Production analysis modeling for product quality detection

Abstract

In at least one example, a method for production analysis modeling for product quality detection includes receiving an inspection request for a product, the inspection request including an identified failure; identifying production analysis data associated with a production process for the product, the production analysis data including data captured during production of the product; providing the production analysis data for the product to a machine learning model for production of the product to compare the production analysis data to a specified range of data for the product; determining data from the production analysis data that is outside of the specified range identified by the machine learning model; and identifying a cause of the identified failure during production of the product based on the determined data.

Description

Technical Field

[0001] This disclosure relates to production analysis modeling for product quality inspection. Such techniques may be particularly useful for predicting quality or quality defects associated with products from a particular manufacturer by modeling production data from multiple different manufacturers. Background Technology

[0002] Artificial neural networks (ANNs) are networks that process information by modeling networks of neurons, such as those in the human brain, to process information sensed in a specific environment (e.g., stimuli). Similar to the human brain, neural networks typically consist of multi-neuron topologies (e.g., which may be referred to as artificial neurons). An ANN operation refers to the operation of using artificial neurons to process input to perform a given task. ANN operations can involve executing various machine learning algorithms to process the input. Example tasks that can be processed by performing ANN operations include machine vision, speech recognition, machine translation, social network filtering, and / or medical diagnosis.

[0003] Multiple manufacturing facilities provide foam formulations and expertise for manufacturing sandwich metal panels with rigid polyurethane or polyisocyanurate cores to dual-belt lamination customers. Proper foam formulation dispensing is crucial for producing high-quality panels and avoiding quality claims. Additionally, foam formulation bubbling can negatively impact panel quality. Furthermore, excessive rolling of the top layer after foam material deposition can occur and negatively affect panel quality. To address this, suppliers have developed their own foam formulation dispensers (also known as feeders), which are provided to panel manufacturers along with the foam formulations. Different foam formulation dispensers provide varying qualities of foam formulation dispensing on different surfaces and / or for different functions of the final product (e.g., panels).

[0004] In addition, proper formulation of the foam is crucial for producing panels free from potential defects that can lead to bubbling over time after panel installation and subsequently trigger claims from downstream customers. The problem is that monitoring of the foam formulation is currently minimal (if any). There are currently no quantitative checks on the foam formulation during production. Summary of the Invention

[0005] This disclosure relates to improvements in using machine learning techniques to predict product properties (e.g., quality, effectiveness, defects, malfunctions, etc.). Image-based visual analytics utilizing machine learning models can be used to monitor and analyze production processes to predict product properties based on production analysis data. Image-based visual analytics can use machine learning models to interpret and extract data from captured images by identifying relevant patterns and features within the image. These features may include, but are not limited to, edges, textures, shapes, colors, and / or other visual attributes. Predictions of product properties can be received from the machine learning model, and these predictions can be used to adjust the production process or to determine whether to reject the product.

[0006] As a concrete example, data can be collected from multiple different manufacturing sites to generate machine learning models that predict the quality of products produced by those sites. These machine learning models can be used to identify production parameters that may be the cause of specific properties. As used herein, production parameters refer to data collected during the production of a particular product. For example, production parameters used for depositing foam materials (e.g., polyurethane (PUR) or polyisocyanurate (PIR) formulations) may include, but are not limited to: foam deposition equipment data, foam deposition flow data, and foam surface distance data. Production parameters used for quality inspection of foam materials may include, but are not limited to: foam surface bubbling data when the product includes foam material, over-rolling data, formed foam edge data, and product size data, as well as other data that may potentially affect the quality of products produced using layers of foam material. Furthermore, the machine learning model can generate a range of production parameters for multiple production parameters, which can be provided to multiple different manufacturing sites to ensure product quality levels.

[0007] The foregoing description of this invention is not intended to describe every disclosed embodiment or every implementation thereof. The following description illustrates exemplary embodiments in more detail. Throughout this application, guidance is provided by a list of examples, which may be used in various combinations. In each case, the enumerated list serves only as a representative group and should not be construed as an exclusive list. Attached Figure Description

[0008] Figure 1 An example of a system for production analysis modeling used for product quality inspection is shown.

[0009] Figure 2 This is an example diagram illustrating a system for producing sandwich panels by depositing foam material onto an application surface through a continuous process.

[0010] Figure 3 Different foam material dispensers are illustrated for depositing foam material onto the application surface.

[0011] Figure 4 An example image illustrating bubbling that occurs during the deposition of foam material onto the applied surface is shown.

[0012] Figure 5 An example of excessive rolling during the deposition of foam material onto the applied surface is illustrated.

[0013] Figure 6 An example of edge dimension measurement of a product utilizing foam material on an applied surface is shown.

[0014] Figure 7 An example of a method for production analysis modeling for product quality inspection is shown.

[0015] Figure 8 An example of a machine-readable medium used for production analysis modeling in product quality inspection is shown.

[0016] Figure 9 An example of equipment used for production analysis modeling in product quality inspection is shown. Detailed Implementation

[0017] This disclosure relates to methods and apparatus for production analysis modeling for product quality inspection, which can utilize machine learning models to predict product properties for one or more products generated using different production analysis data. An example of a machine learning model is an ANN (Advanced Neural Network). An ANN provides learning by forming probabilistic weight associations between inputs and outputs. These probabilistic weight associations can be provided by multiple nodes comprising the ANN. Nodes, along with weights, biases, and / or activation functions, can be used to generate the ANN's output based on its inputs. Multiple nodes of the ANN can be grouped to form layers of the ANN.

[0018] A machine learning model can be a function or equation used to identify patterns in data. A machine learning module can be multiple machine learning models used together to identify patterns in data. In a specific example, a machine learning module can be organized as a neural network. A neural network can include a set of instructions that can be executed to identify patterns in data. Some neural networks can be used to identify underlying relationships in a set of data in a way that mimics the operation of the human brain. Neural networks can adapt to changing or altered inputs, allowing them to produce acceptable results without redesigning the output criteria.

[0019] The production of a specific product can utilize different conditions during the production process. Production conditions during the production process can have varying effects on product quality. Different types of production conditions can have different effects on the quantity of the final product produced. Production conditions can be monitored by different types of equipment. For example, imaging equipment, timing equipment, sensor equipment, and / or other types of equipment can be used to monitor different types of conditions to generate production data specific to the product.

[0020] Production data can be generated from multiple different manufacturing sites and implemented in a uniform format (e.g., production analytics data) to allow different manufacturing sites to provide federated data for generating or training machine learning models. Production analytics data may include image data collected during the production process, production setup data collected during the production process, and / or environmental data collected during the production process. Machine learning models can utilize production data or production analytics data associated with the products being produced, along with corresponding quality data, to generate production ranges that can be used by each of the multiple manufacturing sites to improve product quality. In this way, product reviews for specific products can be generated by machine learning models to identify one or more production data points from the production data attributable to specific defects or specific quality levels of the product to be reviewed by the product review.

[0021] As used herein, unless otherwise expressly stated, the singular forms “a / an” and “the” include both singular and plural indicators. Furthermore, throughout this application, the word “may” is used in a permissible sense (e.g., possibly, able to) rather than a mandatory sense (e.g., must). The term “comprising” and its derivatives mean “including but not limited to”.

[0022] As will be understood, elements shown in the various embodiments herein may be added, interchanged, and / or eliminated to provide multiple additional embodiments of this disclosure. Furthermore, as will be understood, the proportions and relative scales of the elements provided in the figures are intended to illustrate certain embodiments of the invention and should not be construed as limiting.

