A digital assessment and decision-making model for the transfer of the textile and apparel industry
By constructing a digital assessment and decision-making model, the problems of single data collection and static weights in the assessment of textile and apparel industry transfer have been solved, realizing a digital closed loop throughout the entire process and improving the accuracy of assessment and the reliability of decision-making.
Patent Information
- Application Number
- CN202610478527.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-13
- Publication Date
- 2026-11-10
AI Technical Summary
Existing assessment methods for the transfer of textile and apparel industries rely on a single data collection dimension, static weighting of assessment indicators, and a lack of extrapolation and verification methods, resulting in lagging assessment results and high decision-making risks.
A digital assessment and decision-making model for the transfer of the textile and apparel industry is constructed, including a data access module, an indicator system construction module, a dynamic assessment module, and a decision inference module. Through multi-source data processing, dynamic weighting, and digital twin models, a full-process digital assessment is conducted to achieve adaptive and quantifiable decision support.
It has achieved a fully digital closed loop for the assessment of the transfer of the textile and apparel industry, which has improved the accuracy, adaptability and reliability of the assessment, and reduced the lag in assessment results and decision-making risks.
Smart Images

Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] This application relates to the field of digital assessment technology in the textile and apparel industry, and more specifically, to a digital assessment and decision-making model for the transfer of the textile and apparel industry. Background Technology
[0002] With the profound adjustment of the global layout of the textile and apparel industry, industrial relocation has become an important strategic choice for textile companies to cope with rising costs, market changes, and adjustments in trade policies. However, the textile and apparel industry chain is long, requires high levels of process collaboration, and is highly sensitive to order cycles, making industrial relocation decisions face many technical challenges.
[0003] Existing industrial relocation assessment methods suffer from the following technical problems: First, the data collection dimensions are too limited to fully reflect the multiple factors involved in the relocation of the textile and apparel industry, such as capacity adaptation, supply chain restructuring, and order response. Second, the weights of assessment indicators rely heavily on static settings based on expert experience and cannot be dynamically adjusted according to changes in regional data, resulting in assessment results lagging behind the actual situation. Third, relocation plans lack quantifiable methods for extrapolation and verification, making it difficult for companies to predict capacity ramp-up curves, supply chain disruption risks, and order delivery timeliness under different relocation plans, leading to high decision-making risks. Summary of the Invention
[0004] The present invention aims to overcome the shortcomings of the prior art and provide a digital assessment and decision-making model for the transfer of the textile and apparel industry, so as to solve the technical problems of the prior art, such as single assessment dimensions, static weight setting, and lack of extrapolation and verification capabilities.
[0005] To achieve the above technical objectives, the present invention adopts the following technical solution: A digital assessment and decision-making model for the relocation of the textile and apparel industry includes: The data access module is configured to acquire multi-source data related to the transfer of the textile and apparel industry. The indicator system construction module is connected to the data access module and is configured to construct an evaluation indicator system containing multiple evaluation dimensions, wherein the evaluation dimensions correspond to the business areas covered by the multi-source data. The dynamic evaluation module is connected to the data access module and the indicator system construction module, respectively, and is configured to dynamically assign weights to each indicator in the evaluation indicator system based on the multi-source data, and calculate the evaluation score of each candidate region. The decision-making simulation module is connected to the dynamic evaluation module and is configured to build a digital simulation model for candidate areas whose evaluation scores meet a preset threshold, simulate the operational effects under different transfer schemes, and generate optimized transfer decision information based on the simulation results, which is then fed back to the dynamic evaluation module to adjust the evaluation parameters.
[0006] Preferably, the data access module includes an edge computing gateway deployed at the textile enterprise, and the acquired multi-source data includes at least: Manufacturing execution system production data, enterprise resource planning system resource data, textile equipment operating condition data collected by IoT devices, and logistics data in supply chain management system; The edge computing gateway is configured to preprocess and extract features from real-time vibration data, temperature and humidity data, and production data of textile equipment, and upload the compressed feature vector to the dynamic evaluation module.
[0007] Preferably, the evaluation indicator system constructed by the indicator system construction module includes at least: Capacity migration adaptability indicators, supply chain restructuring cost indicators, and order response time indicators; The business areas corresponding to the capacity migration adaptability index are textile equipment and human resources; the business areas corresponding to the supply chain restructuring cost index are raw materials and supporting collaboration; and the business areas corresponding to the order response timeliness index are production cycle and logistics clearance.
