Edge cloud cooperative control method, system and device in automobile general assembly workshop scene and storage medium

Abstract

The invention discloses an edge cloud cooperative control method, system and device in an automobile general assembly shop scene and a storage medium, and relates to the technical field of cooperative manufacturing, and the method comprises the steps that an edge device obtains multi-dimensional data corresponding to a plurality of cooperative processing candidate schemes in the automobile general assembly shop scene and then sends the multi-dimensional data to a cloud server; the cloud end calculates a collaborative efficiency index corresponding to each dimension of the candidate scheme according to the multi-dimensional data corresponding to the candidate scheme; calculating a multi-dimensional coupling degree and a collaborative development index of each candidate scheme according to the collaborative efficiency index; calculating the collaborative score of each candidate scheme based on the multi-dimensional coupling degree, the collaborative development index and the historical execution correction coefficient; judging whether a candidate scheme of which the collaboration score is greater than a collaboration threshold exists or not; and if the pre-estimated execution time corresponding to the determined candidate scheme exists, determining the shortest pre-estimated execution time as a target scheme, and sending the target scheme to the edge device to be executed by the edge device. According to the method, the edge cloud cooperative control efficiency of the automobile assembly workshop can be improved.

Description

Technical Field

[0001] This application relates to the field of collaborative manufacturing technology, and in particular to an edge-cloud collaborative control method, system, equipment and storage medium in the context of an automobile final assembly workshop. Background Technology

[0002] As the automotive industry rapidly transforms towards intelligent, flexible, and large-scale production, the final assembly workshop, as a crucial link in vehicle manufacturing, integrates multiple complex processes such as chassis assembly, interior assembly, powertrain installation, wiring harness installation, and testing and debugging. This involves the coordinated operation of various manufacturing resources, including production equipment, material distribution, personnel scheduling, tooling fixtures, and AGVs (Automated Guided Vehicles). With the industry trend of multi-model mixed-line production and dynamically changing order demands, the final assembly workshop faces higher requirements for real-time allocation of manufacturing resources, seamless process transitions, and overall collaboration.

[0003] Currently, manufacturing resource management and collaboration in traditional automobile assembly workshops largely rely on manual scheduling, fixed process configurations, and independent information system operation modes. Production planning mainly depends on manual scheduling, materials are delivered centrally on a timed and quantitative basis, and resources such as equipment, logistics, and personnel belong to different management modules. Information exchange relies on offline communication or single system input, and each link executes tasks independently according to preset processes. Overall production organization is completed through manual inspection, fixed pace advancement, and post-event statistics, without forming a unified resource collaborative scheduling mechanism.

[0004] Under the aforementioned traditional approach, automobile assembly workshops generally suffer from low efficiency in edge-cloud collaborative control. Summary of the Invention

[0005] This application provides an edge-cloud collaborative control method, system, equipment, and storage medium for automobile final assembly workshops, which can improve the efficiency of edge-cloud collaborative control in automobile final assembly workshops.

[0006] To achieve the above objectives, this application adopts the following technical solution: Firstly, this application provides an edge-cloud collaborative control method for an automotive final assembly workshop scenario, including: Edge devices acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning data, material coordination data, production line equipment data, supply chain coordination data, and historical execution data. The edge device sends multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server. The cloud server calculates the collaborative efficiency index of each candidate solution in each dimension based on the multi-dimensional data corresponding to the candidate solutions; it then calculates the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative efficiency index of each candidate solution in each dimension; based on the multi-dimensional coupling degree, collaborative development index, and historical execution correction coefficient of each candidate solution, it calculates the collaborative score of each candidate solution; it then determines whether there are candidate solutions with a collaborative score greater than the collaborative threshold; if there are candidate solutions with a collaborative score greater than the collaborative threshold, it determines the estimated execution time corresponding to the candidate solutions with a collaborative score greater than the collaborative threshold, and selects the candidate solution with the shortest estimated execution time as the target solution. The cloud server sends the target solution to the edge device; The edge device executes the target scheme.

[0007] Optionally, the method further includes: If there are no candidate solutions with a collaboration score greater than the collaboration threshold, the cloud server determines the reliability scores of the top N candidate solutions in the collaboration score ranking, and selects the candidate solution with the highest reliability score as the target solution.

[0008] Optionally, determining the estimated execution time corresponding to candidate solutions with a collaboration score greater than the collaboration threshold includes: Obtain the total duration of the serial process, the total duration of the parallel process, and the buffer duration for candidate schemes whose collaboration score is greater than the collaboration threshold; The estimated execution time is determined by the sum of the total duration of the serial process, the total duration of the parallel process, and the buffer duration.

[0009] Optionally, determining the reliability score of the top N candidate schemes in the collaborative ranking includes: Obtain the historical similarity, robustness, fault tolerance, and resource redundancy of the top N candidate solutions in the collaborative score ranking. A reliability score is determined based on historical similarity, scheme robustness, fault tolerance, and resource redundancy.

[0010] Optionally, the collaboration threshold is determined in the following way: Obtain the baseline static threshold, production condition complexity adjustment coefficient, system operation status adjustment coefficient, business priority adjustment coefficient, environmental disturbance risk adjustment coefficient, inter-dimensional coupling correction term, and historical feedback iteration term; The collaboration threshold is determined based on the baseline static threshold, the production condition complexity adjustment coefficient, the system operation status adjustment coefficient, the business priority adjustment coefficient, the environmental disturbance risk adjustment coefficient, the inter-dimensional coupling correction term, the historical feedback iteration term, and the preset boundary constraints.

[0011] Optionally, the edge device acquires multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario, including: Edge devices acquire multi-dimensional initial data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario; The edge device performs standardization processing on the multi-dimensional initial data to obtain multi-dimensional data.

[0012] Secondly, this application provides an edge-cloud collaborative control system for an automotive final assembly workshop scenario. The system includes edge devices and cloud servers: The edge device is used to acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning dimension data, material collaboration dimension data, production line equipment dimension data, supply chain collaboration dimension data, and historical execution dimension data. The edge device is used to send multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server; The cloud server is used to calculate the collaborative efficiency index of each candidate solution in each dimension based on the multi-dimensional data corresponding to the candidate solutions; calculate the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative efficiency index of each candidate solution in each dimension; calculate the collaborative score of each candidate solution based on the multi-dimensional coupling degree, collaborative development index and historical execution correction coefficient of each candidate solution; determine whether there is a candidate solution with a collaborative score greater than the collaborative threshold; if there is a candidate solution with a collaborative score greater than the collaborative threshold, determine the estimated execution time corresponding to the candidate solution with a collaborative score greater than the collaborative threshold, and determine the candidate solution with the shortest estimated execution time as the target solution. The cloud server is used to send the target solution to the edge device; The edge device is used to execute the target scheme.

[0013] Thirdly, this application provides a computing device, including a memory and a processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of the first aspects.

