Data collaboration industrial product design method and system

Abstract

The invention provides a data collaboration industrial product design method and system, and relates to the technical field of industrial product design. The method comprises the following steps: firstly, determining a user demand, then collecting and preprocessing historical design data, calculating a design defect probability and various scores by using an expert scoring method, covering design defects, market feedback, user experience, cost control and production feasibility, and calculating the scores of the various data. And constructing a defect prediction model based on a random forest algorithm to realize effective prediction of design defects. In the real-time design data monitoring process, the defect probability is evaluated firstly, and then the comprehensive score is calculated for evaluation, so that adjustment of the design data is completed. According to the method, through multi-dimensional data analysis and a real-time feedback mechanism, the scientificity and flexibility of the design process are improved, and powerful support is provided for optimization and innovation of product design.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial product design, and in particular to a data collaborative industrial product design method and system. Background Art

[0002] Industrial product design is a comprehensive process involving multiple disciplines, aiming to transform creativity and functionality into manufacturable and marketable products. With the rapid development of science and technology, consumer needs are becoming increasingly diversified and personalized, which makes product design face higher challenges and opportunities. Modern industrial product design not only requires products to meet user needs in terms of function and aesthetics, but also considers sustainability, production costs, material selection and production processes.

[0003] In addition, the intensification of global competition has prompted companies to conduct careful market research and analysis during the design stage to ensure that products can adapt to the needs of different markets. Therefore, designers need to work closely with engineers, marketers and users to achieve multidisciplinary collaborative design. In addition, the application of digital technologies, such as computer-aided design, virtual reality and augmented reality, has greatly enriched design tools and methods, making the design process more efficient, intuitive and predictable.

[0004] In the current field of data-based collaborative industrial product design, design teams often face problems of information silos and poor collaboration. Different teams use different design tools and data formats, which makes it difficult to share design data and lags in information transmission. In addition, feedback and iteration in the design process often rely on traditional communication methods, which lack real-time performance, resulting in low design efficiency and uncontrollable product quality. Existing design methods have obvious deficiencies in data integration, interdisciplinary collaboration, and user feedback processing, and cannot effectively meet the requirements of complex industrial product design.

[0005] Deficiencies of existing technologies:

[0006] Existing technologies in industrial product design generally rely on experience and intuitive judgment, lacking systematic data analysis, resulting in defect identification and evaluation in the design process often lagging behind actual needs. This traditional method is difficult to provide real-time feedback on design defects, and has limited predictive power for potential problems, often leading to design rework, increased costs, and poor market adaptability. In addition, existing technologies are also deficient in integrating user feedback and market data, and are unable to fully reflect the performance of products in actual use, thus affecting the scientificity and rationality of design decisions.

[0007] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0008] The purpose of the present invention is to provide a data collaborative industrial product design method and system to solve the problems raised in the above background technology.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A data collaborative industrial product design method, the specific steps comprising:

[0011] Step 1: Determine user requirements, including the main design tools, file formats, data entities, data relationships, and storage formats used, and design a data model to support multiple design file formats;

[0012] Step 2: Collect historical design data, perform preprocessing operations on the historical design data, use expert scoring method to assign design defect probabilities according to design defects in the historical design data, and calculate the scores of each item in the historical design data;

[0013] Step 3: Build a defect prediction model based on the random forest model in machine learning, use historical design data as input, and the corresponding design defect probability as a label to train the defect prediction model;

[0014] Step 4: Input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt will be issued and the design team will make corresponding modifications.

[0015] Step 5: Calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again.

[0016] Furthermore, the specific logic for designing the data model is:

[0017] Select a cloud service provider, use the functions provided by the cloud, build a microservice architecture, create a VPC to ensure data security, and deploy a Kubernetes cluster for container management;

