Systems, methods, and interfaces for predicting coating weathering

Abstract

The computer-implemented method may include identifying initial color data for an initial coating to be applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied; identifying weathered color data for the initial coating on the target object; determining altered colorimetric data between the initial color data and the weathered color data; processing the altered colorimetric data through one or more machine learning algorithms; receiving an end-user selection provided through a graphical user interface of a new coating having a different multiple components; determining a predicted weathering morphology of the new coating upon application of multiple weathering exposure conditions using the one or more machine learning algorithms; and displaying data corresponding to the predicted weathering morphology of the new coating.

Description

[Technical Field]

[0001] The present invention relates to devices, computer-implemented methods, and systems for predicting weathering-based changes to coatings over time. [Background technology]

[0002] Modern coatings serve several important functions in industry and society. Coatings can protect the coated material from corrosion, such as rust. Coatings can also serve an aesthetic function by imparting a particular color and / or texture to the object. For example, most assets, such as automobiles, are coated using paints and various other coatings to protect the metal body of the automobile from the elements and to provide an aesthetic visual effect.

[0003] Given the wide range of applications for various coatings, it is often necessary to identify a target coating composition that can withstand certain weathering conditions, such as maintaining a specific color in a specific location for a specific period of time. For example, it may be necessary to identify a target coating composition that can maintain a blue metallic hue in a hot, humid location, while in other cases it may be desirable to maintain that color in a cool, dry location. However, one or more of the pigments used in such coatings may be more or less stable in various environments. For example, one type of pigment may perform particularly well in a humid environment, but may degrade more rapidly in a cold environment in the absence of a stabilizing component.

[0004] Some conventional systems attempt to address these issues by monitoring how a particular formulation performs over time in a given real-world environment. For example, a conventional system might monitor an object (e.g., an automobile) returned for repainting, monitor relative color fading or other weathering damage, and retain such reports for future reference. In other conventional systems, coating manufacturers might attempt to predict how such a formulation will perform given the known chemical properties of the underlying formulation ingredients. Systems that simply monitor an object over time may not provide relevant, real-time compositional information for future coatings. Future coatings will be continually updated over time with various hues and effect pigments that may not have been previously available. Furthermore, relying primarily or exclusively on the known chemical properties of individual components may ignore the protective or degradative effects of other ingredients in the final coating, especially when placed within a given physical environment.

[0005] Thus, there are many opportunities to quickly identify how a particular coating may perform in various environments. Summary of the Invention

[0006] The present invention provides systems, methods, and computer program products for efficiently and accurately estimating the expected weathering of a coating based on the individual and / or combined contributions from its individual components. For example, the present invention includes a computerized system employing a method for identifying and determining the individual contributions to the actual weathering of a coating and its constituent components.

[0007] The system can further employ artificial intelligence to predict colorimetric changes resulting from future weather-based degradation using one or more of the previously measured components. For example, a computer-implemented method for automatically determining and displaying expected weathering of a coating formulation after weathering exposure includes: identifying initial color data for an initial coating applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied; identifying weathered color data for the initial coating on the target object, the weathered color data including colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to multiple weathering exposure conditions, including at least exposure to heat, light, and water; and determining changed colorimetric data between the initial color data and the weathered color data, the changed colorimetric data identifying (i) each of the coating components and (ii) colorimetric data for the multiple components. a number of weathering exposure conditions; determining a rate of change of the colorimetric data associated with each of the plurality of components; processing the altered colorimetric data through one or more machine learning algorithms, the one or more machine learning algorithms storing in a database an association between each of the plurality of components and the determined rate of change of the colorimetric data upon application of the plurality of weathering exposure conditions; receiving an end-user selection provided through a graphical user interface of a new coating having a different plurality of components; determining a predicted weathering morphology of the new coating upon application of the plurality of weathering exposure conditions using the one or more machine learning algorithms that apply the altered colorimetric data; and displaying data corresponding to the predicted weathering morphology of the new coating on the graphical user interface.

[0008] Additionally, the computerized system can include a database, one or more computer processors, and computer-executable instructions stored on the database, which, when executed, cause the computerized system to identify initial color data for an initial coating applied to the target object, the initial color data identifying (i) a coating having multiple components, and (ii) colorimetric data for the coating as applied; and identify weathered color data for the initial coating on the target object, the weathered color data indicating whether the coating applied to the target object changes over time due to heat, light, and determining changed colorimetric data between the initial color data and the weathered color data, wherein the changed colorimetric data is associated with (i) each of the coating components and (ii) the plurality of weathering exposure conditions; and processing the changed colorimetric data through one or more machine learning algorithms, wherein the one or more machine learning algorithms store in a database an association between each of the plurality of components and the rate of change in the colorimetric data determined upon application of the plurality of weathering exposure conditions.

[0009] Additional features and advantages will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice. The features and advantages may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims and aspects.

[0010] These and other features will become more fully apparent from the following description and appended claims, or may be learned by the practice of the examples set forth hereinafter. To describe how the foregoing and other advantages and features can be obtained, a more particular description of the foregoing briefly set forth will be rendered by reference to specific examples thereof which are illustrated in the accompanying drawings.

[0011] The invention will be described and explained with additional specificity and detail through the use of the accompanying drawings, with the understanding that these drawings are merely illustrative and therefore not to be considered limiting of its scope. [Brief explanation of the drawings]

[0012] [Figure 1A] 1 is a schematic overview of a system trained with multiple blend colorimetric measurements prior to weathering. [Figure 1B] 1 is a diagrammatic overview of a system 100 in which a panel continues to receive learning input after being exposed to weathering. [Figure 1C] 1 is a schematic overview of a client and server system that can be used to identify individual formulation components and their individual impact on the weathering of the formulation. [Figure 1D] This is a graph comparing machine learning predictions with actual data after the system has been trained and validated. [Figure 2] Schematic diagram of end-user access to the system to combine components and identify predicted weathering data in real time. [Figure 3] FIG. 1 illustrates a method including multiple acts in a method for automatically determining and displaying the expected weathering of a coating formulation after weathering exposure. [Figure 4] FIG. 10 illustrates an additional or alternative method performed by a system configured to automatically determine and display the expected weathering of a coating formulation after weathering exposure. DETAILED DESCRIPTION OF THE INVENTION

[0013] The present invention provides systems, methods, and computer program products for efficiently and accurately estimating the expected weathering of a coating based on the individual and / or combined contributions from its individual components. For example, the present invention includes a computerized system employing a method for identifying and determining the individual contributions to the actual weathering of a coating and its constituent components.