[0023] Figure 1 An example of a system 100 for production analysis modeling of product quality inspection is illustrated. System 100 illustrates a system for applying foam material to an application surface. Although system 100 illustrates a specific system for applying foam material to a particular surface, this disclosure is not limited thereto. For example, other types of systems may utilize the functions and / or features described herein in a similar manner for production analysis modeling of product quality inspection of other types of products.

[0024] System 100 includes a dual-belt lamination (DBL) system 102 that dispenses chemical compounds or foam materials that can be used to form rigid polyurethane (PUR) or polyisocyanurate (PIR) core foam materials on a surface. As used herein, foam material refers to a substance comprising air bubbles trapped within a liquid or solid matrix. In a first example, the foam material comprises a substance having air bubbles trapped within a liquid or solid prior to dispensing and / or a substance comprising air bubbles trapped within a liquid or solid after a chemical reaction or after dispensing of the foam material. The foam material can be dispensed onto a surface while it is in a liquid foam state. The liquid foam state can solidify over a period of time and become a solid foam state on the surface.

[0025] DBL system 102 may include a foam dispensing device that can dispense chemical compounds or foam materials onto a surface. (Reference) Figure 2 The DBL system 102 is further described. Generally, the DBL system 102 includes a foam dispersion device that deposits foam material onto a surface (e.g., an application surface, a portion of a surface, etc.) conveyed along a path of the foam dispersion device. In this way, the surface is conveyed through the foam dispersion area to receive foam material from the foam dispensing device.

[0026] Surfaces can be positioned on a delivery system to carry them to different areas, allowing the application of different substances, chemicals, or layers. For example, the delivery system can transport a surface to a first area to receive adhesion-enhancing chemicals and to a second area to receive foam material. In these embodiments, the foam material can be deposited onto the adhesion-enhancing chemicals to ensure adhesion of the foam material to the applied surface.

[0027] The movement of the surface from the first position to the second position can be monitored by a timing device to determine the amount of time it takes for the surface to move from the first position to the second position. (See reference...) Figure 5 As further described, a top application surface can be applied to a foam material on a surface. In these embodiments, a timing device can determine the amount of time between the deposition of the foam material and the application of the top application surface to the foam material. (See reference...) Figure 2As further described herein, system 100 may include additional sensors to monitor the amount of time taken for a surface to move from a location for receiving adhesive-enhancing chemicals from an adhesive dispensing device to a different location for receiving foam material from a foam dispensing device. The chemicals deposited by the adhesive dispensing device may undergo specific chemical reactions. In this way, the amount of time taken when the surface reaches the foam dispensing device can be used to determine the state of the chemical reaction. Additionally, a timing device can be used to determine the amount of time between when the foam material is deposited onto the application surface and when an additional application surface is applied to the foam material. As further described herein, the timing of applying the additional application surface to the foam material can affect the quality of the product.

[0028] System 100 may include imaging device 112. Imaging device 112 may be a device for capturing images of a surface to identify the properties of foam material deposited on the surface by a foam dispensing device. Imaging device 112 may include one or more of the following: a camera, a still image camera, a thermal imaging camera, a hyperspectral camera, and / or other types of devices capable of capturing images and / or video of the surface. Imaging device 112 may be a standardized imaging device that can be utilized by each of multiple manufacturing sites to create federated data from each of the multiple manufacturing sites. Similarly, the position and / or angle of imaging device 112 may be a standardized position and / or angle to capture images of the surface, foam, or panel at the same or similar location from the dual-belt lamination line. In this way, data captured by imaging device 112 can be correlated with data captured at different manufacturing sites.

[0029] In a similar manner, system 100 may include a light source 110. Light source 110 may generate light of different wavelengths to further highlight specific features of the surface. In some embodiments, light source 110 may be selected or changed based on the type of imaging device 112 and / or the type of data to be collected. In some embodiments, light source 110 may be a standardized light source utilized by each of multiple manufacturing sites to create federated data from each of the multiple manufacturing sites. In a similar manner, the position and / or angle of light source 110 may be a standardized position and / or angle to provide light on the surface of the foam or panel at the same or similar distance from the dual-strand lamination line. In this way, data captured by imaging device 112 can be correlated with data captured at different manufacturing sites because the illumination in the images will have the same or similar properties.

[0030] System 100 may include a control panel 106 that can be utilized by user 108. Control panel 106 may be used to display notifications generated by system 100 to inform user 108 when the current conditions or measurements of system 100 exceed a specific range. As further described herein, data collected during the production of surfaces and / or products utilizing surfaces may be used to generate production ranges that can be utilized by system 100.

[0031] Images captured by imaging device 112 can be provided to machine learning model 114 operating on edge computing device 116. Machine learning model 114 can be used to analyze the images received from imaging device 112. In other embodiments, machine learning model 114 can be used to organize data collected by system 100 into a data file that can be associated with the final product utilizing the surface. For example, humidity data, temperature data, foam distribution data of chemical layers, layer uniformity data of foam material, and / or data on the time between the deposition of foam layers and the application of an additional surface to the foam material can be correlated and associated with the product utilizing the specific surface. In this way, the quality of the surface portion over a period of time can be determined and correlated with data associated with the surface.

[0032] System 100 may include an edge computing device 116. In some examples, the edge computing device 116 is a computing device that includes processor resources and machine-readable medium for storing instructions executed by the processor resources to perform a specific function. The edge computing device 116 may be used to communicate with a remote device 124. The remote device 124 may be a cloud device that receives data from multiple manufacturing sites to augment a dataset used to train a machine learning model 114 for a specific product and / or for parts of a product, such as surfaces. The edge computing device 116 may be used to remove data associated with a specific manufacturing site that the site does not wish to share with other manufacturing sites.

[0033] In other embodiments, edge computing device 116 may utilize production ranges (e.g., production condition ranges, condition thresholds, etc.) to monitor data provided by machine learning model 114. When data received by machine learning model 114 exceeds a threshold range, edge computing device 116 may send a notification to control panel 106 to inform user 108. In this way, production data and / or production settings can be changed in real time. When edge computing device 116 determines that production data for a particular product exceeds a data threshold range, the panel may be discarded or marked as a second-class product. As described herein, when remote device 124 is utilizing a machine learning model, the threshold range may be provided by remote device 124.

[0034] As used herein, the production settings can be adjustable settings that define production parameters for how system 100 deposits foam material onto a surface and / or how it produces a product including the foam material. For example, the production settings may include, but are not limited to: the location of the imaging device 112 and / or the light source 110 for altering or influencing the image; the type of foam dispensing device; the flow rate of the foam dispensing device; the back pressure provided by the foam dispensing device; the mixing ratio of the foam material deposited by the foam dispensing device; the amount of components in the foam material; the gelation time of the foam material; the contact time between the foam material and the application surface; the operating duration of the foam dispensing device; the temperature of the foam material when deposited by the foam dispensing device; and / or the linear velocity at which the surface moves from a first position to a second position to alter or influence the dispensing. In this way, system 100 can change or adjust the production settings to alter or adjust how the foam dispensing device applies foam material to the surface of the application surface. In this way, the production settings can be changed to ensure that production data is within threshold ranges provided by a machine learning model.

[0035] Remote device 124 can receive data from edge computing device 116 and data organized in the same manner from multiple other manufacturing sites 120. In this way, remote device 124 can utilize machine learning models to perform advanced data analysis 118 on the data received from edge computing device 116. Remote device 124 can use this data from edge computing device 116 and other manufacturing sites 120 to generate supporting knowledge for control panel 106. As further described herein, supporting knowledge may include product review analysis for specific products manufactured at a specific time and / or a range of production data that can be used by edge computing device 116 to generate real-time notifications to control panel 106.

