EC50, this is merely an example and is not limited thereto, and concentration values ​​for various toxic effects can be considered.

[0090]

[0091] Referring back to Figure 1, various characteristic information can be used to search for hazard reduction substances to replace hazardous substances of concern.

[0092] An explanation of this is provided by way of example in Fig. 5.

[0093] Figure 5 is a table of characteristics of hazardous substances that illustrates exemplary molecular structure-based characteristics considered for exploring hazardous substances that can replace hazardous substances of concern among the components in chemical products.

[0094] Referring to FIG. 5, various characteristics considered for searching for hazard reduction substances may be, for example, the hazard reduction substance characteristic search table illustrated in FIG. 5. The various characteristic information illustrated in FIG. 5 may include, for example, physicochemical characteristics, human hazards, environmental hazards, and uses / functions. The various characteristics considered for searching for hazard reduction substances in the present invention are not limited thereto.

[0095] Referring back to FIG. 1, when a hazardous reduction substance is searched and selected using the hazardous reduction substance characteristic search table of FIG. 5, a hazard assessment is performed on a reconstituted mixture in which the hazardous reduction substance is mixed to replace the hazardous concern substance.

[0096] It goes without saying that the hazard assessment method for the virtual mixture described through Figures 2 and 3 can be applied to the hazard assessment for the reconstituted mixture.

[0097] Finally, a hazard reduction assessment is conducted by comparing the hazard assessment of the mixture containing the hazardous substances of concern with the hazard assessment of the reconstituted mixture containing the hazardous reduction substances that replace the hazardous substances of concern, and the final results are derived.

[0098] For example, if the hazard of a mixture is compared with the hazard of a reconstituted mixture and a hazard reduction assessment is performed, and if the hazard of the mixture is greater, the final result is to design a chemical product with a reconstituted mixture that mixes a hazard reduction substance by replacing the substance of hazard concern. If not, and the hazard of the reconstituted mixture is greater, instead of replacing the substance of hazard concern with a hazard reduction substance, other new hazard reduction substances that can replace the substance of hazard concern are searched for and selected, and a hazard reduction assessment is performed to form a reconstituted mixture. Ultimately, the final result can be to design a chemical product with a reconstituted mixture that replaces the substance of hazard concern with a hazard reduction substance.

[0099]

[0100]

[0101] That is, through a method for predicting reduction in the hazard of a mixture in a chemical product according to some embodiments, hazard assessment can be performed at the mixture level rather than the hazard of individual substances at the chemical product design stage.

[0102] Below, the descriptions of FIGS. 1 to 5 are described in more detail from a device perspective. In addition, any descriptions that overlap with the above descriptions are omitted below.

[0103] FIG. 6 is an exemplary block diagram illustrating a device for predicting the hazard reduction of a mixture in a chemical product according to some embodiments. FIG. 7 is an exemplary flowchart illustrating a method for predicting the hazard reduction of a mixture in a chemical product according to some embodiments.

[0104] Referring to FIGS. 6 and 7, a device (1) for predicting the reduction of hazards of a mixture in a chemical product according to some embodiments receives information about the chemical product through an input unit (100) (S1-1). This will be described in detail with reference to FIG. 8.

[0105] FIG. 8 is an exemplary flowchart for explaining the operation of an input unit (100) of a device for predicting reduction of mixture hazards in chemical products according to some embodiments.

[0106] Referring to FIGS. 6 to 8, information about a chemical product received through the input unit (100) may be, for example, chemical product composition information as described in FIG. 2 (S100). The chemical product composition information may be, for example, mixture composition information, the content of mixture components, etc.

[0107]

[0108] Referring again to FIGS. 6 and 7, hazard information on components within chemical products is collected and hazards are predicted through the hazard information collection and prediction unit (120) (S1-2). At this time, a molecular-based property prediction model (110) and / or a database (130) may be utilized. The molecular-based property prediction model (110) used at this time may be ToxCSM, Deeppk, OECD QSAR (quantitative structure-activity relationship) toolbox, and / or VEGA QSAR. In addition, the database (130) used by the hazard information collection and prediction unit (120) may be, for example, material characteristics (e.g., physicochemical characteristics, hazards, functions, uses, and / or uses, etc.) as described in FIG. 5, and the hazard information collection and prediction unit (120) may utilize product MSDS (Material Safety Data Sheet), CPDat DB (Chemical and Products Database) provided by the US EPA (Environmental Protection Agency), and / or information on the amount and capacity of chemicals (Chemical & Biochemical Products) sold by Sigma-Aldrich, etc. in order to comprehensively consider material characteristics.