[0014] Thus, as will be more fully understood herein, the present invention can provide many advantages to end users, particularly those seeking to optimize the appropriate color for refinishing an article, as well as those seeking to maintain color consistency over time, taking into account local weather patterns and / or durability requirements.

[0015] Additionally, end users, such as asset repair operators and even end customers, can have confidence that the custom color designed through a graphical user interface will age predictably on the finished product. Figure 1A shows a schematic overview of a system employing one or more machine learning algorithms that are trained using colorimetric measurements of multiple formulations prior to weathering. For example, Figure 1A shows that system 100 can include multiple coating formulations 140a, 140b, and 140c, etc., one or more target objects 145(a), and a server 105 configured to operate in the context of input from one or more color measurement steps 150(a).

[0016] FIG. 1A further illustrates that each coating formulation 140a, 140b, and 140c designates coating components and / or subcomponents. For example, FIG. 1A illustrates that coating formulation 140a includes components “A,” “B,” and “C.” As used herein, the terms “subcomponent” and “element” are used interchangeably with the term component. Furthermore, these terms can be understood to refer to specific named components (e.g., a specific formulation of blue pigment in a specific shade), as well as broad classes of such components (pigments or toners of a specific chemical class, e.g., organic / inorganic pigments, binders, solvents, special effect pigments, etc.), or even varying concentrations of the same initial component. Along these lines, FIG. 1A illustrates that coating formulation 140b includes components “B,” “C,” and “D,” coating formulation 140c includes components “C,” “D,” and “E,” etc. The components may include a base coat and / or one or more colorants, toners, pigments, stabilizers, effect pigments, and / or combinations thereof. At least some typical ingredients may include pigments such as titanium dioxide, basecoat components such as polyester polyols, and solvents such as propylene glycol methyl ether and ethylene glycol hexyl ether.

[0017] An operator (or end user) can apply each coating formulation 140a, 140b, 140c to one or more given target objects 145 (e.g., 145a, or others not shown). As shown in FIG. 1A, for example, target object 145a is represented by a flat panel. However, it will be understood that target object 145a may instead comprise a curved or bent metal sheet. Additionally, other target objects 145a of various materials and / or suitable compositions more closely matching the end use are contemplated herein.

[0018] FIG. 1A further illustrates that the target object 145a is sprayed with at least formulation 140a (i.e., formulation “ABC”). As previously discussed, an operator can apply one or more coating formulations directly onto the target object 145a. For example, while FIG. 1A illustrates one panel 145a with one formulation 140a (i.e., formulation “ABC”) applied, the operator can also individually spray other panels (not shown) with corresponding formulations 140b and 140c, etc., or spray multiple different formulations onto the same individual panel. In any event, after curing a given panel(s), FIG. 1A illustrates that the operator can perform an initial (i.e., pre-weathering) color and / or reflectance measurement step of the painted / coated portion(s). As described more fully below, the color measurements can be used to train one or more machine learning modules 110 of the server 105.

[0019] For example, FIG. 1A further illustrates a color measurement step 150a that provides data to the server 105. That is, an operator may measure the target object 145(a) using one or more computerized color or reflectance identification devices, such as a colorimeter, spectrometer, spectrophotometer, and / or image capture device (e.g., a camera). Then, using the associated measurement device, the operator may generate initial color (and / or reflectance) data for each initial coating, such as each coating formulation 140a, 140b, 140c. The operator may then pass the received measurement data for each coating formulation 140a, 140b to the server 105. The server 105 then performs various learning and memory functions to associate and maintain the measurement data associated with the formulation components. In one example, the server 105 uses one or more illustrated machine learning modules 110 to refine the color measurement 150a data for each sprayed coating formulation 140a, 140b, 140c and further associate each of its components with color measurement data. In this manner, the received color data, reflectance data, and / or image data (if applicable) obtained via measurement step 150a can be understood to constitute a “training” set used to train one or more machine learning modules 110 on what the formulation will look like when applied before weathering.

[0020] 1A also shows that server 105 includes one or more storage devices 120, which themselves include various components 125, 130, etc., for storing and associating formula and color measurement data for use in connection with one or more machine learning modules 110. For example, FIG. 1A shows that storage device 120 includes various storage components, in this case in the form of a formulation component 125 and a colorimetric data component 130. Formulation component 125 may be understood as a data structure that stores the formula name, a list of corresponding ingredients, and their various properties (or concentrations) before and after weathering. Colorimetric data component 130, in contrast, may be understood as a data structure that holds measurements obtained in color measurement step 150a. This may include any one or more of image, color, and reflectance data corresponding to the formula in formulation component 125.

[0021] However, such a formal distinction between components and data structures is not necessary and is adopted herein primarily for convenience of description. For example, the various storage components 125, 130 may cooperatively store information regarding the ingredients contained in each formulation 140 (e.g., a-c), and further associations of images and color / reflectance values for each applied formulation and ingredient. In either case, and however configured or partitioned, one or more machine learning modules 110, in conjunction with the storage device 120, learn the user's associations of formulations, colorimetric data, and associated images of the sprayed panels 145.

[0022] For purposes of explanation, the term "module," as used herein, refers to computer-executable code that, when executed by one or more processors in a given computer system (e.g., computer system 100 or server 105), causes the given computer system to perform a particular function. In contrast, the term "component," as used herein, refers to a passive set of instructions or data structures that store, manage, and / or otherwise provide information that is handled or otherwise processed through a given module. However, those skilled in the art will recognize that the distinction between different modules or components is at least partially arbitrary, and that modules or components may be combined and / or divided in other ways and still remain within the scope of the present disclosure. As such, references to a given element as a "module" or a "component" are provided for clarity and explanation only and should not be construed as indicating that any particular structure of computer-executable code and / or computer hardware is required, unless otherwise specified. The terms "component," "agent," "manager," "service," "engine," "virtual machine," and the like may also be used in this description.

[0023] Referring again to FIG. 1A, formulation component 125 can include current and previous information determined and / or stored for any number of given components (or combinations thereof). Such information can include that formulation 140a includes components A, B, and C, formulation 140b includes components B, C, and D, etc. As previously mentioned, alternative components B, D, and C can include adjusted concentrations of the original components A, B, and C, etc. The formulation component further includes analytics associated with each formulation and corresponding component. Such analytics can include association of one or more weathering effects by machine learning module 110 (or an operator applying training to module 110) (e.g., FIG. 1B).

