Failure analysis and identification for assembly lines

Abstract

In at least one example, a method for failure analysis and identification for assembly lines includes receiving a review request that includes an identified failure of a production process of an assembly line during operation, identifying production data associated with the identified failure of the production process during operation of the assembly line, comparing the production data to threshold production data, requesting additional production data based on values of the production data that are outside values of the threshold production data, and identifying a cause of the identified failure based on the production data and the additional production data during operation of the assembly line.

Description

FAILURE ANALYSIS AND IDENTIFICATION FOR ASSEMBLY LINESTechnical Field

[0001] The present disclosure relates to failure analysis and identification for assembly lines. Such techniques can be particularly useful to solve a failure for an assembly line that can cause quality defects associated with a product of a particular manufacturer by modeling production data from a plurality of different manufacturers of the product.Background

[0002] Artificial neural networks (ANNs) are networks that can process information by modeling a network of neurons, such as neurons in a human brain, to process information (e.g., stimuli) that has been sensed in a particular environment. Similar to a human brain, artificial neural networks typically include a multiple neuron topology (e.g., that can be referred to as artificial neurons). An ANN operation refers to an operation that processes, to perform a given task, inputs using artificial neurons. The ANN operation may involve performing various machine learning algorithms to process the inputs. Example tasks that can be processed by performing ANN operations can include machine vision, speech recognition, machine translation, machine transcription, social network filtering, and / or medical diagnosis.

[0003] A plurality of manufacturing sites provide double belt lamination customers with foam formulations and expertise for the fabrication of sandwich metal panels with a rigid polyurethane or polyisocyanurate core. For high index foam formulations, a primer or adhesion promoting layer is distributed over the metal facings before the polyurethane laydown. This is done through a rotating disk, which spreads droplets of the reacting primer on the application surface. These types of products can be produced on an assembly line that can include a plurality of production processes that each generate a part of the product or perform a particular function that is used by a subsequent part of the assembly line.

[0004] Parts of the assembly line can be dependent on other parts of the assembly line. In this way, a particular part of the assembly line may need to be performed within a particular period time to ensure a level of quality. In this way, a failure or technical issue that affects a first part of the assembly line can negatively affect a second part of the assembly line or the final product quality. In this way, providing rapid and accurate customer service to operators of the assembly line can be very important since each minute the assembly line is not operating, or theassembly line is producing low quality productions can be a large expense to the operators. There is currently no remote customer service that utilizes production data from a plurality of different production sites during production of a particular assembly line.Summary of the Disclosure

[0005] The present disclosure is directed to using improvements in machine learning technology to predict a cause of a failure associated with a product generated by an assembly line or other type of production process. Image based visual analysis involving a machine learning model can be utilized to monitor and analyze a production process and / or utilized during a communication session associated with a review request (e.g., failure report, support request, etc.). The image based visual analysis can utilize a machine learning model to interpret and extract data from captured images by identifying relevant patterns and features within the images. These features can include, but are not limited to: edges, textures, shapes, colors, chemical composition, temperature, and / or other visual attributes. A prediction of a failure of the product or production process can be received from the machine learning model and used to adjust the production process or to determine whether to reject the product.

[0006] As a specific example, data can be collected from a plurality of different manufacturing sites to train a machine learning model to predict a failure of a particular mechanical device or process of an assembly line at a particular manufacturing site. The machine learning model can be trained with currently understood data that can be updated as data is collected from a plurality of different assembly lines (e.g., using federated data, etc.). In some embodiments, the machine learning model can utilize data from a database that includes information related to specific information related to a particular customer’s equipment, chemicals purchased, and / or other information that may be specific to the particular customer. The machine learning model can be utilized to identify production parameters that may be a cause of particular failures and / or particular properties that are below thresholds. As used herein, production parameters relate to collected data during a production of a particular product or to data on equipment and chemicals already available on the specific production line or by a specific customer. For example, the production parameters for depositing an adhesive promoting layer can include, but are not limited to: droplet distribution data, layer uniformity data, composition data, mixture ratio data, component’s flow rate, time data between depositing theadhesive promoting layer and depositing a subsequent layer, humidity data, and / or temperature data, among other data that can potentially affect a quality of a produced product utilizing the adhesive promoting layer. As described herein, the application of the adhesive promoting layer can affect the final panel properties or can cause an adhesion failure leading to issues at a cutting station. In this way, failures associated while cutting the panel or a quality of the final panel can be caused by the application of the adhesive layer. With these types of dependencies, the machine learning model can utilize a database of failures that are categorized to identify potential causes for a particular failure at a particular site while the particular site is operating the assembly line.

[0007] The above summary of the present disclosure is not intended to describe each disclosed embodiment or every implementation of the present disclosure. The description that follows more particularly exemplifies illustrative embodiments. In several places throughout the application, guidance is provided through lists of examples, which examples can be used in various combinations. In each instance, the recited list serves only as a representative group and should not be interpreted as an exclusive list.Brief Description of the Drawings

[0008] Figure 1 is a functional block diagram for an example of a system for failure analysis and identification for assembly lines.

[0009] Figure 2 illustrates an example of a system for failure analysis and identification for assembly lines utilizing monitoring devices at the assembly line.

[0010] Figure 3 illustrates an example of a system for failure analysis and identification for assembly lines between a mobile device and a workstation.

[0011] Figure 4 illustrates an example of a method for failure analysis and identification for assembly lines.

[0012] Figure 5 illustrates an example of a machine readable medium for failure analysis and identification for assembly lines.

[0013] Figure 6 illustrates an example of a device for failure analysis and identification for assembly lines.Detailed Description

[0014] The present disclosure relates to methods and devices for failure analysis and identification for assembly lines, which may utilize machine learning models to predict a cause of a failure for one or more products generated with different production analysis data. An example of a machine learning model is an ANN. The ANN can provide learning by forming probability weight associations between an input and an output. The probability weight associations can be provided by a plurality of nodes that comprise the ANN. The nodes together with weights, biases, and / or activation functions can be used to generate an output of the ANN based on the input to the ANN. A plurality of nodes of the ANN can be grouped to form layers of the ANN.

[0015] A machine learning model can include one or more functions or equations for identifying patterns in data. In a specific example, a machine learning model can be organized as an ANN. The ANN can include a set of instructions that can be executed to recognize patterns in data. Some neural networks can be used to recognize underlying relationships in a set of data in a manner that mimics the way that a human brain operates. A neural network can adapt to varying or changing inputs such that the neural network can generate an acceptable result in the absence of redesigning the output criteria.

[0016] Production conditions during the production process can have different effects on a quality of the product. Different types of production conditions can have different effects on the end product produced. The production conditions can be monitored by different types of devices. For example, imaging devices, timing devices, sensor devices, and / or other types of devices can be utilized to monitor the different types of conditions to generate production data for the product. Specific examples herein refer to particular production data and / or conditions of the production process, however other examples can be utilized in a similar way.

[0017] The production data can be generated by a plurality of different manufacturing sites and implemented into a uniform format (e.g., production analysis data, etc.) to allow the different manufacturing sites to provide federated data to be utilized to generate or train a machine learning model. The production analysis data can include image data collected during a production process, production setting data collected during the production process, and / or environmental data collected during the production process. The machine learning model can utilize the production data or production analysis data and corresponding quality data associated with the product produced to generate production ranges or thresholds that can be utilized byeach of the plurality of manufacturing sites to increase a quality of the product. In this way, a product review of a particular product can be generated by the machine learning model to identify one or more of the production data that are attributed to a particular defect or particular quality level of the product to be reviewed by the product review.

[0018] As used herein, the singular forms “a”, “an”, and “the” include singular and plural referents unless the content clearly dictates otherwise. Furthermore, the word “may” is used throughout this application in a permissive sense (e.g., having the potential to, being able to), not in a mandatory sense (e.g., must). The term “include,” and derivations thereof, mean “including, but not limited to.”

[0019] As will be appreciated, elements shown in the various embodiments herein can be added, exchanged, and / or eliminated so as to provide a number of additional embodiments of the present disclosure. In addition, as will be appreciated, the proportion and the relative scale of the elements provided in the figures are intended to illustrate certain embodiments of the present invention and should not be taken in a limiting sense.

[0020] Figure 1 is a functional block diagram for an example of a system 100 for failure analysis and identification for assembly lines. System 100 corresponds to an assembly line or production line that includes a plurality of steps for producing a panel. Although specific references are made to the plurality of steps for producing a panel, additional steps or different steps for generating different products could also be utilized without diverging from the present disclosure.

[0021] In some embodiments, each of the plurality of steps illustrated in Figure 1 can be dependent on a previous step and / or be crucial for subsequent steps to maintain a particular level of product quality. That is, a particular failure or defect associated with a particular step can be related to or be a result of a previous step. In a similar way, a defect associated with a particular step can cause additional defects or failures for subsequent steps. In this way, identifying a failure can be difficult when one or more of the steps of the assembly line or production line may be the cause of the failure or part of the cause of the failure.