[0109]

[0110] Thereafter, a hazard prediction for a virtual mixture containing a portion of the components within a chemical product is performed through a mixture hazard prediction unit (150) (S2-1). As described in FIG. 3, a virtual mixture component table for the virtual mixture can be generated, and the hazard of the virtual mixture can be predicted using the method described in FIG. 4.

[0111] In parallel with step S2-1, the hazardous concern substance selection unit (140) selects hazardous concern substances from among the components in the chemical product (S3-1).

[0112] The operation of the device (1) for predicting reduction of hazards of a mixture in a chemical product according to some of the embodiments described above from steps S1-1 to S3-1 is examined in detail through FIG. 9.

[0113] FIG. 9 is a flowchart specifically explaining a method for predicting the hazard of a mixture component, selecting a substance of concern for hazard, and predicting the hazard of a mixture by using a prediction model and a database, according to some embodiments of a mixture hazard reduction prediction device in a chemical product.

[0114] Referring to FIGS. 6 and 9, chemical product composition information input through the input unit (100) is received (S121). Thereafter, the hazard information collection and prediction unit (120) utilizes the database (130) to search for hazard information on chemical product compositions (S122). At this time, in addition to what is described above through FIG. 6 of the database (130), the ECHA (European Chemicals Agency) DB and / or the CompTox Chemicals Dashboard may be further utilized.

[0115] The hazard information collection and prediction unit (120) collects hazard information on the constituents of chemical products based on the database (S124) if there is data on the constituents of chemical products in the database (130) (Yes). If there is no data on the constituents of chemical products in the database (130) (No), the hazard of the constituents of chemical products is predicted based on the molecular-based characteristic prediction model (110) (S125-2). As described above with reference to FIG. 6, the molecular-based characteristic prediction model (110) may utilize ToxCSM, OECD QSAR (quantitative structure-activity relationship) toolbox, and / or VEGA QSAR.

[0116] Through this, the hazard information of the chemical product components input through the input unit (100) is integrated by the hazard information collection and prediction unit (120) (S126). Thereafter, as described through FIGS. 2 to 5, the hazard of the virtual mixture is predicted by selectively utilizing a mixed toxicity prediction model (e.g., CA, IA, and TSP) for the results grouped by characteristics for the virtual mixture components (S151). In addition, the hazard information collection and prediction unit (120) evaluates the hazard of the components in the chemical product using the integrated information (S141). An example of the evaluation is as described in S141 of FIG. 9. However, the items for evaluating the hazard of the components in the chemical product using the integrated information by the hazard information collection and prediction unit (120) are not limited to this drawing.

[0117] Thereafter, the hazardous concern substance selection unit (140) selects hazardous concern substances among the components in the chemical product based on the component hazard evaluation items of the hazard information collection and prediction unit (120) (S142).

[0118] Referring again to FIG. 6, the operations of the hazardous concern substance selection unit (140) and the mixture hazard prediction unit (150) are performed in parallel, so the operation after the hazardous concern substance selection unit (140) will be described first.

[0119] When the hazardous concern substance selection unit (140) selects a hazardous concern substance, the hazardous reduction substance search unit (170) searches for a hazardous reduction substance to replace the hazardous concern substance (S3-2). At this time, the hazardous reduction substance search unit (170) may search for a hazardous reduction substance using a database (180) and / or a molecular-based characteristic prediction model (160).

[0120] The database (180) may be, for example, Comptox, CPDat, ECHA DB, PubChem, Sigma-Aldrich, and / or a chemical information processing system. In addition, the molecular-based property prediction model (160) may be, for example, ToxCSM, Deeppk, OECD QSAR toolbox, VEGA QSAR, and / or a Fuse prediction model.