[0024] The one or more weathering effects can include not only the type of deterioration (or "failure type") associated with each formulation and ingredient, but also information about various locations for weathering, various facility types, and / or information about devices used to replicate these environmental conditions (e.g., weathering testers), which can be used to help identify defects or other forms of deterioration. For purposes of this specification and claims, "deterioration" means the presence of one or more defects or processes that cause defects, such as lack of pigment durability, fading, gloss change, chemical, electrochemical, biological, and physically based degradation or decomposition processes, fouling (including biological fouling via biomatter growth), pitting, rot, rust, delamination, cratering, cracking, or other surface or coating defects, and the processes that promote them. Thus, the terms "defect," "deterioration," and "failure type" can be understood to be somewhat interchangeable in at least some applications herein regarding weathering effects, especially when referring to the point at which a defect exists.

[0025] As previously described, the formulation component 125 can include storage and / or reference to a wide range of formulas, ingredients, and other related data. For example, in addition to formulation and measurement data, the storage device 120 (via the formulation component 125 and / or colorimetric data component 130) can store hundreds, thousands, or even millions of images of sprayed panels 145 corresponding to each formula, representing pre-weathering, curing, and / or post-weathering conditions, as well as the ingredients of each formula in the system. Similarly, as described more fully below, any of the stored data can be collected before, during, and after application of weathering (see FIG. 1B). The components 125, 130 of the storage device 120 can also include various human or machine-generated determinations regarding qualitative issues in the applied formula. Such determinations can include broad designations of acceptability (e.g., acceptable or “OK” and unacceptable—not OK or “NOK”), or similar such determinations made by one of the one or more machine learning processors 110 after sufficient training.

[0026] For example, the system 100 can employ one or more learning algorithm modules 110 to learn from various training sets and improve its analytical expertise over time. Such machine learning module 110 algorithms can include, without limitation, algorithms understood as supervised learning algorithms, unsupervised learning algorithms, semi-supervised learning algorithms, reinforcement learning algorithms, self-learning algorithms, feature learning algorithms, anomaly detection algorithms, robotic learning algorithms, and / or hybrid versions thereof. In one example, a supervised learning algorithm as used herein can include (i) random forest techniques, (ii) XGBoost tree techniques, or (iii) support vector machine techniques. Any one or more of the foregoing algorithms can be employed in a deep learning neural network to memorize, learn from, and match images of the panel to similar surface features in future images. To improve the class imbalance between the "NOK" and "OK" classes, the analysis method can employ a combination of an oversampling technique (Synthetic Minority Oversampling Technique, or SMOTE) followed by an undersampling technique (Edited Nearest Neighbors, ENN) (cumulatively referred to as SMOTEENN). Broadly speaking, SMOTE is an oversampling technique that generates synthetic samples for minority classes in an imbalanced classification dataset. Using ENN after SMOTE can provide the additional benefit of removing noise in the majority of data points.

[0027] Shown below in Table I are test results showing the improvement in prediction and discrimination before and after using the SMOTEENN technique using a random forest classifier. [Table 1]

[0028] The following outlines an exemplary use of artificial intelligence and machine learning for object and surface identification in accordance with one or more aspects of the present invention. As part of training and / or as final processing of the data provided during color measurement step 150a, one or more machine learning modules 110 can perform image analysis using a "You Only Look Once" (YOLO)-style object detection algorithm on images captured by a user. To enable this use, an operator or user of system 100 can first train machine learning module 110 to process pre-weathered and post-weathered treated panels 145 using various types of image segmentation on training images, e.g., object segmentation, instance segmentation, and semantic segmentation. In general, "object segmentation" can be used to characterize images from a generalized perspective, and system 100 can additionally or alternatively employ "intersection over union" (IOU) in connection with object detection.

[0029] In contrast, "instance segmentation" can involve more specifically defining boundaries (or masks) around various weathering defects. For example, as seen with respect to FIG. 1B, the machine learning module 110 associated with the weathering data component 135 can use instance segmentation to identify specific regions where there are differences in the weathering effect on a sprayed panel. "Semantic segmentation" then involves characterizing each pixel within the boundaries identified in the instance segmentation step. Using semantic segmentation, for example, a weathered panel can be mapped by pixel and associated color value.

[0030] Once continuous training is initiated, the system 100 (i.e., the server 105 and the machine learning module 110) can continue to learn and characterize the identified surfaces. For example, the system 100 can employ training on both previously characterized images and new images fed into the system. As previously described, one or more human experts can label segmented images from a previously weathered training set (not shown) to identify surface defects, as well as the type of surface defect, and provide any other characterizations of the sprayed formulation related to the same. An experienced user can then feed the characterizations for each image back into the neural network of the system 100, allowing the system 100 to self-optimize for future applications.

[0031] Along these lines, the human expert may label the processed images with various geometric features (e.g., labels for cones, rectangles, circles, ellipses, etc.), and may also provide color labels to the extent such labels have not already been identified by the system. Other geometric features may include more specific inputs and corresponding rules; for example, certain types of fading or discoloration may be associated with heat factors compared to exposure to water, as described more fully below. Thus, the human expert(s) may train one or more machine learning modules 110 to understand not only how to identify generalized shapes, but also the context surrounding the identification of objects, which can help ensure that future analysis is contextually accurate.

[0032] FIG. 1B shows a schematic overview of system 100 continuing to receive learning inputs after panels 145 (a, b, c, etc.) have been exposed to weathering. For example, FIG. 1B shows that system 100 further applies weathering to each of the panels (145a, 145b, 145c, etc.) in step 160. Weathering step 160 can be natural, for example, by leaving the panels in an outdoor environment for months or years, or an operator can artificially apply weathering via a weathering tester (not shown). Typically, a weathering tester includes a laboratory-scale device into which an operator inserts a target object (e.g., 145(a, b, c)) and applies simulated weathering phenomena, such as UV light, precipitation, cloud cover, heat index, dew point, other thermal changes, and "exposure to water." As used herein, "exposure to water" refers to precipitation and humidity (naturally occurring or simulated), as well as applied water, such as water applied through a water spray or jet (e.g., by a weathering tester). Weathering testers allow an operator to create weathering effects on a target object 145 over a period of hours, days, or weeks, rather than the more common periods of months or years used to observe natural weathering.

[0033] After or during the weathering step 160, the operator may again collect image and / or color measurement / reflectance data for each panel / target object 145. For example, FIG. 1B illustrates a second color measurement step 150b, in which the operator / user again measures color and / or reflectance data using a colorimeter, spectrometer, or spectrophotometer, etc. Similar to the initial measurement data in step 150a described above, the operator transmits the second color measurements 150b to the server 105, which includes the machine learning processor 110 and storage device 120. As described more fully below, the operator also provides various indications of color acceptability, such as designating various panels as “OK” or “Not OK” (i.e., “NOK”). Such determinations may include, without limitation, determining “Delta E” (or “dE”) acceptance levels at different measurement angles, although ideally, Delta E should be zero at any given angle.