[0036] Figure 2This is an example diagram illustrating a system 202 for depositing foam material 239 onto an application surface 234 via a continuous process to produce a sandwich panel. System 202 may include multiple elements for producing a product comprising foam material 239. For example, system 202 may include elements for generating a panel comprising foam material. In some embodiments, system 202 includes a corona device 231. As used herein, corona device 231 may be a surface treatment apparatus that uses corona discharge to modify the surface properties of the application surface 234. In some embodiments, corona device 231 may utilize corona discharge (e.g., low-temperature plasma, etc.) having reactive ions, electrons, and / or free radicals to improve surface activation. In some embodiments, corona device 231 includes cleaning processes, adhesion improvement processes, and other processes to improve the adhesion of chemicals to the surface of application surface 234. In some embodiments, a delivery system 236 may be used to move the application surface 234 and / or top application surface 235 along a path to allow the application surface 234 and / or top application surface 235 to interact with different components of system 202.

[0037] In some embodiments, the delivery system 236 may move the application surface 234 from the corona treatment device 231 to the adhesive dispensing device 232. The adhesive dispensing device 232 may apply an adhesion enhancement layer (e.g., a primer, etc.) to the application surface 234 that has been treated by the corona treatment device 231. In these embodiments, the delivery system 236 may move the application surface 234 from the adhesive dispensing device 232 to a foam dispensing device 233, which may dispense foam material 239 onto the application surface 234. In some embodiments, the foam dispensing device 233 may deposit the foam material 239 onto the adhesion enhancement layer.

[0038] As described herein, foam material 239 can be used to form a rigid polyurethane (PUR) or polyisocyanurate (PIR) core foam material on application surface 234. Foam material 239 can be deposited as a substance including bubbles trapped within a liquid. Liquid foam material 239 can be cured into a solid foam state between application surface 234 and top application surface 235.

[0039] As further described herein, imaging devices (e.g., such as...) Figure 1 The imaging device 112 (as mentioned above) can be used to monitor the foam dispensing device 233 and the foam material 239 applied to the application surface 234. In this way, the type of foam dispensing device 233 can be determined, the performance of the foam dispensing device 233 can be monitored during the dispensing of the foam material 239, the chemical composition of the foam material 239 can be monitored, and / or the surface properties of the foam material 239 can be monitored as the delivery system 236 moves the application surface 234 and / or the top application surface 235.

[0040] Foam material 239 can expand or increase in height between foam dispensing device 233 and contact point 240. Contact point 240 can be the location where foam material 239 physically contacts the top application surface 235. In some embodiments, the amount of time between the deposition of foam material 239 and contact point 240 can be from 1 second to 40 seconds. More specifically, the amount of time between the deposition of foam material 239 and contact point 240 can be from 10 seconds to 30 seconds. See reference... Figure 5 As further described, the position of contact point 240 can be monitored using a laser or similar sensor device to prevent excessive rolling by adjusting the speed of conveying system 236 using contact point 240. In a specific example, a time-of-flight sensor can be used in conjunction with a laser to determine when the foam material 239 makes physical contact with the top application surface 235. For example, the time-of-flight sensor can determine the amount of time it takes for the laser to reach contact point 240, and this amount of time can be used to determine the distance and / or location of contact point 240.

[0041] In this way, imaging equipment and / or other sensors can be used to monitor the performance of foam dispensing device 233, the surface properties of foam material 239, the amount of time between dispensing of foam material 239 and contact point 240, and / or the chemical composition or chemical ratio of foam material 239. The performance of foam dispensing device 233, the surface properties of foam material 239, the amount of time between dispensing of foam material 239 and contact point 240, and / or the chemical composition or chemical ratio of foam material 239 can be used to generate production analysis data and / or to generate notifications when the data exceeds a threshold provided by a machine learning model.

[0042] In some embodiments, the delivery system 236 moves the foam material 239, the application surface 234, and the top application surface 235 to the cutter 237. The cutter 237 can be used to cut panels of a specific length. As further described herein, imaging equipment can be used to analyze the edge portion 238 at the cut to determine the presence of inherent defects and / or whether the cut edge conforms to specific size and dimensional specifications. In some embodiments, the foam material 239 is allowed to partially solidify before cutting is performed at the cutter 237. As referenced herein... Figure 6 As further described, the edge portion 238 can be further analyzed to include additional production analysis data.

[0043] Figure 3Examples of different foam material dispensers 333-1, 333-2 for depositing foam material 339 onto an application surface 334 are illustrated. As described herein, the application surface 334 may be prepared to receive the foam material 339 using a corona discharge device and / or an application adhesion enhancer layer. The foam material dispensers 333-1, 333-2 may be of different types, applying the foam material 339 to the application surface 334 at different flow rates, angles, and / or distances from the application surface 334. In some embodiments, the type of foam material dispenser may affect the quality of the final product based on its intended use. The type of foam material dispenser may affect the properties of the final product. These different properties may be beneficial or detrimental to the effectiveness of the product depending on how it is used.

[0044] In some embodiments, foam material dispensers 333-1, 333-2 can be used as systems for applying foam material 339 to application surface 334 (e.g., as shown in the image). Figure 1 The DBL system 102 mentioned in the document, such as Figure 2 As part of the system 202 mentioned above. In some embodiments, the imaging device (e.g., such as...) Figure 1 The imaging device (such as 112 mentioned above) can be used to identify the type of foam material dispensers 333-1 and 333-2. In some embodiments, the imaging device can utilize the physical or functional characteristics of the foam material dispensers 333-1 and 333-2 to determine the type of the foam material dispenser. For example, the imaging device can capture images of the size, shape, angle of the deposited foam material 339, flow rate of the deposited foam material 339, and / or other characteristics of the foam material dispensers 333-1 and 333-2, which can indicate a specific model or type of foam material dispenser. In other embodiments, the imaging device can capture codes (e.g., barcodes, QR codes, etc.) to identify the type of foam material dispensers 333-1 and 333-2.

[0045] As described herein, imaging equipment can be used to capture images of foam material dispensers 333-1, 333-2 and deposited foam material 339 during the foam dispensing process. In some embodiments, the imaging equipment can monitor the flow rate of foam material 339 deposited from foam material dispensers 333-1, 333-2. As used herein, the flow rate of foam material 339 refers to the amount of foam material 339 deposited by foam material dispensers 333-1, 333-2 over a period of time. During operation of foam material dispensers 333-1, 333-2, the outlets (e.g., nozzles, etc.) dispensing foam material 339 may become clogged with foam material 339, causing a reduction in flow rate. In some embodiments, the imaging equipment can be used to capture images that can be used to determine when the flow rate drops below a threshold flow rate. As illustrated by foam material dispenser 333-1, the flow rate of the first dispensing flow 342-1 may be less than that of the second dispensing flow 342-2.

[0046] In some implementations, the threshold flow rate may be based on a machine learning model that allows the threshold flow rate to be based on identified flow rates and quality metrics of the corresponding final product from multiple manufacturing sites. That is, the flow rate from the outlet can be monitored and provided as production analytics data. In these implementations, the machine learning model can use the production analytics data along with the corresponding quality data to determine the threshold flow rate. In this way, the threshold flow rate can be updated by the machine learning model as additional data is provided from multiple different manufacturing sites. As described herein, a notification can be generated when the flow rate of the distribution stream from the outlet falls below the threshold flow rate, allowing production to be stopped to clean the outlet, thereby increasing the flow rate before restarting production. In other implementations, when the flow rate of the distribution stream falls below the threshold flow rate during the production of various parts of the final product, those parts can be identified as not conforming to specific manufacturer specifications.

