Furniture production quality evaluation method and system

Abstract

The invention discloses a furniture production quality assessment method and system, and the method comprises the steps: intelligently deciding whether to start, pause or strengthen a special quality assessment measure for a production element according to an analysis result of multi-dimensional dynamic information; if so, lowering or pausing a special evaluation means for the production element; if the analysis result is that the strategy expansion condition is met, actively activating or enhancing a special quality evaluation strategy for the production element; and continuously recording actual quality performance data after strategy adjustment each time, dynamically optimizing various judgment conditions and thresholds by using a statistical analysis technology, and automatically matching and switching a quality evaluation strategy template according to different variables. According to the evaluation method, different quality evaluation strategy templates are automatically matched and switched, so that the intelligent level of furniture production quality management is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of furniture evaluation, and in particular to a furniture production quality evaluation method and system. BACKGROUND

[0002] In the modern furniture production process, in order to ensure that the product quality meets the standard requirements, it is usually necessary to carry out multi-aspect quality evaluation and monitoring on the produced furniture products. However, the traditional furniture production quality evaluation method often adopts a static, unified and relatively extensive evaluation strategy, for example, only relying on fixed appearance inspection, size measurement or sampling detection and the like, and the same or similar quality detection means and input intensity are adopted for all production elements (such as different materials, process links, production teams or equipment, etc.). This method has obvious deficiencies in actual application: on the one hand, it is difficult to accurately identify and differentially control the quality fluctuations under different production states, which easily leads to the key quality problems being ignored or discovered late; on the other hand, the same intensity of evaluation resources is invested in all production links, which not only causes waste of detection resources, but also makes it difficult to achieve efficient quality management in high-yield or complex process scenarios.

[0003] Especially in the production environment of multi-variety, small batch, customization and high requirements (such as export orders, environmentally friendly furniture, children's furniture, etc.), the product quality performance corresponding to different production elements (such as specific boards, connection processes, equipment teams, etc.) is significantly different.

[0004] Based on this, the present application proposes a furniture production quality evaluation method, which automatically matches and switches different quality evaluation strategy templates, thereby forming a complete closed-loop control process from state perception to strategy decision, and then to resource regulation and effect feedback, significantly improving the intelligent level of furniture production quality management. SUMMARY

[0005] The purpose of the present application is to provide a furniture production quality evaluation method and system to solve the problems in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solution: a furniture production quality evaluation method, the evaluation method comprising the following steps: S1: in the production process, real-time collection of multi-dimensional dynamic information related to product quality and production state, based on multi-dimensional dynamic information analysis whether the current state meets the preset two types of conditions, one type is strategy convergence condition, and the other type is strategy unfolding condition; S2: according to the analysis result of the multi-dimensional dynamic information, intelligently deciding whether to enable, suspend or strengthen the special quality evaluation measures for the production elements: S2.1: If the analysis result of the production factor meets the strategy convergence condition, the special evaluation means for the production factor will be adjusted or suspended; S2.2: If the analysis result meets the strategy expansion condition, the special quality evaluation strategy for the production factor will be actively activated or enhanced; S3: The actual quality performance data after each strategy adjustment is continuously recorded, statistical analysis techniques are used to dynamically optimize various judgment conditions and threshold values, and quality evaluation strategy templates are automatically matched and switched according to different variables, forming a control process from state perception to strategy decision.

[0007] In an embodiment disclosed in the present application, the actual quality performance data includes rework rate, defective product rate, and quality inspection pass rate indicators, and the variables include time period, product type, process change, and equipment state.

[0008] In an embodiment disclosed in the present application, the intelligent decision-making is based on a rule engine to realize automatic execution of control logic, including the following steps: Input layer: receiving the strategy convergence condition judgment result or the strategy expansion condition judgment result, while obtaining real-time quality indicators of the production factor, production rate, and risk characteristics of historical similar factors; Rule matching layer: matching the current state according to the preset decision rule library; Instruction generation layer: generating special evaluation control instructions according to the matching result, and issuing them to the corresponding production line unit through a production execution system or a device control system; Feedback loop layer: continuously monitoring the quality performance changes of the production factor after the execution of the control instructions.

[0009] In an embodiment disclosed in the present application, the construction of the quality evaluation strategy template includes the following steps: Pre-storing multiple sets of strategy parameter combinations, and each template is associated with an applicable condition label; Before each strategy decision, real-time collection of variable information of the current production environment is performed, and these variables are mapped to the applicable condition labels of the quality evaluation strategy templates; When it is detected that the current production environment variables match the applicable condition labels of a certain template, the strategy parameters in the template are automatically loaded; If there is no matching template, the template with the highest similarity is selected and manual confirmation is triggered.

[0010] In an embodiment disclosed in the present application, in step S1, the following data dimensions are collected for each monitored production factor: The total number of furniture products produced by the production factor within a time window, reflecting the actual production load and output scale of the factor; The actual performance of quality parameters related to product quality collected during the pre-shipment or post-processing inspection of furniture products related to production factors; The rate at which production factors produce products per unit of time is used to assess the capacity efficiency and operational status of production factors.

[0011] In one embodiment of this application, if the analysis results of the production factors meet the strategy convergence conditions, the special evaluation methods for the production factors will be reduced or suspended, including reducing the frequency of use of the testing equipment, turning off the defect identification unit, or postponing the analysis of material composition.