[0018] Select the CAD model to build a data model to support multiple design file formats and store them in JSON format; building the data model includes the following steps: identifying the format of the file, which includes CAD files, 3D models, documents, and drawings. CAD files include DWG and DXF, 3D models include STL, OBJ, and FBX, documents include PDF and Word, and drawings include PNG and JPEG; defining data entities, identifying key data entities, including design projects, files, users, and versions. Design projects are basic information about projects, including ID, name, description, and creation time. Files are information that stores different design files, including file ID, file name, file type, uploader, and upload time. Users are information that records user information, including user ID, user name, and role. Versions are information that manages file version information, including version number, modification record, and modifier; establishing data relationships, determining the relationships between different entities, including one-to-many relationships and many-to-many relationships. One-to-many relationships mean that a design project has multiple files, and many-to-many relationships mean that a design project has multiple files. The relationship is that multiple users participate in multiple design projects; draw data model diagrams, use ER diagrams and UML class diagram tools to visualize data entities and their relationships; select storage solutions, select the corresponding database type according to the complexity and requirements of the model, the database types include MySQL, PostgreSQL, NoSQL, MongoDB, and create the database structure, create corresponding database tables and fields according to the data model definition, so that the data is effectively stored in the database; database implementation and testing, implement the data model in the selected database, convert entities and relationships into actual database tables and constraints, so that the data model supports the required operations, including insert, update, delete and query, and perform performance testing to verify its efficiency; document writing and maintenance, write documents, that is, record the structure and design decisions of the data model, regular maintenance, that is, define evaluation and update the data model according to user feedback and changes in system requirements; use WebSocket technology for real-time data transmission, and the formula for obtaining the current data is:

[0019] D(t)=D(t-1)+△D

[0020] Among them, D(t) is the data obtained at the current moment, D(t-1) is the data obtained at the previous moment before the current moment, △D is the change in data updated at the current moment, and t is the time variable representing the current moment.

[0021] Furthermore, the specific logic for calculating the scores of each data in the historical design data is as follows:

[0022] The historical design data includes design defects, market feedback, user experience, cost control and production feasibility. The expert scoring method is used to assign the corresponding design defect probability to each set of historical design data based on the identified design defects. The preprocessing operations include removing duplicate data and standardizing the data format to ensure the consistency of each field. The processed historical design data are integrated into a sample set. Market feedback includes the average score of user surveys and the proportion of positive feedback on social media. User experience includes the number of users who have fully experienced the design solution and the total number of users who participated in the test. Cost control includes actual expenditures and budgeted expenditures. Production feasibility includes the number of products successfully tested in production and the total number of products tested.

[0023] The formula for calculating the scores for market feedback, user experience, cost control and production feasibility is:

[0024] The formula used to obtain the market feedback score is:

[0025]

[0026] Among them, M is the market feedback score, F is the average score of the user survey, and P is the proportion of positive feedback on social media;

[0027] The formula used to obtain the user experience score is:

[0028]

[0029] Among them, U is the user experience score, C N is the number of users who fully experience the design solution, and N is the total number of users who participated in the test;

[0030] The formula used to obtain the cost control score is:

[0031]

[0032] Among them, C M Score cost control, A for actual expenditure and B for budgeted expenditure;

[0033] The formula for obtaining production feasibility is:

[0034]

[0035] Among them, T s is production feasibility, S is the number of products successfully tested in production, and T is the total number of products tested.

[0036] Furthermore, the specific logic for building the defect prediction model is as follows:

[0037] The sample set is divided into a training set and a test set with a ratio of 7 to 3. The historical design data in the training set is used as input, and the corresponding design defect probability is used as the label. The defect prediction model is trained based on the random forest model in the machine learning algorithm. The data in the test set is then input into the trained defect prediction model for testing. The mean square error is selected as the loss function. When the mean square error is lower than 0.1, it indicates that the defect prediction model training is completed.

[0038] The specific logic for further real-time detection of the defect probability of design data is as follows:

[0039] The current design data is obtained and input into the trained defect prediction model. The output of the defect prediction model is compared with the preset design defect probability. If the design defect probability predicted by the model is lower than the preset threshold, the model will issue a safety prompt and the design team will continue according to the current plan. If the design defect probability predicted by the model is higher than the preset threshold, the model will issue a warning and the design team will check and modify the design data.

[0040] Furthermore, the specific logic for calculating the comprehensive score is as follows:

[0041] The formula used to calculate the overall score is:

[0042]

[0043] Among them, Q is the comprehensive score, P is the design defect probability predicted by the model, ω 1 ,ω 2 ,ω 3 ,ω 4 ,ω 5 are the weights of each score, 0<ω 4 =ω 5 <ω 3 <ω 2 =ω 1 <1, and ω 1 +ω 2 +ω 3 +ω 4 +ω 5 =1, a, b, c, d, e are the indexes of each score respectively;

[0044] The comprehensive score is compared with the preset score threshold. If the comprehensive score is higher than the preset score threshold, the current design data can be executed without further adjustment. If the comprehensive score is lower than the preset score threshold, a signal is sent to adjust the design data again.