[0047] In some embodiments, once the foam material 339 is applied to the application surface 334, the imaging device can capture a surface image of the foam material 339. In some embodiments, the surface image can be used to determine surface properties associated with each of a plurality of distribution streams. For example, as illustrated by the foam material dispenser 333-2, the surface image captured by the imaging device can determine a first width 343-1 associated with a first distribution stream and a second width 343-2 associated with a second stream. In some embodiments, corresponding widths can be identified for a plurality of additional distribution streams.

[0048] In some embodiments, multiple widths associated with multiple distribution streams can be used to determine the amount of foam material 339 deposited for each of the multiple distribution streams or distribution nozzles. In some embodiments, the multiple widths can be used to ensure that the foam material 339 extends across the entire surface of the application surface 334 without space between the foam material 339 distributed by the first distribution stream and the foam material 339 distributed by the second distribution stream. In some embodiments, the multiple widths associated with a particular surface can be utilized in conjunction with production analysis data provided to a machine learning model. In other embodiments, the multiple widths can be compared to a threshold width provided by a machine learning model.

[0049] As further described herein, surface images captured by an imaging device can also be used to determine the chemical composition and / or mixing ratio associated with foam material 339. For example, the imaging device can be a hyperspectral imaging device. As used herein, a hyperspectral imaging device is an optical device designed to capture and analyze electromagnetic spectra in a large number of narrow, continuous spectral bands or channels. In some embodiments, the hyperspectral imaging device may capture multiple spectral bands spanning the visible, near-infrared, and / or infrared portions of the spectrum. In these embodiments, each band or channel in the captured hyperspectral image corresponds to a specific narrow wavelength range. Spectral information can be used to determine the chemical composition or chemical ratio of foam material 339 during the production process.

[0050] Similar to other production analysis data, the captured hyperspectral images can be included as production analysis data for specific surfaces and / or specific products. Machine learning models can utilize this data, along with corresponding quality data, to generate chemical composition threshold ranges that can be used to monitor the chemical formulation of foam material 339 during production.

[0051] Figure 4 An example image 444 illustrates bubbling 445 that occurs during the deposition of foam material onto an application surface. As described herein, an imaging device can capture an image of the surface of foam material 439. In some embodiments, the flow rate of foam material 439 or other factors associated with the distribution of foam material 439 onto the application surface can cause bubbling 445. Bubbling 445 can be caused by gas (e.g., air, etc.) trapped within and / or between foam material 439 and the application surface.

[0052] In some embodiments, the trapped gas that causes foaming 445 can create defects 446 within the foam material 439 as it is fully formed. For example, the foam material 439 can solidify to the point where the liquid foam material 439 becomes a solid foam material 439. In these embodiments, the solid foam material 439 may include defects 446, which can result in relatively lower quality compared to foam material 439 without defects 446.

[0053] In some examples, the imaging device can capture surface images of the foam material 439, which can be analyzed to determine foaming 445 based on color differences or color patterns along the surface of the foam material 439. For example, foaming 445 can be identified as different colors in the surface images captured by the imaging device. In these examples, surface images can be monitored during the production process to determine when the color difference along the surface of the foam material exceeds a threshold color difference. The threshold color difference can be provided by a machine learning model based on federated data from multiple manufacturing sites. In some implementations, a notification can be generated when the color difference detected during the production process exceeds the threshold color difference. This can allow changes to settings associated with the foam dispersion equipment. For example, the flow rate of the foam material 439 can be changed, the height between the dispensing nozzle and the application surface can be adjusted, and / or other production settings can be changed to reduce foaming 445.

[0054] Figure 5 An example system 502 is illustrated, which undergoes excessive rolling 551 during the deposition of foam material onto an application surface. In some embodiments, system 502 may include... Figure 1 The DBL system 102 illustrated herein and / or as shown Figure 2 The system 502 may contain elements identical or similar to those mentioned in the previous system 202. For example, system 502 may include a dispensing device 533 to deposit foam material 539 onto an application surface 534. Additionally, system 502 may include a conveying system 536 to move the application surface 534 to different locations. In some embodiments, the conveying system 536 may move the application surface 534 with the foam material 539 to a contact point where a top application surface 535 contacts the foam material 539. Furthermore, the conveying system 536 may move the application surface 534, the foam material 539, and the top application surface 535 to a cutter.

[0055] As described herein, imaging devices (e.g., such as...) Figure 1The imaging device 112 (as mentioned above) can be used to monitor the foam dispensing device 533 and the foam material 539 applied to the application surface 534. In this way, the type of foam dispensing device 533 can be determined, the performance of the foam dispensing device 533 can be monitored during the dispensing of the foam material 539, and / or the surface properties of the foam material 539 can be monitored as the conveying system 536 moves the application surface 534 and / or the top application surface 535.

[0056] Foam material 539 can expand or increase in height between the foam dispensing device 533 and the contact point 540 between the foam material 539 and the top application surface 535. The amount of time between the deposition of foam material 539 and the contact point 540 is generally referred to as the contact time. Under some processing conditions, the difference between the gelation time and the contact time can be approximately 5 to 7 seconds. In this way, imaging equipment can be used to monitor the performance of the foam dispensing device 533, the surface properties of the foam material 539, the amount of time between the dispensing of foam material 539 and the contact point 540, the measured difference between the gelation time and the contact time, and / or the chemical composition or chemical ratio of the foam material 539.

[0057] Excessive rolling 551 may occur when the difference between the gel time and the contact time is not well controlled. As used herein, the term excessive rolling 551 refers to the situation where the foam material 539 rolls over the top application surface 535 in a direction opposite to the direction of travel of the surface layer itself. Excessive rolling 551 can lead to impaired mechanical properties of the final product (e.g., a panel, etc.) and / or include the possibility of exhibiting shrinkage or excessively low compressive strength values. Excessive rolling 551 can be controlled by monitoring the difference between the gel time and the contact time. In some embodiments, a laser may be used to calculate or measure the contact time of the foam material 539 to determine when the top application surface interacts with the foam material 539.

[0058] In some implementations, the gelation time and / or contact time may include upper and lower thresholds that can be provided by a machine learning model. In some implementations, defects associated with over-rolling 551 can be used to determine the upper and lower thresholds, such that a notification can be generated when possible over-rolling 551 has occurred at a particular manufacturing location.

[0059] In some embodiments, the delivery system 536 can move foam material 539 between the application surface 534 and the top application surface 535 to a cutter. The cutter can be used to cut edges with specific size and dimensional specifications. (See references herein.) Figure 6 As further described, imaging equipment can be used to analyze the edge portion at the cut section to determine whether there are inherent defects and / or whether the cut edge conforms to specific size and dimension specifications.

[0060] Figure 6 An example of edge dimension measurement of product 660 utilizing foam material 639 on an application surface is illustrated. In some embodiments, product 660 is a panel comprising foam material 639 between a first application surface and a second application surface. For example, foam material 639 may be positioned between a top surface and a bottom surface, such as... Figure 6 As illustrated in the illustration. In some embodiments, the imaging device may capture images of the edges of product 660 to identify specific edge dimensions and / or identify the quality of foam material 639. In these embodiments, foam material 639 may be fully cured or substantially cured, making defects 666, pores, or other potential quality problems observable.

[0061] In some implementations, color anomalies in images captured by the imaging device can be analyzed to identify defects 666 or other anomalies within the foam material 639. The images captured by the imaging device may also include hyperspectral information that can be used to determine the composition of the foam material. The image data can be correlated with production data of product 660, such that defect 666 can be correlated with settings and / or captured images during the production process.

[0062] In some implementations, images captured by the imaging device may be analyzed to determine multiple different size measurements (e.g., measurements of “X”, “Y”, “A”, and / or “B”, etc.) and compared with industry standards or the size specifications of product 660. For example, a reference distance 661 may be compared with a specification reference distance to determine whether the reference distance 661 is within a specific error limit. In some implementations, the specific error limit may be determined based on the specifications of product 660 and / or from a machine learning model.

