Customized furniture intelligent auditing method

Abstract

The invention discloses a customized furniture intelligent auditing method, which comprises the following steps: constructing three types of auditing rule systems of design rationality, process rationality and enterprise product standards, and converting the auditing rule systems into drl scripts executable by a Drools rule engine; building a Kolkhome entertainment platform distributed cluster auditing framework; rule configuration and static and dynamic inspection are completed in a merchant background; the design front end realizes multi-link intelligent auditing and automatic correction of design, order placing and reviewing, order reviewing and quotation; and performing grading processing according to the warning level, and outputting the json format order splitting file. Application results show that the method greatly reduces the labor ratio, greatly improves the order qualification rate, improves the auditing timeliness by multiple times, and provides reliable technical support for digital upgrading of customized furniture enterprises.

Description

A method for intelligent auditing of customized furniture Technical Field

[0001] This invention relates to the field of customized furniture technology, specifically to a method for intelligent verification of customized furniture. Background Technology

[0002] The furniture industry has gradually shifted from a model dominated by traditional finished furniture to a development direction of customized furniture, large-scale customized furniture, and whole-house (whole-house) customized furniture.

[0003] Currently, the integration of design and manufacturing in the customized furniture industry mainly relies on two types of software: one is design software that meets the needs of front-end design visualization and parametric analysis, such as AutoCAD and SolidWorks, used to complete the design and drawing of whole-house customization solutions; the other is order breakdown software that meets the needs of design-to-production conversion, such as ImosWCC, Zaoyi, and DSC, used to parse design models into material lists, processing drawings, and hardware quantities required for production. Although these two types of software achieve a basic connection between design and manufacturing, the entire process relies on manual operation, making it difficult to fully guarantee the standardization and accuracy of design orders. Therefore, design review becomes a key link in improving the product qualification rate under the integrated model, directly affecting the efficiency and quality of order production and delivery.

[0004] However, the existing custom furniture design review process has many problems that urgently need to be addressed, hindering the industry's digital upgrade and efficient development: 1) Currently, the design review of most companies still relies on manual processes. The review results are greatly affected by the reviewers' work experience and professional level, and error feedback is not timely, making it difficult to achieve a balance between review efficiency and quality. Furthermore, the frequent communication and coordination required during manual review further increases the risk of human error. 2) Due to differences in product types, manufacturing levels, order volumes, and target consumer groups among companies, the custom furniture industry has not yet formed a unified design review standard. Most companies' review rules are summarized by reviewers based on long-term order processing experience and company process requirements, lacking a unified format, category division, and clear scope of application. They are highly experience-dependent, with low usability and adaptability in practical application, making it difficult to meet the review requirements of diverse customization needs. 3) The existing design review mainly focuses on two independent manual stages: drawing review and order breakdown, with overlapping processes. Problems found in the drawing review stage need to be fed back to the designer for modification before proceeding to the order breakdown stage; however, if problems found in the order breakdown stage cannot be directly corrected by the reviewers, they need to be fed back to the designer for modification again, after which the review process must restart from the drawing review stage. This repetitive process of separating and splitting drawings and orders leads to a significant increase in the time spent on order review, which directly affects subsequent production scheduling and production cycles, making it difficult to adapt to the development trend of customized furniture companies with increasing order volume, shorter delivery cycles, and increased product complexity.

[0005] From the current application status of intelligent review technology, research and application of intelligent design review can be traced back to the 1960s. In the 21st century, with the development of Building Information Modeling (BIM) technology, BIM-based automatic review methods have been widely researched and applied in industries with high safety requirements, such as architecture and fire protection engineering. Existing research on intelligent review technology mainly focuses on the computability of design drawings and information, and the computability of design rules. Some studies use inference engines such as RuleML, Jess, and Drools to model review rules. However, these studies have significant industry limitations, mainly concentrated in the field of architectural engineering. Research on intelligent review technology for design and manufacturing industries such as customized furniture is still relatively scarce, and mature solutions have not yet been formed.

[0006] In addition, although some existing order splitting software (such as WCC) attempts to embed auditing plugins to achieve automatic auditing, the auditing rules need to be written manually and generally only contain a small number of core rules, resulting in extremely low rule coverage. Most of the auditing content still needs to be completed by the order splitting personnel by comparing models, drawings and material lists, which fails to fundamentally solve the problems of low auditing efficiency and high error rate.