[0014] Fourthly, this application provides a computer-readable storage medium for storing a computer program for performing the method as described in any one of the first aspects.

[0015] As can be seen from the above technical solution, this application has at least the following beneficial effects: In this application, by collecting data from all dimensions of production planning, material coordination, production line equipment, supply chain coordination, and historical execution, and combining the coordination efficiency index, coupling degree, and coordination development index to construct a quantitative coordination score, resources such as production equipment, material distribution, and personnel scheduling that were originally belonging to different management modules are incorporated into a unified coordination evaluation system. This promotes the transformation of each link from independent execution to collaborative operation, achieves the global optimal matching of manufacturing resources, and significantly improves the overall resource coordination efficiency of the final assembly workshop.

[0016] Furthermore, by setting collaboration thresholds for a two-tiered selection logic, when the collaboration score meets the standard, the shortest estimated execution time is used as the criterion to ensure production cycle time and execution efficiency; when no solution meets the standard, the reliability score is the core, combined with multi-dimensional indicators such as historical similarity and robustness to select the optimal solution, taking into account both production continuity and risk resistance, thus getting rid of the subjectivity and experience of traditional manual scheduling, and making the formulation of resource collaboration solutions more in line with the actual production scenarios of mixed production lines of multiple vehicle models and dynamic changes in orders.

[0017] Furthermore, the collaboration threshold is not a fixed value, but rather a combination of a baseline static threshold and multi-dimensional adjustment coefficients, including production condition complexity, system operating status, business priority, and environmental disturbance risk. Closed-loop optimization is achieved through dimensional coupling correction terms, superimposed multi-factor influences, and historical feedback iteration terms. Simultaneously, preset boundary constraints allow the threshold to be dynamically adjusted in real time according to changes in the production scenario. This ensures the admission standards for collaboration solutions while avoiding situations where no solutions are available under extreme conditions, making resource collaboration control more flexible and practical.

[0018] Finally, through comprehensive data collection, standardized processing, quantitative evaluation, and scientific selection, the traditional static production organization model of timed and quantitative delivery, fixed-rhythm progress, and post-event statistics has been replaced. This has enabled dynamic resource allocation throughout the entire process, from production planning to material distribution, production line operation, and supply chain linkage. It can quickly respond to production needs such as order changes and changes in working conditions, effectively reduce problems such as idle resources, material backlog, and poor process connection, reduce production and operating costs, and improve the intelligent production level and market responsiveness of the automobile assembly workshop.

[0019] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0020] Figure 1 A flowchart illustrating an edge-cloud collaborative control method in an automotive assembly workshop scenario, provided as an embodiment of this application; Figure 2 A schematic diagram of an edge-cloud collaborative control system in an automobile assembly workshop scenario provided in this application embodiment; Figure 3 This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0021] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0022] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first: The automobile final assembly workshop is a key link in the whole vehicle manufacturing process. It integrates many complex processes such as chassis assembly, interior assembly, powertrain installation, wiring harness installation, testing and debugging. It is a centralized and coordinated operation scene for various manufacturing resources such as production equipment, material distribution, personnel scheduling, tooling fixtures, and AGV carts. It is also the final link in realizing the whole vehicle from parts assembly to finished product roll-off. Its operational efficiency directly determines the production capacity and delivery capability of the whole vehicle manufacturing.

[0024] Manufacturing resource collaboration refers to the overall coordination, real-time linkage, and efficient connection of various production factors such as production equipment, materials, personnel, tooling, logistics equipment, and supply chain resources in the automobile final assembly workshop. This enables the matching and coordination of various resources, processes, and links, ensuring the smooth progress of the production process according to the rhythm, and ultimately achieving the resource management model with the best overall production efficiency.

[0025] The main technical problem addressed in this application is the low efficiency of edge-cloud collaborative control and poor collaborative scheduling of manufacturing resources in automobile assembly workshops. With the rapid transformation of the automotive industry towards intelligence, flexibility, and large-scale production, and the dynamic changes in order demands across multiple vehicle models on mixed production lines, automobile assembly workshops face higher requirements for real-time allocation of manufacturing resources, precise process connection, and efficient global collaboration. Traditional management and collaboration models are no longer adequate to meet these demands, specifically manifesting as lagging resource allocation, poor process connection, resource imbalance, and slow response to anomalies. This directly leads to disrupted production rhythms, high unplanned downtime rates, and increased production and operating costs, making it difficult to meet the capacity and delivery requirements of vehicle manufacturing.

[0026] The main reasons for the above-mentioned major technical problems are multifaceted and overlapping. The main reason is that the traditional model lacks a scientific collaborative system adapted to the needs of modern production: on the one hand, it adopts an outdated management model of manual scheduling and modular control. Various resources such as production equipment, materials, and personnel are divided into different management modules and operate independently. There is no unified global collaborative scheduling mechanism. Moreover, the information systems of each module are independent of each other, and data interaction relies on offline communication or manual entry, forming information silos, resulting in a lack of real-time and accurate data support for resource collaboration. On the other hand, the formulation of collaborative solutions relies entirely on human experience, lacking a quantitative evaluation and selection system. At the same time, it adopts a static production organization mode with fixed process configuration and fixed pace, and the criteria for judging collaborative solutions lack clear quantitative thresholds or are only fixed static values. It cannot be dynamically adjusted according to actual scenarios such as production conditions, system status, and business priorities, resulting in a serious lack of accuracy, flexibility, and dynamic adaptability of resource collaboration, which ultimately restricts the collaborative efficiency of manufacturing resources in the final assembly workshop.

[0027] In view of this, embodiments of this application provide an edge-cloud collaborative control method in the context of an automobile final assembly workshop.

[0028] This application addresses the technical problems of low efficiency in edge-cloud collaborative control and poor performance in collaborative scheduling of manufacturing resources in automotive assembly workshops. Firstly, it collects multi-dimensional data on production planning, material coordination, production line equipment, supply chain coordination, and historical execution to construct a collaborative performance evaluation model that integrates a collaborative efficiency index, multi-dimensional coupling degree, and collaborative development index. This model scientifically quantifies and rates candidate collaborative processing solutions. Simultaneously, it sets a dynamic collaborative threshold as the admission standard for solutions and designs a two-layer selection logic prioritizing efficiency and reliability. When the collaborative score meets the standard, the solution with the shortest estimated execution time is selected to ensure production efficiency. When no solution meets the standard, solutions are selected based on reliability scores such as historical similarity and robustness to ensure production continuity. Furthermore, through multi-dimensional coefficient correction and historical feedback, the collaborative threshold is dynamically and adaptively adjusted. Ultimately, this forms a full-process resource collaborative control system from data collection, quantitative evaluation, scientific selection to solution execution. This system achieves global coordination, real-time linkage, and dynamic optimization of manufacturing resources in automotive assembly workshops, breaking down information barriers and management gaps. It makes the formulation and execution of resource collaboration solutions more aligned with actual production scenario needs, fundamentally improving the efficiency of manufacturing resource collaboration.