[0045] The present invention further provides a data collaborative industrial product design system, which is used to implement the above-mentioned data collaborative industrial product design method, including:

[0046] User requirements analysis module, used to determine user requirements, including the main design tools, file formats, data entities, data relationships and storage formats used, and design data models to support multiple design file formats;

[0047] The historical data collection module is used to collect historical design data, perform preprocessing operations on the historical design data, assign design defect probabilities based on design defects in the historical design data using an expert scoring method, and calculate the scores of various data in the historical design data;

[0048] Defect prediction module building module, used to build a defect prediction model based on the random forest model in machine learning, with historical design data as input and the corresponding design defect probability as label to train the defect prediction model;

[0049] The design defect early warning module is used to input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt is issued and the design team makes corresponding modifications;

[0050] The comprehensive score evaluation module is used to calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again.

[0051] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0052] The efficiency and quality of industrial product design have been significantly improved through a data-driven comprehensive evaluation mechanism. Through systematic user demand analysis, accurate processing of historical data, and the construction of a defect prediction model, this method can identify potential problems in the design in real time, thereby providing timely feedback and modification suggestions to the design team. At the same time, the calculation method of the comprehensive score takes into account multiple key factors, such as market feedback and user experience, making design decisions more scientific and reasonable. This flexible evaluation system not only reduces the risks in the design process, but also effectively reduces development costs, improves the market competitiveness of products, promotes design innovation and continuous improvement, and provides strong support for enterprises in a rapidly changing market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0054] Figure 2 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0055] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0056] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0057] Example:

[0058] See also Figure 1 , the present invention provides a technical solution:

[0059] A data collaborative industrial product design method, the specific steps comprising:

[0060] Step 1: Determine user requirements, including the main design tools, file formats, data entities, data relationships, and storage formats used, and design a data model to support multiple design file formats;

[0061] In this embodiment, the specific logic based on which the data model is designed is:

[0062] Select a cloud service provider, use the functions provided by the cloud, build a microservice architecture, create a VPC to ensure data security, and deploy a Kubernetes cluster for container management;

[0063] Select the CAD model to build a data model to support multiple design file formats and store them in JSON format; building the data model includes the following steps: identifying the format of the file, which includes CAD files, 3D models, documents, and drawings. CAD files include DWG and DXF, 3D models include STL, OBJ, and FBX, documents include PDF and Word, and drawings include PNG and JPEG; defining data entities, identifying key data entities, including design projects, files, users, and versions. Design projects are basic information about projects, including ID, name, description, and creation time. Files are information that stores different design files, including file ID, file name, file type, uploader, and upload time. Users are information that records user information, including user ID, user name, and role. Versions are information that manages file version information, including version number, modification record, and modifier; establishing data relationships, determining the relationships between different entities, including one-to-many relationships and many-to-many relationships. One-to-many relationships mean that a design project has multiple files, and many-to-many relationships mean that a design project has multiple files. The relationship is that multiple users participate in multiple design projects; draw data model diagrams, use ER diagrams and UML class diagram tools to visualize data entities and their relationships; select storage solutions, select the corresponding database type according to the complexity and requirements of the model, the database types include MySQL, PostgreSQL, NoSQL, MongoDB, and create the database structure, create corresponding database tables and fields according to the data model definition, so that the data is effectively stored in the database; database implementation and testing, implement the data model in the selected database, convert entities and relationships into actual database tables and constraints, so that the data model supports the required operations, including insert, update, delete and query, and perform performance testing to verify its efficiency; document writing and maintenance, write documents, that is, record the structure and design decisions of the data model, regular maintenance, that is, define evaluation and update the data model according to user feedback and changes in system requirements; use WebSocket technology for real-time data transmission, and the formula for obtaining the current data is:

[0064] D(t)=D(t-1)+△D

[0065] Among them, D(t) is the data obtained at the current moment, D(t-1) is the data obtained at the previous moment before the current moment, △D is the change in data updated at the current moment, and t is the time variable representing the current moment.