[0022] In some embodiments, the system 100 can perform an adhesive promoting layer (APL) distribution check 101, a foam distribution check 102, a foam rise check 103, a voids check 104, a panel joint check 105, a foam quality check 106, a thickness / flatness check 107, and a finished panel check 108. In some embodiments, each of the plurality of checks can beperformed during and / or after an application of the material being checked. For example, the foam distribution check 102 can be performed during distribution of the foam material and / or after the foam material is distributed onto an application surface.

[0023] As used herein, the APL distribution can include application and distribution of an APL on a first surface of a panel. As used herein, an adhesive promoting layer refers to a specialized coating or treatment applied to a surface to enhance the bonding strength between that surface and a different material (e.g., foam, adhesive, etc.). The adhesive promoting layer is crucial in many manufacturing and repair processes, where secure and durable adhesion between materials with potentially different mechanical and chemical properties is required. The adhesive promoting layer acts at the molecular or mechanical level to increase the surface energy of the substrate, making it more receptive to adhesion to the different material. Adhesive promoting layers are designed to improve adhesion on a wide range of substrates, including metals, plastics, glass, and composites. They are particularly valuable when bonding materials that naturally exhibit low surface energy, such as polyethylene or polypropylene, which are challenging to bond without special surface treatment. In some embodiments, the adhesive promoting layer can introduce functional groups to the substrate's surface, facilitating stronger chemical bonds between the adhesive and the substrate.

[0024] In some embodiments, the APL distribution step can be followed by the APL distribution check 101. In some embodiments, the APL distribution check 101 can be monitored through a monitoring system. In some embodiments, the monitoring system can include imaging devices that can capture images of a distribution unit and / or the applied APL as the panels move through the distribution unit. In some embodiments, the captured images can be analyzed to determine droplet distribution of the APL on the surface, temperature of the surface, temperature of the APL material, functionality of the distribution unit, and / or chemical composition and curing state of the APL material.

[0025] In this way, multiple factors of the APL distribution check 101 can be monitored in real-time as the panel moves from the APL distribution check 101 to the next step of the production process. That is, the APL distribution can be monitored during operation of the production process or assembly line. For example, the panel may have to move onto the foam distribution such that the foam is applied to the APL within a threshold quantity of time. For example, if the foam material is applied after the threshold quantity of time, the foam materialmay not bind to the APL properly such that the end panel (e.g., end product, etc.) includes particular quality metrics.

[0026] Once the foam material is applied to the APL, the system 100 can perform a foam distribution check 102. The foam distribution check 102 can utilize a monitoring system similar to the monitoring system to monitor the APL distribution check 101. As described herein, the monitoring system can utilize imaging devices to monitor features of the foam material that is distributed on the APL while the foam material is being applied to the APL and / or while the surface is provided to the next step of the production process. In this way, the foam material can be monitored during operation of the production process or assembly line.

[0027] Once the foam material is applied and checked at the foam distribution check 102, the panel with the APL and distributed foam material can move to the foam rise check 103. The foam rise check 103 can be performed by a monitoring system to monitor a rise of the foam material after the foam material is applied to the APL. As used herein, the foam rise check 103 refers to a quality control test specifically designed for assessing the expansion properties of the foam material. This test can ensure that the foam material will perform as expected, providing a uniform and defect-free final foam, with the final properties inside specification. The foam rise check 103 can verify that the foam material expands properly. Proper expansion affects the thermal resistance (R-value), adherence to substrates, and their ability to fill cavities and gaps completely, eliminating voids potentially leading to mechanical and aesthetical issues.

[0028] In some embodiments, the foam rise check 103 can include observing the foam material after application and / or identify bubbling that can result in voids under a surface of the foam. During the observation, the foam's expansion can be monitored by a monitoring system that includes imaging devices to capture image data of the foam’s expansion. The contact time (how long it takes for the foam to touch the top facing) can be recorded by the monitoring system. In some embodiments, the foam rise check 103 can determine the quality of the foam's cell structure, the foam’s uniformity, and the foam’s adherence to the substrate and / or APL. In these embodiments, the monitoring system can perform the foam rise check 103 during operation of the assembly line such that the panel can continue to the voids check 104.

[0029] The voids check 104 can include monitoring the foam for voids or areas that are supposed to be occupied by foam but lack foam material. In some embodiments, the voids check 104 can be performed by a monitoring system that includes imaging devices to capture images ofthe foam after the foam rise check 103 to determine if voids exist in the foam material. In some embodiments, the imaging device of the monitoring system can utilize machine learning to identify dark spots on a surface of the foam to identify voids that exist. In these embodiments, the monitoring system can be utilized to perform the voids check 104 during operation of the assembly line.

[0030] In some embodiments, the voids check 104 can be coupled to the panel joint check 105. The panels usually includes joints that can be utilized to connect a first panel to a second panel. In order for a panel to meet specifications, the panel must be able to be connected to other panels through the panel joints. In some embodiments, a monitoring system can include imaging devices that can capture images of the panel joints. In these embodiments, the images can be utilized to determine the dimensions of the joints to determine if the dimensions are within threshold dimensions. The threshold dimensions can ensure that a particular panel can be coupled to other panels during installation.

[0031] The assembly line can include a foam quality check 106. The foam quality check 106 can be an overall check of the foam quality. In some embodiments, a plurality of factors can be applied to determine the overall quality of the foam material during the foam quality check 106. In some embodiments, the foam quality check 106 can be an inspection at the joints to identify particular defects of the foam. For example, the foam quality check 106 can include a monitoring system that can capture images of the exposed foam portions to identify different surface properties of the foam that is exposed at the joints.

[0032] The assembly line can move the panel from the foam quality check 106 to the thickness and flatness check 107. The thickness and flatness check 107 can include thickness distances of the panel across a surface of the panel. This measurement or distance can relate to the flatness of the surface of the panel. In some embodiments, the flatness of the surface or the thickness of the panel can correspond to a quality of the panel since a panel that is relatively flatter or smoother can be viewed by a customer as being of higher quality versus a panel that is relatively wavy or not as flat. In this way, the panel can be coupled to a different panel and appear as a single panel. Without flat features, the panel can be below a quality threshold that will likely cause a customer not to be satisfied with the product. In some embodiments, the thickness and flatness check 107 can be performed by a monitoring system that includes an imaging device to capture a plurality of images of the panel to determine a flatness level and / orthickness of the panel. Tn this way, the thickness can be compared to a threshold thickness and the flatness level can be compared to a threshold flatness level to ensure that the panel is within thickness and flatness metrics.

[0033] In some embodiments, the assembly line can move the panel from the thickness and flatness check 107 to the finished panel check 108. In some embodiments, the finished panel check 108 can be a calculation utilizing the plurality of metrics from each of the previous steps to determine when the overall quality of the panel is greater than a threshold quality level. In other embodiments, the finished panel check 108 can include a monitoring system that includes imaging devices and / or sensors that can determine properties of the finished panel to determine whether the metrics identified by the sensors are within threshold metrics for a finished panel.

[0034] Figure 1 illustrates how an assembly line can include a plurality of monitoring systems to monitor each part of the system 100. In these embodiments, the data from the monitoring systems of the system 100 can be combined and provided to a centralized user such as a centralized support provider. In some embodiments, the centralized support provider can be provided with the data such that the centralized support provider can utilize the data to help or maintain other assembly lines utilizing the collected data without disclosing sensitive information, trade secrets, or other intellectual property associated with the collected data. In some embodiments, this collection of data from a plurality of different sources that each utilize similar data collection techniques can be referred to as federated data.

[0035] As used herein, federated data refers to a distributed data management approach where data is kept separate across multiple locations or systems, but is still accessible and usable in a unified manner. In a federated data system, each data source retains control over its own data, but the data can be queried, accessed, and integrated as if it were stored in a single centralized location. In this way, the data collected from the system 100 can be utilized for remote systems that perform similar steps. For example, the data collected from the system 100 can be utilized to identify a cause of a failure associated with a completely different assembly line associated with a different organization. By federating the data collection process, the data collected by the system 100 can be kept secret from competitors while also allowing a centralized organization to utilize the data for solving issues for other systems.

[0036] Figure 2 illustrates an example of a system 200 for failure analysis and identification for assembly lines. The system 200 can be utilized for applying an adhesivepromoting layer on an application surface 211. Although the system 200 illustrates a specific system for applying an adhesive promoting layer on a particular surface 211, the disclosure is not so limited. For example, other types of systems can utilize the functions and / or features described herein in a similar way for production analysis modeling for product quality detection of other types of products. For example, the application of the adhesive promoting layer could be replaced by one or more steps provided by the system 100 as illustrated in Figure 1. Specific systems will be described associated with a particular assembly line (e.g., system 100, production line, etc.), but the disclosure is not so limited.

[0037] The system 200 includes a primer distribution device 210 that can dispense a chemical compound that is utilized as an adhesive promoting layer (e.g., primer, etc.) on a surface 211. The primer distribution device 210 can include a distribution disk that can rotate to distribute the adhesive promoting layer onto a surface. The distribution disk can rotate at different speeds to spread or dispense the chemicals that comprise the adhesive promoting layer. In some examples, the adhesive promoting layer can comprise a plurality of chemicals that are supplied to the distribution disk. In some examples, the plurality of chemicals can be supplied to the distribution disk at different flow rates to generate a particular chemical ratio between the plurality of chemicals.