[0012] In one embodiment of this application, if the analysis result meets the strategy deployment conditions, the special quality assessment strategy for the production factor will be actively activated or enhanced, including enabling image recognition technology, introducing artificial intelligence defect classification models, detecting the chemical composition of materials, increasing the sampling inspection frequency, or implementing real-time monitoring of process nodes.

[0013] In one embodiment disclosed in this application, the multi-dimensional dynamic information includes the quantity of furniture products corresponding to a certain production factor, the quantitative performance of furniture products in terms of quality indicators, and the production rate of the corresponding products.

[0014] In one embodiment disclosed in this application, the production factors include materials, process steps, production teams or equipment, and the quantitative performance of the furniture products in terms of quality indicators includes dimensional qualification rate, surface defect rate and structural stability.

[0015] This application also provides a furniture production quality assessment system, including a data acquisition and analysis module, a strategy output module, and a dynamic optimization module; Data acquisition and analysis module: During the production process, it collects multi-dimensional dynamic information related to product quality and production status in real time. Based on the multi-dimensional dynamic information, it analyzes whether the current status meets two preset conditions: one is the strategy convergence condition, and the other is the strategy deployment condition. Strategy Output Module: Based on the analysis results of multi-dimensional dynamic information, intelligently decide whether to enable, suspend or strengthen special quality assessment measures for production factors. If the analysis results of production factors meet the strategy convergence conditions, the special assessment measures for that production factor will be reduced or suspended. If the analysis results meet the strategy deployment conditions, the special quality assessment strategy for that production factor will be actively activated or strengthened. Dynamic optimization module: continuously records the actual quality performance data after each strategy adjustment, uses statistical analysis technology to dynamically optimize various judgment conditions and thresholds, and automatically matches and switches quality assessment strategy templates according to different variables, forming a control process from state perception to strategy decision-making.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This application achieves comprehensive perception and real-time monitoring of various potential quality-influencing factors in the production process by collecting multi-dimensional dynamic information related to product quality and production status in real time. This solves the problems of one-sided information collection and lagging status perception in traditional furniture production quality assessment methods. Multi-dimensional dynamic information includes, but is not limited to, the quantity of products corresponding to different production elements (such as specific materials, process steps, production lines, or work groups), quantitative indicators of quality inspection, production flow, and product priority. Real-time acquisition of this data lays a solid foundation for subsequent intelligent analysis and strategy formulation.

[0017] 2. This application can proactively reduce or suspend the investment in detailed assessment of this element, such as reducing the frequency of use of high-precision testing instruments, decreasing the activation of specific defect identification, or delaying in-depth analysis of material composition. This effectively saves testing costs and human and material resources while ensuring basic quality control. Conversely, when the analysis results show that the strategy deployment conditions are met (such as high quality feedback intensity, large production flow, or belonging to high-priority or special target product types), this method will proactively activate or further enhance relevant specific quality assessment strategies, strengthen material physicochemical property analysis, or increase sampling and testing frequency. This ensures that key links receive high-intensity quality control commensurate with their risks, truly achieving dynamic adaptation of tightening where necessary and loosening where appropriate, thereby improving the efficiency and targeting of overall quality control.

[0018] 3. This application utilizes automatic matching and switching of the most suitable quality assessment strategy template. This adaptive optimization mechanism endows the entire quality assessment system with long-term evolution capabilities, enabling it to move beyond static judgments in the current state and continuously self-correct and upgrade in practice, forming a dynamic response capability for sustainable optimization. In summary, this solution constructs an efficient, accurate, and adaptive furniture production quality assessment method by dynamically sensing production status, intelligently adjusting assessment resources, and combining feedback data for self-optimization. This not only improves the intelligence level and resource utilization of quality management but also provides effective technical support and implementation paths for the digital transformation and refined quality control of the furniture manufacturing industry. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0020] Figure 1 This is a flowchart of the evaluation method of the present invention.

[0021] Figure 2 This is a timing diagram of the evaluation method of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example: This example provides a method for evaluating the quality of furniture production. Please refer to [link / reference]. Figure 1 and Figure 2 As shown, the evaluation method includes the following steps: S1: During the production process, multi-dimensional dynamic information related to product quality and production status is collected in real time. This information includes, but is not limited to: the quantity of furniture products corresponding to a specific production factor (such as a specific material, process step, production team, or equipment), the quantitative performance of these furniture products in key quality indicators (such as dimensional pass rate, surface defect rate, structural stability, etc.), and the production rate of the corresponding products. Based on the above multi-dimensional data, further analysis is conducted to determine whether the current status meets two preset conditions: one is the strategy convergence condition, which is that when a certain production factor currently has no associated product output, or although there is output, the quantity is small and the overall quality performance is good, or although the quality is acceptable, the output is low, it is considered that there is no need to invest too many special quality assessment resources in this factor; the other is the strategy deployment condition, which is that when the quality performance of a certain type of product fluctuates, the quality indicators exceed the normal range, or the product flow is large, it is considered that it is necessary to strengthen the depth of quality assessment and monitoring of this production factor.