[0029] To make the technical solution of this application clearer and easier to understand, the following describes, in conjunction with the accompanying drawings, an edge-cloud collaborative control method for an automotive assembly workshop scenario provided by an embodiment of this application. Figure 1 As shown, this figure is a flowchart of an edge-cloud collaborative control method in an automobile assembly workshop scenario provided by an embodiment of this application. The method includes: S201. Edge devices acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning dimension data, material collaboration dimension data, production line equipment dimension data, supply chain collaboration dimension data, and historical execution dimension data.

[0030] The automobile final assembly workshop is the main operation scenario for vehicle manufacturing. It integrates multiple complex processes such as chassis assembly, interior assembly, powertrain installation, testing and debugging. It involves the coordinated operation of various manufacturing resources such as production equipment, material distribution, AGV carts, and personnel scheduling. It is also a real production application scenario for multi-model mixed-line production and dynamic order adjustment.

[0031] The candidate solutions for collaborative processing are a variety of pre-defined resource collaboration execution strategies to achieve efficient collaboration of manufacturing resources in the automotive final assembly workshop and complete predetermined production goals (such as production schedule execution, response to order changes, and resolution of material shortages). Each solution covers the resource allocation and execution logic of the entire process, including production plan adjustment, material distribution methods, production line equipment scheduling, and supply chain linkage.

[0032] Production planning data includes various data related to workshop production scheduling, such as order priority, daily production quantity, vehicle model launch queue, single vehicle model production cycle time, and order change probability / quantity. Material coordination dimension data includes various data related to the supply and distribution of production materials in the workshop, including material BOM level, material completeness rate, line-side inventory capacity, material delivery route, JIT / JIS response time, supplier delivery cycle, etc. Production line equipment dimension data includes various data related to the final assembly line and various operating equipment, such as equipment utilization rate, equipment failure probability, equipment maintenance time, workstation operation time, production line bottleneck cycle time, and line buffer capacity. Supply chain collaboration dimension data includes various data related to upstream supply chain resources in the workshop, such as the capacity of major suppliers, the time for cross-plant material transfer, logistics capacity, emergency replenishment cycle, and supplier supply guarantee rate. Historical execution dimension data includes various data related to the past execution of similar collaborative solutions, such as historical execution duration, number of unplanned line stoppages, material deviation rate, anomaly handling time, and solution execution success rate.

[0033] The specific process for obtaining multi-dimensional data is as follows: First, the edge device acquires multi-dimensional initial data corresponding to multiple collaborative processing candidate solutions in the automotive assembly workshop scenario; the edge device then standardizes the multi-dimensional initial data to obtain multi-dimensional data.

[0034] Standardization is a process of normalizing and homogenizing multi-dimensional initial data by using a unified mathematical method to address the original characteristics of the data. The goal is to eliminate differences in dimensions (such as time, quantity, and ratio) and magnitudes between different data, as well as to unify positive / negative indicator attributes, so that the original data that could not be directly compared or calculated can be transformed into standardized data that can be uniformly analyzed and calculated.

[0035] Specifically, the edge device first connects to various information systems such as workshop MES, LES, WMS, equipment management system, and supply chain management system, as well as hardware nodes such as production line sensors and data acquisition terminals. In the actual production scenario of the automobile final assembly workshop, it collects raw data corresponding to five dimensions of all pre-defined collaborative processing candidate solutions, i.e., multi-dimensional initial data. Because these initial data have different units (e.g., production cycle time in "seconds", material availability rate in "%"), large differences in magnitude (e.g., supplier capacity in "thousands of units / month", workstation operation time in "minutes / unit"), and different indicator attributes (e.g., equipment utilization rate is a positive indicator, failure probability is a negative indicator), they cannot be directly used for subsequent quantitative calculations. Therefore, the edge device then performs standardization processing on the acquired multi-dimensional initial data, eliminating the above differences through a unified mathematical method, and transforming it into multi-dimensional data with a unified format that can be directly calculated and analyzed. This provides accurate data support for subsequent calculation of quantitative indicators such as collaborative efficiency index, multi-dimensional coupling degree, and collaborative score of each candidate solution.

[0036] The standardized expression for a positive indicator is:

[0037] The standardized expression for the negative indicator is:

[0038] in, Let represent the standardized values ​​of the i-th candidate solution, the j-th dimension, and the k-th metric. Let represent the initial data for the i-th candidate solution, the j-th dimension, and the k-th metric. This represents the initial data set for the j-th dimension and k-th metric among all candidate solutions. Represents a set The minimum value among all candidate solutions, that is, the minimum original value of this index. Represents a set The maximum value in the range is the maximum original value of the index among all candidate solutions.

[0039] S202, The edge device sends multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server.

[0040] A cloud server is a high-performance computing and data processing server deployed in the cloud. It has powerful computing and storage capabilities and is responsible for complex algorithm calculations, collaborative evaluation, scheme selection, and decision-making. It is the decision center in the edge-cloud collaborative architecture.

[0041] After completing the collection and standardization of multi-dimensional initial data on site, the edge devices upload the resulting multi-dimensional data, which can be directly used for calculation, to the cloud server through the workshop network. This provides a complete and unified data source for the cloud server to carry out subsequent collaborative efficiency index calculation, multi-dimensional coupling degree calculation, collaborative sub-calculation, and optimal solution decision-making, thereby realizing a collaborative control architecture of edge collection and cloud decision-making.

[0042] S203. The cloud server calculates the collaborative efficiency index of the candidate solution in each dimension based on the multi-dimensional data corresponding to the candidate solution.

[0043] The synergy effectiveness index is a numerical value used to quantitatively evaluate the synergy effect of candidate solutions in a single dimension. The value ranges from [0,1]. The larger the value, the better the synergy performance of the solution in that dimension.

[0044] Specifically, the cloud server performs weighted or average calculations on all standardized feature indicators of each candidate solution within a certain dimension to obtain a comprehensive value, namely the collaborative effectiveness index of the candidate solution in that dimension. The purpose of this step is to reduce the dimensionality of the multi-indicator data under each dimension, forming a single, comparable quantitative indicator. This provides a foundation for subsequent calculations of multi-dimensional coupling degree, collaborative development index, and collaborative score, thereby achieving an accurate and objective assessment of the collaborative capabilities of candidate solutions.

[0045] The formula for calculating the synergy effectiveness index is:

[0046] in, This represents the synergistic effectiveness index of the i-th candidate solution in the j-th dimension. This represents the number of indicators in the j-th dimension. This represents the standardized values ​​of the i-th candidate solution, the j-th dimension, and the k-th indicator.

[0047] S204. The cloud server calculates the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative effectiveness index corresponding to each dimension.