[0034] FIG. 1B further illustrates that storage device 120 may include one or more additional data structures in the form of weathering data component 135 that can be used to store or otherwise adjust the measured data from step 150b. Again, the specific distinction between components 125, 130, and 135 is somewhat arbitrary and made for convenience of illustration. In any event, FIG. 1B illustrates that weathering data 135 includes at least colorimetric data for each component of coating formulations 140a, 140b, 140c, as well as any other weathering data for each component (e.g., OK / NOK, specific degradation points). Weathering data for formulation components in component 135 is denoted as "[A']" for component "[A]." Similarly, weathered measurement data for component "[B]" is denoted as "[B']," weathering data for component "[C]" is denoted as "[C']," and so on. As previously mentioned, in at least one case, the difference between "[A]" and "[A']" can be expressed in terms of Delta E colorimetric difference. The differences shown as [A'], [B'], and [C'] can correspond to image-based assessments showing various defects (e.g., pigment durability, gloss changes, delamination, rust, or other forms of degradation) and any measurements (color, reflectance, etc.) that can be correlated to those effects.

[0035] For example, as previously described, an operator can provide a qualitative assessment for each weathered panel 145a', 145b', and 145c'. The qualitative assessment can include a rough characterization (e.g., "OK" or "NOK") based on a particular tolerance range for Delta E, or observations regarding the type of observed discoloration, gloss change, color intensity, delamination, or other defects. The qualitative assessment can further include specific identification of deterioration on the panel, such as points of human identification of discoloration, bubbling, rust, or other forms of observed deterioration. The operator's qualitative assessment for each panel 145a', 145b', 145c', etc. can be stored with other data in the weathering data component 135 through one or more machine learning modules 110. To the extent that OK and NOK determinations result in an imbalanced data set, the server 105 can further employ SMOTEENN to oversample the minority data set. These techniques, in essence, also enable one or more machine learning modules 110 to learn to associate specific dataset-wide variations in image data, colorimetric (e.g., Delta E), and / or reflectance data across the surface of any given weathered panel 145 (e.g., a'-c') with both qualitative and numerical assessments, thus helping to locate the contributions of specific components.

[0036] Resulting determinations by the one or more machine learning modules 110 can include determining the rate of change in gloss, fading, or other appearance metrics, including the composition of a coating on an object in various forms due to weathering. For example, the one or more machine learning modules 110 may determine that component or element C of formula 140a tends to fade, induce gloss changes, or delaminate at a higher or lower rate than another component or element in the presence of humidity, or UV light, or in some combination with other weathering factors. Similarly, the one or more machine learning modules 110 may determine that component B of formula 140b appears to have a stabilizing effect on element C of the same formula, and that the combination of components B and C tends to cause a formulation including both components B and C to withstand degradation in the same type of environment, or perhaps accelerate degradation in a different type of environment. The same analysis can be applied by simply varying the concentration of component B, or the concentrations of B and / or C.

[0037] FIG. 1C shows a schematic overview of a client and server system that can be used to identify individual formula ingredients and their correlations or contributions to weathering data. For example, FIG. 1C shows a server 105 in communication with one or more client systems 200 via a network 205. The network 205 can include any local, wide, regional, or global area network connection anywhere from a direct wired connection to a connection via the Internet. The network 205 can also, or additionally, include local short-range networks, including Bluetooth and WIFI networks. In any event, FIG. 1C shows that the client systems 200 interact with the data from the server 105 to assist in training one or more machine learning modules, evaluate their progress, and obtain predictions corresponding to a particular formula ingredient and its response to the weathering step 160.

[0038] For example, FIG. 1C illustrates a client system 200 communicating with the server 105, such as via a network 205. The client system 200 can include several different types of end-user terminals. For example, without limitation, it can be a traditional personal computing device or a mobile device, such as a laptop, tablet, mobile phone, or other form of network-connected computing device. However, a network-connected client system is not required; an operator may function in direct connection with the server 105. In other words, the server 105 and database 120 may be interpreted as standalone or other database applications installed on the client system 200. In any event, the operator can interact with the server 105 through several different means to input data and train one or more machine learning modules 110 with pre- and post-weathering measurements (e.g., analysis 215). Such operator interaction (e.g., via one or more analysis messages 215) can further include monitoring the progress of the one or more machine learning modules 110 in terms of their ability to provide consistent or accurate predictions and / or adjustments to the predictions (e.g., after application of SMOTEENN technology).

[0039] Along these lines, FIG. 1D shows various example result charts (225a, 225b) illustrating the typical progress of machine learning (in this case, using a random forest classifier) predictions given various training sets provided to the server 105 by an operator. The charts in FIG. 1D plot the percentage of correctly identified "OK" (225a) and "NOK" (225b) against the measured angle (e.g., steps 150a, 150b) using a "Delta E" measurement for every angle. As shown, the one or more machine learning modules 110 can use realistic supervised class distributions to generate analyses and predictions that perform better than a random baseline (i.e., the lower line in chart 225a).

[0040] For example, charts 225a and 225b show the results of test data used to validate how well the machine learning algorithms outlined herein perform in real-world scenarios using weathering data. The data shown in the charts display both recall and precision metrics traditionally used to evaluate the accuracy of computer predictions. For example, chart 225a displays the percentage of "not OK" (NOK) results at each measurement angle, with "NOK predictions" referring to the proportion of correct NOK predictions (computer-generated data using one or more machine learning algorithms) among all NOK computer-generated predictions (i.e., precision), while "identified NOK" refers to the amount of actual NOK samples that the model was able to correctly label as NOK (i.e., recall). Along these lines, chart 225a demonstrates that the computer predictions exceed the accuracy of random controls. Furthermore, chart 225a shows that the line showing the rate of correct NOK identifications is roughly the same shape and size as the line showing the percentage of correct NOK identifications, indicating an overall high predictive power / accuracy of the model.

[0041] In contrast, chart 225b compares the % of correctly identified OKs (OK recall) with the % of correct OK predictions (i.e., using computer-generated predictions based on one or more machine learning algorithms) (OK precision). As with the NOK measurements in chart 225a, the OK measurements and predictions in chart 225b show that the computer predictions perform well at most angles, failing only at low angles where variation from the measurements creates noise for the model. In other words, the chart shows that one or more machine learning algorithms can use Delta E data to predict weathering degradation and come very close to actual observations. That is, the fact that both the precision and recall metrics of the computer predictions for both OK and NOK are high (typically above the 75% threshold for precision) confirms that the computer-aided models and algorithms outlined herein can correctly distinguish between OK and NOK characteristics in most cases, thus validating the component-specific approach used to form the predictions.