[0024] S2: Based on the continuous operation of basic and global quality monitoring methods (such as routine appearance inspection, basic dimensional measurement, functional testing, etc.), and according to the analysis results of multi-dimensional dynamic information in step S1, the system intelligently decides whether to activate, suspend, or strengthen specific quality assessment measures for specific production factors. If the analysis results meet the strategy convergence conditions, the specific assessment methods for that factor will be appropriately reduced or suspended. For example, the frequency of use of high-precision inspection equipment will be reduced, specific defect identification units will be turned off, or in-depth analysis of material composition will be postponed, thereby avoiding unnecessary waste of resources and concentrating limited assessment capabilities on more critical aspects. If the analysis results meet the strategy deployment conditions, the system will proactively activate or enhance the specific quality assessment strategy for that production factor. For example, high-resolution image recognition technology will be enabled, an artificial intelligence defect classification model will be introduced, rapid chemical composition detection of key materials will be performed, sampling inspection frequency will be increased, or real-time monitoring of core process nodes will be implemented to ensure that the production quality of this part is more accurately and strictly controlled.

[0025] S3: To further enhance the intelligence and long-term applicability of the quality assessment method, this approach also includes a key innovative element: establishing an adaptive optimization and dynamic adjustment mechanism for the quality assessment strategy. This enables the entire assessment system to learn and continuously evolve. Specifically, the system continuously records actual quality performance data after each strategy adjustment, including key indicators such as rework rate, defect rate, and quality inspection pass rate. Based on this actual quality performance data, statistical analysis techniques are used to dynamically optimize various judgment conditions and thresholds (such as the upper and lower limits of quality feedback intensity). Simultaneously, it can automatically match and switch the most suitable quality assessment strategy template based on variables such as different time periods, product types, process changes, and equipment status, thus forming a control process from status perception to strategy decision-making. Through this mechanism, the furniture production quality assessment method can not only flexibly adjust the assessment focus according to the current production status but also continuously optimize itself as the production environment and product demands evolve, ultimately building a continuously evolving quality control system and effectively improving production efficiency and product quality stability.

[0026] This embodiment also provides a furniture production quality assessment system, including a data acquisition and analysis module, a strategy output module, and a dynamic optimization module; Data acquisition and analysis module: During the production process, it collects multi-dimensional dynamic information related to product quality and production status in real time. Based on the multi-dimensional dynamic information, it analyzes whether the current status meets two preset conditions: one is the strategy convergence condition and the other is the strategy deployment condition. The analysis results are sent to the strategy output module, and the multi-dimensional dynamic information is sent to both the strategy output module and the strategy output module. Strategy Output Module: Based on the analysis results of multi-dimensional dynamic information, intelligently decide whether to enable, suspend or strengthen special quality assessment measures for production factors. If the analysis results of production factors meet the strategy convergence conditions, the special assessment measures for that production factor will be reduced or suspended. If the analysis results meet the strategy deployment conditions, the special quality assessment strategy for that production factor will be actively activated or strengthened, and the strategy adjustment results will be sent to the dynamic optimization module. Dynamic optimization module: continuously records the actual quality performance data after each strategy adjustment, uses statistical analysis technology to dynamically optimize various judgment conditions and thresholds, and automatically matches and switches quality assessment strategy templates according to different variables, forming a control process from state perception to strategy decision-making.

[0027] The following details several steps involved in this application: During the production process, multi-dimensional dynamic information related to product quality and production status is collected in real time. This information includes, but is not limited to: the quantity of furniture products corresponding to a specific production factor (such as a specific material, process step, production team, or equipment), the quantitative performance of these furniture products in key quality indicators (such as dimensional pass rate, surface defect rate, structural stability, etc.), and the production rate of the corresponding products. Based on the above multi-dimensional data, the current status is further analyzed to determine whether it meets two preset conditions: one is the strategy convergence condition, which is that when a certain production factor currently has no associated product output, or although it has output, the quantity is small and the overall quality performance is good, or although the quality is acceptable, the output is low, it is considered that there is no need to invest too many special quality assessment resources in this factor; the other is the strategy deployment condition, which is that when the quality performance of a certain type of product fluctuates, the quality indicators exceed the normal range, or the product flow is large, it is considered that it is necessary to strengthen the depth of quality assessment and monitoring of this production factor.

[0028] Throughout the entire manufacturing process, multi-dimensional dynamic information closely related to product quality and production status is acquired in real time by deploying sensors at key nodes of the production line (including but not limited to sensor networks, MES system interfaces, PLC data acquisition ports, visual inspection equipment outputs, and manual data entry terminals). Specific production factors refer to production elements that have a significant impact on product quality or can serve as anchor points for quality control, including but not limited to: a specific type of material (such as a certain type of wood, board type, or surface coating type), a specific process (such as drying, painting, tenon and mortise processing, and sanding and polishing), a specific production team (such as woodworking team A and painting team B), and a specific piece of production equipment (such as CNC machining centers, automatic painting lines, and CNC sawing machines).