[0038] As used herein, a flow rate describes a rate at which a quantity of chemical is supplied to the distribution disk. When the primer distribution device 210 dispenses the chemical on the surface 211, the chemical can be dispersed in a particular droplet distribution and / or a particular layer uniformity. As used herein, the droplet distribution can be a quantity of droplets that are deposited within a sub-area within a particular area. Each sub-area can have a different droplet distribution representing a respective quantity of droplets for each of the plurality of subareas. The different droplet distributions can be utilized to identify when particular sub-areas have a relative greater or lesser quantity of droplets compared to other sub-areas within the particular area.

[0039] In a similar way, the particular layer uniformity can represent a layer thickness of the adhesive promoting layer for the plurality of sub-areas over the particular area. In this way, the particular layer uniformity can identify when a first sub-area has a different layer thickness of the chemical than a second sub-area. A fluctuation of the quantity of chemical dispersed on the surface 211 of a material can be captured and utilized as production data and / or productionanalysis data since the droplet distribution and / or layer uniformity can affect a quality of an end product. In some embodiments, production settings data can be included with the image data such that the productions settings utilized to produce the particular layer uniformity can be used as production analysis data.

[0040] The surface 211 can be a metal surface that includes the adhesive promoting layer deposited by the primer distribution device 210. The system 200 can include a light source 214 that can be directed to the surface 211. The light source 214 can be a visible light source and / or another type of light source based on the type of images to be captured. The location or angle of the light source 214 can be altered over a period of time to capture a plurality of different types of images of the surface 211 over the period of time.

[0041] The surface 211 can be positioned on a transportation system to bring the surface 211 to a different area to allow a different substance or chemical to be deposited on the surface 211. In these embodiments, the light source 214 can be positioned at first location and a different light source can be positioned at a second location. In this way, the surface 211 can be captured with different light sources as it moves from a location to receive a deposit from the primer distribution device 210 to a different distribution device.

[0042] Transporting the surface portion from the first location to the second location can be monitored by a timing device to determine the quantity of time it takes the surface portion to move from the first location to the second location or by a device monitoring the distance and calculating the time needed based on the line speed. The system 200 can include additional sensors to monitor a quantity of time it takes for the surface 211 to move from the location to receive the chemical deposit from the primer distribution device 210 to a different distribution device (e.g., foam material distribution device, etc.). The chemical deposited by the primer distribution device 210 can undergo a particular chemical reaction. In this way, the quantity of time can be utilized to determine a state of the chemical reaction when the surface 211 reaches the different distribution device.

[0043] In some embodiments, the system 200 can be a double belt lamination (DBL) system that can represent a portion of a larger system (e.g., system 100 as referenced in Figure 1, etc.). The system 200 can include a foam distribution device that can be different in shape than the primer distribution device 210, but functionally used in a similar manner to distribute a chemical on to a surface. The foam distribution device can receive the surface 211 from theprimer distribution device 210 such that the foam distribution device can dispense a chemical compound or foam material that can be utilized to form a rigid polyurethane (PUR) or polyisocyanurate (PIR) core foam material on the surface 211. As used herein, a foam material refers to a substance that includes gas bubbles trapped within a liquid or solid matrix. In a first example, the foam material includes a substance with gas bubbles trapped within the liquid or solid prior to distribution and / or a substance that includes gas bubbles trapped and / or formed within the liquid after a chemical reaction or after distribution of the foam material. The foam material can be distributed onto the adhesive promoting layer when the foam material is in a liquid foam state. The liquid foam state can solidify and become a solid foam state on the surface over a period of time.

[0044] The foam distribution device can distribute the chemical compound or foam material onto a surface that includes the adhesive promoting layer (e g., surface 211). As described herein, the surface 211 can be positioned on a transportation system to bring the surface 211 to different areas of a larger assembly line. In this way, the transportation system can move the surface 211 from the primer distribution device 210 to the foam distribution device. In addition, the transportation system can move the surface portion to other steps or areas of the assembly line. For example, the transportation system can move the surface 211 to inspection areas to monitor particular features of the surface 211.

[0045] The system 200 can include a control panel 212 that can be utilized by a user 213. The control panel 212 can be utilized to display notifications generated by the system 200 to notify the user 213 when current conditions or metrics of the system 200 are outside a particular range. As described further herein, the data collected during the production of the surface 211 and / or a product utilizing the surface 211 can be utilized to generate production ranges that can be utilized by the system 200.

[0046] The images captured by the imaging device 215 can be provided to a machine learning model 216 operating on the edge computing device 217. The machine learning model 216 can be utilized to analyze the received images from the imaging device 215. In other embodiments, the machine learning model 216 can be utilized to organize the data collected by the system 200 into a data file that can be correlated to an end product utilizing the surface 211. For example, humidity data, temperature data, droplet distribution data of a chemical layer, layer uniformity data of the chemical layer, and / or time data between depositing the chemical layerand depositing a different layer on the chemical layer can be correlated together and associated with a product that utilized the particular surface 211. In this way, a quality of the surface 211 can be determined over a period of time and correlated to the data associated with the surface 211.

[0047] The system 200 can include an edge computing device 217. In some examples, the edge computing device 217 is a computing device that includes a processor resource and a machine readable medium to store instructions that are executed by the processor resource to perform particular functions. The edge computing device 217 can be utilized to communicate with a remote device 221. The remote device 221 can be a cloud device that can receive data from a plurality of manufacturing sites to increase the data set used to train the machine learning model 216 for a particular product and / or for portions of a product such as the surface 211. The edge computing device 217 can be utilized to remove data associated with the particular manufacturing site that the site does not want to be shared with other manufacturing sites.

[0048] In other embodiments, the edge computing device 217 can utilize production ranges (e.g., production condition ranges, condition thresholds, etc.) to monitor the data provided by the machine learning model 216. When data received by the machine learning model 216 is outside a threshold range, the edge computing device 217 can send a notification to the control panel 212 to notify the user 213. In this way, alterations can be made to the production data and / or production settings in real time. The final panel produced using the surface 211 can be discarded or marked as lower quality when the edge computing device 217 determines the production data were outside a threshold range of data. As described herein, the threshold ranges can be provided by the remote device 221 when the remote device 221 is utilizing a machine learning model.

[0049] As used herein, production settings can be adjustable settings that can define production parameters of how the system 200 deposits the chemical layer on the surface 211, how the foam material layer is deposited on the surface 211, and / or how other functions of the system 200 are executed. For example, the production settings can include, but are not limited to: position of the imaging device 215 and / or light source 214 to alter or influence the images, the position of the primer distribution device 210, the rotational speed of the primer distribution device 210, flow rates of the chemicals, temperature of the chemicals, temperature of the primer distribution device 210, line speed of the surface 211 moving from a first location to a secondlocation to alter or influence the distribution, curing time of the chemical or primer and / or chemical ratio of the chemical or primer. In this way, the system 200 can alter or adjust the production settings to alter or adjust how the primer distribution device 210 applies the chemical layer to the surface 211. In this way, the production settings can be altered to ensure that production data is within the threshold range of data provided by the machine learning model.

[0050] The remote device 221 can receive data from the edge computing device 217 and receive data 219 organized in the same way from a plurality of other manufacturing sites. In this way, the remote device 221 can utilize a machine learning model to perform advanced data analytics 218 on the data received from the edge computing device 217. The remote device 221 can utilize this data from the edge computing device 217 and the other manufacturing sites to generate support knowledge to the control panel 212. As described further herein, the support knowledge can include product review analysis for a specific product produced at a particular time and / or production data ranges that can be utilized by the edge computing device 217 to generate real time notifications to the control panel 212.

[0051] Figure 3 illustrates an example of a system 325 for failure analysis and identification for assembly lines. In some embodiments, the system 325 can be utilized to provide information to a mobile device 327 to help a customer 326 identify a solution to a particular issue or failure associated with an assembly line. In some embodiments, the system 325 can include a workstation 329 utilized by a support 328 (e.g., support organization, etc.). In these embodiments, the workstation 329 can include federated data that can be collected from a plurality of different sites operating assembly lines independently. For example, the plurality of different assembly lines can be associated with different organizations that may not want to or be able to communicate directly (e.g., share information related to their respective site, etc.). In these embodiments, the data from the plurality of different sites can be provided to the workstation 329 or computing device associated with the workstation 329.

[0052] In some embodiments, the customer 326 can be a human user located at a particular assembly line. In this way, the customer 326 can be attempting to fix or determine a cause of an issue or failure associated with the particular assembly line. In some embodiments, the customer 326 may receive the mobile device 327 from an organization associated with the support 328. For example, the organization can employ the support 328 to provide a secure connection through a network 330 between the mobile device 327 and the workstation 329. Asdescribed herein, information associated with the organization of the customer 326 can utilize trade secrets and / or other secure information with the particular assembly line. In these embodiments, the information shared between the mobile device 327 and the workstation 329 can remain confidential and / or secret from organizations associated with different assembly lines.