[0121] When the hazardous reduction material search unit (170) searches for hazardous reduction materials, the hazardous reduction material selection unit (190) selects hazardous reduction materials from the searched hazardous reduction materials.

[0122] The hazardous reduction material search and selection operations of the hazardous reduction material search unit (170) and the hazardous reduction material selection unit (190) described above through FIG. 6 will be examined in detail through FIG. 10.

[0123] FIG. 10 is a flowchart specifically explaining a method for a chemical product mixture hazard reduction prediction device according to some embodiments to search for and select a hazard reduction substance.

[0124] Referring to FIGS. 6 and 10, the hazardous reduction substance search unit (170) receives hazardous substances from the hazardous concern substance selection unit (140) (S171-1). Thereafter, the hazardous reduction substance search unit (170) utilizes the database (180) (S171-3) to search for the characteristics of the hazardous concern substances (e.g., physicochemical properties, hazards, functions, uses, and / or costs, etc., as described in FIG. 5) (S171-2).

[0125]

[0126] A device for predicting the reduction of hazards in a mixture of chemical products according to some embodiments can significantly improve search efficiency by utilizing an AI-based optimization algorithm in the process of searching for hazard reduction substances.

[0127]

[0128] If data on substances of concern regarding hazardous substances is found as a result of searching using the database (180) (Yes), characteristic information on substances of concern regarding hazardous substances is collected based on the database (180) (S173).

[0129] If, instead, the hazardous substance search unit (170) utilizes the database (180) to search and finds no data on hazardous substances (No), then molecular structure-based characteristics are predicted using a prediction model (S174). The molecular-based characteristic prediction model (160) may be, for example, models as illustrated in step S174-2, but the present invention is not limited thereto.

[0130] The hazardous substance exploration unit (170) integrates the characteristic information of hazardous substances of concern through a molecular-based characteristic prediction model (160) (S175).

[0131] Thereafter, the hazardous reduction material search unit (170) compares molecular structure-based characteristics of hazardous reduction material candidates (S176). At this time, the hazardous reduction candidate material database (S176-2) can be utilized.

[0132]

[0133] According to an embodiment, in step S176, the hazardous substance search unit (170) can search for and / or select hazardous substance by comparing various characteristics (physicochemical characteristics, human / environmental hazards, structure-based characteristics, function / use, etc.) based on molecular structure (S176, S191 to S193).

[0134] The hazardous substance detection unit (170) can map hazardous substances of concern and hazardous substance reduction candidate substances onto a multidimensional characteristic space. For example, the hazardous substance detection unit (170) can map hazardous substances of concern and hazardous substance reduction candidate substances onto a multidimensional characteristic space constructed based on various characteristic information of chemical substances.

[0135] A hazard reduction substance can refer to a substance with similar properties to a substance of concern. For example, a hazard reduction substance can refer to a substance with similar properties to a substance of concern if the distance is relatively close.

[0136] According to an embodiment, in steps S176, S191 and / or S192, the distance between each substance may be calculated based on at least one piece of information among various pieces of information such as molecular structure similarity, physicochemical property information (molecular weight, polarity, solubility, LogP value, etc.), function / use, etc., so that substances with relatively close distances can be searched.

[0137] In step S193, substances with relatively low hazards can be searched for and / or selected while maintaining functional similarity to substances of concern. For example, during the search, various characteristic information on human and environmental hazards can be mapped to space to calculate distances. In step S193, by mapping various characteristic information on human and environmental hazards to space and calculating distances during the search, substances with relatively distant locations in the multidimensional characteristic space can be selected as substances with low hazards.

[0138]

[0139]

[0140] The hazardous substance selection unit (190) can select a substance with similar physicochemical properties to the hazardous substance candidates searched by the hazardous substance search unit (170) as a hazardous substance (S191). Any portions that overlap with the above description will be omitted.

[0141]

[0142]

[0143] According to an embodiment, the hazardous reduction material selection unit (190) may select a material having a similar function and use as a hazardous reduction material candidate (S192).

[0144] Afterwards, the hazardous reduction material selection unit (190) compares the hazards of the hazardous reduction materials selected through steps S191 and S192 and selects a material with low hazard as the final hazardous reduction material (S193).