[0029] In this embodiment of the application, the following three data dimensions are collected for each monitored production factor: The total quantity of furniture products produced by a factor of production within a specific time window (such as per minute, per hour, or per shift) is used to reflect the actual production load and output scale of that factor. The actual performance of a series of core quality parameters directly related to product quality, collected during the pre-shipment or post-processing inspection of furniture products related to production factors, typically includes but is not limited to: dimensional accuracy pass rate (such as the proportion of products within the length, width and height tolerance range), surface quality defect rate (such as the proportion of surface defects such as scratches, dents, color difference, and particles), and structural stability indicators (such as the firmness of connectors, load-bearing test results, and the pass rate of gaps in joints). These indicators are usually collected by automated testing instruments, image recognition systems, physical performance testing benches, etc., and recorded in numerical or graded form. The output rate of a production factor per unit of time is usually expressed in units such as "pieces / hour", "pieces / shift" or "pieces / minute". It is used to evaluate the capacity efficiency and operating status of the factor. When combined with quality data, its "quality-efficiency" balance characteristics can be further analyzed.

[0030] In this embodiment, after the real-time aggregation of the aforementioned multi-dimensional data is completed, the current production and quality status of each production factor is intelligently analyzed and classified based on pre-set strategy condition judgment logic, thereby determining whether to trigger the corresponding quality assessment strategy adjustment mechanism. This strategy condition judgment is divided into the following two mutually exclusive and comprehensive logical branches: (I) Logic for Determining Policy Convergence Conditions When the multidimensional data analysis results for a specific production factor show any of the following situations, the system will determine that the factor is currently in a low-risk, low-concern state, thus meeting the strategy convergence condition. This means that there is no need to invest too many special quality assessment resources in it, in order to avoid unnecessary quality control costs and resource waste: Scenario 1: No product output or zero correlation In other words, within the current time window, this production factor has not participated in the production process of any furniture products, or the system has failed to detect any output records associated with it (such as equipment downtime, work shifts not being scheduled, specific materials not being used, etc.), therefore there is no corresponding product to be evaluated.

[0031] Scenario 2: Product output is relatively low but overall quality is excellent. This means that although the production factor produces products, the quantity of output is lower than the system's preset "concern threshold" (e.g., less than 5 pieces per hour, less than 20 pieces per shift, etc., which can be dynamically set according to different production factor types, historical data distribution, process complexity, etc.). Furthermore, within this small output, the actual performance of all key quality indicators is better than or equal to the system's preset "quality baseline" (e.g., dimensional pass rate ≥99%, surface defect rate ≤0.5%, structural stability 100% compliance, etc.). This indicates that although production activities exist, their scale is limited and the quality risk is extremely low.

[0032] Scenario 3: Quality performance is acceptable, but product flow rate is low. In other words, although the output of this production factor has not reached an extremely low level, it is still in a low range overall. At the same time, although the performance of its key quality indicators has not reached the top level, it is still stable within an acceptable range (for example, all indicators are within the preset "normal fluctuation range" and there is no obvious abnormal trend). The comprehensive assessment believes that the current contribution of this factor to the overall quality fluctuation is small, and there is no need to prioritize the allocation of special assessment resources.

[0033] For any production factor that meets any of the above conditions, the system will automatically classify it as a strategy convergence object and trigger a resource optimization mechanism, such as reducing the frequency of special inspections of the factor, reducing the scheduling of high-precision inspection equipment, suspending the operation of specific defect identification units, and postponing or delaying in-depth analysis of material composition, thereby achieving rational allocation of quality control resources and cost optimization.

[0034] (ii) Logic for Determining the Conditions for Strategy Deployment Conversely, when the system analyzes multidimensional data of a production factor and finds any one or a combination of the following situations, it will determine that the factor is currently in a high-risk, high-concern state, thus meeting the conditions for strategy implementation. That is, it is necessary to immediately strengthen the depth of quality assessment and monitoring of the factor to prevent the spread of potential quality problems and improve the control accuracy of key links: Scenario 1: Product quality performance fluctuates significantly. This means that the furniture products associated with this production factor show abnormal changes in one or more key quality indicators that exceed the historical baseline range or normal fluctuation range (e.g., the size qualification rate drops sharply from 98% to 95%, the surface defect rate rises from 0.3% to 1.2%, and the number of qualified structural stability indicators drops significantly). Such fluctuations may be caused by factors such as drift in process parameters, batch differences in raw materials, changes in operators, and deterioration of equipment condition, and timely intervention and analysis are required.

[0035] Scenario 2: Key quality indicators exceed the normal range threshold. If the actual test results of one or more core quality parameters exceed the system's preset upper limit or qualified standard range (e.g., dimensional deviation exceeds ±0.5mm limit, surface defect rate exceeds 1% warning line, structural test fails, etc.), it directly indicates that there is a clear quality risk in the current production status, and monitoring and evaluation must be strengthened.

[0036] Scenario 3: Product flow (output quantity) is relatively large This means that the production factor produces a large number of furniture products per unit of time (e.g., more than 20 pieces per hour, more than 100 pieces per shift, etc., with specific thresholds set according to the scale of the production line and the type of products). Even if there are no obvious abnormalities in the current quality performance, due to its high output characteristics, once a quality problem occurs, it may lead to a large number of defective products. Therefore, it is necessary to strengthen quality assessment and process control in advance to implement preventive management.

[0037] For any production factor that meets any of the above conditions, the system will automatically classify it as a strategy deployment object and trigger a special quality assessment enhancement mechanism. For example, it will enable a high-resolution image recognition system to conduct real-time detection of product surfaces, conduct rapid chemical composition or physical property testing of key raw materials, increase the frequency and coverage of sampling inspections, and implement real-time monitoring and feedback control of process parameters for core process nodes (such as gluing, pressing, heat treatment, etc.), thereby achieving precise and in-depth quality control of high-concern factors.