[0008] Preferably, the capacity migration adaptability index includes the textile equipment compatibility index; The dynamic evaluation module uses graph neural network algorithms to calculate the process connection similarity and control system protocol matching degree between the equipment to be transferred and the existing equipment in the target area, based on the textile equipment operating condition data and equipment model data in the multi-source data, in order to determine the textile equipment compatibility index.
[0009] Preferably, the supply chain restructuring cost indicators include the completeness of outsourcing support; The dynamic evaluation module, based on the target region enterprise distribution data in the multi-source data, determines the completeness of the external cooperation by analyzing the enterprise distribution density and network connectivity in the industrial chain links. The aforementioned industrial chain links include at least spinning, weaving, dyeing and finishing, printing, embroidery and finishing.
[0010] Preferably, the order response timeliness indicators include at least the potential for reducing the cycle from order acceptance to first shipment, the flexible production adjustment capability for fast-response orders, and the efficiency of cross-border logistics and customs clearance; The dynamic evaluation module calculates various sub-indicators of the order response timeliness index based on the production history data, order fulfillment data, and logistics timeliness data from the multi-source data.
[0011] Preferably, the dynamic evaluation module uses an improved entropy-weighted analytic hierarchy process to dynamically assign weights to the evaluation index system; The improved entropy-weighted analytic hierarchy process includes: using the analytic hierarchy process to determine subjective weights, using the entropy weight method to calculate objective weights based on the degree of variation of each indicator in the multi-source data, and solving for the optimal combined weights by constructing a combined weight optimization model.
[0012] Preferably, the digital inference model constructed by the decision inference module is a digital twin model; The digital twin model maps the textile equipment layout, logistics network nodes, and supporting enterprise distribution of the target area based on the target area's geographic information, equipment layout data, and supporting enterprise distribution data from the multi-source data. The decision-making simulation module is also configured to introduce a supply chain resilience assessment submodule into the digital twin model, which simulates the supply chain recovery time and capacity loss range when a sudden event occurs in the target area based on historical supply chain disruption data from the multi-source data, and uses the supply chain resilience score as a correction factor for the decision output.
[0013] Preferably, the model further includes a feedback optimization module, which is connected to the data access module and the decision inference module respectively. It is configured to continuously collect actual operating data of the transferred area after the enterprise completes part of the capacity transfer, compare and analyze the data with the inference results of the decision inference module, calculate the prediction deviation, and adjust the evaluation parameters of the dynamic evaluation module and the inference parameters of the decision inference module.
[0014] Preferably, the dynamic evaluation module is further configured to preset corresponding indicator weight thresholds and evaluation benchmarks according to different sub-sectors of the textile and apparel industry; the sub-sectors include at least cotton textiles, wool textiles, garments and industrial textiles.
[0015] This invention provides a digital assessment and decision-making model for the transfer of the textile and apparel industry. By constructing a complete technical architecture including a data access module, an indicator system construction module, a dynamic assessment module, and a decision deduction module, it achieves a fully digital closed-loop assessment process for the textile and apparel industry transfer. Specifically, the data access module preprocesses and extracts features from multi-source heterogeneous data through edge computing gateways deployed at textile enterprises, uploading only compressed feature vectors, significantly reducing bandwidth consumption for cross-regional data transmission and solving the technical challenge of remotely aggregating massive amounts of real-time data from textile equipment. The indicator system construction module constructs an assessment indicator system corresponding to the multi-source data business domain. The dynamic assessment module uses an improved entropy-weighted analytic hierarchy process to dynamically assign weights to various indicators based on real-time collected multi-source data, enabling the assessment to be more comprehensive and effective. The assessment results can adapt to changes in regional characteristics, overcoming the technical shortcomings of traditional static weight settings, which are highly subjective and cannot be dynamically adjusted with regional data. The decision-making simulation module constructs digital twin models for candidate regions whose assessment scores meet preset thresholds and introduces a supply chain resilience assessment sub-module to achieve visualized simulations of capacity ramp-up curves, supply chain disruption risks, and order delivery timeliness under different relocation schemes, providing enterprises with quantifiable decision verification methods. The feedback optimization module continuously collects actual operational data after relocation, compares and analyzes it with the simulation results, and automatically adjusts model parameters to form a closed-loop self-optimization mechanism. This comprehensively solves the technical problems of single assessment dimensions, static weight settings, and lack of simulation verification capabilities in existing technologies, significantly improving the accuracy, adaptability, and decision reliability of industrial relocation assessment.