[0048] Multidimensional coupling degree is a numerical value used to quantify the degree of synergy and interaction between a single candidate solution and five dimensions. The value range is [0,1]. The larger the value, the better the linkage and adaptability between the dimensions, and the less the problem of disconnection between dimensions or imbalance in resource matching.

[0049] The collaborative development index is a comprehensive value obtained by weighting the collaborative efficiency index of all dimensions of a single candidate solution by combining the actual weight of each dimension in the resource collaboration of the automobile assembly workshop. The value range is [0,1]. The larger the value, the better the overall collaborative efficiency of the candidate solution.

[0050] After obtaining the collaboration efficiency indexes for each candidate solution across five dimensions, including production planning and material coordination, the cloud server performs two calculations based on these indices. First, it calculates the multi-dimensional coupling degree of each candidate solution to measure the degree of cooperation, linkage, and adaptation among the five dimensions under the solution, avoiding situations where a single dimension is highly efficient but the other dimensions are disconnected. Second, it calculates the collaboration development index of each candidate solution by weighting the collaboration efficiency indices of each dimension according to the actual importance of the five dimensions in the resource coordination of the automotive assembly workshop, thus quantifying and evaluating the overall collaboration efficiency level of the solution.

[0051] The expression for calculating multidimensional coupling degree is:

[0052] in, This represents the multi-dimensional coupling degree of the i-th candidate solution. This represents the total number of evaluation dimensions in this application. These correspond to five dimensions: production planning, material coordination, production line equipment, supply chain coordination, and historical execution. Let represent the synergistic effectiveness index of the i-th candidate solution in the j-th dimension.

[0053] The formula for calculating the coordinated development index is:

[0054] in, Let represent the collaborative development index of the i-th candidate scheme. This represents the weight coefficient of the j-th dimension.

[0055] These two indices provide a more comprehensive and in-depth quantitative analysis of the collaborative capabilities of candidate solutions from two different dimensions: inter-dimensional linkage and overall collaborative foundation. Compared with a single-dimensional efficiency index, they are more in line with the actual production needs of multi-resource linkage and multi-stage connection in automobile assembly workshops, and provide a scientific and comprehensive quantitative basis for subsequent calculation of the collaborative score that reflects the overall collaborative capability of the solution.

[0056] S205. The cloud server calculates the collaborative score of each candidate solution based on the multi-dimensional coupling degree, collaborative development index, and historical execution correction coefficient of each candidate solution.

[0057] The historical execution correction coefficient is a correction value calculated based on the actual execution effect (such as execution success rate, anomaly rate, downtime, etc.) of similar collaborative solutions in historical production. The value range is [0,1]. It is used to correct the evaluation results of the current solution, avoid evaluation bias caused by relying solely on current data, and improve the reliability and historical adaptability of solution evaluation.

[0058] The synergy score refers to the final value obtained by integrating multi-dimensional coupling degree, synergy development index and historical execution correction coefficient, which is used to comprehensively and quantitatively evaluate the overall manufacturing resource synergy capability of candidate solutions. The value range is [0,1], and the larger the value, the stronger the overall synergy capability of the solution.

[0059] After obtaining the multi-dimensional coupling degree (measuring the degree of linkage and matching between dimensions) and the collaborative development index (measuring the basis of comprehensive collaborative effectiveness) of each candidate solution, the cloud server combines the historical execution correction coefficient obtained from the historical execution of the same type of solution, and performs fusion calculation through a preset quantitative model to obtain the collaborative score of each candidate solution.

[0060] The formula for calculating the collaborative score is:

[0061] in, This represents the collaborative score of the i-th candidate solution. Indicates the weighting coefficient. This represents the multi-dimensional coupling degree of the i-th candidate solution. Let represent the collaborative development index of the i-th candidate scheme. This represents the historical execution correction factor.

[0062] This step organically integrates three key elements: interdimensional linkage, comprehensive collaborative foundation, and historical execution reliability. It avoids the one-sidedness of evaluating a single indicator and enables the collaborative score to comprehensively, objectively, and accurately reflect the overall resource collaboration capability of candidate solutions in the actual production scenario of the automobile assembly workshop. This provides a scientific and reliable quantitative basis for subsequent judgment on whether a solution meets the standard and for selecting the optimal target solution.

[0063] S206. The cloud server determines whether there are candidate solutions with a collaboration score greater than the collaboration threshold. If there are candidate solutions with a collaboration score greater than the collaboration threshold, the estimated execution time corresponding to the candidate solution with a collaboration score greater than the collaboration threshold is determined, and the candidate solution with the shortest estimated execution time is determined as the target solution.

[0064] The collaboration threshold is a dynamic quantitative judgment standard adapted to the current production scenario of the automobile final assembly workshop. It is obtained by combining the baseline static threshold with multi-dimensional coefficients such as production condition complexity, system operation status, and business priority. It is the critical value for screening candidate solutions with basic collaboration capabilities.

[0065] The collaboration threshold can be determined in the following ways: First, the cloud server obtains the baseline static threshold, production condition complexity adjustment coefficient, system operation status adjustment coefficient, business priority adjustment coefficient, environmental disturbance risk adjustment coefficient, inter-dimensional coupling correction term, and historical feedback iteration term.

[0066] The baseline static threshold is the basic anchor point and initial reference value of the collaborative threshold. It is a fixed value preset by combining the normal production scenario of the automobile final assembly workshop, industry general standards and actual production needs. It provides a benchmark framework for the dynamic adjustment of the collaborative threshold and takes the value of a fixed constant, such as 0.75.

[0067] The production condition complexity adjustment coefficient is a coefficient used to quantify and correct the impact of the complexity of the production scenario on the collaboration threshold. Its value changes dynamically with factors such as the number of mixed-line models in the final assembly workshop, production cycle time, order change rate, and process complexity. The more complex the condition, the larger the coefficient value, which is used to improve the collaboration threshold to raise the solution access standard.

[0068] The system operation status adjustment coefficient is a correction coefficient calculated based on the operational stability of the production line equipment, material system, data interaction, cloud-edge collaboration, and other systems in the final assembly workshop. The worse the system operation status (such as low equipment utilization rate or insufficient material availability rate), the larger the coefficient value. The threshold is adjusted to adapt to the actual operation of the system.

[0069] The business priority adjustment coefficient is a correction coefficient calculated by combining business requirements such as order priority, delivery urgency, and quality control level in the final assembly workshop. The higher the business importance and urgency (such as strategic orders and urgent delivery orders), the larger the coefficient value, which is used to increase the collaboration threshold to match the collaboration requirements of high-priority businesses.

[0070] The environmental disturbance risk adjustment coefficient is a correction coefficient calculated for environmental disturbance factors such as external supply chain risks, equipment failure risks, and logistics and distribution risks in the final assembly workshop. The higher the level of external risk, the larger the coefficient value. By adjusting the threshold, collaborative solutions with stronger risk resistance are selected.