[0145] If there are multiple hazardous reduction substances selected in this way, the hazardous reduction substance selection unit (190) can select the substance with the lowest cost as the hazardous reduction substance (S194).

[0146]

[0147]

[0148]

[0149] Referring again to Figure 6, the hazard reduction prediction unit (200) predicts the hazard reduction when a hazard reduction substance is mixed to replace a hazardous substance of concern in a chemical product (S3-3).

[0150] Thereafter, the hazard reduction evaluation department (210) evaluates the degree of hazard reduction of the chemical product when a hazardous concern substance is replaced with a hazard reduction substance based on the results derived through steps S2-1 and S3-3 (S4-1).

[0151] For example, whether hazard is reduced is evaluated through the hazard reduction evaluation unit (210), and in the process of evaluating whether hazard is reduced between the existing mixture and the reconstituted mixture containing the hazard reduction substance, a relative evaluation may be possible based on the 50% hazard concentration indicated by the dotted line in FIG. 4e and the emphasized thick line in FIG. 4f (for example, if the toxicity concentration of the reconstituted mixture containing the hazard reduction substance for 50% hazard is higher than that of the existing mixture, it may be possible to evaluate that the toxicity is reduced).

[0152] According to one embodiment, the device for predicting the hazard reduction of a mixture in a chemical product may further include an AI-based optimization module to improve the efficiency and accuracy of the hazard reduction prediction unit of the mixture.

[0153] For example, the mixture hazard reduction prediction unit (200) is connected to an AI-based optimization module and can perform an intelligent mixture combination ratio search function through CA, IA, and TSP prediction model algorithms and Bayesian optimization.

[0154] In one embodiment, the AI-based optimization module may include an AI-based prediction model selection engine that groups the toxicity of multiple components included in a mixture during the mixture toxicity prediction process and selects an appropriate prediction model among the CA, IA, and TSP models.

[0155] In one embodiment, the AI-based optimization module includes a Bayesian optimization engine, capable of exploring combinations of mixture components that reduce toxicity within a limited number of experiments. The Bayesian optimization engine intelligently determines the next search point by learning from previous search results, enabling efficient exploration of the entire search space and rapid convergence to an optimal solution.

[0156] In one embodiment, the Bayesian optimization engine uses a Gaussian Process model to predict the hazard reduction effect of an unknown mixture combination and determines the next mixture combination to be evaluated using an acquisition function. This allows for efficient discovery of optimal mixture combinations even under limited computational resources.

[0157] In one embodiment, the AI-based optimization module may include a combination ratio optimization engine that determines the optimal combination ratio of selected hazard-reducing substances. The combination ratio optimization engine can derive an optimal combination ratio that maximizes the hazard-reducing effect of the overall mixture while maintaining functional performance by varying the mixing ratio of each substance.

[0158] Combination ratio optimization is defined as a constrained optimization problem. The objective function is the toxicity reduction of the mixture, and constraints can include concentration limits for each component and conditions for maintaining functional performance. To solve this multi-objective optimization problem, combination ratio optimization engines can utilize metaheuristic algorithms such as genetic algorithms, particle swarm optimization, or multi-objective Bayesian optimization.

[0159] In one embodiment, combination ratio optimization is performed on a toxicity value (e.g., EC 50 , LC 50 ) and it can be evaluated that the higher the concentration that shows the same toxic effect (i.e., the higher the toxicity value), the lower the toxicity. Based on this inverse correlation between toxicity and concentration, the combination ratio optimization engine can search for the optimal mixing ratio that maximizes the toxicity value of the mixture (minimizes toxicity).

[0160] In one embodiment, the combination ratio optimization engine can simulate the expected toxicity value change of a mixture depending on the mixing ratio of each substance. For example, the concentration that produces a 50% toxic effect (EC 50 ) indicates that the higher the EC, the lower the toxicity of the mixture. 50 The mixing ratio that maximizes the value can be selected as the optimal solution.

[0161] In some embodiments, the toxicity value calculation utilizes the mathematical model described in Figures 4c and 4d, generating a toxicity value change curve according to a change in the mixing ratio to find an optimal point. The optimization process can simultaneously consider constraints such as functional performance, cost, and availability in addition to toxicity value.