[0042] FIG. 2 shows a schematic diagram of an end user using client system 200 to combine ingredients and identify expected weathering data for a new formulation in real time. For example, FIG. 2 shows an overview of client system 200, including a graphical user interface 250 with various sequences of input / output interfaces (220a / 220b and 230a / 230b, respectively). Through first input interface 220, the user inputs a set of ingredients, for example, based on formula 140a, but replaces component D with component C or changes its concentration. For example, component D can be an alternative concentration of component C. Computer system 200 then transmits the new formula to server 105 via one or more messages 20a. For example, a user may be interested in using a combination of a particular color with a new effect pigment (element D, such as "mica"), but system 100 only has prior measurements for the color using a different effect pigment (element C, such as "aluminum"). Client 200 transmits the input to server 105 via network 205. One or more machine learning modules 110 in server 105 can process the data from one or more messages 260a. In one example, one or more machine learning algorithms 110 extrapolate stored weathering data associated with each ingredient to ascertain their respective contribution to degradation (e.g., as measured from formulations 140b and 140c) in one or more other / new formulations.

[0043] That is, one or more machine learning modules 110 can then combine data from measurements related to formula 140a with the contribution of element D from before / after measurements of formulas 140b and 140c (and possibly others) to prepare a predicted weathering effect. Thus, in response to an input request at interface 220a, user interface 250 may provide output 230a (transmitted via one or more messages 260b) indicating that the new formula ABD is predicted to succeed, improve, or fail by a certain amount, or a certain percentage change, relative to the original baseline formula 140a, in response to a particular type of weathering. For example, output interface 230a may indicate that the requested formula ABD will fade, change in adhesion or gloss, or degrade by 5%, or otherwise degrade by 20% in different types of accelerated testing (e.g., different weatherometer conditions) within 9 years, compared to 11 years for the original formula. Alternatively, output interface 230a may indicate that the input formula of ABD will last longer (e.g., 12 years) than the expected duration (11) without such defects in one or more of those same factors by changing one of the components to a new formula (e.g., changing formula ABD to ABE).

[0044] Along these lines, in the next illustrated example, the user similarly inputs another new formula BCE into input interface 220b, where the new formula BCE is a variation of original formula 140b (i.e., altering formula BCD by component E). Computer system 200 transmits the user's input to server 105 via one or more input messages 260b, indicating that the input is an adjustment to formula 140b by inserting component E in place of the original component D. FIG. 2 then shows that server 105 generates one or more forecasts for the modified formula and returns one or more weathering forecasts 265b. Computer system 200 then displays the forecast information in output screen 230b.

[0045] Accordingly, Figures 1-2 provide several components, modules, and schematic diagrams as part of a system for automatically determining and predicting the expected weathering effects of various individual components of a coating / composition. The present invention can also be described in terms of one or more methods for achieving a particular result. Along these lines, Figures 3 and 4 illustrate various methods for automatically determining and displaying the expected weathering of a coating formulation after weathering exposure. The acts and steps of Figures 3 and 4 are described below with reference to the system components and modules of Figures 1A-2.

[0046] For example, Figure 3 shows that a method 300 for automatically determining and displaying expected weathering of a coating formulation after weathering exposure can include act 310 of identifying initial color data for an initial coating to be applied to a target object. Act 310 includes identifying initial color data for an initial coating to be applied to the target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied. For example, Figure 1A shows that various formulations as applied to target object 145(a), their corresponding components, and their initial color measurements (via step 150a) can be passed to server 105 via storage device 120 for storage.

[0047] FIG. 3 also illustrates that method 300 can include act 320 of identifying weathered color data for the initial coating. Act 320 includes identifying weathered color data for the initial coating on the target object, where the weathered color data includes colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to a plurality of weathering exposure conditions, including at least exposure to heat, light, and water. For example, FIG. 1B illustrates an operator performing second color measurements 150b on various target objects after the applied weathering step 160. As previously discussed, weathering exposure conditions can include natural and artificially applied (e.g., via a weathering tester) weathering in the form of exposure to heat, light, and water (in the form of precipitation and humidity, and water applied through a water spray or jet).

[0048] 3 illustrates that method 300 can include act 330 of determining altered colorimetric data between the initial coating and the weathered color data. Act 330 includes determining altered colorimetric data between the initial color data and the weathered color data, the altered colorimetric data being associated with (i) each of the coating components and (ii) a plurality of weathering exposure conditions. For example, FIG. 1B illustrates that server 105 can identify component-specific associations (e.g., stored within component 135) with various weatherings (applied or natural) at least in part by comparing the color measurements made in steps 150a (initial measurements) and 150b (weathered measurements).

[0049] FIG. 3 further illustrates that method 300 can include act 340 of processing the altered colorimetric data through artificial intelligence. Act 340 can include processing the altered colorimetric data through one or more machine learning algorithms, where the one or more machine learning algorithms store, in a database, an association between each of a plurality of ingredients and a determined rate of change in the colorimetric data upon application of a plurality of weathering exposure conditions. For example, FIGS. 1A-1B illustrate that one or more machine learning algorithms process pre- and post-weathering data correlating with various formulations and store the associations in storage device 120. As part of the processing, the one or more machine learning algorithms can identify various associations between individual ingredients and identified changes in color measurements (e.g., delta E or "dE" between steps 150a and 150b).

[0050] Furthermore, Figure 3 shows that method 300 can include act 350 of receiving an end user selection for a new coating having new components. Act 350 can include receiving an end user selection provided through a graphical user interface of a new coating having different components. For example, Figure 2 shows a user inputting an adjustment to formula 140a through input interface 220a, such as by replacing component C with component D in original formula 140a.

[0051] FIG. 3 also illustrates that method 300 can include act 360 of predicting the weathering morphology of the new component using a machine learning algorithm. Act 360 can include using one or more machine learning algorithms that apply the altered colorimetric data to determine the predicted weathering morphology of the new coating upon application of multiple weathering exposure conditions. For example, FIG. 2 illustrates that in response to user input (i.e., message 260a) requesting predicted weathering of adjusted formula 140a, server 105 sends a response via message 265a containing a weathering prediction for the requested adjusted formula. Message 265a can include, for example, a Delta E prediction made by one or more machine learning modules 110 for a coating formula made at various angles and given based on certain assumptions about expected weathering.

[0052] Additionally, FIG. 3 illustrates that method 300 can include act 370 of displaying data for the predicted weathering morphology. Act 370 can include displaying data corresponding to the predicted weathering morphology of the new coating on a graphical user interface. For example, FIG. 2 illustrates that input / output interfaces (220a / b, 230a / b, respectively) can be provided and displayed on display device 250 of client computer system 200. As previously discussed, the predicted weathering morphology for the new coating can include predicted changes to the new coating based on multiple components of the new coating and multiple weathering exposure conditions. The predicted changes can include changes in color, adhesion, gloss, and / or other appearance metrics (e.g., fouling).