[0038] Based on the continuous operation of fundamental and comprehensive quality monitoring methods (such as routine appearance inspection, basic dimensional measurement, and functional testing), and according to the analysis results of multi-dimensional dynamic information in step S1, the system intelligently decides whether to activate, suspend, or strengthen specific quality assessment measures for particular production factors. If the analysis results meet the strategy convergence conditions, the specific assessment methods for that factor will be appropriately reduced or suspended. For example, the frequency of use of high-precision inspection equipment will be reduced, specific defect identification units will be turned off, or in-depth analysis of material composition will be postponed, thereby avoiding unnecessary waste of resources and concentrating limited assessment capabilities on more critical aspects. If the analysis results meet the strategy deployment conditions, the system will proactively activate or enhance the specific quality assessment strategy for that production factor. For example, high-resolution image recognition technology will be enabled, an artificial intelligence defect classification model will be introduced, rapid chemical composition detection of key materials will be performed, sampling inspection frequency will be increased, or real-time monitoring of core process nodes will be implemented to ensure that the production quality of that part is more accurately and rigorously controlled.

[0039] Maintaining the continuous operation of basic quality monitoring employs general-purpose detection technologies and standardized processes (e.g., basic identification of surface scratches / color differences using visual sensors, basic qualification judgment of key dimensions (such as hole diameter and plate thickness) using measuring tools, and verification of basic performance by simulating user scenarios using functional test benches). Its characteristics include broad coverage, low execution cost, and coarse-grained risk identification. Simultaneously, based on the results of strategy condition judgments (i.e., whether a production factor belongs to the strategy convergence object or the strategy deployment object), specific assessment and control instructions for that factor are dynamically generated.

[0040] In this embodiment, when the analysis result of step S1 determines that a certain production factor meets the strategy convergence condition (specific situations include: the factor has no product output, the output quantity is lower than the system's preset low attention threshold and the overall quality performance is excellent or the quality is acceptable but the output is low), the system determines that the factor's current contribution to the overall quality fluctuation is extremely low. If high-cost special evaluation resources are continued to be invested, it will lead to resource redundancy and efficiency loss. At this time, a special evaluation downgrade or suspension mechanism will be triggered. The specific control logic is as follows: Reduce the depth and frequency of specialized assessments, decrease the application of resource-intensive testing methods, and allocate limited assessment capabilities to high-risk processes. Check the testing task queue associated with current production factors. If the number of products to be inspected for that factor is below a threshold (e.g., ≤5 pieces per hour), automatically reduce the priority of high-precision testing equipment (such as laser 3D scanners, X-ray internal defect detectors, etc.), or reassign the testing time slot originally allocated to that factor to other high-risk factors. If the minimum testing requirement is not met for multiple consecutive time windows (e.g., 3 consecutive hours), directly suspend the specialized testing tasks for that equipment. For specialized assessments relying on specific algorithms (e.g., machine vision-based surface micro-defect recognition units, structural stress concentration area detection, etc.), if the surface defect rate of that production factor is consistently below the baseline (e.g., ≤0.3%) and shows no abnormal fluctuation trend, disable the real-time analysis function for that factor and replace it with basic visual inspection. For critical materials that require verification through laboratory-level testing (such as material composition spectral analysis, adhesive durability testing, etc.), if the batch of materials currently used in this production factor is a historically stable batch (e.g., the test results of the last 10 batches all meet the standards) and the structural stability indicators of related products continue to meet the standards, then the in-depth chemical analysis of the batch of materials will be temporarily suspended, and only the rapid screening of basic physicochemical indicators will be retained.

[0041] In this embodiment of the application, when the analysis results determine that a certain production factor meets the strategy deployment conditions (specific situations include: significant fluctuations in product quality performance (e.g., key indicators exceeding the historical normal fluctuation range), key quality indicators exceeding the preset normal range threshold (e.g., size qualification rate <95%, surface defect rate >1%, structural test failure, etc.), or product flow (output quantity) exceeding the system's preset "high load threshold" (e.g., ≥20 pieces per hour, ≥100 pieces per shift)), the system determines that the factor currently has a high quality risk or potential large-scale defective product risks, and it is necessary to strengthen special assessment methods to achieve precise and strict quality control. The specific control logic is as follows: To improve the accuracy, frequency, and coverage of specialized assessments, high-sensitivity detection technologies and intelligent analysis tools are introduced to implement process-level monitoring of key processes, ensuring early detection and intervention of risks. For furniture components with high surface quality requirements (such as cabinet door panels and carved decorative parts), the system activates a high-resolution industrial camera (resolution ≥ 50 million pixels) combined with multispectral light sources (visible light + infrared / ultraviolet) to perform micron-level defect detection on the product surface (such as cracks and pinholes smaller than 0.1 mm), and transmits the images in real time to an AI analysis server for defect location and classification. Based on deep learning models (such as convolutional neural networks CNN) trained on historical quality data, the detected surface or structural anomalies are automatically classified (such as scratches / dents / deformation / poor bonding), and associated with specific process parameters or operational steps (e.g., a certain type of scratch is related to excessively high sanding machine speed), providing targeted basis for subsequent process optimization. For core materials that may affect structural stability or durability (such as adhesives, coatings, and metal connectors), the system adds a rapid on-site detection step using a portable spectrometer (such as XRF) or near-infrared spectrometer to the conventional physical performance testing. This allows for real-time acquisition of the material's composition ratio (such as formaldehyde content, resin curing agent ratio, etc.) and comparison with preset safety thresholds to avoid quality problems caused by batch differences in materials.