[0063] In other embodiments, the top distance 662 may be the distance between multiple ridges, and the ridge height 663, the first edge dimension 664, and / or the second edge dimension 665 may be determined using images from an imaging device. In these embodiments, measurements from the images may be compared to dimensional specifications of the product 660. In these embodiments, specific features or measurements associated with the first edge dimension 664 and / or the second edge dimension 665 may be compared to dimensional specifications to ensure that the product 660 is within acceptable tolerance limits of the dimensional specifications.

[0064] In other embodiments, images from an imaging device can be used to identify potential defects visible on the top or bottom surface of the final product. For example, the top or bottom surface may include ripples or visual defects that may occur in the presence of different defects associated with the foam material 639, the application layer, and / or the adhesive reinforcement layer. In these embodiments, ripples or visual defects can be identified from the image by color changes or color patterns on the surface. In these embodiments, the identified defects may be correlated with data collected during the production of product 660.

[0065] Figure 7 An example of a method 770 for production analysis modeling for product quality inspection is illustrated. In some examples, method 770 may be performed by a computing device as described herein. Method 770 may be used to identify specific production data that may be the cause of a particular defect and / or to identify threshold ranges of specific production data that can be utilized by a particular manufacturing site.

[0066] At step 771, method 770 may include receiving a review request for the product, which includes identified defects related to the product's foam material. As used herein, the review request may be a request to analyze a specific product. The review request may include an identifier that can be used to identify the product's production analysis data. As used herein, the identifier may be an indicator or a unique identifier that allows the product to be associated with production analysis data. In this way, specific production analysis data utilized during the production of the product (e.g., image data, calculations performed from image data, environmental data, production setup data, etc.) can be compared with other production analysis data of other products. The identified defect may be an indication of a quality level below a threshold quality level. For example, the identified defect may be part of the product that is visually or functionally below the product's threshold quality level.

[0067] The review request may be provided by one of several different users. For example, the review request may be provided by the product manufacturer, the product's customer, the product's seller, the product's distributor, or other users associated with the product. A physical product may not be required to identify the product's production analysis data. For example, an identifier may be provided within the review request and used to identify the product's production analysis data. The identified faults can be used to (further) train a machine learning model as a known output to the product's production analysis data as input. In this way, the product's production analysis data can be associated with the identified fault data.

[0068] At step 772, method 770 may include identifying production analysis data associated with the manufacturing process for applying foam material to a product. Production analysis data includes multiple data points captured during the production of the product. In some embodiments, this includes applying a top layer during foam application, after cutting, and / or after the foam material has dried, as well as at other stages of the manufacturing process. For example, production analysis data may include multiple data points captured during the application of foam material to the application surface of the product. In another example, production analysis data may include multiple data points captured when the top layer is applied to the surface of the foam material. Furthermore, production analysis data may include multiple data points captured after the foam material has dried. The combination of production analysis data at different stages or processes can allow for the identification of the root cause of defects or identified failures.

[0069] Identifying production analytics data for a specific product may include using product identifiers to extract production analytics data associated with the production process of that product. By identifying production analytics data associated with the production process of a product, data collected during the production of a specific product can be compared with data collected during the production of similar or the same type of product.

[0070] Production analysis data for a product may include, but is not limited to: thermal imaging data or hyperspectral image data for determining the surface texture of the foam material layer of the product; hyperspectral imaging data for determining the chemical formulation data of the foam material layer; process timing data associated with the time of foam material layer deposition; foam deposition equipment data; foam deposition flow data; foam surface distance data; foam surface bubbling data; over-rolling data; formed foam edge data; and product size data when the product includes foam material; and / or environmental data when the product is manufactured at a specific production site. As described herein, identifying production analysis data for a product may include using identifiers to determine a specific date and time corresponding to when the product or a portion of the product was manufactured by a specific manufacturer. In this way, data from a specific manufacturer can be used to identify production analysis data during a specific date and time of production of the product or a portion of the product.

[0071] As described herein, production analytics data may include production data and analytics data during the production of a product and / or multiple parts of the product. In this way, the failure points of identified faults (e.g., the part of the product where the fault occurred) can be identified by machine learning models, and the part of the product where the fault occurred can be further analyzed.

[0072] At step 773, method 770 may include providing production analysis data of the product to a machine learning model used for producing the product, to compare multiple data points from the production analysis data with specified areas for applying foam material to the product. Providing the production analysis data to the machine learning model may include using the machine learning model to compare the product's production analysis data with production analysis data of multiple other products. These other products may be produced in a similar manner to the product and may be manufactured at the same or different manufacturing sites. For example, the same type of product may be generated or produced at multiple different manufacturing sites. In this way, the machine learning model may identify other products that include the same or similar production analysis data to determine when production conditions exceed threshold production conditions used for manufacturing the product. Additionally, the machine learning model may use the production analysis data to identify the cause of identified failures. For example, the machine learning model may identify one or more production data values ​​within the production analysis data that exceed a threshold range that can be determined or calculated by the machine learning model.

[0073] Training the machine learning model may also include identifying a first plurality of products, which includes production analysis data falling within a specific range of production analysis data for products including the identified faults or defects. Identifying the first plurality of products may include comparing production analysis data for multiple similar products or products of a specific type. The first plurality of products may be identified as similar manufactured products produced under conditions similar to those of the product (e.g., the product to be analyzed). The first plurality of products may be products manufactured under the same or similar production conditions. In this way, the first plurality of products can be analyzed or used to train a machine learning model to identify products including those with identified faults that are the same as or similar to those of the product. The first plurality of products may have different degrees of identified faults, which can be used to determine the cause of the identified faults by comparing the degree of the identified faults with the production analysis data of the product and the first plurality of products.

[0074] In these examples, training the machine learning model also includes identifying a second plurality of products that include the identified faults. Identifying the second plurality of products may include comparing the identified faults of a particular product (e.g., the product to be analyzed, etc.) with defects or faults associated with a plurality of other products. The second plurality of products may be identified as similar faulty products that include the same or similar types of faults as the particular product to be analyzed. The machine learning model is able to identify the second plurality of products based on the characteristics of the identified faults. For example, “bubbles” or “bumps” in the surface of a product may have specific characteristics or properties. In this example, the machine learning model may identify a second plurality of products that include similar “bubbles” or “bumps.” In these embodiments, the machine learning model may identify production analysis data associated with the second plurality of products to identify potential faults in the products. For example, production analysis data of products similar to the second plurality of products may be identified as causes of faults. The identified faults and / or the identified causes of faults may be used to update and / or train the machine learning model.

[0075] Method 770 may include comparing an identified fault in a product with a fault state in a first plurality of products. In this manner, the identified fault can be provided to a machine learning model, which can then compare the identified fault with the fault state in the first plurality of products. In some embodiments, a portion of the first plurality of products may be identified as having the same or similar production analysis data and the same or similar defects as a product being analyzed that has the identified fault. As described herein, quality data or fault states of the first plurality of products may be compared with the identified fault to determine whether the same or similar type of fault exists between the product and the first plurality of products.

[0076] As used herein, a fault status may include an indication of whether the product has or does not have an identified fault (e.g., an identified defect, etc.). In other embodiments, when an identified fault is present, the fault status may include an indication of the type of fault. In these embodiments, production analysis data for the product is compared with production analysis data corresponding to a second or more products. As described herein, production analysis data or production conditions for the second or more products may be compared with the production analysis data or production conditions for the product to identify similarities. The identification of similarities may be further analyzed to determine whether these production conditions are a cause or potential cause of the identified fault.