[0053] The predicted weathering morphology for a new coating may also include a prediction of whether the new coating will fail (e.g., change in gloss or adhesion beyond a predetermined value) after a certain amount of exposure to weathering exposure conditions. The predicted weathering morphology for a new coating may also include a prediction of what type of failure the new coating will experience (e.g., change in gloss and / or adhesion). The prediction may further include expected weathering variations based on specific climate conditions in a particular geographic location. For example, a response from server 205 may include that the new formulation will degrade unacceptably within 10 years in a cold, dry climate and will degrade unacceptably within 8 years in a wet, hot climate.

[0054] 4 illustrates that an additional or alternative method 400 for automatically determining and displaying expected weathering of a coating formulation after weathering exposure can include act 410 of identifying initial color data for an initial coating applied to a target object. Act 410 includes identifying initial color data for an initial coating applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied. For example, FIG. 1A illustrates an operator applying one or more formulations including multiple components to a panel and then performing color measurement step 150a to identify at least an initial colorimetric measurement for the applied formulation (i.e., before weathering).

[0055] FIG. 4 also illustrates that method 400 can include act 420 of determining weathered color data for the initial coating. Act 420 includes determining weathered color data for the initial coating on the target object, where the weathered color data includes colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to a plurality of weathering exposure conditions, including at least exposure to heat, light, and precipitation. For example, FIG. 1B illustrates the application of a weathering step 160 (i.e., naturally or via a weathering tester). The weathering step can include the application of light, heat, water, humidity, and other forms of weathering in various combinations as outlined herein. An operator can then perform a second or subsequent color measurement step 150b to determine at least slight differences in the colorimetric data due to the applied weathering.

[0056] Additionally, FIG. 4 illustrates that method 400 can include act 430 of determining altered colorimetric data between the initial coating and weathered color data. Act 430 includes determining altered colorimetric data between the initial color data and the weathered color data, where the altered colorimetric data is associated with (i) each of the coating components and (ii) a plurality of weathering exposure conditions. For example, FIG. 1B illustrates that a second color measurement step 150b is performed after the applied weathering step 160. An operator can pass the second color measurement data for a given panel to server 105, where the color measurements can be linked to various formulations.

[0057] FIG. 4 further illustrates that method 400 can include act 440 of processing the altered colorimetric data through artificial intelligence. Act 440 includes processing the altered colorimetric data through one or more machine learning algorithms, which store, in a database, an association between each of a plurality of components and a determined rate of change in the colorimetric data upon application of a plurality of weathering exposure conditions. For example, server 105 can include one or more machine learning modules 110, which apply any number of one or more supervised learning algorithms (e.g., random forest techniques, XGBoost tree techniques, and support vector machine techniques). The one or more machine learning modules 110 can determine the various contributions of individual components of a formula by analyzing one or more formulas containing the components, as described above, and determine how the various formulas responded to weathering step 160. The one or more machine learning modules 110 can process the change in color measurements from step 150b relative to step 150a and determine the change in colorimetric data, reflectance, etc., applied to each individual component of each formula.

[0058] Although not shown, computer-implemented methods 300 and 400 may further include recommending one or more ingredient substitutes. The one or more ingredient substitutes may be optimized by one or more machine learning modules 110 to minimize changes in appearance in response to multiple weathering exposure conditions. That is, the one or more ingredient substitutes may be optimized by one or more machine learning algorithms employed by module 110 to withstand expected changes due to multiple weathering exposure conditions.

[0059] 1A-4 thus provide a number of systems, components, and modules that can be used to significantly improve both the speed and accuracy of individual component contributions to weathering responses. Part of this improvement can be obtained by using various types of weathering, including employing both natural and artificial weathering conditions. Part of the improvement can also be obtained by using one or more machine learning algorithms to process and correlate changes in instrumental measurement data and associate these changes with individual components. Thus, the principles outlined in this disclosure can provide end users with significantly more efficient and accurate tools for obtaining immediate feedback on the weathering degradation effects of new formulations containing new components.

[0060] The present invention may include or utilize computer hardware, such as one or more processors and system memory, as described in more detail below, including special-purpose or general-purpose computer systems. The scope of the present invention also includes physical and other computer-readable media for carrying or storing computer-executable instructions and / or data structures. Such computer-readable media can be any available media accessible by a general-purpose or special-purpose computer system. Computer-readable media that store computer-executable instructions and / or data structures are computer storage media. Computer-readable media that carry computer-executable instructions and / or data structures are transmission media. Thus, by way of example, and not limitation, the present invention can include at least two distinctly different types of computer-readable media: computer storage media and transmission media.

[0061] Computer storage media are physical storage media that store computer-executable instructions and / or data structures, including computer hardware such as RAM, ROM, EEPROM, solid-state drives ("SSD"), flash memory, phase-change memory ("PCM"), optical disk storage, magnetic disk storage or other magnetic storage devices, or any other hardware storage device(s) that can be used to store program code in the form of computer-executable instructions or data structures that can be accessed and executed by a general-purpose or special-purpose computer system to implement the disclosed functions of the present invention.

[0062] Transmission media can be used to convey program code in the form of computer-executable instructions or data structures and can include networks and / or data links accessible by a general-purpose or special-purpose computer system. A "network" is defined as one or more data links that enable the transport of electronic data between computer systems and / or modules and / or other electronic devices. Information can be transferred or provided to a computer system over a network or another communications connection (either wired, wireless, or a combination of wired or wireless), and the computer system can view the connection as a transmission medium. Combinations of the above should also be included within the scope of computer-readable media.

[0063] Furthermore, program code in the form of computer-executable instructions or data structures may be automatically transferred from transmission media to computer storage media (or vice versa) upon reaching various computer system components. For example, computer-executable instructions or data structures received over a network or data link may be buffered in RAM within a network interface module (e.g., a "NIC") and eventually transferred to computer system RAM and / or less volatile computer storage media within the computer system. Thus, it should be understood that computer storage media may be included within computer system components that also (or even primarily) utilize transmission media.

[0064] Computer-executable instructions include, for example, instructions and data that, when executed on one or more processors, cause a general-purpose computer system, special-purpose computer system, or special-purpose processing device to perform a certain function or group of functions. Computer-executable instructions may be, for example, binaries, intermediate format instructions such as assembly language, or even source code.