[0042] Based on the fluctuation range of product flow and quality, the system automatically calculates and adjusts the sampling ratio (for example, the sampling ratio is 1% to 3% under normal circumstances, and increases to 5% to 10% when the quality fluctuation exceeds ±5%; if three consecutive defective products are found, the system is upgraded to full inspection mode) to ensure that the coverage of abnormal samples is sufficient to reflect the overall quality status. For key process links with concentrated quality risks (such as hot pressing temperature control, coating thickness adjustment in the coating line, and gluing pressure monitoring of mortise and tenon structures), the system collects real-time process parameters (sampling frequency ≥ 1 time / second) through sensors deployed on the equipment (such as temperature sensors, pressure sensors, flow sensors, etc.), and combines them with preset process windows (such as temperature ±2℃, pressure ±0.5MPa) to immediately alarm for parameters deviating from the window and trace the related product batches, thereby achieving closed-loop control of process quality.

[0043] In this embodiment of the application, intelligent decision-making is based on a rule engine to automate the execution of control logic. The processing logic is as follows: Input layer: Receives the results of strategy convergence condition determination or strategy expansion condition determination, and at the same time obtains auxiliary information such as real-time quality indicators (e.g., current size pass rate, defect rate), production rate (pieces / hour), and risk characteristics of similar historical factors (e.g., the frequency of quality fluctuations of similar shifts / equipment in the past 30 days).

[0044] Rule matching layer: Based on a pre-defined decision rule base (e.g., "If the strategy convergence condition is met and the high-precision equipment utilization rate is <10%, then suspend the special inspection of this equipment"); "If the strategy expansion condition is met and the defect rate is >1%, then activate the AI ​​defect classification model and increase the sampling ratio to 5%), the current state is matched. The rule base supports dynamic updates and can adjust thresholds and response strategies according to changes in the production environment (such as the introduction of new materials or the launch of new processes).

[0045] Instruction generation layer: Based on the matching results, specific special assessment and control instructions (such as "shut down unit X", "start equipment Y", "adjust sampling frequency from 2% to 8%)) are generated and issued to the corresponding production line units through the production execution system (MES) or equipment control system (PLC).

[0046] Feedback closed-loop layer: After the control command is executed, the system continuously monitors the changes in the quality performance of the production factor (such as whether the defect rate drops after the strengthening measures are activated, and whether the quality rebounds after the measures are suspended), and uses the feedback data for subsequent strategy optimization.

[0047] To further enhance the intelligence and long-term applicability of the quality assessment method, this approach incorporates a key innovation: establishing an adaptive optimization and dynamic adjustment mechanism for the quality assessment strategy. This enables the entire assessment system to learn and continuously evolve. Specifically, the system continuously records actual quality performance data after each strategy adjustment, including key indicators such as rework rate, defect rate, and quality inspection pass rate. Based on this actual performance data, statistical analysis techniques are used to dynamically optimize various judgment conditions and thresholds (such as the upper and lower limits of quality feedback intensity). Simultaneously, it can automatically match and switch the most suitable quality assessment strategy template based on variables such as different time periods, product types, process changes, and equipment status, thus forming a control process from status perception to strategy decision-making. Through this mechanism, the furniture production quality assessment method can not only flexibly adjust the assessment focus according to the current production status but also continuously optimize itself as the production environment and product demands evolve, ultimately building a continuously evolving quality control system that effectively improves production efficiency and product quality stability.

[0048] Traditional quality assessment strategies are usually based on fixed experience thresholds (such as "size pass rate ≥ 95% is normal" and "defect rate > 1% requires intervention") and preset rules (such as "initiate enhanced testing when a certain equipment produces more than 20 pieces"). These strategies are difficult to adapt to the dynamic changes in the production environment (such as equipment aging and process upgrades), product characteristics (such as the application of new materials and complex structural designs), and market demands (such as an increase in customized orders).

[0049] This application continuously collects actual quality performance data after strategy adjustment, and dynamically optimizes the core judgment conditions (such as quality fluctuation threshold and resource scheduling trigger point) and threshold parameters (such as rework rate warning line and defect rate upper limit) in the strategy based on statistical analysis technology. At the same time, it automatically matches the optimal strategy template by combining multi-dimensional variables of the production scenario (such as time period, product type, process parameters and equipment status).