[0162] Finally, if the final result output unit (220) receives a judgment from the hazard reduction evaluation unit (210) that the hazard of the chemical product has been reduced by replacing the hazardous concern substance with a hazardous reduction substance or by changing the mixture combination ratio, the final result output unit (220) outputs the final result that the hazardous concern substance has been replaced with a hazardous reduction substance or that the mixture combination ratio has been changed. Otherwise, if the final result output unit (220) receives a judgment from the hazard reduction evaluation unit (210) that the hazard of the chemical product has not been reduced, the final result output unit (220) may output the final result that the hazardous reduction substance search unit (170) will search for a hazardous reduction substance again or the mixture hazard reduction prediction unit (200) will predict the optimal mixture combination ratio without replacing the hazardous concern substance with a hazardous reduction substance or changing the combination ratio.

[0163] This is examined in detail through Figure 11.

[0164] FIG. 11 is a flowchart specifically explaining a method for predicting and evaluating hazard reduction when a hazard reduction substance is mixed into a chemical product by a device for predicting hazard reduction in a chemical product according to some embodiments.

[0165] Referring to FIGS. 6 and 11, when a hazardous reduction substance is searched for through a hazardous reduction substance search unit (S170) and a hazardous reduction substance is selected through a hazardous reduction substance selection unit (S190), the mixture hazard reduction prediction unit (200) collects reduced substance mixture information for a product (mixture) containing the hazardous reduction substance (S201). Thereafter, the mixture hazard reduction prediction unit (200) utilizes the mixture toxicity prediction model (CA, IA, and TSP) described through FIG. 4 for the results grouped by the characteristics of the product (mixture) components to predict the hazard of the reduced substance mixture (S202). Thereafter, the mixture hazard reduction prediction unit (200) derives a hazard prediction result for a mixture in which a hazard reduction substance is included as a component by replacing a hazard concern substance with a hazard reduction substance (S203), the mixture hazard prediction unit (150) derives a hazard prediction result for a virtual mixture in which a hazard concern substance is included (S151), and the hazard reduction evaluation unit (210) evaluates the degree of hazard reduction by comparing hazards (toxicity values) based on the hazard prediction values ​​derived through steps S203 and S151 (S152).

[0166] Through this, if the final result output unit (220) determines that the hazard of the chemical product has been reduced as a result of replacing the hazardous substance with a hazardous reduction substance, it outputs the final result to replace the hazardous substance with the hazardous reduction substance (Yes). If the final result output unit (220) determines that the hazard of the chemical product has not been reduced as a result of replacing the hazardous substance with a hazardous reduction substance, it outputs the final result to not replace the hazardous substance with the hazardous reduction substance, and to have the hazardous reduction substance search unit (170) search for a hazardous reduction substance to replace the hazardous substance again (No).

[0167] Although the embodiments of the present disclosure have been described with reference to the attached drawings, those skilled in the art will appreciate that the present disclosure can be implemented in other specific forms without altering the technical spirit or essential features thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of protection of the present disclosure should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included within the scope of the rights of the present disclosure.

Claims

1. An input section for receiving information on a chemical product containing multiple components; A hazard information collection and prediction unit that collects hazard information on the plurality of components and collects and predicts hazards on the plurality of components; A mixture hazard prediction unit that predicts the hazard of a mixed mixture composed of the above-mentioned multiple components and generates a mixture hazard prediction result value; A hazardous substance selection unit that receives hazardous information on the plurality of components and selects a hazardous substance from among the plurality of components; A hazardous substance exploration department that explores a group of hazardous substance candidates to replace the above hazardous substances of concern; A hazardous reduction material selection unit that selects at least some of the above-explored hazardous reduction material candidates as hazardous reduction materials to replace the hazardous concern materials; A mixture hazard reduction prediction unit that predicts the hazard of a mixture containing a reduced substance mixed in the chemical product by replacing the above hazardous concern substance with the above hazard reduction substance; and A device for predicting the hazard reduction of a mixture in a chemical product, comprising a hazard reduction evaluation unit that evaluates the degree of hazard reduction of the mixture by comparing the prediction result generated through the mixture hazard prediction unit with the prediction result of a mixture containing a reduction substance generated through the mixture hazard reduction prediction unit.