[0053] In some embodiments, the mobile device 327 can be provided to the customer 326 such that the customer 326 can have a secure connection with the support 328. In some embodiments, the assembly line may not allow mobile devices without secure connections to maintain a level of security. For example, the assembly line may utilize trade secrets that can be protected through protocols that limit access to mobile devices that can transmit data collected on the assembly line. In some embodiments, the mobile device 327 can be configured to establish a secure connection with the workstation 329 without allowing the mobile device 327 to transfer data to other devices. For example, the mobile device 327 can be configured to include security settings to allow the mobile device 327 to exclusively connect to the workstation 329 and / or cloud-based mobile device management 331 through a secure network 330. In this way, the security protocol for the particular site can be maintained while also being able to be provided with customer support and also allowing the customer 326 to provide data from the particular assembly line to the support 328.

[0054] In some embodiments, the mobile device 327 can include an imaging device that can be utilized to capture images of the assembly line. In some embodiments, the imaging device can be a still image camera, video camera, a hyperspectral camera, or other type of imaging device. In this way, the mobile device 327 can be utilized to capture images of particular features of the assembly line during the communication with the support 328. In some embodiments, the mobile device 327 can receive instructions from the cloud-based mobile device management 331. In some embodiments, the cloud-based mobile device management 331 can provide the instructions based on data or information received from the mobile device 327. For example, the mobile device 327 can provide an identified failure or technical issue associated with the assembly line. In this example, the identified failure or technical issue can be utilized by the cloud-based mobile device management 331 to determine instructions to provide to the mobile device 327.

[0055] In some embodiments, the instructions generated by the cloud-based mobile device management 331 can be provided to the mobile device 327 to instruct the mobile device 327 and / or customer 326 to capture data or provide data to the cloud-based mobile device management 331. For example, the instructions provided to the mobile device 327 can be a description or location to capture images of the assembly line, product generated by the assembly line, or portion of the product generated by the assembly line. In this way, cloud-based mobile device management 331 can request additional information from the mobile device 327 based on the type of identified failure or technical issue.

[0056] As described further herein, the requested additional information can include, but is not limited to: device names, images of devices, images of products, images of outcomes associated with the failure, process data, among other information that can be associated with the identified failure. In some embodiments, the cloud-based mobile device management 331 can extract information from images or data that is received from the mobile device 327. For example, the images of the devices and / or products can be utilized by the cloud-based mobile device management 331 to extract device information and / or product information for the specific device or specific product.

[0057] In some embodiments, the mobile device 327 can provide location information. The location information can be utilized to identify a plurality of devices that are utilized at a particular manufacturing site, particular assembly line, and / or particular portion of the assembly line. In this way, the cloud-based mobile device management 331 can utilize the location information to identify devices utilized by a particular site to determine possible reasons for a failure or technical issues. For example, a particular model of a foam dispensing device or application layer dispensing device can have different types of mechanisms that can have different types of failures associated with them. In addition, the cloud-based mobile device management 331 can utilize federated data associated with the specific type of mechanism and / or specific model to help identify the type of failure that may be causing a particular issue.

[0058] In some embodiments, the mobile device 327 and the workstation 329 can establish a two-way communication channel through the network 330. As described herein, the mobile device 327 can establish a secure connection with the workstation 329 through the network 330. In some embodiments, the mobile device 327 may be restricted from communicating with devices other than the workstation 329 and / or the cloud-based mobiledevice management 331. As described further herein, the two-way secure communication can be monitored by the cloud-based mobile device management 331 to identify potential issues in real time during the communication session between the mobile device 327 and the workstation 329.

[0059] Figure 4 illustrates an example of a method 440 for failure analysis and identification for assembly lines. In some examples, the method 440 can be executed by a computing device as described herein. In some examples, the method 440 can be executed by a mobile device management device or system. For example, the method 440 can be executed by the cloud-based mobile device management 331 as referenced in Figure 3. In this way, the method 440 can be performed to monitor and / or utilize data associated with a communication session between a mobile device and a workstation.

[0060] The method 440 can be utilized to identify particular production data that can be a cause of a particular defect and / or identify a threshold range for a particular production data that can be utilized by a particular manufacturing site. In some embodiments, the particular defect can be identified by a user or device at the assembly line of a production site. In some embodiments, the method 440 can include receiving image data from monitoring devices associated with the assembly line. For example, the mobile device management can receive image data from an imaging device that is utilized to monitor a quality of a product being produced by the assembly line.

[0061] In other embodiments, the mobile device management can receive a request from a mobile device with a description and / or data associated with a particular defect. In some examples, a user can initiate a secure communication session with a workstation and the mobile device management can monitor the secure communication session to utilize the information from the secure communication. For example, the mobile device management can extract information provided from the mobile device to identify devices and / or portions of the assembly line that may be a cause of the particular defect. In addition, the identified devices can be utilized by the mobile device management to compare with federated data from other production sites and / or other assembly lines that utilize the same or similar devices.

[0062] At step 441, the method 440 can be executed to receive a review request that includes an identified failure of a production process of an assembly line during operation. In some embodiments, the review request can be a request from a remote device that is configured to provide a secure connection with a particular workstation or mobile device management. Forexample, the remote device can be a remote device provided by a support organization that can have access within the assembly line area since the mobile device is only allowed to provide the secure connection with the particular workstation or mobile device management. That is, the assembly line area can be a secure area that would normally not allow mobile devices or communication devices to be utilized on the assembly line.

[0063] In some embodiments, the mobile device can be configured to comply with a security policy of the assembly line area. For example, the mobile device can be configured to prevent communication with unauthorized devices. In this example, the unauthorized devices can be any other device that is not associated with the support organization. In this way, the support organization can be responsible for configuring the device such that it is not able to communicate with devices other than an authorized mobile device management or workstation. By having the support organization configure the device and provide the device to the assembly line, the mobile device will have greater security compared to devices that have been utilized for other purposes prior to being utilized to transfer sensitive data. For example, an unsecure mobile device can have software or firmware installed that may be able to operate even when the unsecure mobile device is put into a secure mode or operating with a secure application. For this reason, the support organization can ensure that the mobile device was not utilized for other purposes prior to configuring the mobile device to be utilized by the assembly line.

[0064] In some embodiments, the method 440 includes transcribing and performing metadata tagging to the review request, wherein the review request includes communication data from an operator of the assembly line. In some embodiments, the mobile device can be utilized by an assembly line operator or assembly line user. In these embodiments, the assembly line operator can generate a review request with the workstation and / or mobile device management. In some embodiments, the review request can be an audio and / or textual request for assistance. The review request can include, but is not limited to: images captured by the mobile device, audio descriptions captured by the mobile device, and / or textual descriptions provided to the mobile device. The review request can be provided to the workstation and / or mobile device management through a secure network connection established by the mobile device.

[0065] In some embodiments, the mobile device management can receive the review request information and transcribe the audio descriptions captured by the mobile device. The audio descriptions can be transcribed and stored by the mobile device management for easierstorage and retrieval when attempting to utilize the review request for information at a later date (e g., for a different review request received at a later date, etc.). In some embodiments, the transcribed audio description can be tagged with metadata by the mobile device management prior to storing the transcribed audio description. As used herein, tagging the review request with metadata can include adding metadata information associated with the particular assembly line, particular devices utilized by the assembly line, particular chemicals utilized by the assembly line, and / or category of identified failure to portions of the review request.

[0066] In some embodiments, the portions of the review request can be tagged with different metadata since each of the portions may or may not utilize the same devices, chemicals, or have other properties. In this way, the mobile device management can more easily categorize the review request and utilize only relevant portions of the review request for future review requests by the assembly line or future review requests from different assembly lines. In some embodiments, tagging the metadata can also include adding metadata information to each of a plurality of portions of the review request to identify security information. In some embodiments, the assembly line can be a secure site that utilizes trade secrets or other types of sensitive information. In this way, the mobile management device can tag information with a particular security level to ensure that the sensitive information or trade secrets are protected by preventing this information from being provided to other assembly lines or unauthorized users.

[0067] In some embodiments, the review request can include textual descriptions. The textual descriptions can be tagged with metadata in a similar way as the transcribed descriptions. Furthermore, the images that are received through the review request can be tagged with metadata. For example, portions of the image can be embedded with metadata to identify particular devices within the image, particular chemicals or compounds within the image, security information associated with the different portions of the image, and / or other site information associated with the assembly line operator that captured the image. In this way, the text data and / or image data can be stored and categorized based on the metadata and the information can be more easily retrieved to be utilized for future review requests.

[0068] At step 442, the method 440 can be executed to identify production data associated with the identified failure of the production process during operation of the assembly line. As described herein, production data can include data collected during production of a product at the assembly line. The production data can be different for each stage or portion of theassembly line. For example, the production data for an adhesive promoting layer application can be different than the production data for a foam material layer. The production data can include, but is not limited to: foam height rates, foam distribution rates, surface application rates, panel thickness, among other rates or data. For example, the data can describe how a surface can be coated with an adhesive promoting layer, how a foam is applied on the surface, how many voids are visible after the foam has completely cured, the dimensions of a finished product, among other descriptions that can be associated with a quality of an end product.