[0016] The further effects of the aforementioned non-conventional preferred method will be explained below in conjunction with specific embodiments. Attached Figure Description
[0017] To more clearly illustrate the embodiments of the present invention or the existing technical solutions, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A schematic diagram illustrating the composition of a digital assessment and decision-making model for the transfer of the textile and apparel industry, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the data access module in a digital assessment and decision-making model for the transfer of the textile and apparel industry, provided in an embodiment of the present invention. Figure 3 A schematic diagram illustrating the composition of a decision deduction module in a digital assessment and decision-making model for the transfer of the textile and apparel industry, provided as an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.
[0019] The attached diagram is labeled as follows: 100-Data access module, 110-Edge computing gateway; 200-Indicator system construction module; 300-Dynamic evaluation module; 400-Decision inference module, 410-Digital twin model construction unit, 420-Supply chain resilience assessment sub-module; 500-Feedback optimization module; 600-Electronic device, 601-Processor, 602-Memory, 603-Computer program. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0021] like Figure 1 As shown in the figure, this embodiment provides a digital assessment and decision-making model for the transfer of the textile and apparel industry, including: a data access module 100, an indicator system construction module 200, a dynamic assessment module 300, a decision deduction module 400, and a feedback optimization module 500.
[0022] The data access module 100 is used to acquire multi-source data involved in the transfer of the textile and apparel industry.
[0023] like Figure 2 As shown, in one specific embodiment, the data access module 100 includes an edge computing gateway 110 deployed at the textile enterprise. The edge computing gateway 110 is an ARM-based embedded device deployed at the textile enterprise's production site. It connects to the manufacturing execution system, IoT devices, and supply chain management system via OPC UA protocol, Modbus TCP protocol, and API interface, respectively. The edge computing gateway 110 preprocesses (including data cleaning and outlier removal) and extracts features (including extracting spectral features from vibration data using fast Fourier transform, extracting time-domain statistical features from temperature and humidity data, and extracting efficiency change features from production data using differential processing) from real-time vibration data, temperature and humidity data, and production output data of the textile equipment. The compressed feature vector is then uploaded to the dynamic evaluation module 300 via a 4G or 5G network.
[0024] In one specific embodiment, the multi-source data acquired by the data access module 100 includes at least: manufacturing execution system production data (equipment uptime, operating efficiency, output data), enterprise resource planning system resource data (raw material inventory, work-in-process quantity, order status), textile equipment operating condition data collected by IoT devices (vibration, temperature, speed, energy consumption), and logistics data from the supply chain management system (goods in transit, transportation timeliness, customs clearance status).
[0025] The indicator system construction module 200 is connected to the data access module 100 and configured to construct an evaluation indicator system containing multiple evaluation dimensions, with the evaluation dimensions corresponding to the business areas covered by the multi-source data.
[0026] In one specific embodiment, the evaluation indicator system constructed by the indicator system construction module 200 includes at least: capacity migration adaptability indicator, supply chain restructuring cost indicator, and order response timeliness indicator. The business areas corresponding to the capacity migration adaptability indicator are textile equipment and human resources; the business areas corresponding to the supply chain restructuring cost indicator are raw materials and supporting collaboration; and the business areas corresponding to the order response timeliness indicator are production cycle and logistics clearance.
[0027] The dynamic evaluation module 300 is connected to the data access module 100 and the indicator system construction module 200 respectively. It is configured to dynamically assign weights to each indicator in the evaluation indicator system based on multi-source data and calculate the evaluation score of each candidate region.
[0028] In one specific embodiment, the capacity migration adaptability index includes the textile equipment compatibility index. The dynamic evaluation module 300, based on textile equipment operating condition data and equipment model data from multiple sources, uses a graph neural network algorithm to calculate the process connection similarity and control system protocol matching degree between the equipment to be transferred and the existing equipment in the target area, in order to determine the textile equipment compatibility index.