[0066] First, by establishing a data model that supports multiple design file formats, good compatibility between different design tools and file formats can be achieved, which not only simplifies the process of data exchange and sharing, but also enables the design team to integrate and use various types of design data more efficiently, improving work efficiency. Secondly, the design of cloud services and microservice architecture ensures data security and scalability. The container management of the Kubernetes cluster enables the system to flexibly adjust resources according to demand to meet the needs of different projects while reducing maintenance costs. In addition, the use of WebSocket technology to achieve real-time data transmission ensures that the design team can obtain the latest data instantly, enhancing the real-time and effectiveness of collaboration. Furthermore, the construction process of the data model promotes communication and collaboration between teams through clear data entities and relationships. The clear definition of design projects, files, users, and versions not only lays the foundation for subsequent data management, but also provides team members with clear roles and responsibilities, which helps to reduce misunderstandings and improve overall design quality. Finally, regular document writing and maintenance enable the data model to be continuously optimized and updated according to user feedback and changes in system requirements, thereby maintaining the adaptability and flexibility of the system. This dynamic adjustment capability ensures that the system can always meet the actual needs of users and improve user satisfaction. In summary, the technical effects and advantages of implementing this step are not only reflected in improving the efficiency and accuracy of the design process, but also in enhancing team collaboration, data security and system flexibility, providing a solid foundation and guarantee for industrial product design.

[0067] Step 2: Collect historical design data, perform preprocessing operations on the historical design data, use expert scoring method to assign design defect probabilities according to design defects in the historical design data, and calculate the scores of each item in the historical design data;

[0068] In this embodiment, the specific logic for calculating the score of each data in the historical design data is as follows:

[0069] The historical design data includes design defects, market feedback, user experience, cost control and production feasibility. The expert scoring method is used to assign the corresponding design defect probability to each set of historical design data based on the identified design defects. The preprocessing operations include removing duplicate data and standardizing the data format to ensure the consistency of each field. The processed historical design data are integrated into a sample set. Market feedback includes the average score of user surveys and the proportion of positive feedback on social media. User experience includes the number of users who have fully experienced the design solution and the total number of users who participated in the test. Cost control includes actual expenditures and budgeted expenditures. Production feasibility includes the number of products successfully tested in production and the total number of products tested.

[0070] The formula for calculating the scores for market feedback, user experience, cost control and production feasibility is:

[0071] The formula used to obtain the market feedback score is:

[0072]

[0073] Among them, M is the market feedback score, F is the average score of the user survey, and P is the proportion of positive feedback on social media;

[0074] Design a questionnaire with multiple questions to collect user feedback on various aspects of the product. The questionnaire can cover multiple dimensions, such as product functions, user experience, appearance design, cost performance, etc. Each question is scored on a scale of 0 to 1, where 0 represents the worst and 1 represents the best. Social media here include Weibo, WeChat, Douyin, Facebook, YouTube, etc.

[0075] The formula used to obtain the user experience score is:

[0076]

[0077] Among them, U is the user experience score, C N is the number of users who fully experience the design solution, and N is the total number of users who participated in the test;

[0078] The formula used to obtain the cost control score is:

[0079]

[0080] Among them, C M Score cost control, A for actual expenditure and B for budgeted expenditure;

[0081] The formula for obtaining production feasibility is:

[0082]

[0083] Among them, T s is production feasibility, S is the number of products successfully tested in production, and T is the total number of products tested.

[0084] By collecting historical design data, the diversity and integrity of data samples are ensured, covering multiple aspects such as design defects, market feedback, user experience, cost control and production feasibility. This comprehensiveness ensures the accuracy of subsequent analysis and model training. Preprocessing operations to remove duplicate data and standardize data formats ensure the consistency and reliability of data. This is the basis of data analysis and modeling, which can reduce noise and errors and improve the predictive ability of the model. The expert scoring method is used to assign design defect probabilities to historical design data, introduce professional knowledge and industry experience, and improve the scientificity and rationality of the scoring. This method can effectively identify and quantify design defects and provide high-quality labels for subsequent defect prediction models. The formulas for each score are clear and can calculate effective scores based on specific data. This quantitative approach makes the evaluation process transparent and controllable, facilitating subsequent analysis and decision-making.