[0069] In some embodiments, the production data can be provided to the mobile device management or workstation by the mobile device. For example, the production data can be input into the mobile device and sent through the secure communication path. In other embodiments, the mobile device can receive the production data from a control panel or other computing device (e.g., control panel 212, etc.). For example, the mobile device can be securely connected to the control panel that receives the production data from an edge computing device (e.g., edge computing device 217 as referenced in Figure 2, etc.). In other embodiments, the edge computing device 217 can provide the production data to the mobile device management or workstation in response to receiving a signal that a review request has been provided by the mobile device. In this way, the production data can be provided to the mobile device management or workstation securely. In addition, the production data can be provided to the mobile device management or workstation such that the production data corresponds to the review request being submitted and / or while the assembly line is still functioning.

[0070] At step 443, the method 440 can be executed to compare the production data to threshold production data. As described herein, the threshold production data can be determined utilizing federated data from the assembly line as well as federated data from other assembly lines. The threshold production data can be value ranges for the production data such that production data within the value ranges typically produces higher quality end products compared to production data outside the value ranges. In some embodiments, the threshold production data can refer to value ranges for a combination of different values. For example, the threshold production data can refer to a difference in time between an application of a first material and a second material and also refer to a composition of the first material or the second material. In this example, a particular composition of a material can require a different time threshold value.

[0071] At step 444, the method 440 can be executed to request additional production data based on values of the production data that are outside values of the threshold production data. In some embodiments, the production data that is provided can be an indication of a particular issue or a plurality of potential issues that are causing identified failure of the review request. In some embodiments, the mobile device management and / or the workstation can send a request for the additional production data based on the values of the received production data. In this example, the production data may be outside the values of the threshold production data or the values of the production data may indicate a potential issue if another production value is within or outside a particular range of values. For this reason, the mobile device can receive the request for additional production data.

[0072] In some embodiments, the production data that is outside the range of the threshold production value can indicate a particular failure, but in order to confirm the particular failure, additional production data may be needed by the workstation and / or mobile device management. For example, the production data can indicate that one or more foam dispensers are clogged or not providing the foam at a rate that will cover a designated area on the surface. In this example, the workstation and / or mobile device management can send instructions to the mobile device to provide additional data associated with the foam dispensers. In a specific example, the instructions to the mobile device can indicate that an image (e.g., still image, hyperspectral image, video image, etc.) is to be captured by the mobile device or other device and the image is to be sent to the workstation and / or mobile device management. In this way, the image that is captured can be analyzed and utilized to determine if there is a failure associated with one or more of the foam dispensers. In addition, the image can be analyzed and / or utilized to provide a solution for fixing the determined failure associated with the one or more foam dispensers.

[0073] In some embodiments, the method 440 can include sending image instructions to capture an image of a particular portion of the assembly line during operation. As described herein, the mobile device can receive instructions to capture an image of a particular portion of the assembly line and / or a particular portion of a product during the operation of the assembly line. Shutting down the assembly line may not be an option due to time needed to restart. Thus, the images may have to be captured from a specific angle and / or under specific lighting conditions in order to be analyzed by the mobile device management or workstation. In this way,the image instructions can include details for how to capture the image of the particular portion of the assembly line.

[0074] In some embodiments, the image instructions can be displayed while attempting to capture the image. The image instructions can be utilized to instruct a user where to capture the image, settings to be utilized, and / or where to move the mobile device to capture the correct image. For example, the user can direct the camera of the mobile device at a particular portion of a dispenser device. In this example, the image instructions can instruct the user to move the camera closer to the dispenser device, add lighting to the area of the dispenser device, and / or alter other settings of the mobile device before capturing the image. In some embodiments, the image instructions can instruct the mobile device to capture the image when the camera is positioned at the correct location with the correct lighting and / or correct settings. This can ensure that the captured image can be analyzed to provide accurate results.

[0075] In some embodiments, the image instructions include a location and type of imaging device to be utilized to capture the image. As described herein, the image instructions can be utilized to identify a particular location of the assembly line to capture an image. In some embodiments, the location can be a particular area of the assembly line or a particular device of the assembly line. In some embodiments, the image instructions include instructions to utilize a hyperspectral camera to capture a hyperspectral image of a material layer to determine a mixing ratio of the material layer.

[0076] As used herein, a hyperspectral image is a type of image that captures a wide spectrum of light beyond the visible light spectrum. For example, a hyperspectral image can capture light in the infrared and / or ultraviolet regions. Unlike traditional photography or even multispectral imaging, which might capture images in three (RGB) or several distinct wavelength bands, hyperspectral imaging divides the spectrum into many more bands, often hundreds or thousands. This technique can provide very detailed information about the spectral properties of the materials in the image. In some embodiments, each pixel in a hyperspectral image contains a continuous spectrum of the light intensity, offering a unique spectral signature for the materials present. This information allows for the identification and differentiation of objects, materials, or processes that would not be distinguishable in a regular photo or even in multispectral images. In this way, the hyperspectral image can be utilized to identify a composition of materials utilized to generate a production utilizing the assembly line.

[0077] In some embodiments, the method 440 can include requesting a type of the production process and a type of assembly line. In these embodiments, the type of production process identifies a product generated by the production process and the type of assembly line identifies a plurality of steps and mechanical devices utilized by the assembly line. As described herein, the workstation and / or mobile device management can request information related to the assembly line associated with an organization that has submitted the review request. In some embodiments, this information can be provided by the mobile device through the secure connection. In some embodiments, the mobile device can be configured with this information since the mobile device can be configured specifically for the particular assembly line. In some embodiments, the review request can include the information related to the assembly line through an identification that is provided to the mobile device. For example, the organization associated with the assembly line can utilize a particular identification that can correspond to their assembly line, which can be different for different assembly lines.

[0078] At step 445, the method 440 can be executed to identify a cause of the identified failure based on the production data and the additional production data during operation of the assembly line. As described herein, the method 440 can be executed while the assembly line continues to operate. In this way, identifying the failure in a relatively short period of time can be very important to an organization associated with the assembly line. In some embodiments identifying the cause of the identified failure can be based on the production data and the additional production data provided by the mobile device while the devices associated with the assembly line continue to operate. In some embodiments, the cause of the identified failure can be tagged to the production data and the additional production data and stored for use in additional review requests.

[0079] In some embodiments, identified cause can be tagged to specific features of the production data and / or additional production data. For example, the identified cause can be tagged to production data associated with a particular device and not tagged to an image or production data of a particular chemical composition. In this way, the identified cause can be tagged to specific features of the production process to categorize the production data and / or additional production data more accurately such that the data can be extracted more quickly from a database based on future production data associated with a future review request.

[0080] In some embodiments, the method 440 can include altering a setting of the assembly line based on the cause of the identified failure. In some embodiments, the setting alters how the production process is performed, which results in production data of the production process to result in an altered value that is within the values of the threshold production data. In some embodiments, the setting of the assembly line can affect how the devices associated with the assembly line perform or function. For example, a setting can refer to temperature settings, flow rate settings, material rate settings, mixing settings associated with compositions, among other settings that control how the devices of the assembly function. In this way, stings of the assembly line can be altered and resulting production can be monitored to determine when the production data is within the production data thresholds.

[0081] In some embodiments, the setting alterations can be tagged as metadata to the data associated with the review request. For example, the setting alterations can be tagged as metadata associated with a particular device and / or the identified failure. In this way, the same or similar identified failure associated with a different review request can be identified through the stored metadata. As described herein, the assembly can continue to operate and generate products that may not be above a quality level threshold. In this way, it can be important to identify the failure faster and / or more efficiently to enable the assembly line to begin producing products that are above the quality level threshold.

[0082] In some embodiments, the metadata can be utilized to categorize the data collected during the review request. That is, the mobile device management can continuously collect the data provided by the mobile device. In this way, there can be a relatively large quantity of data collected. In order to utilize the categorized data for subsequent review request, the data can be categorized utilizing the metadata. As described herein, the data collected during the review request can be utilized for subsequent review requests associated with the same assembly line or utilized for a subsequent review request associated with a different assembly line that may be associated with a different organization.

[0083] In some embodiments, the data can be categorized or organized based on the features of the particular assembly line. The features of the assembly line can include, but are not limited by the: devices, processes, chemicals, and / or substances utilized by a particular assembly line. In this way, the metadata can be utilized to categorize the data collected during a review request to be utilized for a plurality of different assembly lines based on the features of theassembly line. For example, the metadata can be tagged for a particular device from a first assembly line and utilized for a review request from a second assembly line that utilizes the particular device. In this way, information from the first assembly line can be utilized to identify causes of failures for the second assembly line. In these embodiments, the first assembly line may be a competitor of the second assembly line and may not be willing or able to share data. In this way, the support organization can help identify causes of failures for a plurality of assembly lines while maintaining confidentiality between the plurality of assembly lines.