[0071] The interdimensional coupling correction term is used to quantify the correction value of the coupling effect of the four dimensions of production conditions, system status, business priority, and environmental disturbance risk that are superimposed and interact with each other. When multiple adverse factors exist at the same time, this correction term will amplify the threshold adjustment range, reflecting the impact of multiple factors superimposed on the admission criteria of collaborative solutions.

[0072] The historical feedback iteration item is an iterative correction value calculated based on the execution effect of the collaborative scheme in the past production shifts of the final assembly workshop (such as execution success rate, unplanned downtime, deviation between actual collaborative score and threshold, etc.). This achieves closed-loop optimization of the collaborative threshold, allowing the threshold to be dynamically adapted to actual production based on historical execution experience, thereby improving the scientific and rational nature of the threshold setting.

[0073] Then, the cloud server determines the collaboration threshold based on the baseline static threshold, the production condition complexity adjustment coefficient, the system operation status adjustment coefficient, the business priority adjustment coefficient, the environmental disturbance risk adjustment coefficient, the inter-dimensional coupling correction term, the historical feedback iteration term, and the preset boundary constraints.

[0074] Preset boundary constraints are hard constraints set at upper and lower limits (such as a lower threshold of 0.6 and an upper threshold of 0.95) to prevent threshold overflow from the reasonable industrial range under extreme operating conditions. This ensures that the collaborative threshold is always within the feasible industrial range, avoiding both the failure of the solution due to an excessively low threshold and the lack of usable solutions due to an excessively high threshold.

[0075] After obtaining the baseline static threshold, the adjustment coefficients of the four dimensions, the inter-dimensional coupling correction term, and the historical feedback iteration term, the cloud server performs comprehensive calculations on these parameters using a preset nonlinear fusion formula to obtain a preliminary threshold result. Subsequently, the result is compared with the preset boundary constraints. If it exceeds the upper or lower limit, it is truncated and corrected to finally determine the dynamic collaborative threshold that is suitable for the current production scenario.

[0076] The expression for calculating the collaboration threshold is:

[0077] in, Indicates the collaboration threshold. Indicates the baseline static threshold. This represents the adjustment factor for production condition complexity. This represents the system operating status adjustment coefficient. This represents the business priority adjustment factor. This represents the environmental disturbance risk adjustment coefficient. This represents the weighting coefficient of the dimensional coupling effect. This represents the inter-dimensional coupling correction term. This represents the historical feedback iteration weight coefficient. This represents the historical feedback iteration term corresponding to the t-th production shift.

[0078] The calculation expression is:

[0079] in, This indicates that the complexity of the operating conditions affects the weight. This represents the number of SKUs in production at time t1, i.e., the number of different models or configurations currently being produced in the final assembly line. Indicates the number of baseline SKUs. This represents the power coefficient of the change in the number of SKUs. Indicates the baseline production cycle time. This represents the actual production cycle time at time t1. The power coefficient representing the change in production cycle time. This represents the order change rate at time t1. The power factor representing the order change rate. The process complexity factor at time point t1 (such as the introduction of a new process, the addition of a process step, etc.) represents the process complexity factor at time point t1.

[0080] The calculation expression is:

[0081] in, This indicates the weight of the influence of the system state. This represents the first weighting coefficient. This represents the overall efficiency of the equipment at time point t1. This represents the second weighting coefficient. This represents the material availability rate at time t1. This represents the third weighting coefficient. This indicates the availability of data interaction at time point t1. This represents the fourth weighting coefficient. This represents the edge computing / cloud-edge collaboration availability at time point t1.

[0082] The calculation expression is:

[0083] in, This indicates that business priority affects the weight. express The weighting coefficients, This indicates the order priority (e.g., strategic orders, urgent orders) at time point t1. express The corresponding weighting coefficients, This indicates the urgency of delivery at time t1. express The corresponding weighting coefficients, This indicates the quality control level at time point t1. This represents the power coefficient of the overall business score.

[0084] The calculation expression is:

[0085] in, Indicates the weight of environmental risk impact. express The weighting coefficients, This indicates the supply chain risk level at time point t1. express The weighting coefficients, This indicates the equipment failure risk level at time point t1. express The weighting coefficients, This indicates the risk level of logistics delivery at time point t1.

[0086] The calculation expression is:

[0087]

[0088] in, A standardized correction value representing the complexity of production conditions. Standardized correction values ​​representing the system's operating status. A standardized adjustment value representing business priority. Standardized correction values ​​representing the risk of environmental disturbance.

[0089] The calculation expression is:

[0090] in, This indicates the weighting coefficient of historical feedback influence. Indicates recent The average success rate of the collaborative solution execution within each shift. Indicates the length of the time window for historical feedback. Indicates recent The average production loss time caused by abnormal execution of the collaboration plan within each shift. A benchmark reference value representing production loss. Indicates recent Within each shift, the actual average collaboration score of the collaboration plan. Indicates recent The average collaborative threshold actually used within each shift. This represents the baseline static threshold.

[0091] The expression for the preset boundary constraints is:

[0092] in, Indicates the lower limit of the threshold. This indicates the upper limit of the threshold.

[0093] The purpose of this step is to make the collaboration threshold no longer a fixed static value, but a dynamic standard that can fully respond to real-time changes such as production complexity, system operating status, business priority, and environmental disturbance risks, and to perform closed-loop optimization in combination with historical execution experience. This provides a suitable basis for the scientific selection of subsequent candidate solutions, ensuring that the solution admission standards are highly matched with actual production needs.

[0094] The estimated execution time is the total estimated time from the start of execution to the completion of all collaborative tasks, calculated by a quantitative model based on the resource allocation logic of the candidate solution. It comprehensively covers the serial links, parallel links, and buffer redundancy time, and is an indicator for measuring the execution efficiency of the solution.

[0095] The specific process for determining the estimated execution time is as follows: First, the cloud server obtains the total duration of the serial process, the total duration of the parallel process, and the buffer duration for candidate solutions whose collaboration scores are greater than the collaboration threshold.

[0096] The total time of sequential steps refers to the total time spent on all steps in a candidate solution that must be executed in a fixed order, such as order breakdown, plan issuance, BOM verification, and exception closure. These steps cannot be carried out in parallel, and their total time directly constitutes the basic part of the solution execution time.

[0097] The total duration of parallel processes refers to the execution time of the longest process among the multiple processes that can be carried out simultaneously in the candidate solution. For example, processes such as material distribution, production line adjustment, equipment maintenance, and supplier inventory preparation can be carried out in parallel, and the total duration is determined by the longest process.

[0098] Buffer time refers to the redundant time reserved to cope with possible sudden anomalies in the production process (such as material delays, minor equipment failures, temporary personnel reassignment, etc.). It is used to ensure the continuity and stability of the plan execution and avoid the overall execution timeout caused by minor disturbances.