[0085] Step 3: Build a defect prediction model based on the random forest model in machine learning, use historical design data as input, and the corresponding design defect probability as a label to train the defect prediction model;

[0086] In this embodiment, the specific logic for constructing the defect prediction model is:

[0087] The sample set is divided into a training set and a test set with a ratio of 7 to 3. The historical design data in the training set is used as input, and the corresponding design defect probability is used as the label. The defect prediction model is trained based on the random forest model in the machine learning algorithm. The data in the test set is then input into the trained defect prediction model for testing. The mean square error is selected as the loss function. When the mean square error is lower than 0.1, it indicates that the defect prediction model training is completed.

[0088] The random forest model is an integrated learning method with strong resistance to overfitting and high prediction accuracy. By combining multiple decision trees, random forests can capture complex patterns in the data, thereby improving the accuracy of defect prediction. Using the mean square error (MSE) as a loss function helps quantify the deviation between the model's predicted value and the true value. By monitoring the changes in the mean square error, the design team can effectively identify the convergence of the model and confirm that the model training is complete when it reaches a preset threshold (such as 0.1). This setting can ensure the effectiveness and practicality of the model.

[0089] Step 4: Input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt will be issued and the design team will make corresponding modifications.

[0090] In this embodiment, the specific logic for real-time detection of the defect probability of the design data is as follows:

[0091] The current design data is obtained and input into the trained defect prediction model. The output of the defect prediction model is compared with the preset design defect probability. If the design defect probability predicted by the model is lower than the preset threshold, the model will issue a safety prompt and the design team will continue according to the current plan. If the design defect probability predicted by the model is higher than the preset threshold, the model will issue a warning and the design team will check and modify the design data.

[0092] Step 5: Calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again;

[0093] In this embodiment, the specific logic for calculating the comprehensive score is:

[0094] The formula used to calculate the overall score is:

[0095]

[0096] Among them, Q is the comprehensive score, P is the design defect probability predicted by the model, ω 1 ,ω 2 ,ω 3 ,ω 4 ,ω 5 are the weights of each score, 0<ω 4 =ω 5 <ω 3 <ω 2 =ω 1 <1, and ω 1 +ω 2 +ω 3 +ω 4 +ω 5 =1, a, b, c, d, e are the indices of each score, and Z is the normalization factor used to ensure that the overall score is within the range of 0-100;

[0097] The comprehensive score is compared with the preset score threshold. If the comprehensive score is higher than the preset score threshold, the current design data can be executed without further adjustment. If the comprehensive score is lower than the preset score threshold, a signal is sent to adjust the design data again.

[0098] Market feedback score: M, user experience score: U, cost control score: C M and production feasibility score T sare all positively correlated with the overall score, indicating that their increase will lead to an improvement in the overall score, while an increase in the probability of design defects will lead to a decrease in the overall score, and the two are inversely proportional. The weight of each score ω reflects their importance in the total score. By adjusting the weights, certain indicators can be emphasized according to the specific needs of the project. Design defects directly affect the quality and reliability of the product. A higher probability of design defects may cause the product to fail in the market, so it is reasonable to occupy a larger proportion in the overall score. Market feedback reflects the consumer's acceptance of the product and is usually directly related to the sales of the product. Good market feedback can increase the sales of the product, so its weight is equal to that of design defects. User experience directly affects customer satisfaction and loyalty. In a highly competitive market, good user experience can significantly improve the market performance of the product, so it is given a higher weight. Cost control is crucial to the profitability of the project, but its importance is lower than product quality and market acceptance. Although cost management must be considered, overly strict cost control may lead to quality degradation during the product design stage, so its weight is appropriately reduced. Production feasibility ensures that the design can be successfully implemented in actual production. Although important, its impact is usually reflected in the later stages, so a relatively low weight is reasonable. ω 1 +ω 2 +ω 3 +ω 4 +ω 5 =1 This constraint ensures that the composite score is a normalized indicator, ensuring that the total contribution of the weighted combination of the five factors is 100%.