[0029] Let the set of equipment to be transferred be... The existing equipment set in the target area is Build a device compatibility diagram. , where the node set edge set This represents the process connections between equipment. A graph convolutional network is used to learn the embedding representation of nodes. Then the equipment and The compatibility index is:
[0030] , in, This represents vector concatenation; MLP stands for Multilayer Perceptron. It is the Sigmoid activation function. The control system protocol matching degree (1 indicates a perfect match, 0 indicates a mismatch). The equipment to be transferred. The overall compatibility index is the average compatibility with all devices in the target area:
[0031] , In one specific embodiment, the cost indicators for supply chain restructuring include the completeness of outsourcing. The dynamic evaluation module 300, based on enterprise distribution data in the target region from multi-source data, determines the completeness of outsourcing by analyzing the enterprise distribution density and network connectivity of each link in the industrial chain. The industrial chain links include at least spinning, weaving, dyeing and finishing, printing, embroidery, and finishing processes.
[0032] Let the set of links in the industrial chain be For the process The number of enterprises in this region is The enterprise distribution density is: , Where A represents the area of the target region. Construct a supply chain supporting network topology diagram. Nodes represent enterprises, and edges represent collaborative relationships between enterprises. (Link / Segment) The network connectivity is:
[0033] , in, For the link The number of nodes in the connected component. The completeness of the external support is the weighted sum of the density and connectivity of each component:
[0034] , in, As the weight of the process, The normalized firm density , This is the balance coefficient.
[0035] In one specific embodiment, the order response timeliness index includes at least the potential for reducing the cycle from order acceptance to first shipment, the flexible production adjustment capability for fast-response orders, and the efficiency of cross-border logistics and customs clearance. The dynamic evaluation module 300 calculates various sub-indicators of the order response timeliness index based on historical production data, order fulfillment data, and logistics timeliness data from multiple sources.
[0036] Cycle compression potential Calculated using a time series prediction model: , in, This is the current average delivery cycle. This represents the average delivery cycle predicted based on post-transfer resource allocation. Flexible production capacity. The order response elasticity coefficient is used to characterize:
[0037] , in, For the i-th fast-response order batch, For average batch size, For order response time. Logistics and customs clearance efficiency. Average customs clearance time calculated based on historical customs clearance data The reciprocal:
[0038] .
[0039] In one specific embodiment, the dynamic evaluation module 300 employs an improved entropy-weighted analytic hierarchy process to dynamically assign weights to the evaluation index system. For n candidate regions and m indicators, an original data matrix is constructed. The data is standardized using the extreme value method:
[0040] (Positive indicator) (Negative indicator) Calculating objective weights using the entropy weight method : , , .
[0041] Determining Subjective Weights Using the Analytic Hierarchy Process Construct a combined weight optimization model to minimize the combined weights. The target is the deviation from the subjective and objective weights.
[0042] , , in, These are the preference coefficients. Solving for the optimal combination weights yields the solution. .
[0043] The comprehensive potential index of candidate region i is: The comprehensive potential index is a specific form of evaluation score.
[0044] like Figure 3As shown, the decision-making simulation module 400 is connected to the dynamic evaluation module 300, and includes a digital twin model construction unit 410 and a supply chain resilience evaluation submodule 420. The digital twin model construction unit 410 is used to construct a specific implementation of the digital simulation model—the digital twin model. The supply chain resilience evaluation submodule 420 is used to evaluate supply chain resilience. The digital twin model construction unit 410 is configured to construct a digital simulation model for candidate regions whose evaluation scores meet a preset threshold. This model maps the textile equipment layout, logistics network nodes, and supporting enterprise distribution of the target region based on the geographical information, equipment layout data, and supporting enterprise distribution data of the target region from multi-source data. The supply chain resilience evaluation submodule 420 is configured to simulate the supply chain recovery time and capacity loss range when a sudden event occurs in the target region based on historical supply chain disruption data from multi-source data, and uses the supply chain resilience score as a correction factor for the decision output.
[0045] In one specific embodiment, the digital twin model building unit 410 employs 3D modeling technology to construct a 3D scene including topography, road network, and factory layout based on the digital elevation model of the target area, satellite imagery, and industrial park planning map. On this basis, the BIM model of the textile equipment is imported, placed according to the actual layout, and associated with the equipment's operating parameters and status data. Simultaneously, logistics data from the supply chain management system is coupled with the road network model to construct a logistics transportation simulation model.