[0077] Machine learning models can be trained using production analytics data from multiple data sources associated with different production sites (e.g., manufacturing sites) that produce components of a product or similar components associated with the product. In some examples, different production sites may manufacture or produce a portion of the product. In these examples, production analytics data can be generated in a unified manner to allow for federated use of the data. That is, data from multiple different production sites can be used to generate and / or train machine learning models.

[0078] As described herein, production analysis data from multiple data sources includes foam surface distance data, foam surface foaming data, and / or over-rolling data when the product includes foam material. Foam surface distance data, foam surface foaming data, and / or over-rolling data can be used to determine the uniformity of a particular foam material at a specific location on the application surface. In this way, a product may have been manufactured using a specific portion or section of the application surface, and the uniformity of the foam material layer within that specific portion or section can be identified and associated with the product, while other unused portions or sections can be associated with other products. Production analysis data may include environmental data and process timing data associated with the time of foam material layer deposition. Environmental data may include, but is not limited to: humidity at the manufacturing site, temperature at the manufacturing site, concentration of specific substances at the manufacturing site, and / or other site characteristics related to the environment or conditions of the manufacturing site.

[0079] As described herein, process timing data or contact time can refer to the amount of time between the deposition of the foam material and the application of the top application surface. In some examples, the application of the top application surface may be timed to be applied during a specific stage of the foam material's curing process, such as the gel time of the foam material. As described herein, the gel time of the foam material can be monitored and can be affected by the mixing ratio and / or environmental conditions. Process timing data, together with curing time data, can be used to determine which stage of curing the foam material layer is in when the top application surface is applied.

[0080] At step 774, method 770 may include determining data from production analysis data that exceeds a specified range identified by the machine learning model. In some examples, the production analysis data may be specific data comparing the product with multiple other products that include the same or similar identified faults. In other examples, the specified range identified by the machine learning model may be generated based on previously collected data related to other products similar to the product.

[0081] At step 775, method 770 may include identifying the cause of a fault identified during the production of the product based on the determined data. The determined data may be outside the range defined by the machine learning model. For example, the machine learning model may generate a range of production analysis data based on historical production analysis data and / or the identified faults. In this example, the machine learning model may determine that one or more data points fall outside these generated ranges, and determine that the cause of the fault is due to one or more data points falling outside the generated range.

[0082] Figure 8 An example of a machine-readable medium 880 for production analysis modeling of product quality inspection is illustrated. The machine-readable medium 880 is communicatively connected to a processor resource 881 via a communication path 882. In some examples, the communication path 882 may include a wired or wireless connection that allows communication between devices and / or components within a single device. As used herein, the processor resource 881 may include, but is not limited to: a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a metal programmable cell array (MPCA), a semiconductor-based microprocessor, or other combinations of circuitry and / or logic for coordinating the execution of instructions 876, 877, 878, 879. In a specific example, the processor resource 881 utilizes a non-transitory computer-readable medium storing instructions 876, 877, 878, 879, which, when executed, cause the processor resource 881 to perform the corresponding function.

[0083] Machine-readable storage medium 880 can be an electronic, magnetic, optical, or other physical storage device that stores executable instructions. Therefore, a non-transitory machine-readable medium (MRM) (e.g., machine-readable medium 880) can be, for example, a non-transitory MRM including random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), a storage drive, an optical disk, etc. Machine-readable medium 880 can be disposed within a controller and / or computing device. In this example, executable instructions 876, 877, 878, and 879 can be "installed" on the device. Additionally and / or alternatively, machine-readable medium 880 can be a portable, external, or remote storage medium, for example, allowing a computing system to download instructions 876, 877, 878, and 879 from a portable / external / remote storage medium. In this case, the executable instructions can be part of an "installation package".

[0084] Machine-readable medium 880 includes instructions 876 for training a machine learning model for applying foam material to the production process of a certain type of product using production analysis data captured at multiple different locations during the production process of that type of product. In these examples, the machine learning model correlates the production analysis data of applying foam material with the quality level of the foam material for the product type. As described herein, this type of product may include an application layer or application surface having a layer of foam material applied during the production process.

[0085] Machine-readable medium 880 may include instructions for generating a machine learning model for the production process of a product using production analytics data captured at multiple different locations during the production process. The machine learning model can correlate the production analytics data with the quality level of the product. As described herein, the machine learning model can be generated or trained using production data provided by multiple different manufacturing sites. Different manufacturing sites may be instructed to collect data in the same or similar manner. For example, different manufacturing sites may be instructed to collect the same type of data (e.g., infrared images, hyperspectral images, laser measurement data, etc.). In some examples, different manufacturing sites may be instructed to collect data at specific times (e.g., the amount of time after depositing a specific layer, etc.). Additionally, different manufacturing sites may be instructed to collect different environmental data during the production of the product.

[0086] Data collected from different manufacturing sites can be organized into production analytics data, which can be provided as federated data to train or generate machine learning models. Different manufacturing sites can be instructed to analyze raw data and provide specified analyses to the machine learning model. For example, different manufacturing sites can be instructed to calculate foam surface distance data for a specific portion of the substrate as it moves along a conveyor system. In this way, foam surface distance data can be provided to the machine learning model without having to provide raw data that may include sensitive information. In this way, different manufacturing sites can each provide data to the machine learning model without risking the provision of sensitive information. In some implementations, raw data refers to unprocessed, unfiltered, and unmodified data collected directly from the source. For example, raw data could be image data collected directly from an imaging device without modification or updating. This type of raw data may include sensitive information that may not be wanted to be disclosed to external sources.

[0087] Machine-readable medium 880 includes instructions 877 for providing site-specific production analysis data to a machine learning model. In these examples, the production analysis data includes image data and environmental data over a period of time during the application of foam material for that type of product at the site. In some examples, production analysis data from different manufacturing sites can be used to generate or train the machine learning model. The trained machine learning model can be used to analyze the production of that type of product at the site. The site can be one of several manufacturing sites; however, it can also be a different site that does not provide production analysis data to train the machine learning model. The production analysis data from the site can be compared with production analysis data from different sites to determine whether the site is producing a product within the range calculated by the machine learning model.

[0088] Machine-readable medium 880 includes instructions 878 for identifying data of a production process based on site-specific production analysis data, which includes values ​​beyond the range determined by a machine learning model. Multiple data measures captured at various locations during the production process may include thermal imaging data to determine foam distribution and surface uniformity. As described herein, data measures captured at various locations during the production process may include production settings utilized during the production process. For example, the production settings may be adjustable settings that define production parameters for mechanical components used to distribute layers of foam material onto a surface. Alternatively, the production settings may be settings used to collect image data or other data to be used as production analysis data.

[0089] Multiple data metrics captured at multiple different locations during the production process may include hyperspectral imaging data to determine chemical formulation data for foam material layers. These multiple data metrics can also be federated data received from multiple different locations. As used herein, federated data refers to a method in which data from multiple sources or entities remain separate and distributed, while still allowing for collaborative analysis and insights to be derived from the combined data. Instead of centralizing data in a single location or organization, a federated data system enables data to remain distributed and fragmented across various independent entities or locations. In some implementations, each entity or organization retains control and ownership of its own data, thereby protecting data privacy and security. The federated approach does not involve directly sharing or aggregating data, but rather involves implementing protocols that enable data analysis and processing to be performed across distributed data sources without requiring data movement or direct access.