[0007] In conclusion, under the development background of integrated design and manufacturing of customized furniture, the existing design review model can no longer meet the industry's development needs for high efficiency, high quality and low cost. There is an urgent need to develop an intelligent review method that is adapted to the characteristics of the customized furniture industry, relies on a mature digital platform and integrates an efficient rule engine, so as to solve the core pain points such as high dependence on manual labor, inconsistent rules and redundant processes, and promote the digital transformation and intelligent upgrading of customized furniture enterprises. Summary of the Invention

[0008] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0009] Therefore, the purpose of this invention is to provide a method for intelligent auditing of customized furniture to solve the problems mentioned in the background art.

[0010] To solve the above-mentioned technical problems, according to one aspect of the present invention, the present invention provides the following technical solution: a method for intelligent review of customized furniture, comprising the following steps: S1, constructing an intelligent review rule system for customized furniture, and converting the review rules into rule scripts that can be recognized and executed by a rule engine, wherein the review rule system includes design rationality rules, process rationality rules, and enterprise product standard rules; S2, building an intelligent review architecture based on the Kujiale platform, wherein the review architecture embeds a rule engine, performs distributed cluster layered processing on customized furniture orders, and matches corresponding detection clusters according to the order data volume; S3, completing the configuration and verification of review rules in the merchant backend of the Kujiale platform, wherein the configuration includes setting the rule effective scope, warning level, and application stage, and the verification includes static syntax detection and dynamic performance detection; S4, implementing multi-stage intelligent review in the Kujiale design frontend, including real-time detection and automatic correction during the design process, rule detection during order submission, problem investigation in the order review stage, and final verification in the quotation stage; S5, performing graded processing according to the warning level of the review results, blocking subsequent business at the error level, only prompting and archiving at the warning level, and outputting data files that can be connected to order splitting software after the review is passed.

[0011] As a preferred embodiment of the intelligent auditing method for customized furniture described in this invention, in step S1, the design rationality rules include material matching detection rules and design error prevention detection rules. The material matching detection rules are used to verify the mixing of materials, the consistency of the base material of cabinet components, and the uniformity of hardware brands. The design error prevention detection rules are used to verify the accuracy of cabinet structure, component parameters, and cabinet placement angle and position.

[0012] As a preferred embodiment of the intelligent auditing method for customized furniture described in this invention, in step S1, the process rationality rules include hardware hole position detection rules, component conflict detection rules, component adaptation detection rules, anti-deformation detection rules, and component process parameter detection rules, which are used to ensure that the customized furniture design meets the technical standards of the production equipment.

[0013] As a preferred embodiment of the intelligent auditing method for customized furniture described in this invention, in step S1, the enterprise product standard rules include product model library matching rules, material and color validity rules, product off-shelf status detection rules, and order remarks information verification rules, which are used to verify the consistency between the ordered products and the enterprise's product positioning and product catalog.

[0014] As a preferred embodiment of the intelligent review method for customized furniture described in this invention, in step S1, the rule engine is the Drools rule engine, the rule script is written in drl format, and the script structure includes a rule body identified by the keyword "rule", a condition judgment part guided by the keyword "when", and an execution logic part guided by the keyword "then". The execution logic includes model highlighting, text prompt output, and automatic correction operation.

[0015] As a preferred embodiment of the intelligent auditing method for customized furniture described in this invention, in step S2, the distributed cluster layered processing specifically involves: allocating internal model detection to the real-time detection cluster, allocating routine order scheme detection to the routine rule detection cluster, and allocating schemes with data exceeding 100MB to the Huge cluster.

[0016] As a preferred embodiment of the intelligent review method for customized furniture described in this invention, in step S3, the scope of the rule includes the user department and the review scope, the application process includes order submission and review, order review and drawing and quotation output, the static syntax detection is used to check for script writing errors, and the dynamic performance detection verifies the rule execution efficiency, conflict situation and the validity of the review results by capturing real order schemes.

[0017] As a preferred embodiment of the intelligent review method for customized furniture described in this invention, in step S4, the real-time detection and automatic correction specifically refers to: when the designer performs model dragging or modification operations, the system triggers rule detection in real time, automatically performs parameter correction for problems that meet the correction conditions, and provides real-time prompts for problems that cannot be automatically corrected.

[0018] As a preferred embodiment of the intelligent review method for customized furniture described in this invention, in step S4, the rule detection during order submission specifically involves: after the designer submits the order, the system performs a full rule scan on the design scheme, highlights models with errors, and categorizes and displays the problem types and warning levels in the review list.