[0046] Supply chain resilience assessment submodule 420 uses the Monte Carlo simulation method to define a set of emergency event types. And its probability distribution P(e). For each transfer scheme s, N simulations are performed, and the supply chain recovery time is recorded in each simulation. and capacity loss Supply chain resilience score Defined as:
[0047] , in, To minimize the possible loss of production capacity, This represents the time decay coefficient. The decision-making simulation module 400 simulates the capacity ramp-up curves, supply chain disruption risk probabilities, and order delivery delay days under different transfer schemes based on the digital twin model. It also uses the supply chain resilience score as a correction factor to generate optimized transfer decision information, which is then fed back to the dynamic evaluation module 300 to adjust the evaluation parameters.
[0048] The feedback optimization module 500 is connected to the data access module 100 and the decision inference module 400 respectively. It is configured to continuously collect the actual operation data of the transferred area after the enterprise completes part of the capacity transfer, compare and analyze the data with the inference results of the decision inference module 400, calculate the prediction deviation, and adjust the evaluation parameters of the dynamic evaluation module 300 and the inference parameters of the decision inference module 400.
[0049] In one specific embodiment, the feedback optimization module 500 employs a Bayesian optimization method, using the deviation between the actual operating data and the projection results as the optimization objective. Let the projection results be... Actual operating data is Define the loss function:
[0050] , in, These are the model parameters. A surrogate model of the loss function is constructed using Gaussian process regression. A sampling function (such as the desired improvement in EI) is used to select the next set of parameters for evaluation. The parameters are iteratively updated until convergence, thus achieving closed-loop self-optimization of the model.
[0051] In one specific embodiment, the dynamic evaluation module 300 is further configured to preset corresponding indicator weight thresholds and evaluation benchmarks based on different sub-sectors of the textile and apparel industry. These sub-sectors include at least cotton textiles, wool textiles, apparel, and industrial textiles. When a user selects the apparel sector, the system automatically increases the weight threshold of the order response timeliness indicator and adjusts the evaluation benchmark of the capacity migration adaptability indicator to focus on sewing equipment compatibility; when a user selects the cotton textile sector, the system increases the weight threshold of the capacity migration adaptability indicator and adjusts the evaluation benchmark of the supply chain restructuring cost indicator to focus on cotton raw material synergy.
[0052] The technical solution of the present invention will be described in detail below with reference to specific application scenarios.
[0053] Scenario 1: Data Access and Edge Computing In this embodiment, the data access module 100 is deployed at the production site of a garment processing enterprise. The enterprise has 200 weaving machines and 500 sewing machines, with brands including STOLL (Germany), Shima Seiki (Japan), and various domestic models. The edge computing gateway 110 is an ARM-based embedded device deployed in the workshop.
[0054] The edge computing gateway 110 accesses the MES system of the weaving equipment via the OPC UA protocol to obtain equipment uptime, operating efficiency, and production data, with a data sampling frequency of 1 time / minute; it collects vibration sensor and temperature and humidity sensor data of the sewing equipment via the Modbus TCP protocol, with a vibration sensor sampling frequency of 100Hz and a temperature and humidity sensor sampling frequency of 1Hz; and it calls the raw material inventory data (updated daily) of the enterprise ERP system and the in-transit logistics data (updated in real time) of the supply chain management system through the API interface.
[0055] The edge computing gateway 110 performs a fast Fourier transform on the collected vibration data, extracting the energy of the first 10 spectral orders as feature vectors; it performs a sliding window average (window length 1 minute) on the temperature and humidity data, extracting the time-domain mean and variance as statistical features; and it performs differential processing on the production data to extract the production efficiency change rate feature. After feature extraction, the original data stream (vibration data approximately 200KB / s, temperature and humidity data approximately 2KB / s) is compressed into a feature vector of 128 bytes / s, which is then uploaded to the cloud dynamic evaluation module 300 via a 4G network.
[0056] Scenario 2: Construction and Dynamic Weighting of Indicator System This embodiment focuses on a scenario where a textile group plans to transfer production capacity to Southeast Asia, constructing an evaluation index system and dynamically assigning weights.
[0057] The indicator system construction module 200 constructs a three-dimensional evaluation indicator system based on the business areas covered by multi-source data, including capacity migration adaptability indicators, supply chain restructuring cost indicators, and order response timeliness indicators.