[0090] Machine-readable medium 880 includes instructions 879 for generating notifications to change production settings for a site-specific production process to alter values ​​within a range determined by a machine learning model. The notification may be an instruction to change specific production settings of the production process. In other embodiments, the instruction may be to perform the change without human interaction. Changing the production process settings may include sending a notification to a site-specific control panel instructing the site to change specific settings of the production process (e.g., foam dispensing flow rate, foam dispensing equipment cleaning, chemical composition of the foam material, etc.), which may be adjusted by the control panel or by a human user. In some examples, the foam material layer may include multiple chemicals that can be dispensed by the foam dispensing equipment. The flow rate of the dispensing equipment may change the foam surface distance, foam surface foaming, and / or excessive rolling of the foam material. In other examples, changing the settings may include updating the production process range utilized by the site. For example, the site may utilize a first operating duration of the foam dispensing equipment, and the machine learning model may indicate that the site should instead utilize a second operating duration of the foam dispensing equipment. As described herein, the foam dispensing equipment may become clogged with foam material after a certain amount of time. Updating the operation duration allows the foam dispensing device to be used for a longer period of time, or ensures that the flow rate of the foam dispensing device does not drop below the threshold flow rate during operation.

[0091] In other examples, changing production process settings may include modifying the production process at a specific location. For instance, a control panel may be accessible, and production process settings can be changed. In this example, foam dispersion equipment settings may be changed to alter the flow rate to a value within the range specified by the machine learning model. Other types of settings or processes may be changed based on the identified cause of the malfunction.

[0092] Machine-readable medium 880 may include instructions to update a machine learning model using production analytics data captured at a specific location. The production analytics data from that specific location can be used to update the machine learning model. The updated machine learning model can be used to modify production processes at multiple different locations. For example, the data range of a specific piece of production analytics data may change based on the production analytics data provided by the specific location. In this example, the updated range may be provided to multiple different locations.

[0093] Figure 9 An example of a device 901 for production analysis modeling of product quality inspection is illustrated. In some examples, device 901 is a computing device that includes processor resources 981 and machine-readable medium 980 for storing instructions 991, 992, 993, 994, 995 that are executed by processor resources 981 to perform specific functions. Figure 9This illustrates how a computing device can execute instructions to perform the functions described herein. Device 901 can be a machine learning model operating on a computing device.

[0094] Device 901 can be communicatively coupled to imaging device 912 via communication path 982. As described herein, when foam material is applied to application surface 934, imaging device 912 can capture an image of application surface 934. The captured image from imaging device 912 can be transmitted to device 901 via communication path 982, where the captured image can be analyzed.

[0095] Device 901 includes instructions 991 stored by machine-readable medium 980, which are executed by processor resource 981 to monitor received image data of application surface 934 during the application of a foam layer onto application surface 934 in a production environment. Image data can be received from imaging device 912 via communication path 982. Image data can be infrared image data, hyperspectral image data, visual image data, and other types of image data. It can be generated by machine learning models (e.g., such as...) Figure 1 The machine learning models mentioned in the document 114, etc.) and / or edge computing devices (e.g., such as Figure 1 The edge computing device 116 mentioned in the text is used to analyze image data.

[0096] The device 901 includes instructions 992 stored in the machine-readable medium 670, which are executed by the processor resource 981 to calculate production analysis data associated with applying a foam layer to an application surface based on image data. The production data may include, but is not limited to: production setup, environmental characteristics of different locations over a period of time, foam distribution, surface uniformity of the foam layer, and the amount of time between depositing the foam layer and applying the top application surface to the foam layer, as well as other data that may be collected during the application of the foam layer.

[0097] The device 901 includes instructions 993 stored by a machine-readable medium 980, which are executed by a processor resource 981 to provide production analysis data to a machine learning model for comparison with threshold data of the machine learning model. In these embodiments, the machine learning model utilizes production data and quality data from multiple different production sites that produce a product including an application surface 934 and a foam layer on the application surface 934. In some embodiments, the machine learning model may correlate the production analysis data with multiple quality data associated with the product including the application surface 934. As described herein, multiple production data and / or production analysis data may be correlated with multiple quality data. In this way, production data can be combined with quality data such that specific quality data for a particular product can be utilized together with production data for that particular product. Production data and quality data may be correlated under a specific product identifier to associate a particular product with the production data and quality data.

[0098] Device 901 includes instructions 994 stored by machine-readable medium 980, which are executed by processor resource 981 to identify when production data exceeds a threshold range identified by a machine learning model. As described herein, production data can be monitored in real time and over a period of time, which might be impossible without an imaging device such as imaging device 912. For example, a surface area with an applied layer of foam material can be conveyed to have an additional layer deposited or attached, such that the additional layer is applied within a threshold time period. In this way, monitoring production data using a machine learning model can prevent products from being delivered to customers.

[0099] As described herein, multiple production data points and multiple quality data points can be provided to a machine learning model as production analytics data. Production analytics data can be used as federated data from multiple different production sites. The multiple production data points and multiple quality data points provided to the machine learning model may lack the raw data collected by imaging device 912 (e.g., raw data is not included). As used herein, raw data refers to data collected during the production process. Production data, quality data, and / or production analytics data may not include the actual data collected, but rather include calculations using the actual data collected. In this way, a higher level of security is provided for multiple production sites. Multiple production data points and multiple quality data points can be provided to a machine learning model to train the machine learning model.

[0100] Device 901 includes instructions 995 stored by machine-readable medium 980, which are executed by processor resource 981 to generate a notification for changing the production settings for applying the foam layer based on identified production data. As described herein, device 901 may generate a notification that the settings for applying the foam layer exceed specific production data and may be defective. In some embodiments, device 901 may identify a manufactured product or a portion of a manufactured product as defective and send a notification that the manufactured product or a portion of a manufactured product is defective. In some cases, this may prevent the identified product from being shipped or removed from the production of another product.

[0101] Device 901 includes instructions stored by machine-readable medium 980, which are executed by processor resource 981 to receive a product review from a machine learning model. This product review includes changes to the application of a foam layer to an application surface to alter production data and improve the quality level of the product, including quality levels. The product review may include analysis of quality and production data for a specific product or product range. The product review may include analysis of the expected quality of the product based on production data for a specific product. In other embodiments, the product review may include an analysis of the identified causes of product failures based on production data. In this way, the product review can compare the product's production data with production and quality data from different products.

[0102] Device 901 may include instructions stored by machine-readable medium 980, which are executed by processor resource 981 to provide multiple production data and multiple quality data to update a machine learning model associated with a product including the applied surface 934. In these embodiments, the machine learning model may utilize production data and quality data from multiple different production sites that produce the product including the applied surface 934 and the foam layer. In other embodiments, device 901 may provide the machine learning model with a product review request for a product including a quality level. As described herein, a product review request may be a request to analyze the cause of an identified failure. However, a product review request may be a request to analyze a particular product or product range to determine whether the product or product range is manufactured within the current set of production ranges determined by the machine learning model.

[0103] Device 901 may include instructions stored by machine-readable medium 980, which are executed by processor resource 981 to receive multiple threshold ranges based on multiple production data and multiple quality data provided by multiple different production sites. As described herein, the multiple threshold data may include threshold ranges that can be applied to multiple production data during the production of the product. The multiple threshold ranges may be updated or changed at one or more of the multiple different production sites to improve the quality data of the produced product. In specific examples, the multiple threshold ranges may include, but are not limited to: temperature thresholds, environmental thresholds, gelation time thresholds, contact time thresholds, foaming thresholds, and other thresholds.

[0104] Device 901 may include instructions stored by machine-readable medium 980, which are executed by processor resource 981 to monitor received image data applied during the application of a foam layer to an application surface 934 in a production setting. As described herein, imaging device 912 may be used to monitor production data in real time during the production of a product. In these examples, device 901 may monitor production data of the foam layer based on the received image data. Monitoring production data may include comparing the production data with production thresholds identified by a machine learning model.