[0050] In this embodiment, after each strategy adjustment (including the activation, suspension, or enhancement of a special assessment triggered in S2), subsequent quality performance data associated with the strategy is collected in real time through the interfaces of the Production Execution System (MES), Quality Management System (QMS), and testing equipment. This data serves as a direct evaluation basis for the strategy's effectiveness. This data is stored in a structured format in a quality database and indexed by time series and production factor dimensions, including: Rework rate (the percentage of products that need to be repaired due to quality issues), defect rate (the percentage of products that are ultimately deemed unqualified), quality inspection pass rate (the percentage of products that pass the inspection on the first attempt), customer complaint rate (the percentage of feedback caused by quality issues after delivery), etc. The specific type of this adjustment (e.g., "activating the AI ​​defect classification model for a certain work group" or "suspending in-depth material analysis of a certain equipment"), the adjustment time, the strategy parameters before and after the adjustment (e.g., the original sampling frequency was 2%, now increased to 5%), and the associated production element identifiers (e.g., material number, process code, equipment ID). Auxiliary variables include the time period during strategy execution (e.g., weekdays / holidays, day shifts / night shifts), product type (e.g., solid wood dining tables / panel wardrobes), process version (e.g., V1.2 coating process), and equipment operating status (e.g., equipment load rate, fault alarm records).

[0051] The complete record of the above multi-dimensional data provides a sufficient chain of causal evidence for subsequent strategy optimization analysis.

[0052] In this embodiment, the system periodically (e.g., daily / weekly / monthly, with the specific cycle set according to the production rhythm) analyzes the stored actual quality performance data, identifies the reasonableness deviation between the strategy judgment conditions and thresholds, and dynamically adjusts the parameters through processing logic. The logic of this process is as follows: For each production factor or strategy type (such as the defect rate under the strategy deployment conditions of a certain material), the system aggregates key quality indicators (such as calculating the average rework rate and the standard deviation of the defect rate) by time window (such as the last 30 days), and identifies abnormal situations such as continuous deterioration of quality performance (such as the average defect rate rising by more than 10% compared with the historical baseline) or excessive investment of resources (such as the quality inspection pass rate being higher than 99% for a long time but the frequency of special inspections still remaining high) under the current strategy.

[0053] In this embodiment of the application, the judgment conditions and thresholds preset in the strategy (e.g., the upper limit of the quality feedback intensity is a defect rate of 1%, and the trigger condition for strategy deployment is a rework rate of >5%) are evaluated by comparing the degree of matching between the actual observed value and the threshold: if the actual quality fluctuates (e.g., the defect rate is stable at 0.3% to 0.5% for a long time, far below the 1% threshold) but frequently triggers special assessment reinforcement, it is determined that the current threshold setting is too sensitive and there is a risk of wasting resources; if the actual quality has deteriorated significantly (e.g., the defect rate is >6% for 5 consecutive days) but the strategy adjustment is not triggered, it is determined that the threshold setting is too lenient and there is a risk of missed detection.

[0054] In this embodiment of the application, the threshold and conditions are iteratively adjusted based on the evaluation results: For thresholds that are too sensitive (such as the original defect rate threshold of 1% corresponding to an actual long-term performance of 0.3%), the thresholds should be raised to a reasonable range (e.g., adjusted to 0.5% to 0.8%) to reduce the frequency of special assessments in low-risk scenarios. If the threshold is too lenient (e.g., the original rework rate threshold of 5% corresponds to the actual recent average of 6.5%), the threshold should be lowered to a more stringent level (e.g., adjusted to 4% to 4.5%) to detect potential quality risks in advance. For strategies with multiple conditions (such as initiating enhanced detection when product traffic is high and defect rate is >0.8%), analyze the independent contribution of each condition (e.g., it is found that high traffic alone leads to a higher increase in defect rate), and dynamically adjust the weight or priority of the conditions (e.g., reduce the trigger threshold for high traffic from 20 pieces / hour to 15 pieces / hour).

[0055] The furniture production environment is highly variable (e.g., differences in equipment stability during the day / night, varying process sensitivities between solid wood and panel furniture, and differing quality fluctuation characteristics between new and skilled worker teams). Therefore, a single, fixed strategy template is difficult to apply universally across all scenarios. By establishing a quality assessment strategy template (containing predefined strategy combinations for different time periods, product types, process states, and equipment configurations), and combining this with real-time production environment variables, the optimal template is automatically selected and switched. The processing logic is as follows: In this embodiment of the application, the process for constructing the quality assessment strategy template is as follows: The template library pre-stores multiple sets of strategy parameter combinations (e.g., "Weekday Day Shift Solid Wood Furniture Strategy Template" includes high-frequency sampling inspection, AI defect classification model activation, and real-time monitoring of key equipment; "Holiday Night Shift Panel Furniture Strategy Template" includes basic sampling inspection, enhanced routine appearance inspection, and postponement of in-depth material analysis). Each template is associated with clear applicable condition tags (e.g., time period = weekday 8:00-20:00, product type = solid wood, process version = V2.0, equipment load rate <80%).

[0056] Before each strategy decision, the system collects variable information of the current production environment in real time (such as the current time is 22:00 (night shift), the product type is panel wardrobe, the process is V1.5 coating process, and the load rate of a certain equipment is 85%), and maps these variables to the applicable condition labels of the quality assessment strategy template through processing logic (for example, "night shift + panel furniture + load rate > 80%" matches "holiday night shift high load strategy template").

[0057] When the system detects that the current production environment variables completely match the applicable condition label of a certain template, it automatically loads the strategy parameters in the template (such as increasing the sampling frequency from the usual 3% to 5%, enabling a special detection unit for splicing gaps in panel furniture, and pausing in-depth chemical analysis of high-load equipment) to ensure a high degree of adaptation between the strategy and the scenario. If there is no completely matching template, the template with the highest similarity is selected and manual confirmation is triggered (or the optimal parameter combination is recommended based on historical similar scenario data).