[0099] After selecting candidate solutions with a collaboration score greater than the collaboration threshold, the cloud server extracts and calculates three types of key time data from the execution flow of each solution: the total duration of all sequential steps that must be executed in order; the duration of the longest-running step among all concurrently executable steps; and the buffer time reserved to handle unexpected anomalies. These three types of data serve as input for subsequent calculations of the estimated execution time of the solution. By acquiring this data, the cloud server can lay the foundation for evaluating the execution efficiency of the solution and selecting the most efficient target solution.

[0100] The expression for calculating the total duration of the serial link is:

[0101] in, This represents the total duration of the serial steps in the i-th candidate scheme. This represents the standard execution time of the p-th serial step in the i-th candidate scheme. Indicates the risk coefficient. Let represent the time variation coefficient of the p-th serial step in the i-th candidate scheme. This indicates the total number of serial links.

[0102] The expression for calculating the total duration of parallel processes is:

[0103] in, This represents the total time of the m-th parallel step in the i-th candidate solution. This represents the baseline execution time of the m-th parallel step in the i-th candidate solution. Let m represent the parallelism of the m-th parallel step in the i-th candidate scheme. This represents the coefficient affecting the coverage rate. This represents the resource guarantee rate of the m-th parallel step in the i-th candidate scheme.

[0104] The expression for calculating the buffer duration is:

[0105] in, This represents the buffer duration for the i-th candidate solution. Indicates the buffer coefficient. This represents the total duration of the serial steps in the i-th candidate scheme. Let represent the total duration of the m-th parallel step in the i-th candidate scheme.

[0106] Then, the sum of the total duration of the serial stage, the total duration of the parallel stage, and the buffer duration is determined as the estimated execution time. The expression for calculating the estimated execution time is:

[0107] in, Let represent the estimated execution time of the i-th candidate solution. This represents the total time of the first parallel step in the i-th candidate solution. This represents the total time of the second parallel step in the i-th candidate solution. This represents the total time of the m-th parallel step in the i-th candidate solution. This represents the buffer duration for the i-th candidate solution.

[0108] When the cloud server determines that there are candidate solutions whose collaboration score exceeds the dynamic collaboration threshold, it means that there are solutions that meet the basic collaboration requirements in the current production scenario. At this time, the cloud server will calculate the corresponding estimated execution time for each such qualified solution based on its process design. Finally, with maximizing production efficiency as the main selection criterion, the solution with the shortest estimated execution time will be selected from these qualified solutions and determined as the final target solution.

[0109] This logic aligns with the automotive assembly plant's need for rhythmic and efficient production. It ensures the basic collaborative capabilities of the solution through collaboration thresholds and guarantees that the solution can fit the production rhythm to the maximum extent through execution time sequencing, avoiding resource idleness and process delays, and achieving optimal efficiency in manufacturing resource collaboration.

[0110] If there are no candidate solutions with a collaboration score greater than the collaboration threshold, the cloud server determines the reliability score of the top N candidate solutions in terms of collaboration score, and selects the candidate solution with the highest reliability score as the target solution.

[0111] The specific process for determining the reliability score is as follows: First, the cloud server obtains the historical similarity, robustness, fault tolerance, and resource redundancy of the top N candidate solutions in the collaborative score ranking.

[0112] The top N candidate schemes in the collaboration score ranking are selected from the top N schemes (N is a preset positive integer) after the previous collaboration score calculation and threshold comparison. When there are no schemes that meet the collaboration score standard, these schemes are ranked from high to low. Although these schemes do not reach the collaboration threshold, their collaboration ability is relatively optimal among all candidate schemes, and they are the objects of reliability selection.

[0113] Historical similarity refers to the degree of overlap between the current candidate solution and the collaborative solutions previously executed in the automotive assembly workshop in terms of resource allocation logic, production condition adaptability, and business requirement matching. The value ranges from [0,1]. The larger the value, the more similar the solution is to historical successful cases and the lower the execution risk.

[0114] Solution robustness refers to the ability of a candidate solution to maintain stable collaborative functions and avoid overall failure when faced with minor fluctuations in the production scenario (such as minor order adjustments, minor delays in material delivery, or short-term equipment shutdowns). It is an indicator for measuring the solution's resistance to interference.

[0115] Anomaly tolerance refers to the ability of a candidate solution to respond to and mitigate sudden anomalies during the production process (such as shortage of critical materials, AGV malfunctions, or temporary staff shortages). Specifically, it is reflected in the effectiveness of the solution's built-in anomaly handling procedures and alternative resource allocation. The higher the value, the stronger the solution's ability to cope with emergencies.

[0116] Resource redundancy refers to the degree of surplus of manufacturing resources (such as material inventory, spare equipment, mobile personnel, and logistics capacity) in a candidate solution relative to the minimum requirements for solution execution. The value range is [0,1]. A solution with reasonable redundancy can quickly replenish resources when they are lost, ensuring the smooth execution of the solution.

[0117] When the cloud server determines that there are no candidate solutions with a collaboration score greater than the collaboration threshold, it will switch to a reliability-first selection logic. First, it will sort the candidate solutions from high to low collaboration scores and select the top N candidate solutions with the best collaboration capabilities. Then, the cloud server will retrieve four types of evaluation data corresponding to these N solutions by connecting to the workshop's historical execution database, solution design documents, resource management system and other data sources, respectively: historical similarity, solution robustness, anomaly tolerance capability and resource redundancy.

[0118] The expression for calculating historical similarity is:

[0119] in, Indicate the historical similarity of the i-th candidate scheme. This represents the total number of evaluation dimensions. This represents the total number of indicators across all dimensions, and the number of indicators for each dimension. Summary This represents the number of indicators in the j-th dimension. Let represent the standardized values ​​of the i-th candidate solution, the j-th dimension, and the k-th metric. This represents the standardized reference value of the j-th dimension and the k-th indicator in the historically optimal collaborative solution.

[0120] The formula for calculating the robustness of the scheme is:

[0121] in, Indicates the robustness of the i-th candidate solution. This represents the total number of evaluation dimensions. Let denote the partial derivative term, representing the synergistic efficacy index of the j-th dimension. The k-th standardized index under this dimension The degree of sensitivity, This represents the historical standard deviation of the j-th dimension and the k-th indicator.

[0122] The formula for calculating fault tolerance capability is:

[0123] in, This represents the fault tolerance capability of the i-th candidate solution. This indicates the total number of preset abnormal scenarios. This represents the collaborative performance loss value of the i-th candidate solution under the ab-th anomalous scenario. Let represent the collaborative score of the i-th candidate solution under normal operating conditions.

[0124] The formula for calculating resource redundancy is:

[0125] in, This represents the resource redundancy of the i-th candidate solution. This indicates the total number of resource types included in the redundancy assessment. This represents the actual allocation amount of the u-th type of resource in the i-th candidate scheme. This represents the minimum threshold requirement for the i-th candidate solution to execute the u-th resource. This represents the theoretical maximum capacity of the i-th candidate scheme to execute the u-th type of resource.