[0084] As described herein, the plurality of assembly lines can be associated with different organizations that may or may not share information. The data associated with the review request can be tagged with metadata indicating a security level of the data and / or indicating a security level for a plurality of portions of data. For example, first portions of data can be identified as allowable such that the first portions can be provided to a particular organization. As another example, second portions of the data can be identified as not allowable such that the second portions cannot be provided to the particular organization. In some embodiments, the data may still be useable to help the particular organization even when the data is not provided directly to the particular organization. In this way, the support can provide more efficient support for each of the plurality of assembly line organizations while still providing data security.

[0085] Figure 5 illustrates an example of a machine readable medium 550 for failure analysis and identification for assembly lines. The machine readable medium 550 can be communicatively connected to a processor resource 551 by a communication path 552. In some examples, a communication path 552 can include a wired or wireless connection that can allow communication between devices and / or components within a single device. As used herein, the processor resource 551 can 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 combination of circuitry and / or logic to orchestrate execution of instructions 553, 554, 555, 556, 557, 558, 559. In a specific example, the processor resource 551 utilizes a non-transitory computer- readable medium storing instructions 553, 554, 555, 556, 557, 558, 559, that, when executed, cause the processor resource 551 to perform corresponding functions.

[0086] The machine readable medium 550 may be electronic, magnetic, optical, or other physical storage device that stores executable instructions. Thus, a non-transitory machine- readable medium (MRM) (e.g., machine readable medium 550) may be, for example, a non- transitory MRM comprising Random-Access Memory (RAM), read-only memory (ROM), an Electrically Erasable Programmable ROM (EEPROM), a storage drive, an optical disc, and the like. The machine readable medium 550 may be disposed within a controller and / or computing device. In this example, the executable instructions 553, 554, 555, 556, 557, 558, 559, can be “installed” on the device. Additionally, and / or alternatively, the machine readable medium 570 can be a portable, external, or remote storage medium, for example, which allows a computing system to download the instructions 553, 554, 555, 556, 557, 558, 559, from the portable / external / remote storage medium. In this situation, the executable instructions may be part of an “installation package”.

[0087] The machine readable medium 550 includes instructions 553 to receive a review request that includes audio messages that describe an identified failure of a production process of an assembly line during operation of the assembly line. As described herein, a review request can be a communication received from a designated mobile device. As described herein, the mobile device can be a device provided to an assembly line organization by a support organization. In some examples, the mobile device can be utilized to establish a secure communication session with a workstation and / or mobile device management when a user indicates that a failure or malfunction has occurred at the assembly line.

[0088] The failure or malfunction can be determined by the user at the assembly line and / or a monitoring system that notifies the user of the failure or malfunction. In some embodiments, a failure or malfunction may be associated with a particular quality level of a final product produced by the assembly line. That is, the user or monitoring system can identify that an end product of the assembly line is below a quality threshold and at that time the user or monitoring system can send the review request to determine which device or portion of the assembly line is causing the product to be below the quality threshold.

[0089] The machine readable medium 550 includes instructions 554 to generate a transcription of the audio messages and tag information within the transcription with metadata. As used herein, metadata is data that provides information about other data. Metadata can help in organizing, finding, and understanding data by describing its characteristics, origins, usage, andstructure. As described herein, the metadata can be tagged to the transcribed audio messages of the identified failure. In some embodiments, the metadata can be utilized to categorize and / or retrieve information related to the identified failure. For example, the metadata can be tagged to the transcription with information that can allow the transcription data to be retrieved when a similar identified failure is reported in a future review request.

[0090] In some embodiments, the metadata can include a timestamp that indicates when the review request with the audio messages was received from a mobile device associated with a particular assembly line. In a similar way, the metadata can include location information that identifies the particular assembly line. In some embodiments, a specific portion of the transcription can be tagged with metadata that identifies security information or a security level of the specific portion of the transcription. In some embodiments, the metadata can be tagged with device information and / or chemical information that is utilized at the assembly line. In this way, the metadata can be utilized to categorize the transcription and / or utilized to categorize a plurality of portions of the transcription. In this way, specific information can be retrieved to help solve future identified failures from the assembly line and / or other assembly lines.

[0091] In some embodiments, the machine readable medium 550 includes instructions to identify triggers within the transcription to categorize the identified failure category of the production process. As used herein, triggers can be terms or phrases that are included in the transcription that can be utilized to identify a type of device, a type of chemical, a type of product, a type of identified failure, and / or other identifiers. In this way, the transcription or portion of the transcription can be retrieved utilizing the triggers that are tagged with metadata. Identified triggers within the transcription can be utilized to search for data related to the identified triggers from a database that includes related information and / or previous review requests and responses. As described herein, the identified triggers within the transcription can be utilized to categorize the identified failure and tag the transcription with metadata based on the categorization.

[0092] The machine readable medium 550 includes instructions 555 to compare the metadata of the transcription to determine a failure category. As described herein, the metadata of the transcription can be tagged to identified triggers. The metadata can be utilized to determine a failure category associated with the transcription. In some embodiments, the failure category can be utilized to identify data stored within a database that includes metadata tagged toidentifiable triggers. As described herein, the metadata can be embedded or tagged descriptions of the data. In this way, data related to the determined failure category can be retrieved from a database utilizing the tagged metadata.

[0093] In some previous examples, it can be difficult to retrieve relevant data associated with an identified failure without utilizing metadata tagged to identified triggers included within the data. For example, a description of an identified failure can include information related to an outcome or quality of a product. However, the identified failure may not include specific information related to the devices, chemicals utilized, and / or current performance of the assembly line. Since stopping an assembly line to solve an identified failure may be costly, identifying the cause of the identified failure rapidly can be advantageous for assembly lines. In this way, the present disclosure can be utilized to more rapidly provide relevant information utilizing the metadata tagged to information from the particular assembly line that generated the review request as well as metadata tagged to information from a plurality of additional assembly lines.

[0094] The machine readable medium 550 includes instructions 556 to identify production data associated with the identified failure category of the production process based on the metadata during operation of the assembly line. In some embodiments, the identified failure category can be associated with particular production data. For example, a failure in the adhesion between the foam and the facing can be a result of production data including, but not limited to: wrong facing temperature, wrong conveyor temperature, low final foam density, mixing issue, among other factors that can be associated with the failure category.

[0095] In some embodiments, the production data associated with the identified failure category can be identified from a database based on tagged metadata. In this example, the tagged metadata can correspond to triggers of the identified failure category. For example, the failure category can be a foam distribution failure. Identifying production data associated with the foam distribution failure category can include searching for metadata related to problems associated with foam distribution, metadata related to the particular device utilized in the production process, and / or metadata related to other triggers within the review request.

[0096] The machine readable medium 550 includes instructions 557 to compare the production data to threshold production data. Comparing the production data identified based on the metadata associated with the triggers of the review request to threshold production data canidentify one or more factors of the production data of the assembly line that is outside the threshold production data. As described herein, the threshold production data can be value ranges for monitored data of the assembly line that can be utilized to identify failures of particular devices, compounds, or other features of the assembly line. In this way, the identified production data that is associated with the identified failure can be analyzed to determine if any of the identified production data are outside the threshold production data.

[0097] In some embodiments, the identified production data can be further analyzed to determine when one or more of the values of the production data are relatively close to a corresponding threshold production data value. Although the production data value may not be outside a designated range of the threshold production data values, being relatively close to the threshold production data value range may be an indication of a potential cause of the identified failure. In other embodiments, the identified production data values can be utilized in combination without considering other production data values not associated with the identified failure to determine when a particular combination of production data values result in the identified failure. That is, a relationship between a portion of the production data values can be utilized instead of utilizing individual production data values. In some embodiments, the relationship between the portion of the production data can be utilized to identify a cause of the identified failure.

[0098] The machine readable medium 550 includes instructions 558 to request additional production data based on values of the production data that are outside values of the threshold production data. In some embodiments, production data values that are outside values of the threshold production data value range can be utilized to determine unknown production data values associated with the assembly line. In these embodiments, the unknown production data can be data that was not provided in the review request. In some embodiments, the unknown production data can be captured by a mobile device.

[0099] As described herein, the mobile device can receive instructions to capture the additional production data. In some embodiments, the additional production data can be images captured by the mobile device. For example, instructions can be sent to the mobile device to identify a location and / or portion of a device to capture an image. As described herein, the mobile device can establish a secure connection to provide images and / or other data to a mobile device management or workstation of a support organization. In these examples, the mobiledevice can receive the instructions to capture a particular image with particular settings and send the captured image to the mobile device management and / or the workstation of the support organization.

[0100] In some embodiments, the additional production data can be additional data not originally provided during the review request. In some embodiments, the additional production data can be provided by the mobile device or provided from a control panel associated with the assembly line. For example, the mobile device can receive instructions from a user of the control panel to send the additional data to the mobile device management and / or the workstation of the support organization.