[0065] As will be appreciated by those skilled in the art, the present invention may be practiced in a network computing environment having many types of computer 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, cellular phones, PDAs, tablets, pagers, routers, switches, and the like. The present invention may also be practiced in a distributed system environment where tasks are performed by both local and remote computer systems linked through a network (either by wired data links, wireless data links, or a combination of wired and wireless data links). Thus, in a distributed system environment, a computer system may include multiple component computer systems. In a distributed system environment, program modules may be located in both local and remote memory storage devices.

[0066] Those skilled in the art will also appreciate that the present invention may be implemented in a cloud computing environment. A cloud computing environment may be distributed, but this is not required. If distributed, the cloud computing environment may be distributed internationally within an organization and / or may have components held across multiple organizations. In this description 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 intended to be limiting to any of the many other benefits that may be obtained when such a model is properly deployed.

[0067] Cloud computing models can consist of various characteristics, such as on-demand self-service, wide network access, resource pooling, rapid elasticity, measured service, etc. Cloud computing models can also be offered in various service models, such as software as a service ("SaaS"), platform as a service ("PaaS"), and infrastructure as a service ("IaaS"), etc. Cloud computing models can also be deployed using various deployment models, such as private clouds, community clouds, public clouds, hybrid clouds, etc.

[0068] A cloud computing environment or cloud computing platform may include a system that includes one or more hosts, each capable of running one or more virtual machines. During operation, the virtual machines emulate a running computing system that supports an operating system and possibly one or more other applications. Each host may include a hypervisor that emulates virtual resources for the virtual machine using physical resources abstracted from the virtual machine's view. The hypervisor also provides appropriate isolation between the virtual machines. Thus, from the perspective of any given virtual machine, the hypervisor provides the illusion that the virtual machine is functioning in connection with physical resources, even though the virtual machine is functioning in connection with only the appearance of physical resources (e.g., virtual resources). Examples of physical resources include processing power, memory, disk space, network bandwidth, media drives, etc.

[0069] In view of the foregoing, the present invention can be embodied in a number of different configurations, as outlined above and exemplified by the following.

[0070] For example, one configuration is a computer-implemented method for automatically determining and displaying expected weathering of a coating formulation after weathering exposure, comprising: identifying initial color data for an initial coating applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied; identifying weathered color data for the initial coating on the target object, the weathered color data including colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to multiple weathering exposure conditions, including at least exposure to heat, light, and water; and determining changed colorimetric data between the initial color data and the weathered color data, the changed colorimetric data identifying (i) each of the coating components and (ii) colorimetric data for the multiple components. and processing the altered colorimetric data through one or more machine learning algorithms, the one or more machine learning algorithms storing in a database an association between each of the plurality of components and the determined rate of change of the colorimetric data upon application of the plurality of weathering exposure conditions; receiving an end-user selection provided through a graphical user interface of a new coating having a different plurality of components; determining a predicted weathering morphology of the new coating upon application of the plurality of weathering exposure conditions using the one or more machine learning algorithms that apply the altered colorimetric data; and displaying data corresponding to the predicted weathering morphology of the new coating on the graphical user interface.

[0071] In a further or alternative configuration, the weathering conditions include exposing the target object to natural weathering in a natural environment for a period of multiple months. The computer-implemented method of any preceding claim, wherein the weathering conditions include simulating weathering conditions through a weathering tester. In a further or alternative configuration, the method may further include identifying a historical dataset for the initial coating, the historical dataset including changed colorimetric data associated with multiple weathering exposure conditions and one or more different applications of the initial coating to different target objects exposed to the different multiple weathering exposure conditions. In a further or alternative configuration, the method may further include using one or more machine learning algorithms applying the historical dataset to determine expected changes in the colorimetric data for each of the components associated with each weathering condition across the multiple weathering exposure conditions and the different multiple weathering exposure conditions. In a further or alternative configuration, the method may further include determining a damage type for the initial coating after application of the multiple weathering exposure conditions, the damage type including a color change.

[0072] In an additional or alternative configuration, the method may further include determining a failure type for the initial coating after application of the plurality of weathering exposure conditions, the failure type including a change in gloss or a change in adhesion. In an additional or alternative configuration, the method may further include receiving a user selection reflecting a change in one or more components of the new coating and determining a change in weathering of the new coating upon application of any of the plurality of weathering exposure conditions. In an additional or alternative configuration, the method may further include displaying, on a graphical interface, predicted changes to the new coating after a plurality of time intervals simulating application of the plurality of weathering exposure conditions.

[0073] In additional or alternative configurations, the predicted change to the new coating indicates a change in adhesion. In additional or alternative configurations, the predicted change to the new coating indicates a change in coloration. In additional or alternative configurations, the one or more machine learning algorithms employ a supervised learning algorithm to calculate the predicted weathering of the new coating formulation. In additional or alternative configurations, the supervised learning algorithm includes one or more of: (i) a random forest technique; (ii) an XGBoost tree technique; or (iii) a support vector machine technique. In additional or alternative configurations, the method may further include displaying predicted colorimetric values for the selected coating formulation and displaying one or more changes in the predicted colorimetric values for the selected coating formulation after one or more future time intervals.

[0074] In an additional or alternative configuration, the present disclosure provides a computerized system having a database, one or more computer processors, and computer-executable instructions stored on the database, the computer-executable instructions, when executed, causing the computerized system to: identify initial color data for an initial coating applied to a target object, the initial color data identifying (i) a coating having multiple components, and (ii) colorimetric data for the coating as applied; and identify weathered color data for the initial coating on the target object, the weathered color data indicating whether the coating applied to the target object changes color as a result of heat, light, and determining changed colorimetric data between the initial color data and the weathered color data, wherein the changed colorimetric data is associated with (i) each of the coating components and (ii) the plurality of weathering exposure conditions; and processing the changed colorimetric data through one or more machine learning algorithms, wherein the one or more machine learning algorithms store in a database an association between each of the plurality of components and the determined rate of change in the colorimetric data upon application of the plurality of weathering exposure conditions.

[0075] In additional or alternative configurations, the weathered color data is measured by at least a spectrophotometer. In additional or alternative configurations, the weathered color data represents a plurality of measurements of the initial coating on the target object over a multi-year period. In additional or alternative configurations, the weathered color data represents a plurality of measurements of the initial coating on the target object over a multi-month period. In additional or alternative configurations, the weathered color data represents a plurality of measurements of colorimetric data for the initial coating at a plurality of different measurement angles. In additional or alternative configurations, the system is further configured to receive an end user selection provided through the graphical user interface of a new coating having a plurality of different components, the different components varying by either one of the base coats or one or more components in the initial coating.