[0099] Introducing indexes a, b, c, d, and e can make the model more sensitive to changes in each score. For example, the index distribution can be set as follows: a = 1.2, b = 1.2, c = 1.5, d = 1.0, and e = 1.0. Since design defects have a very large impact on product success, setting the index to 1.2 makes it have a higher nonlinear impact in the comprehensive score. A slight increase in its index can effectively highlight its risks. Market feedback is closely related to the market performance of the product, so choosing an index of 1.2 can enhance its role in the comprehensive score and ensure that changes in market feedback are taken seriously. User experience is increasingly regarded as a key factor in product success. Setting its index to 1.5 reflects its significant impact on the final score, making it more prominent in the overall evaluation. Cost control is usually a linear relationship, and setting the index to 1.0 means that its impact on the comprehensive score is direct and will not be amplified. Production feasibility is also usually a linear relationship, and setting it to 1.0 means that it has a direct impact on the comprehensive score and does not require additional enhancement. This formula not only takes into account the independence of each score, but also the relative relationship between each score, making the comprehensive score more representative and scientific. Finally, the comprehensive score is compared with the preset score threshold to achieve a clear decision-making standard. The threshold here can be set to 85 points. When the comprehensive score is lower than the threshold, a signal can be sent in time to prompt the design team to make adjustments and reduce the probability of later problems.

[0100] See also Figure 2 The present invention further provides a data collaborative industrial product design system, which is used to implement the above-mentioned data collaborative industrial product design method, including:

[0101] User requirements analysis module, used to determine user requirements, including the main design tools, file formats, data entities, data relationships and storage formats used, and design data models to support multiple design file formats;

[0102] The historical data collection module is used to collect historical design data, perform preprocessing operations on the historical design data, assign design defect probabilities based on design defects in the historical design data using an expert scoring method, and calculate the scores of various data in the historical design data;

[0103] Defect prediction module building module, used to build a defect prediction model based on the random forest model in machine learning, with historical design data as input and the corresponding design defect probability as label to train the defect prediction model;

[0104] The design defect early warning module is used to input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt is issued and the design team makes corresponding modifications;

[0105] The comprehensive score evaluation module is used to calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again.

[0106] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0107] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0108] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0109] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A data collaborative industrial product design method, characterized in that: The specific steps include: Step 1: Determine user requirements, including the main design tools, file formats, data entities, data relationships, and storage formats used, and design a data model to support multiple design file formats; Step 2: Collect historical design data, perform preprocessing operations on the historical design data, use expert scoring method to assign design defect probabilities according to design defects in the historical design data, and calculate the scores of each item in the historical design data; Step 3: Build a defect prediction model based on the random forest model in machine learning, use historical design data as input, and the corresponding design defect probability as a label to train the defect prediction model; Step 4: Input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt will be issued and the design team will make corresponding modifications. Step 5: Calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again.

2. A data collaborative industrial product design method according to claim 1, characterized in that: The specific logic for designing the data model is: Select a cloud service provider, use the functions provided by the cloud, build a microservice architecture, create a VPC to ensure data security, and deploy a Kubernetes cluster for container management; Select CAD models to build data models to support multiple design file formats and store them in JSON format; Building a data model includes the following steps: identifying the format of the file, which includes CAD files, 3D models, documents, and drawings. CAD files include DWG and DXF, 3D models include STL, OBJ, and FBX, documents include PDF and Word, and drawings include PNG and JPEG; defining data entities, identifying key data entities, including design projects, files, users, and versions. Design projects are basic information about projects, including ID, name, description, and creation time. Files are information that stores different design files, including file ID, file name, file type, uploader, and upload time. Users are records of user information, including user ID, user name, and role. Versions are version information that manages files, including version number, modification records, and modifiers; establishing data relationships, determining the relationships between different entities, including one-to-many relationships and many-to-many relationships. One-to-many relationships mean that a design project has multiple files, and many-to-many relationships mean that multiple users participate in multiple design projects; drawing data According to the model diagram, use ER diagram and UML class diagram tools to visualize data entities and their relationships; select storage solutions, select the corresponding database type according to the complexity and requirements of the model, including MySQL, PostgreSQL, NoSQL, MongoDB, and create the database structure, create corresponding database tables and fields according to the data model definition, so that data can be effectively stored in the database; database implementation and testing, implement the data model in the selected database, convert entities and relationships into actual database tables and constraints, so that the data model supports the required operations, including insert, update, delete and query, and perform performance testing to verify its efficiency; document writing and maintenance, write documents, that is, record the structure and design decisions of the data model, and perform regular maintenance, that is, define, evaluate and update the data model according to user feedback and changes in system requirements; use WebSocket technology for real-time data transmission, and the formula for obtaining the current data is: D(t)=D(t-1)+△D Among them, D(t) is the data obtained at the current moment, D(t-1) is the data obtained at the previous moment before the current moment, △D is the change in data updated at the current moment, and t is the time variable representing the current moment.