[0019] As a preferred embodiment of the intelligent review method for customized furniture described in this invention, in step S5, the data file that can be connected to the order splitting software is a JSON format file, and the hierarchical processing also includes a manual repair step. For problems that cannot be solved by automatic correction, relevant personnel manually modify them according to the prompts and then re-trigger the review.

[0020] Compared with existing technologies, the beneficial effects of this invention are as follows: 1. By deeply integrating the Drools rule engine with the Kujiale platform, this invention transforms 253 custom furniture review rules into automatically executable rule scripts, realizing the intelligent and automated design review. Compared with the traditional 100% manual review mode, it can significantly reduce the proportion of manual operations in orders and significantly reduce the intensity of manual operations. At the same time, through a unified rule system and automated detection logic, it avoids human errors caused by differences in experience and high communication frequency in manual review, greatly improving the order qualification rate and ensuring the stability and consistency of review results.

[0021] 2. This invention is the first to summarize the customized furniture review rule system into three categories: design rationality, process rationality, and enterprise product standards. This forms a clear and comprehensive rule system, solving the problems of inconsistent industry review standards, messy rule formats, and low usability. At the same time, relying on the dynamic configuration advantages of the Drools rule engine, enterprises can flexibly adjust, add, or delete review rules according to their own product type, manufacturing level, and order volume without recompiling the code. It is adapted to various application scenarios such as whole-house customization and kitchen and bathroom customization, meeting the personalized review needs of different enterprises and significantly improving the scalability and industry adaptability of the rule system.

[0022] 3. The multi-stage review process is integrated into the Kujiale platform, achieving a closed-loop review process encompassing real-time detection during the design process, full rule scanning during order submission, precise screening during order review, and final verification during quotation. This eliminates process overlap and redundant reviews. Furthermore, leveraging Kujiale's self-developed MicroTask framework and distributed cluster layered processing mechanism, the invention allocates internal model detection, routine scheme detection, and large file scheme detection to real-time detection clusters, routine rule detection clusters, and Huge clusters respectively. This efficiently solves the problems of large 3D model data volume and high resource consumption, improving resource utilization efficiency in the review process. Simultaneously, automated review reduces the time and communication costs of manual review, lowering production rework costs due to review errors; its standardized intelligent review process and framework... Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 is a definition object scope diagram provided by an embodiment of the present invention; Figure 2 is a flowchart of the Kujiale platform review architecture provided by an embodiment of the present invention; Figure 3 is a flowchart of the intelligent review process provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the rule configuration interface provided by an embodiment of the present invention; Figure 5 is a schematic diagram of the detection settings of different stages provided by an embodiment of the present invention; Figure 6 is a schematic diagram of the rule creation interface provided by an embodiment of the present invention; Figure 7 is a schematic diagram of the rule detection process provided by an embodiment of the present invention; Figure 8 is a schematic diagram of the design process review provided by an embodiment of the present invention; Figure 9 is a schematic diagram of the order review provided by an embodiment of the present invention; Figure 10 is a schematic diagram of the order review provided by an embodiment of the present invention; Figure 11 is a schematic diagram of the quotation review provided by an embodiment of the present invention. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0027] This invention provides a method for intelligent auditing of customized furniture, comprising the following steps: S1. Constructing an intelligent auditing rule system for customized furniture, and converting the auditing rules into rule scripts that can be recognized and executed by a rule engine. The auditing rule system includes design rationality rules, process rationality rules, and enterprise product standard rules. The design rationality rules include material matching detection rules and design error prevention detection rules. The material matching detection rules are used to verify the mixing of materials, the consistency of cabinet component base materials, and the uniformity of hardware brands. The design error prevention detection rules are used to verify the accuracy of cabinet structure, component parameters, and cabinet placement angles and positions. The process rationality rules include hardware hole position detection rules and component conflict detection rules. The system includes component compatibility testing rules, anti-deformation testing rules, and component process parameter testing rules to ensure that customized furniture designs meet the technical standards of production equipment. The enterprise product standard rules include product model library matching rules, material and color validity rules, product off-shelf status detection rules, and order remarks information verification rules to verify the consistency between ordered products and the enterprise's product positioning and product catalog. The rule engine is the Drools rule engine, and the rule scripts are written in drl format. The script structure includes a rule body identified by the `rule` keyword, a conditional judgment section guided by the `when` keyword, and an execution logic section guided by the `then` keyword. The execution logic includes model highlighting, text prompt output, and automatic correction operations.