[0126] Then, the cloud server determines the reliability score based on historical similarity, solution robustness, fault tolerance, and resource redundancy. The formula for calculating the reliability score is:

[0127] in, This represents the reliability score of the i-th candidate solution. Weighting coefficients representing historical similarity The weighting coefficients represent the robustness of the solution. The weighting coefficients represent the fault tolerance capability. The weighting coefficient represents the resource redundancy.

[0128] When the cloud server determines that there are no candidate solutions with a collaboration score exceeding the dynamic collaboration threshold, it indicates that under the current production scenario (such as extreme order changes or multiple equipment failures), no solution can simultaneously meet the requirements of basic collaboration capabilities and efficient execution. At this point, the selection logic will switch to reliability-first. The cloud server first sorts the collaboration scores from high to low and selects the top N candidate solutions with relatively optimal collaboration capabilities. Then, it calculates the reliability score of each of these N solutions using a preset weighted formula. Finally, based on the criterion of maximizing production continuity, the candidate solution with the highest reliability score is determined as the final target solution.

[0129] This logic is adapted to the production needs under abnormal operating conditions in the automobile final assembly workshop. It abandons the priority of execution efficiency and instead ensures the risk resistance and execution stability of the solution, avoiding production line shutdowns and order delays due to solution failure, and achieving optimal reliability of manufacturing resource coordination.

[0130] S207. The cloud server sends the target solution to the edge device.

[0131] The target solution refers to the solution that has undergone multiple rounds of screening: either the most efficient solution with a collaboration score greater than the collaboration threshold and the shortest estimated execution time; or the most robust solution with the highest reliability score when no solution meets the target.

[0132] After completing all calculations and determining the optimal target solution, the cloud server sends the target solution to the field edge devices in the form of control commands through the workshop network. This provides the edge devices with a clear and unified basis for execution, preparing for the actual implementation of the collaborative solution in the final assembly workshop.

[0133] S208, Edge devices execute target scheme.

[0134] After receiving the target plan from the cloud server, the edge device no longer performs decision-making operations such as data calculation, plan evaluation, and optimal plan selection. Instead, it strictly follows the determined target plan and issues corresponding execution instructions to various execution units in the automobile assembly workshop, such as production equipment, automated guided vehicles (AGVs), material distribution systems, and production control systems. This officially starts the manufacturing resource collaboration task and drives the collaborative operation of all execution units throughout the process to ensure that the target plan is implemented and the established production collaboration task is successfully completed.

[0135] Based on the above description, this application has the following beneficial effects: In this application, by collecting data from all dimensions of production planning, material coordination, production line equipment, supply chain coordination, and historical execution, and combining the coordination efficiency index, coupling degree, and coordination development index to construct a quantitative coordination score, resources such as production equipment, material distribution, and personnel scheduling that were originally belonging to different management modules are incorporated into a unified coordination evaluation system. This promotes the transformation of each link from independent execution to collaborative operation, achieves the global optimal matching of manufacturing resources, and significantly improves the overall resource coordination efficiency of the final assembly workshop.

[0136] Furthermore, by setting collaboration thresholds for a two-tiered selection logic, when the collaboration score meets the standard, the shortest estimated execution time is used as the criterion to ensure production cycle time and execution efficiency; when no solution meets the standard, the reliability score is the core, combined with multi-dimensional indicators such as historical similarity and robustness to select the optimal solution, taking into account both production continuity and risk resistance, thus getting rid of the subjectivity and experience of traditional manual scheduling, and making the formulation of resource collaboration solutions more in line with the actual production scenarios of mixed production lines of multiple vehicle models and dynamic changes in orders.

[0137] Furthermore, the collaboration threshold is not a fixed value, but rather a combination of a baseline static threshold and multi-dimensional adjustment coefficients, including production condition complexity, system operating status, business priority, and environmental disturbance risk. Closed-loop optimization is achieved through dimensional coupling correction terms, superimposed multi-factor influences, and historical feedback iteration terms. Simultaneously, preset boundary constraints allow the threshold to be dynamically adjusted in real time according to changes in the production scenario. This ensures the admission standards for collaboration solutions while avoiding situations where no solutions are available under extreme conditions, making resource collaboration control more flexible and practical.

[0138] Finally, through comprehensive data collection, standardized processing, quantitative evaluation, and scientific selection, the traditional static production organization model of timed and quantitative delivery, fixed-rhythm progress, and post-event statistics has been replaced. This has enabled dynamic resource allocation throughout the entire process, from production planning to material distribution, production line operation, and supply chain linkage. It can quickly respond to production needs such as order changes and changes in working conditions, effectively reduce problems such as idle resources, material backlog, and poor process connection, reduce production and operating costs, and improve the intelligent production level and market responsiveness of the automobile assembly workshop.

[0139] The above text combined Figure 1 The edge-cloud collaborative control method in the automotive assembly workshop scenario provided in this application embodiment has been described in detail. The system and equipment provided in this application embodiment will be described below with reference to the accompanying drawings.

[0140] like Figure 2 As shown in the figure, this is a schematic diagram of an edge-cloud collaborative control system in an automobile assembly workshop scenario provided by an embodiment of this application. The system includes edge devices and a cloud server: The edge device is used to acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning dimension data, material collaboration dimension data, production line equipment dimension data, supply chain collaboration dimension data, and historical execution dimension data. The edge device is used to send multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server; The cloud server is used to calculate the collaborative efficiency index of each candidate solution in each dimension based on the multi-dimensional data corresponding to the candidate solutions; calculate the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative efficiency index of each candidate solution in each dimension; calculate the collaborative score of each candidate solution based on the multi-dimensional coupling degree, collaborative development index and historical execution correction coefficient of each candidate solution; determine whether there is a candidate solution with a collaborative score greater than the collaborative threshold; if there is a candidate solution with a collaborative score greater than the collaborative threshold, determine the estimated execution time corresponding to the candidate solution with a collaborative score greater than the collaborative threshold, and determine the candidate solution with the shortest estimated execution time as the target solution. The cloud server is used to send the target solution to the edge device; The edge device is used to execute the target scheme.

[0141] Optionally, the cloud server is further configured to determine the reliability scores of the top N candidate solutions in terms of collaboration score if there are no candidate solutions with a collaboration score greater than the collaboration threshold, and to determine the candidate solution with the highest reliability score as the target solution.

[0142] Optionally, the cloud server is specifically used to obtain the total duration of the serial process, the total duration of the parallel process, and the buffer duration corresponding to the candidate schemes whose collaborative scores are greater than the collaborative threshold. The estimated execution time is determined by the sum of the total duration of the serial process, the total duration of the parallel process, and the buffer duration.

[0143] Optionally, the cloud server is specifically used to obtain the historical similarity, robustness, fault tolerance and resource redundancy of the top N candidate solutions in the collaborative score ranking. A reliability score is determined based on historical similarity, scheme robustness, fault tolerance, and resource redundancy.