[0058] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0059] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for evaluating the quality of furniture production, characterized in that: The evaluation method includes the following steps: S1: During the production process, multi-dimensional dynamic information related to product quality and production status is collected in real time. Based on the multi-dimensional dynamic information, it is analyzed whether the current status meets two preset conditions: one is the strategy convergence condition, and the other is the strategy deployment condition. S2: Based on the analysis results of multi-dimensional dynamic information, intelligent decision-making determines whether to activate, suspend, or strengthen specific quality assessment measures for production factors. S2.1: If the analysis results of the production factors meet the strategy convergence conditions, the specific assessment methods for that production factor will be reduced or suspended. S2.2: If the analysis results meet the conditions for strategy implementation, the specific quality assessment strategy for this production factor will be actively activated or enhanced. S3: Continuously record the actual quality performance data after each strategy adjustment, use statistical analysis techniques to dynamically optimize various judgment conditions and thresholds, and automatically match and switch quality assessment strategy templates according to different variables to form a control process from state perception to strategy decision-making.

2. The furniture production quality assessment method according to claim 1, characterized in that: The actual quality performance data includes rework rate, defect rate, and quality inspection pass rate. The variables include time period, product type, process changes, and equipment status.

3. The furniture production quality assessment method according to claim 2, characterized in that: The intelligent decision-making system, based on a rule engine, automates the execution of control logic, including the following steps: Input layer: Receives the results of policy convergence condition determination or policy expansion condition determination, and simultaneously obtains real-time quality indicators, production rates, and risk characteristics of similar historical factors of production. Rule matching layer: Matches the current state according to a preset decision rule base; Instruction generation layer: Generates special assessment and control instructions based on the matching results, and issues them to the corresponding production line units through the production execution system or equipment control system; Feedback closed-loop layer: After the control instructions are executed, the quality performance of production factors is continuously monitored.

4. The furniture production quality assessment method according to claim 3, characterized in that: The construction of the quality assessment strategy template includes the following steps: Multiple sets of strategy parameter combinations are pre-stored, and each template is associated with applicable condition tags; Before each strategic decision, collect variable information of the current production environment in real time and map these variables to the applicable condition labels of the quality assessment strategy template; When it is detected that the current production environment variable matches the applicable condition tag of a template, the strategy parameters in that template are automatically loaded. If no matching template is found, the template with the highest similarity will be selected and manual confirmation will be triggered.

5. The furniture production quality assessment method according to claim 1, characterized in that: In step S1, for each monitored production factor, the following data dimensions are collected: The total quantity of furniture products produced by a factor of production within a time window reflects the actual production load and output scale of that factor. The actual performance of quality parameters related to product quality collected during the pre-shipment or post-processing inspection of furniture products related to production factors; The rate at which production factors produce products per unit of time is used to assess the capacity efficiency and operational status of production factors.

6. The furniture production quality assessment method according to claim 2, characterized in that: If the analysis results of the production factors meet the strategy convergence conditions, the special assessment methods for that production factor will be reduced or suspended, including reducing the frequency of use of testing equipment, shutting down the defect identification unit, or postponing the analysis of material composition.

7. The furniture production quality assessment method according to claim 6, characterized in that: If the analysis results meet the conditions for strategy implementation, the specific quality assessment strategy for that production factor will be actively activated or enhanced, including enabling image recognition technology, introducing artificial intelligence defect classification models, conducting chemical composition testing on materials, increasing the frequency of sampling inspections, or implementing real-time monitoring of process nodes.

8. The furniture production quality assessment method according to claim 1, characterized in that: The multi-dimensional dynamic information includes the quantity of furniture products corresponding to a certain production factor, the quantitative performance of furniture products in terms of quality indicators, and the production rate of the corresponding products.

9. A method for evaluating the quality of furniture production according to claim 8, characterized in that: The production factors include materials, processes, production teams or equipment, and the quantitative performance of the furniture products in terms of quality indicators includes dimensional qualification rate, surface defect rate and structural stability.

10. A furniture production quality assessment system, used to implement the assessment method according to any one of claims 1-9, characterized in that: It includes a data acquisition and analysis module, a strategy output module, and a dynamic optimization module; Data acquisition and analysis module: During the production process, it collects multi-dimensional dynamic information related to product quality and production status in real time. Based on the multi-dimensional dynamic information, it analyzes whether the current status meets two preset conditions: one is the strategy convergence condition, and the other is the strategy deployment condition. Strategy Output Module: Based on the analysis results of multi-dimensional dynamic information, intelligently decide whether to enable, suspend or strengthen special quality assessment measures for production factors. If the analysis results of production factors meet the strategy convergence conditions, the special assessment measures for that production factor will be reduced or suspended. If the analysis results meet the strategy deployment conditions, the special quality assessment strategy for that production factor will be actively activated or strengthened. Dynamic optimization module: continuously records the actual quality performance data after each strategy adjustment, uses statistical analysis technology to dynamically optimize various judgment conditions and thresholds, and automatically matches and switches quality assessment strategy templates according to different variables, forming a control process from state perception to strategy decision-making.

Images