[0028] S2. Build an intelligent review architecture based on the Kujiale platform. The review architecture embeds a rule engine to perform distributed cluster layered processing on customized furniture orders. Match the corresponding detection cluster according to the order data volume. Specifically, the distributed cluster layered processing is as follows: the internal detection of the model is assigned to the real-time detection cluster, the regular detection of the order scheme is assigned to the regular rule detection cluster, and the scheme with a data volume of more than 100MB is assigned to the Huge cluster.

[0029] S3. Configure and verify the review rules in the merchant backend of the Kujiale platform. The configuration includes setting the rule's effective scope, warning level, and application stage. The verification includes static syntax detection and dynamic performance detection. The rule's effective scope includes the department using it and the scope of review. The application stage includes order submission for review, order review, and drawing and quotation generation. The static syntax detection is used to check for script writing errors. The dynamic performance detection verifies the rule execution efficiency, conflict situation, and validity of the review results by capturing real order scenarios.

[0030] S4. Implement multi-stage intelligent review in the Kujiale design front end, including real-time detection and automatic correction during the design process, rule detection during order submission, problem investigation during the order review stage, and final verification during the quotation stage. Specifically, the real-time detection and automatic correction are as follows: when the designer drags or modifies the model, the system triggers rule detection in real time, automatically corrects parameters for problems that meet the correction conditions, and provides real-time prompts for problems that cannot be automatically corrected. Specifically, the rule detection during order submission is as follows: after the designer submits the order, the system performs a full rule scan of the design scheme, highlights models with errors, and categorizes and displays the problem types and warning levels in the submission list.

[0031] S5. Based on the warning level of the audit results, the data is processed in a tiered manner. Error-level data blocks subsequent business, while warning-level data only provides a notification and is archived. After the audit is passed, a data file that can be integrated with order splitting software is output. The data file that can be integrated with order splitting software is a JSON format file. The tiered processing also includes a manual repair step. For problems that cannot be resolved by automatic correction, relevant personnel manually modify them according to the prompts and then re-trigger the audit.

[0032] To verify the technical effect of the present invention, the following specific embodiments are provided, and the specific steps are as follows: Step 1: Based on the core requirements of customized furniture design and production, integrate the review standards of enterprises S, M, and L, and construct an intelligent review rule system that includes three major categories: design rationality rules, process rationality rules, and enterprise product standard rules.

[0033] Design rationality ensures that the product design possesses structural stability and aesthetic appeal. This category primarily includes material matching testing (whether materials are mixed, whether the base materials of cabinet components are consistent, and whether multiple brands of hardware appear in the same product design), and design error prevention testing (cabinet structure testing, component parameter testing, and whether the cabinet placement angle and position are accurate). Process rationality ensures that the product design meets production requirements and aligns with equipment technical standards. The review rules mainly include the following: hardware hole positions, whether there are conflicts between components, whether components are compatible, deformation prevention testing, and the accuracy of component process parameters. Enterprise product standards ensure that the product meets the company's product positioning and product catalog content. The main review directions include whether the product model exists in the company's product library, whether the material and color of the product model are still in production and sales, whether the product has been discontinued, and whether there are manually checked notes on order products. During the summarization process, to facilitate subsequent management and retrieval of the rules, the review rules are collected by name, category, description, review scope, warning level, and user department. Some examples are shown in Table 1.

[0034]

[0035] Step 2: Convert the audit rules into a rule script that the rule engine can recognize and execute. The rule engine is the Drools rule engine. The rule script is written in drl format. The script structure includes the rule body identified by the keyword "rule", the condition judgment part guided by the keyword "when", and the execution logic part guided by the keyword "then". The execution logic includes model highlighting, text prompt output, and automatic correction operation. Here, "rule" represents the beginning of the rule and is identified by a unique rule name, usually written in the form of "rule name". "When" represents the rule condition part. The object defined in "when" will have one or more constraints, and multiple constraints are connected by constraint connectors. Once the defined object satisfies the constraints, the rule will continue to execute the then part. Typically, an object is first defined in the form of "$" + any lowercase letter, using an object within the custom model data structure (as shown in Figure 1) as the defined object. Then, constraints are established within "()". Various keywords such as from, eval, collect, and contain can be used to establish the condition constraint language. then represents the rule result part. The content in then mainly contains the logic to be executed after the condition in when is met. It can be written in Java code form. The structure of the Drools rule engine is shown in Table 2.