[0105] Device 901 can monitor environmental data at the production site during the application of the foam layer and generate a notification when either the production data or environmental data exceeds a threshold range provided by a machine learning model. As described herein, an edge computing device for the production site can generate a notification when either the environmental data or the production data exceeds a threshold range identified by the machine learning model. Device 901 can also generate a notification when a combination of environmental and production data exceeds a combined threshold range. For example, environmental data such as external temperature may exceed a specific threshold that later alters a threshold for the foam's curing or gelling time. In this example, the combination of environmental data with a specific curing or gelling time may exceed a combined threshold for both environmental and production data. Other combinations of environmental and production data can be used to generate notifications.

[0106] Device 901 may include instructions stored by machine-readable medium 980, which are executed by processor resource 981 to receive a federated data instruction set from a machine learning model. This federated data instruction set indicates parameters for calculating multiple production data and collecting image data. The federated instruction set may be instructions from the machine learning model regarding how to generate federated data that can be provided to the machine learning model. As described herein, product analysis data can be received from multiple production sites, each generating its own production analysis data according to specifications from the machine learning model. In this way, the machine learning model can be used to collect and compare data. Furthermore, the collected data can protect the privacy of the production sites providing the data.

[0107] In this manner, federated data instructions may include instructions on how to configure imaging device 912 to capture similar portions of applied surface 934 by multiple different production sites. Similarly, lighting or other settings for capturing data may be standardized across multiple different production sites. Calculations used to determine production data and / or quality data may be standardized across multiple different production sites via federated data instructions. In these embodiments, the set of federated data instructions includes instructions for positioning imaging device 912 relative to applied surface 934 and for positioning light sources relative to applied surface 934. Other types of federated instructions may be provided to different production sites to ensure that the data collected and provided to machine learning models is standardized.

[0108] Although specific embodiments have been described above, these embodiments are not intended to limit the scope of this disclosure, even where only a single embodiment is described with respect to a particular feature. Unless otherwise stated, the examples of features provided in this disclosure are intended to be illustrative and not restrictive. The above description is intended to cover such alternatives, modifications, and equivalents that will be obvious to those skilled in the art who benefit from this disclosure.

[0109] The scope of this disclosure includes any feature or combination of features disclosed herein (express or implicit), or any generalization thereof, whether or not it alleviates any or all of the problems addressed herein. Various advantages of this disclosure have been described herein; however, embodiments may provide some, all, or none of these advantages, or may provide other advantages.

[0110] In the foregoing specific embodiments, for the purpose of simplifying this disclosure, some features are combined in a single embodiment. This approach of the disclosure should not be construed as reflecting an intention that the disclosed embodiments of the disclosure must use more features than are expressly recited in each claim. Rather, as reflected in the following claims, the subject matter of the invention does not consist of all features of a single disclosed embodiment. Therefore, the following claims are hereby incorporated into the specific embodiments, wherein each claim exists independently as a separate embodiment.

Claims

1. A method, the method comprising: Receive a request for review of the product, the request for review including identified defects related to the foam material of the product; Identify and associate production analysis data with the production process of applying the foam material to the product, wherein the production analysis data includes data captured during the production of the product; The production analysis data of the product is provided to a machine learning model used to produce the product to compare the production analysis data with a specified data range for applying the foam material to the product; Data that exceeds the specified range identified by the machine learning model is determined from the production analysis data; as well as The causes of malfunctions identified during the production of the product are determined based on the data determined.

2. The method according to claim 1, further comprising: The machine learning model is trained using production analysis data from multiple data sources associated with multiple different production sites that produce the components of the product or similar components associated with the product.

3. The method of claim 2, wherein the production analysis data from the plurality of data sources includes data associated with depositing a layer of foam material on the application surface.

4. The method of claim 3, wherein the production analysis data includes environmental data and process timing data associated with the time of deposition of the foam material layer.

5. The method of claim 1, wherein when the product comprises a foam material, the production analysis data for the product having a foam material layer includes foam deposition equipment data, foam deposition flow data, foam surface distance data, foam surface bubbling data, over-rolling data, formed foam edge data, and product size data.

6. The method of claim 1, further comprising training the machine learning model by identifying a first plurality of products corresponding to the production analysis data included in the scope of the production analysis data of the product as products with similar manufacturing conditions.

7. The method of claim 6, further comprising training the machine learning model by identifying a second plurality of products, including the identified fault, as products with similar faults.

8. The method of claim 7, further comprising training the machine learning model by: The identified fault of the product is compared with the fault status of the first plurality of products to identify products from the first plurality of products that include the same fault type as the identified fault; and The production analysis data of the product is compared with the production analysis data corresponding to the second plurality of products to identify products from the second plurality of products that include the production analysis data within the range of the production analysis data of the product.

9. A machine-readable medium storing machine-readable instructions that, when executed by a processor of a device, cause the processor to: A machine learning model for applying foam material to the production process of a certain type of product is trained using production analysis data captured at multiple different locations during the production process of that type of product, wherein the machine learning model correlates the production analysis data of applying the foam material with the quality level of the foam material for the product of that type. The production analysis data for a specific location is provided to the machine learning model, wherein the production analysis data includes image data and environmental data during a period of time during which the foam material for the type of product is applied at the specific location; Identify data in the production analysis data for the specific location that includes values ​​outside the range defined by the machine learning model; and Generate a notification to change the production settings of the production process for the specific location to change the value to within the defined range of the machine learning model.

10. The machine-readable medium of claim 9, wherein the production setup of the production process comprises at least one of the following: Types of foam dispensing equipment; The back pressure provided by the foam dispensing device; The amount of components in the foam material; The flow rate of the foam dispensing device; The mixing ratio of the foam material deposited by the foam dispensing device; The rate at which the surface moves through the distribution area of ​​the foam dispensing device; The contact time between the foam material and the top application surface; and The operating duration of the foam dispensing device.

11. The machine-readable medium of claim 9, wherein the production analysis data captured at the plurality of different locations during the production process includes imaging data for determining the surface texture of the foam material layer of the product of the type.

12. The machine-readable medium of claim 11, wherein the production analysis data captured at the plurality of different locations during the production process includes hyperspectral imaging data for determining chemical formulation data of the foam material layer.

13. The machine-readable medium of claim 9, wherein the production analysis data captured at the plurality of different locations during the production process is federated data received from the plurality of different locations.

14. The machine-readable medium of claim 9, wherein the machine-readable medium includes instructions for updating the machine learning model using production analysis data captured at the particular location.

15. A system comprising: An imaging device for capturing image data of the application surface during the application of a foam layer to the application surface at a production site; The device is configured to: During the application of the foam layer to the application surface at the production site, the received image data of the application surface is monitored; Based on the image data, production analysis data associated with applying the foam layer to the application surface is calculated; The production analysis data is provided to a machine learning model to compare the production analysis data with threshold data of the machine learning model, wherein the machine learning model utilizes production data and quality data from multiple different production sites that produce products including the applied surface and a foam layer on the applied surface. The system identifies when the production data in the production analysis data exceeds a threshold range identified by the machine learning model. and A notification is generated to change the applied production settings of the foam layer based on the identified production data.

16. The system of claim 15, wherein the device is configured to receive a set of instructions indicating parameters for calculating the production analysis data.

17. The system of claim 16, wherein the instruction set includes instructions for positioning the imaging device relative to the application surface and for positioning the light source relative to the application surface.

18. The system of claim 15, wherein the production data and the quality data provided to the machine learning model lack raw data collected from the plurality of different production sites.

19. The system of claim 15, wherein the device is configured to receive a plurality of updated threshold data based on the production data and the quality data provided by the plurality of different production sites.

20. The system of claim 15, wherein the device is configured to: Environmental data of the production site are monitored during the application of the foam layer; and The notification is generated when the environmental data exceeds the threshold range provided by the machine learning model.

Images