3. A data collaborative industrial product design method according to claim 2, characterized in that: The specific logic for calculating the scores of each data in the historical design data is as follows: The historical design data includes design defects, market feedback, user experience, cost control and production feasibility. The expert scoring method is used to assign the corresponding design defect probability to each set of historical design data based on the identified design defects. The preprocessing operations include removing duplicate data and standardizing the data format to ensure the consistency of each field. The processed historical design data are integrated into a sample set. Market feedback includes the average score of user surveys and the proportion of positive feedback on social media. User experience includes the number of users who have fully experienced the design solution and the total number of users who participated in the test. Cost control includes actual expenditures and budgeted expenditures. Production feasibility includes the number of products successfully tested in production and the total number of products tested. The formula for calculating the scores for market feedback, user experience, cost control and production feasibility is: The formula used to obtain the market feedback score is: Among them, M is the market feedback score, F is the average score of the user survey, and P is the proportion of positive feedback on social media; The formula used to obtain the user experience score is: Among them, U is the user experience score, C N is the number of users who fully experience the design solution, and N is the total number of users who participated in the test; The formula used to obtain the cost control score is: Among them, C M Score cost control, where A is actual expenditure and B is budgeted expenditure; The formula for obtaining production feasibility is: Among them, T s is the production feasibility, S is the number of products successfully tested in production, and f is the total number of products tested.

4. A data collaborative industrial product design method according to claim 3, characterized in that: The specific logic for building the defect prediction model is: The sample set is divided into a training set and a test set with a ratio of 7 to 3. The historical design data in the training set is used as input, and the corresponding design defect probability is used as the label. The defect prediction model is trained based on the random forest model in the machine learning algorithm. The data in the test set is then input into the trained defect prediction model for testing. The mean square error is selected as the loss function. When the mean square error is lower than 0.1, it indicates that the defect prediction model training is completed.

5. A data collaborative industrial product design method according to claim 4, characterized in that: The specific logic for real-time detection of the defect probability of design data is as follows: The current design data is obtained and input into the trained defect prediction model. The output of the defect prediction model is compared with the preset design defect probability. If the design defect probability predicted by the model is lower than the preset threshold, the model will issue a safety prompt and the design team will continue according to the current plan. If the design defect probability predicted by the model is higher than the preset threshold, the model will issue a warning and the design team will check and modify the design data.

6. A data collaborative industrial product design method according to claim 5, characterized in that: The specific logic for calculating the comprehensive score is as follows: The formula used to calculate the overall score is: Among them, Q is the comprehensive score, P is the design defect probability predicted by the model, ω1, ω2, ω3, ω4, and ω5 are the weights of each score, 0<ω4=ω5<ω3<ω2=ω1<1, and ω1+ω2+ω3+ω4+ω5=1, a, b, c, d, and e are the indices of each score; The comprehensive score is compared with the preset score threshold. If the comprehensive score is higher than the preset score threshold, the current design data can be executed without further adjustment. If the comprehensive score is lower than the preset score threshold, a signal is sent to adjust the design data again.

7. A data collaborative industrial product design system, characterized in that: The data collaborative industrial product design system is used to implement the data collaborative industrial product design method according to any one of claims 1 to 6, comprising: User requirements analysis module, used to determine user requirements, including the main design tools, file formats, data entities, data relationships and storage formats used, and design data models to support multiple design file formats; The historical data collection module is used to collect historical design data, perform preprocessing operations on the historical design data, assign design defect probabilities based on design defects in the historical design data using an expert scoring method, and calculate the scores of various data in the historical design data; Defect prediction module building module, used to build a defect prediction model based on the random forest model in machine learning, with historical design data as input and the corresponding design defect probability as label to train the defect prediction model; The design defect early warning module is used to input the current design data into the trained defect prediction model. If the design defect probability predicted by the model is higher than the preset threshold, a corresponding prompt is issued and the design team makes corresponding modifications; The comprehensive score evaluation module is used to calculate the comprehensive score of the current design data in combination with the defect probability predicted by the model to evaluate whether the current design data needs to be adjusted again.

Images