[0144] Optionally, the cloud server is specifically used to obtain the baseline static threshold, production condition complexity adjustment coefficient, system operation status adjustment coefficient, business priority adjustment coefficient, environmental disturbance risk adjustment coefficient, inter-dimensional coupling correction term, and historical feedback iteration term. The collaboration threshold is determined based on the baseline static threshold, the production condition complexity adjustment coefficient, the system operation status adjustment coefficient, the business priority adjustment coefficient, the environmental disturbance risk adjustment coefficient, the inter-dimensional coupling correction term, the historical feedback iteration term, and the preset boundary constraints.

[0145] Optionally, the edge device is specifically used to acquire multi-dimensional initial data corresponding to multiple collaborative processing candidate solutions in the automotive assembly workshop scenario; and to standardize the multi-dimensional initial data to obtain multi-dimensional data.

[0146] The edge-cloud collaborative control system in the automotive assembly workshop scenario according to the embodiments of this application can correspondingly execute the method described in the embodiments of this application, and the other operations and / or functions of each module / unit of the edge-cloud collaborative control system in the automotive assembly workshop scenario are respectively for implementing Figure 1 For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0147] This application also provides a computing device. For example... Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0148] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0149] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0150] The communication interface 703 is used for external communication.

[0151] Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0152] The memory 704 stores executable code, and the processor 702 executes the executable code to perform the aforementioned edge-cloud collaborative control method in the automobile assembly workshop scenario.

[0153] Specifically, in achieving Figure 2 In the case of the illustrated embodiment, and Figure 2 In the embodiment, when the modules or units of the edge-cloud collaborative control system in the automotive assembly workshop scenario are implemented through software, the execution... Figure 2 The software or program code required for the functions of each module / unit can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704, and executes the aforementioned edge-cloud collaborative control method in the automotive assembly workshop scenario.

[0154] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the edge-cloud collaborative control method described above in the automotive assembly workshop scenario.

[0155] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0156] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0157] When the computer program product is executed by a computer, the computer executes any one of the methods of the aforementioned edge-cloud collaborative control method in the automobile assembly workshop scenario. The computer program product can be a software installation package; when any one of the aforementioned edge-cloud collaborative control methods in the automobile assembly workshop scenario needs to be used, the computer program product can be downloaded and executed on the computer.

[0158] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0159] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for edge-cloud collaborative control in an automobile assembly workshop setting, characterized in that, The method includes: Edge devices acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning data, material coordination data, production line equipment data, supply chain coordination data, and historical execution data. The edge device sends multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server. The cloud server calculates the collaborative efficiency index of each candidate solution in each dimension based on the multi-dimensional data corresponding to the candidate solutions; it then calculates the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative efficiency index of each candidate solution in each dimension; based on the multi-dimensional coupling degree, collaborative development index, and historical execution correction coefficient of each candidate solution, it calculates the collaborative score of each candidate solution; it then determines whether there are candidate solutions with a collaborative score greater than the collaborative threshold; if there are candidate solutions with a collaborative score greater than the collaborative threshold, it determines the estimated execution time corresponding to the candidate solutions with a collaborative score greater than the collaborative threshold, and selects the candidate solution with the shortest estimated execution time as the target solution. The cloud server sends the target solution to the edge device; The edge device executes the target scheme.

2. The method according to claim 1, characterized in that, The method further includes: If there are no candidate solutions with a collaboration score greater than the collaboration threshold, the cloud server determines the reliability scores of the top N candidate solutions in the collaboration score ranking, and selects the candidate solution with the highest reliability score as the target solution.

3. The method according to claim 1, characterized in that, The determination of the estimated execution time corresponding to candidate solutions with a collaboration score greater than the collaboration threshold includes: Obtain the total duration of the serial process, the total duration of the parallel process, and the buffer duration for candidate schemes whose collaboration score is greater than the collaboration threshold; The estimated execution time is determined by the sum of the total duration of the serial process, the total duration of the parallel process, and the buffer duration.

4. The method according to claim 2, characterized in that, The reliability score for determining the top N candidate schemes in the collaborative ranking includes: Obtain the historical similarity, robustness, fault tolerance, and resource redundancy of the top N candidate solutions in the collaborative score ranking. A reliability score is determined based on historical similarity, scheme robustness, fault tolerance, and resource redundancy.

5. The method according to any one of claims 1-4, characterized in that, The collaboration threshold is determined in the following way: Obtain the baseline static threshold, production condition complexity adjustment coefficient, system operation status adjustment coefficient, business priority adjustment coefficient, environmental disturbance risk adjustment coefficient, inter-dimensional coupling correction term, and historical feedback iteration term; The collaboration threshold is determined based on the baseline static threshold, the production condition complexity adjustment coefficient, the system operation status adjustment coefficient, the business priority adjustment coefficient, the environmental disturbance risk adjustment coefficient, the inter-dimensional coupling correction term, the historical feedback iteration term, and the preset boundary constraints.

6. The method according to claim 1, characterized in that, The edge device acquires multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive assembly workshop scenario, including: Edge devices acquire multi-dimensional initial data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario; The edge device performs standardization processing on the multi-dimensional initial data to obtain multi-dimensional data.

7. An edge-cloud collaborative control system for an automobile assembly workshop scenario, characterized in that, The system includes edge devices and cloud servers: The edge device is used to acquire multi-dimensional data corresponding to multiple collaborative processing candidate solutions in the automotive final assembly workshop scenario. The multi-dimensional data includes production planning dimension data, material collaboration dimension data, production line equipment dimension data, supply chain collaboration dimension data, and historical execution dimension data. The edge device is used to send multi-dimensional data corresponding to multiple collaborative processing candidate solutions to the cloud server; The cloud server is used to calculate the collaborative efficiency index of the candidate solutions in each dimension based on the multi-dimensional data corresponding to the candidate solutions; calculate the multi-dimensional coupling degree and collaborative development index of each candidate solution based on the collaborative efficiency index of each candidate solution in each dimension; and calculate the collaborative score of each candidate solution based on the multi-dimensional coupling degree, collaborative development index and historical execution correction coefficient of each candidate solution. Determine if there are candidate solutions with a collaboration score greater than the collaboration threshold; if there are candidate solutions with a collaboration score greater than the collaboration threshold, determine the estimated execution time of the candidate solutions with a collaboration score greater than the collaboration threshold, and determine the candidate solution with the shortest estimated execution time as the target solution; The cloud server is used to send the target solution to the edge device; The edge device is used to execute the target scheme.

8. The system according to claim 7, characterized in that, The cloud server is also used to determine the reliability scores of the top N candidate solutions in terms of collaboration score if there are no candidate solutions with a collaboration score greater than the collaboration threshold, and to determine the candidate solution with the highest reliability score as the target solution.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 6.

Images