[0101] The machine readable medium 550 includes instructions 559 to identify a cause of the identified failure based on the production data and the additional production data during operation of the assembly line. As described herein, the cause of an identified failure can be identified utilizing the production data and / or the additional production data while the assembly line continues to operate. In this way, the assembly line may not have to be shut down while the cause of the identified failure is determined.

[0102] In some embodiments, the machine readable medium 550 includes instructions to adjust a production setting of the assembly line based on the cause of the identified failure. In some embodiments, the production setting of the assembly line includes a distribution disk position, a distribution disk rotational speed, a chemical temperature, and a chemical flow rate. As described herein, the production setting of the assembly line can refer to altering a function that is performed by the assembly line. For example, the distribution disk of an application promoting layer can rotate at a particular speed to deposit a chemical on a surface. In this example, the particular speed can be controlled by a setting. In this way, the setting of the distribution disk can be altered to change the speed of the distribution disk that will alter how the chemical is deposited on the surface.

[0103] As used herein, the distribution disk position can refer to a distance or angle between the distribution disk and the surface. In addition, the distribution disk rotational speed can refer to a speed (e.g., rotations per minute, etc.) at which the distribution disk rotates. In some embodiments, the chemical temperature can refer to a temperature of the chemical or substance when deposited by the distribution disk. Furthermore, the chemical flow rate can refer to a rate that the chemical or substance is distributed to the surface. These are specific examplesof particular devices or portions of devices that can be altered by setting changes to alter a performance of the device or assembly line.

[0104] In some embodiments, the machine readable medium 550 includes instructions to provide a first set of data associated with the cause of the identified failure to a first user with a first set of credentials and provide a second set of data associated with the cause of the identified failure to a second user with a second set of credentials. As described herein, the credentials can indicate a level of security for receiving or viewing particular information. In some embodiments, the first user can be associated with a support organization and the second user can be associated with a particular assembly line organization. As described herein, the assembly line can utilize sensitive information and / or trade secrets that they may not want to share with non-authorized individuals. In this way, the first set of credentials can identify the first user as having permission to view information from a plurality of assembly lines and the second set of credentials can identify the second user as having permission to view information from only a particular assembly line.

[0105] In some embodiments, the first set of data includes historical data for a plurality of different assembly lines and the second set of data includes data associated with only the assembly line of the second user. As described herein, the data stored by a database can include historical data from previous review requests and / or failure reports. In some embodiments, the previous review requests can be embedded or tagged with metadata to identify portions of the review request or historical data that include sensitive information. In this way, particular data can be utilized by the first user to help the second user determine a cause of the identified failure without providing the data to the second user.

[0106] Figure 6 illustrates an example of a system 680 for failure analysis and identification for assembly lines. In some embodiments, the system 680 can include a device 660. In some embodiments, the device 660 is a computing device that includes a processor resource 651 and a machine readable medium 650 to store instructions 661, 662, 663, 664, 665, 666, 667, 668, 669 that are executed by the processor resource 651 to perform particular functions. Figure 6 illustrates how a computing device can execute instructions to perform functions described herein. The device 660 can be a machine learning model operating on a computing device.

[0107] The device 660 can be communicatively coupled to an imaging device 615 through a communication path 670-2. As described herein, the imaging device 615 can capture images of an assembly line 600. For example, the imaging device 615 can capture an image of a surface of an application surface when a chemical layer (e.g., adhesive promoting layer, foam material, etc.) is applied to the application surface. The captured images from the imaging device 615 can be sent to the device 660 through the communication path 670-2 where the captured images can be analyzed.

[0108] In some embodiments, the system 680 can include a mobile device 671. As described herein, the mobile device 671 can be a device that is provided by a support organization to establish a secure communication (e.g., communication path 670-1, etc.) with the device 660. In some embodiments, the device 660 can be a workstation and / or mobile device management as described herein. In some embodiments, the mobile device 671 can capture data from the assembly line 600 and transmit the data through the communication path 670-1 to the device 660. For example, the mobile device 671 can include a camera that can be utilized to capture images from specific locations of the assembly line 600 and transmit the captured images to the device 660.

[0109] The device 660 includes instructions 661 stored by the machine readable medium 650 that are executed by the processor resource 651 to receive a review request that includes audio messages from a mobile device 671 during operation of the assembly line 600. As described herein, the review request can be initiated by the mobile device 671. In some embodiments, the mobile device 671 can be utilized by a user associated with the assembly line 600. In these embodiments, the user can initiate the review request utilizing the mobile device 671 in response to an identified failure of the assembly line 600. In these embodiments, audio messages can be received or monitored during a telecommunication between the mobile device 671 and the device 660. In some embodiments, the mobile device 671 can receive instructions to capture images of the assembly line 600. In these embodiments, the pictures can be transferred to the device 660 through the communication path 670-1.

[0110] In some embodiments, the machine readable medium 650 can include instructions executable by the processor resource 651 to access the imaging device 615 to capture images during operation of the assembly line 600 in response to receiving the review request from the mobile device 671. In some embodiments, the device 660 can access the imaging device 615and / or receive image data captured by the imaging device 615 in response to receiving the review request from the mobile device 671. In this way, the device 660 can monitor the assembly line 600 utilizing the monitoring system of the assembly line 600. In these embodiments, the device 660 may only have access to the imaging device 615 during a review request session. For example, a user can start a review request session by submitting a review request to the device 660 utilizing the mobile device 671 and end the review request session utilizing the mobile device 671 to send an end session notification to the device 660. In some embodiments, the device 660 can provide instructions to the imaging device 615 to capture images of the assembly line 600.

[0111] The device 660 includes instructions 662 stored by the machine readable medium 650 that are executed by the processor resource 651 to generate a transcription of the audio messages and tag information within the transcription with metadata. As described herein, the transcription can be generated from audio data transmitted to the device 660 through the communication path 670-1. In some embodiments, the transcription can be tagged with metadata to describe the context of the transcription. For example, the transcription can be tagged to identify additional information related to the equipment and / or chemicals utilized by the assembly line 600. In some embodiments, additional information provided within the metadata can be utilized to categorize the transcription information. For example, the transcription information can be tagged with metadata such that the transcription information can be extracted from a database based on the categorization.

[0112] The device 660 includes instructions 663 stored by the machine readable medium 650 that are executed by the processor resource 651 to compare the metadata of the transcription to failure data stored at a failure database to determine a failure category. As described herein, a failure category can be a type of failure, type of mechanical device, particular quality that is below a quality threshold, type of chemical utilized by the assembly line 600, and / or other category type that can be utilized to search for stored failure data in the failure database. In this way failure data can be searched and extracted to be utilized during the review request session. As described herein, the assembly line 600 can continue to operate during the review request session, which can make it important to extract relevant failure data from the failure database quickly.

[0113] The device 660 includes instructions 664 stored by the machine readable medium 650 that are executed by the processor resource 651 to identify production data associated with the identified failure category of the production process based on the metadata during operation of the assembly line. In some embodiments, the failure category can be utilized to determine what type of production data to request from the mobile device 671. For example, a first type of failure category can be associated with a first set of production data and a second type of failure category can be associated with a second set of production data. In this way, the device 660 can send instructions to the mobile device 671 to provide a particular set of production data based on the identified failure category.

[0114] The device 660 includes instructions 665 stored by the machine readable medium 650 that are executed by the processor resource 651 to compare the production data to threshold production data. In some embodiments, the mobile device 671 can provide the production data to the device 660 in response to a request from the device 660. As described herein, the device 660 can request specific production data from the mobile device 671 based on the type of failure category. The production data can be compared to threshold production data. In some embodiments, the device 660 can have access to production data thresholds that are based on federated data from a plurality of different assembly lines including the assembly line 600. In this way, the threshold production data can be determined based the production data and corresponding product quality metrics from the plurality of assembly lines and the assembly line 600.

[0115] The device 660 includes instructions 666 stored by the machine readable medium 650 that are executed by the processor resource 651 to request additional production data from the mobile device based on values of the production data that are outside values of the threshold production data. In some embodiments, one or more of the production data provided to the device 660 can be outside the threshold values for the corresponding production data. In some embodiments, the production data value that is outside a corresponding threshold value can indicate that other production data values may be outside corresponding threshold values. In some embodiments, the additional production data can include additional image data. For example, the production data value that is outside a corresponding threshold value can indicate that a particular device or portion of the device is malfunctioning. In this example, instructions can be provided to the mobile device 671 to capture an image of the particular device or portionof the device. In this example, the instructions to capture the image of the device can include a specific area of the device and / or area of the assembly line 600 associated with the device.

[0116] In some embodiments, the additional production data can be utilized to confirm a cause of the failure. For example, the initial production data can indicate that one of a plurality of possible causes are responsible for the defect or failure associated with the review request. In these embodiments, the additional production data can further limit the number of possible causes of the defect associated with the review request. In some embodiments, the additional production data can be utilized to identify the cause of the failure.