2. In paragraph 1, When the hazard reduction evaluation department receives a judgment that the hazard of the chemical product has been reduced by replacing the hazardous concern substance with the hazardous reduction substance, the final result is output as the degree of hazard reduction of the mixture and the hazardous concern substance to be replaced with the hazardous reduction substance. When the hazard reduction evaluation department receives a judgment that the hazard of the chemical product has not been reduced by replacing the hazardous concern substance with a hazardous reduction substance, the hazardous concern substance is not replaced with the hazardous reduction substance, and the hazardous reduction substance search department outputs the final result that it will search for the hazardous reduction substance candidate group again. A device for predicting the reduction of hazards of mixtures in chemical products, further comprising a final result output section.

3. In paragraph 1, A device for predicting the reduction of mixture hazards in a chemical product, wherein the information about the chemical product received by the input unit includes mixture composition information.

4. In paragraph 1, The above hazard information collection and prediction unit is a device for reducing the hazard of a mixture in a chemical product that collects and predicts the hazard of the plurality of components by utilizing a molecular-based characteristic prediction model and / or database.

5. In paragraph 1, The above-mentioned hazardous reduction substance exploration unit is a device for predicting the hazard reduction of a mixture in a chemical product that explores the above-mentioned hazardous reduction substance candidate group by utilizing a molecular-based characteristic prediction model and / or database.

6. A step of inputting information on a chemical product containing multiple components through an input unit; A step of collecting hazard information on the plurality of components through a hazard information collection and prediction unit, and collecting and predicting hazard information on the plurality of components; A step of predicting the hazard of a virtual mixture in which at least some of the plurality of components are virtually mixed through a mixture hazard prediction unit, thereby generating a virtual mixture hazard prediction result value; A step of receiving hazard information on the plurality of constituents through a hazardous concern substance selection unit and selecting a hazardous concern substance from among the plurality of constituents; A step of exploring a group of candidate hazardous reduction substances to replace the hazardous substances of concern through the hazardous reduction substance exploration department; A step of selecting at least some of the searched candidate hazard reduction substances as hazard reduction substances to replace the hazardous concern substances through a hazard reduction substance selection unit; A step of predicting the hazard of a mixture containing a reduced substance mixed in the chemical product by replacing the hazardous concern substance with the hazard reduction substance through a mixture hazard reduction prediction unit; and A method for predicting the hazard reduction of a mixture in a chemical product, comprising the step of evaluating the degree of hazard reduction of the mixture by comparing the prediction result generated through the mixture hazard prediction unit with the prediction result of the mixture containing the reduced substance generated through the mixture hazard reduction prediction unit.

7. In paragraph 6, Through the final result output section, When the hazard reduction evaluation department receives a judgment that the hazard of the chemical product has been reduced by replacing the hazardous concern substance with the hazardous reduction substance, the final result is output as the degree of hazard reduction of the mixture and the hazardous concern substance to be replaced with the hazardous reduction substance. A method for predicting reduction of hazard of a mixture in a chemical product, further comprising a step of outputting a final result that, when a judgment is received from the hazard reduction evaluation unit that the hazard of the chemical product has not been reduced by replacing the hazardous concern substance with a hazard reduction substance, the hazardous concern substance is not replaced with the hazard reduction substance, and the hazard reduction substance search unit will search for a group of hazard reduction substance candidates again.

8. In paragraph 6, A method for predicting reduction of mixture hazards in a chemical product, wherein the information about the chemical product received by the input unit includes mixture composition information.

9. In paragraph 6, The above hazard information collection and prediction unit is a method for predicting the reduction of hazard of a mixture in a chemical product by using a molecular-based characteristic prediction model and / or database to predict the hazard of the plurality of components.

10. In paragraph 6, The above-mentioned hazardous reduction substance search unit is a method for predicting the hazard reduction of a mixture in a chemical product, which searches for a group of candidate hazardous reduction substances by utilizing a molecular-based characteristic prediction model and / or database.

Images