[0036]

[0037] Step 3: Using the Kujiale platform as a carrier, embed the Drools rule engine to build a three-level review architecture as shown in Figure 2, which includes data input, rule processing, and result output. The review architecture embeds the rule engine and performs distributed cluster layered processing on customized furniture orders. The corresponding detection cluster is matched according to the order data volume. Specifically, the distributed cluster layered processing is as follows: the internal detection of the model is assigned to the real-time detection cluster, the regular detection of the order scheme is assigned to the regular rule detection cluster, and the scheme with a data volume of more than 100MB is assigned to the Huge cluster.

[0038] Step 4: Based on the Kujiale platform, the process of configuring rules for intelligent review of customized furniture in the merchant backend of the Kujiale platform is mainly divided into two parts: the first part is to set the review rules and check the rationality of the rule usage in the merchant backend; the second part is to implement intelligent review in the design process, order submission, manual order review, and quotation in the design frontend. This method optimizes the review of drawings and models in the early stage to one link in Kujiale, reducing the error rate after connecting production data before production. The specific process is shown in Figure 3. In the merchant backend of the Kujiale platform, enter the rule detection configuration interface, as shown in Figure 4. Here, you can complete the backend settings such as rule creation, maintenance, scope of effect, folder management, and warning level. In the rule list, you can set the effective department (using department) and effective detection items (review scope) of the rule. If these two items are not set, the rule will not be effective in the design tool. At the same time, you can set whether to use rule detection for review in the order submission, order review, and quotation stages, as shown in Figure 5. Creating a rule is shown in Figure 6. Here you can edit the rule script content, name, and set the alarm level, etc. When writing rule scripts, backend developers need to consider not only the applicability and accuracy of the rules, but also to avoid conflicts between rules. For example, rule A requires that cabinets >1200mm need reinforcement, and rule B requires that reinforcement parts be disabled. This conflict can be avoided by filtering out models that meet the requirements of rule A in the `when` statement when writing rule B. Alert levels are divided into two types: error level and warning level. An error level means that once the rule is triggered in scenarios such as drawing output, review, or quotation, it will block subsequent business operations. The error must be resolved and the rule must be approved before continuing. A warning level only provides a notification and does not affect subsequent processes, but the system will still archive the warning.

[0039] As shown in Figure 7, during the rule creation process in the merchant backend, the Kujiale platform performs intelligent checks on uploaded scripts to prevent rule conflicts and errors. This process is mainly divided into static and dynamic checks: static checks examine the script code syntax, identifying errors and prompting the user to correct them; dynamic checks, based on rules that pass the static checks, automatically retrieve three real-world implementations executed by the merchant within the past month for review, checking for rule performance issues. As long as the script doesn't time out, doesn't cause conflicts, and generates review results normally, it passes. If no suitable implementation is found, the merchant's backend staff will be prompted to manually add a correct implementation that complies with the rule's script review process. If the rule reports an error message for that implementation, it passes.

[0040] Step 5: Once the rules have been verified, the order plan can proceed to the intelligent review application stage. On the Kujiale design front end, designers will automatically trigger real-time rule detection and automatic correction prompts during the order design process to avoid structural, process, and other errors. When clicking "Place Order for Review," the review rules will be automatically triggered, and designers will handle the issues according to their level. For example, as shown in Figure 8: If there are back panels on the top and bottom of the active shelf, errors in the vertical panel will be displayed on the left panel of the design interface, reminding the designer of the problem. After entering the review stage, a second review will be conducted. If problems are found, the model with incorrect panels will be highlighted, and a reminder will be displayed again in the review list on the right (as shown in Figure 9).

[0041] Step 6: When an order enters the order review stage, rule checks will be triggered again (as shown in Figure 10). This automatically helps the reviewers list the problems and warning types in the solution, facilitating their troubleshooting and modification. Finally, when the order enters the quotation stage, another round of rule checks will be performed as the final review to ensure that the design and process rationality of the order will not affect quotation errors and the finance department's receipt of payment (as shown in Figure 11). After the quotation is verified to be correct, the order solution can be exported as a JOSN format file that can be integrated with order splitting software, and the order can enter the pre-production stage. The intelligent review process ends here.

[0042] In the intelligent review system based on the Kujiale platform, the rule engine realizes large-scale, standardized, and step-by-step intelligent review of standard parts order solutions. When the detection results are output, a warning will be issued according to the warning level set in the enterprise's backend. This process still requires manual judgment and processing by the enterprise. However, compared with the previous completely manual review, which relied heavily on experience, had easily repetitive and redundant steps, and had a high error rate, this intelligent review has made certain improvements in terms of the proportion of manual review for orders, the order pass rate, and the timeliness of order review. The data after the actual application of Kujiale intelligent review by enterprises L and S are shown in Table 3.