[0117] The device 660 includes instructions 667 stored by the machine readable medium 650 that are executed by the processor resource 651 to send instructions to the mobile device 671 to provide additional image data to be captured by the mobile device. As described herein, the mobile device 671 can include an imaging device. In these embodiments, the mobile device 671 can be utilized to capture images of the assembly line 600. As described herein, the device 660 can send instructions to the mobile device 6 1 to instruct a location to capture images of the assembly line 600. In this way, the device 660 can receive image data related to specific portions of the assembly line 600 based on the review request.

[0118] In some embodiments, the machine readable medium 650 can include instructions executable by the processor resource 651 to request information from the mobile device 671 that includes a type of the production process and a type of assembly line 600. In these embodiments, the type of production process identifies a product generated by the production process and the type of assembly line 600 identifies a plurality of steps and mechanical devices utilized by the assembly line 600. In some embodiments, the information requested from the mobile device 671 can be provided from stored information associated with the mobile device 671. For example, the support organization associated with the device 660 can provide the mobile device 671 to the organization associated with the assembly line 600. In this way, the support organization can configure the mobile device 671 with information associated with the assembly line 600. In these embodiments, the support organization and / or the assembly line organization can periodically update the mobile device 671 with information associated with the assembly line.

[0119] In other embodiments, the mobile device 671 can be utilized to capture images of a device or process of the assembly line 600. The captured images can be provided to the device 660 and the device 660 can analyze the received images to determine the type of devices and / orchemicals utilized by the assembly line 600. Tn some embodiments, the device 660 can extract information from the received images to identify properties of the assembly line 600. For example, the type or model of devices can be identified, chemical composition of a foam material or application promoting layer can be determined, and / or other information can be identified utilizing the captured images.

[0120] In some embodiments, the instructions to the mobile device 671 include a location to capture the additional image data and a type of image to be captured at the location. In some embodiments, the instructions to the mobile device 671 include instructions to utilize a hyperspectral camera to capture a hyperspectral image of a material layer to determine a mixing ratio of the material layer. As described herein, the additional image data can be captured at particular locations of the assembly line 600. In this way, the mobile device 671 can be instructed to capture images at particular locations, at particular angles, and / or under particular lighting conditions.

[0121] In some embodiments, the mobile device 671 can be instructed to capture a particular type of image. For example, the mobile device 671 can be instructed to capture video data of a mechanical device during operation. In this way, the device 660 can analyze the captured video to determine if the mechanical device is functioning properly. In other embodiments, the mobile device 671 can be instructed to capture the hyperspectral image of an applied material layer (e.g., foam material layer, chemical layer, application promoting layer, etc ). The hyperspectral image can be provided to the device 660 and the device 660 can analyze the hyperspectral image to determine a chemical composition of the material layer.

[0122] The device 660 includes instructions 668 stored by the machine readable medium 650 that are executed by the processor resource 651 to identify a cause of the identified failure based on the production data and the additional production data during operation of the assembly line 600. In some embodiments, the cause of the identified failure is one of a chemical process or mechanical process of the assembly line 600. As described herein, the cause of the identified failure can be a result of a mechanical device not functioning within a manufacturing specification, a chemical composition that is outside a threshold, and / or a combination thereof. In some embodiments, the identified failure can be utilized to determine how to fix the failure. For example, the identified failure can be utilized to determine settings associated with the assembly line 600 that may be changed to alter a function of the assembly line 600.

[0123] The device 660 includes instructions 669 stored by the machine readable medium 650 that are executed by the processor resource 651 to send setting changes to be applied to the assembly line 600 to the mobile device 671 based on the identified failure. In some embodiments, the device 660 can provide instructions to the mobile device 671 to instruct a user of the mobile device 671 to alter settings associated with the assembly line 600. In some embodiments, the device 660 can be utilized to alter settings associated with the assembly line 600 through a direct connection with a control panel of the assembly line 600.

[0124] In some embodiments, the machine readable medium 650 can be executed by the processor resource 651 to alter a setting of the assembly line 600 based on the cause of the identified failure and the setting changes sent to the mobile device 671. In some embodiments, the settings of the assembly line 600 can be changed or altered in response to sending the setting changes to the mobile device 671. In some embodiments, the setting changes can correspond to altering the production data that are outside the threshold production data values.

[0125] In some embodiments, the machine readable medium 650 can be executed by the processor resource 651 to compare updated production data of the production process to the threshold production data. In these embodiments, the updated production data is collected after the setting of the assembly line 600 is altered by the device 660 or altered by a control panel of the assembly line 600. As described herein, the settings of the assembly line can be altered and the production data can be monitored after the settings have been altered to determine if the altered settings affected the production data. In some embodiments, the updated production data can be monitored to determine if the updated production data is within the threshold production data. In these embodiments, the settings can be further altered in response to the updated production data not being within the threshold production data value. In some embodiments, the updated production data can be utilized to determine an updated identified failure when the setting change does not alter the updated production to a value that is within the threshold production values.

[0126] In some embodiments, the machine readable medium 650 can be executed by the processor resource 651 to provide a first set of data associated with the cause of the identified failure to a first user with a first set of credentials and provide a second set of data associated with the cause of the identified failure to a second user with a second set of credentials. In these embodiments, the first set of data includes historical data for a plurality of different assemblylines and the second set of data includes data associated with only the assembly line 600. As described herein, the data stored by the device 660 and / or the support organization can be sensitive information that may not be shared with other organizations. In this way, some data received from the assembly line 600 may not be shared with other organizations associated with different assembly lines. Although the information can be utilized by the support organization and / or the device 660, the information may be utilized to prevent specific information from being provided to unauthorized users.

[0127] For example, images captured by a different mobile device can be utilized by the device 660 to identify a particular failure associated with a different assembly line. In this example, the device 660 can compare the captured images of the different mobile device to captured images from the mobile device 671. In this way, the images form the mobile device 671 can be utilized to determine a failure associated with the assembly line 600. However, the images captured by the different mobile device may not be provided to a user of the mobile device 671 since the user of the mobile device 6 1 may not have authorization for the different assembly line. Thus, the historical data from the different assembly line may not be provided to the organization of the assembly line 600.

[0128] Although specific embodiments have been described above, these embodiments are not intended to limit the scope of the present disclosure, even where only a single embodiment is described with respect to a particular feature. Examples of features provided in the disclosure are intended to be illustrative rather than restrictive unless stated otherwise. The above description is intended to cover such alternatives, modifications, and equivalents as would be apparent to a person skilled in the art having the benefit of this disclosure.

[0129] The scope of the present disclosure includes any feature or combination of features disclosed herein (either explicitly or implicitly), or any generalization thereof, whether or not it mitigates any or all of the problems addressed herein. Various advantages of the present disclosure have been described herein, but embodiments may provide some, all, or none of such advantages, or may provide other advantages.

[0130] In the foregoing Detailed Description, some features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the disclosed embodiments of the present disclosure have to use more features than are expressly recited in each claim. Rather, as thefollowing claims reflect, inventive subject matter lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment.

Claims

ClaimsWhat is claimed is:

1. A device, comprising: an imaging device to capture image data of a production process of an assembly line during operation of the assembly line; and a computing device communicatively coupled to the imaging device, the computing device configured to: receive a review request that includes audio messages from a mobile device during operation of the assembly line; generate a transcription of the audio messages and tag information within the transcription with metadata; compare the metadata of the transcription to failure data stored at a failure database to determine a failure category; identify production data associated with the identified failure category of the production process based on the metadata during operation of the assembly line; compare the production data to threshold production data; request additional production data from the mobile device based on values of the production data that are outside values of the threshold production data; send instructions to the mobile device to provide additional image data to be captured by the mobile device; identify a cause of the identified failure based on the production data and the additional production data during operation of the assembly line; and send setting changes to be applied to the assembly line to the mobile device based on the identified failure.

2. The device of claim 1, wherein the computing device is further configured to alter a setting of the assembly line based on the cause of the identified failure and the setting changes sent to the mobile device.

3. The device of claim 2, wherein the computing device is further configured to compare updated production data of the production process to the threshold production data, wherein the updated production data is collected after the setting of the assembly line is altered by the computing device.

4. The device of claim 1, wherein the computing device is configured to access the imaging device to capture images during operation of the assembly line in response to receiving the review request from the mobile device.

5. The device of claim 1, wherein the instructions to the mobile device include a location to capture the additional image data and a type of image to be captured at the location.

6. The device of claim 1, wherein the instructions to the mobile device include instructions to utilize a hyperspectral camera to capture a hyperspectral image of a material layer to determine a mixing ratio of the material layer.

7. The device of claim 1, wherein the computing device is configured to request information from the mobile device that includes a type of the production process and a type of assembly line, wherein the type of production process identifies a product generated by the production process and the type of assembly line identifies a plurality of steps and mechanical devices utilized by the assembly line.

8. The device of claim 1, wherein the computing device is configured to provide a first set of data associated with the cause of the identified failure to a first user with a first set of credentials and provide a second set of data associated with the cause of the identified failure to a second user with a second set of credentials.

9. The device of claim 8, wherein the first set of data includes historical data for a plurality of different assembly lines and the second set of data includes data associated with only the assembly line.

10. The device of claim 1, wherein the cause of the identified failure is one of a chemical process or mechanical process of the assembly line.

Images