[0076] In an additional or alternative configuration, the system is further configured to use one or more machine learning algorithms to apply the altered colorimetric data to determine a predicted weathering morphology of the new coating upon application of the plurality of weathering exposure conditions, and display data corresponding to the predicted weathering morphology of the new coating on the graphical user interface. In an additional or alternative configuration, the system is further configured to display one or more component substitution recommendations for the new coating, the one or more component substitution recommendations being optimized by the one or more machine learning algorithms to minimize changes in appearance in response to the plurality of weather conditions. In an additional or alternative configuration, the one or more displayed component substitution recommendations include a change in concentration of any of the plurality of components.

[0077] While the present subject matter has been described in language specific to structural features and / or methodological acts, and specific embodiments of the present disclosure have been described above for purposes of illustration, it will be apparent to those skilled in the art that many variations in the details of the present disclosure can be made without departing from the scope of the appended claims. It is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above, or to the order of acts described above. Rather, the described features and acts are disclosed as exemplary forms of implementing the claims.

Claims

1. 1. A computer-implemented method for automatically determining and displaying expected weathering of a coating formulation after weathering exposure, comprising: identifying initial color data for an initial coating to be applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied; identifying weathered color data for the initial coating on the target object, the weathered color data including colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to a plurality of weathering exposure conditions, including at least exposure to heat, light, and water; determining altered colorimetric data between the initial color data and the weathered color data, the altered colorimetric data being associated with (i) each of the coating components and (ii) the plurality of weathering exposure conditions; processing the altered colorimetric data through one or more machine learning algorithms, wherein the one or more machine learning algorithms store, in a database, an association between each of the plurality of components and the determined rate of change of the colorimetric data upon application of the plurality of weathering exposure conditions; receiving an end user's selection provided through a graphical user interface of a new coating having different components; using the one or more machine learning algorithms that apply the altered colorimetric data to determine a predicted weathering morphology of the new coating upon application of the plurality of weathering exposure conditions; and and displaying on the graphical user interface data corresponding to the predicted weathering morphology of the new coating.

2. The computer-implemented method of claim 1 , wherein the weathering conditions include exposing the target object to natural weathering in a natural environment for a period of multiple months.

3. 10. A computer-implemented method according to any preceding claim, wherein the weathering conditions include simulating weathering conditions through a weathering tester.

4. further comprising identifying a historical data set for the initial coating; 10. The computer-implemented method of claim 1, wherein the historical data set includes the changed colorimetric data associated with the plurality of weathering exposure conditions and one or more different applications of the initial coating to different target objects exposed to different weathering exposure conditions.

5. 10. The computer-implemented method of any preceding claim, further comprising: using the one or more machine learning algorithms applying the historical dataset to determine an expected change in colorimetric data for each of the components associated with each weathering condition across the plurality of weathering exposure conditions and the different plurality of weathering exposure conditions.

6. determining a damage type for the initial coating after application of the plurality of weathering exposure conditions; 10. A computer-implemented method according to any preceding claim, wherein the damage type comprises a change in colour.

7. determining a damage type for the initial coating after application of the plurality of weathering exposure conditions; 10. The computer-implemented method of any preceding claim, wherein the damage type comprises a change in gloss or a change in adhesion.

8. receiving a user selection reflecting one or more changes in the composition of the new coating; 10. The computer-implemented method of any preceding claim, further comprising: determining a change in weathering of the new coating when subjected to any of the plurality of weathering exposure conditions.

9. 10. The computer-implemented method of claim 1, further comprising displaying on the graphical interface predicted changes to the new coating after multiple time intervals simulating application of the multiple weathering exposure conditions.

10. The computer-implemented method of claim 9 , wherein the predicted change to the new coating indicates a change in adhesion.

11. 11. The computer-implemented method of claim 9, wherein the predicted change to the new coating indicates a change in coloration.

12. 10. The computer-implemented method of any preceding claim, wherein the one or more machine learning algorithms employ supervised learning algorithms to calculate the expected weathering of the new coating formulation.

13. 13. The computer-implemented method of claim 12, wherein the supervised learning algorithm comprises one or more of: (i) a random forest technique; (ii) an XGBoost tree technique; or (iii) a support vector machine technique.

14. displaying predicted colorimetric values for the selected coating formulation; 10. The computer-implemented method of any preceding claim, further comprising: displaying one or more changes in predicted colorimetric values for the selected coating formulation after one or more future time intervals.

15. 1. A computerized system comprising: a database; and one or more computer processors. and computer executable instructions stored on said database, said computer executable instructions, when executed, causing said computerized system to: identifying initial color data for an initial coating to be applied to a target object, the initial color data identifying (i) a coating having multiple components and (ii) colorimetric data for the coating as applied; identifying weathered color data for the initial coating on the target object, the weathered color data including colorimetric data corresponding to the initial coating after the coating applied to the target object has been subjected to a plurality of weathering exposure conditions, including exposure to heat, light, and precipitation; determining altered colorimetric data between the initial color data and the weathered color data, the altered colorimetric data being associated with (i) each of the coating components and (ii) the plurality of weathering exposure conditions; and processing the changed colorimetric data through one or more machine learning algorithms, the one or more machine learning algorithms causing a database to store and process an association between each of the plurality of components and the determined rate of change of the colorimetric data upon application of the plurality of weathering exposure conditions.

16. 16. The computerized system of claim 15, wherein the weathered color data is measured by at least a spectrophotometer.

17. 17. The computerized system of claim 15 or 16, wherein the weathered color data represents multiple measurements of the initial coating on the object over a multi-year period.

18. 18. The computerized system of any preceding claim 15 to 17, wherein the weathered color data represents multiple measurements of the initial coating on the object over a multiple month period.

19. A computerized system according to any preceding claim 15 to 18, wherein the weathered colour data represents a plurality of measurements of colourimetric data for the initial coating at a plurality of different measurement angles.

20. The system further comprises:

20. The computerized system of any of the preceding claims 15 to 19, configured to receive an end-user selection provided through a graphical user interface of a new coating having a plurality of different components, the plurality of different components differing from any of the base coats or one or more components in the initial coating.

21. The system further comprises: using the one or more machine learning algorithms that apply the altered colorimetric data to determine a predicted weathering morphology of the new coating upon application of the plurality of weathering exposure conditions; and and displaying on the graphical user interface data corresponding to the predicted weathering morphology of the new coating.

22. The system further comprises: configured to display one or more ingredient substitution recommendations for the new coating; 22. The computerized system of claim 20 or 21, wherein the one or more ingredient substitution recommendations are optimized by the one or more machine learning algorithms to minimize changes in appearance in response to multiple weather conditions.

23. 23. The computerized system of claim 22, wherein the displayed one or more ingredient substitution recommendations comprise a change in concentration of any of the plurality of ingredients.

Images