[0043]

[0044] In summary, this paper analyzes the problems existing in the design review process of customized furniture companies and summarizes and classifies commonly used review rules. Based on the Kujiale design platform and utilizing the Drools rule engine, a process and method for intelligent review of customized furniture is proposed. By simply writing the actual review rules into Drools-driven rule scripts and uploading, configuring, and maintaining them in the merchant backend of the Kujiale platform, intelligent multi-stage review of design schemes can be achieved. The feasibility and efficiency of this method are verified through actual operation and enterprise application data in various stages of the Kujiale platform. This significantly reduces the proportion of manual review and the number of review steps, lowering the error rate and time cost of review. This study comprehensively and thoroughly elaborates on the intelligent review method based on the Kujiale platform, which has profound significance for the intelligent development of customized furniture design review.

[0045] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for intelligent verification of customized furniture, characterized in that, The steps are as follows: S1. Construct a smart review rule system for customized furniture, transforming the review rules into rule scripts that can be recognized and executed by the rule engine. The review rule system includes rules for design rationality, rules for process rationality, and enterprise product standard rules. S2. Build a smart review architecture based on the Kujiale platform. This architecture embeds a rule engine and performs distributed cluster-based layered processing on customized furniture orders, matching corresponding detection clusters based on order data volume. S3. Configure and verify the review rules in the Kujiale platform merchant backend. The configuration includes setting the rule's effective scope, warning level, and application stage. The verification includes static syntax detection and dynamic performance detection. S4. Implement multi-stage smart review in the Kujiale design frontend, including real-time detection and automatic correction during the design process, rule detection during order submission, problem investigation during the order review stage, and final verification during the quotation stage. S5. Process the review results according to the warning level. Error-level results block subsequent business, while warning-level results only provide a notification and are archived. After approval, output a data file that can be integrated with order splitting software.

2. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S1, the design rationality rules include material matching detection rules and design error prevention detection rules. The material matching detection rules are used to verify the mixing of materials, the consistency of the base material of cabinet components, and the uniformity of hardware brands. The design error prevention detection rules are used to verify the accuracy of cabinet structure, component parameters, and cabinet placement angle and position.

3. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S1, the process rationality rules include hardware hole position detection rules, component conflict detection rules, component adaptation detection rules, anti-deformation detection rules, and component process parameter detection rules, which are used to ensure that the customized furniture design meets the technical standards of the production equipment.

4. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S1, the enterprise product standard rules include product model library matching rules, material and color validity rules, product off-shelf status detection rules, and order remarks information verification rules, which are used to verify the consistency between the ordered products and the enterprise's product positioning and product catalog.

5. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S1, the rule engine is the Drools rule engine, and the rule script is written in drl format. The script structure includes the rule body identified by the keyword "rule", the condition judgment part guided by the keyword "when", and the execution logic part guided by the keyword "then". The execution logic includes model highlighting, text prompt output, and automatic correction operation.

6. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S2, the distributed cluster layered processing specifically involves: allocating internal model detection to the real-time detection cluster, allocating routine order scheme detection to the routine rule detection cluster, and allocating schemes with data exceeding 100MB to the Huge cluster.

7. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S3, the scope of the rule includes the department using it and the scope of review. The application process includes order submission and review, order review and drawing and quotation. The static syntax detection is used to check for script writing errors. The dynamic performance detection verifies the rule execution efficiency, conflict situation and validity of the review results by capturing real order scenarios.

8. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S4, the real-time detection and automatic correction specifically means that when the designer performs model dragging or modification operations, the system triggers rule detection in real time, automatically corrects parameters for problems that meet the correction conditions, and provides real-time prompts for problems that cannot be automatically corrected.

9. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S4, the rule detection during order submission and review is as follows: After the designer submits the order, the system performs a full rule scan on the design scheme, highlights the models with errors, and displays the problem types and warning levels in the review list.

10. The method for intelligent verification of customized furniture according to claim 1, characterized in that, In step S5, the data file that can be connected to the order splitting software is a JSON format file. The hierarchical processing also includes a manual repair step. For problems that cannot be solved by automatic correction, relevant personnel will manually modify them according to the prompts and then re-trigger the review.

Images