[0058] The dynamic evaluation module 300 collects the latest multi-source data from Vietnam, Cambodia, and Bangladesh. Taking the textile equipment compatibility index as an example: the equipment to be transferred is the Stoll CMS530 series computerized flat knitting machine, and the existing equipment in the target area, Vietnam, includes models such as the Stoll CMS502 and Shima Seiki SIR123. The dynamic evaluation module 300 constructs an equipment compatibility map using a graph neural network algorithm, calculates a process connection similarity of 0.85, a control system protocol matching degree of 0.75, and substitutes these values into the formula to calculate a textile equipment compatibility index of 0.82.
[0059] After collecting data on various indicators from the three countries, an original data matrix was constructed. The entropy weight method was used to calculate the entropy value and objective weight of each indicator. The indicator of "accessibility of skilled workers" showed significant differences across the three countries, resulting in a lower entropy value (0.52) and a higher objective weight (0.15). Conversely, the indicator of "transportation cost of semi-finished products" showed less difference across the three countries, resulting in a higher entropy value (0.91) and a lower objective weight (0.06). Simultaneously, the analytic hierarchy process (AHP) was employed, with industry transfer experts conducting pairwise comparisons of each indicator to construct a judgment matrix and calculate the subjective weight vector.
[0060] The optimal weight combination was obtained by solving the combined weight optimization model. Based on the optimal weight combination and the standardized values of each indicator, the evaluation scores of the three countries were calculated: Vietnam 0.86, Cambodia 0.72, and Bangladesh 0.68. Vietnam will proceed to the next round of simulation.
[0061] Scenario 3: Digital Twin Inference and Decision Optimization For Vietnam, which scored the highest in the evaluation, the decision-making simulation module 400 constructed a digital twin model for simulation.
[0062] Digital twin model construction unit 410 constructs a digital twin model based on geographic information data of Dong Nai Province, Vietnam, equipment layout data of 50 textile enterprises in an industrial park, and distribution data of supporting enterprises such as dyeing and embroidery in the surrounding area. The model includes the layout location of 200 weaving machines of the enterprise to be relocated within the park, the logistics route from the park to Ho Chi Minh Port, and the distribution nodes of 8 dyeing and embroidery enterprises and 12 embroidery enterprises in the surrounding area.
[0063] The decision-making simulation module 400 simulates two relocation plans in a digital twin model: Plan A involves relocating all 500 sewing machines at once, while Plan B involves relocating them in two phases (300 machines in the first phase, and the remaining 200 machines plus weaving equipment in the second phase). The supply chain resilience assessment submodule 420, based on historical supply chain disruption data, sets up three emergency scenarios: port strikes, pandemic lockdowns, and power rationing, and conducts 500 Monte Carlo simulations.
[0064] The simulation results show that Option A has a steep but volatile capacity ramp-up curve, while Option B has a gentler and more stable capacity ramp-up. In the event of a simulated port strike, Option A's raw material inventory can support the workforce for 15 days, while Option B can support it for 28 days. Under pandemic lockdown scenarios, Option A's recovery time is approximately 45 days with a capacity loss of approximately 3 million pieces, while Option B's recovery time is approximately 30 days with a capacity loss of approximately 1.8 million pieces. The decision-making simulation module 400 generates an optimized transfer path decision based on the simulation results: Option B is recommended, initially transferring 300 sewing machines while retaining weaving and key fabric production at the original factory, and establishing strategic partnerships with surrounding dyeing and printing enterprises. This decision information is fed back to the dynamic evaluation module 300, increasing the supply chain resilience weight from 0.12 to 0.18.
[0065] Scenario 4: Feedback Optimization and Model Iteration After the company completed the first phase of relocation according to the decision-making recommendations, the feedback optimization module 500 continuously collected actual operational data from the Vietnam factory. In the third month of production, the actual production capacity reached 92% of the projected forecast, but the actual number of supply chain disruptions was two more than predicted (mainly due to delivery delays from local dyeing and printing companies).
[0066] The feedback optimization module 500 compared and analyzed the deviations between actual data and projected results, finding that the prediction errors mainly stemmed from insufficient assessment of dependence on local raw material suppliers. A Bayesian optimization method was used to adjust the delivery cycle parameters of local suppliers in the digital twin model, and the weight of "outsourcing completeness" in the supply chain restructuring cost index of the dynamic evaluation module 300 was adjusted from 0.3 to 0.35. The adjusted model resulted in more accurate evaluations in subsequent transfer decisions.
[0067] Scenario 5: Differentiated Configuration in Niche Markets A textile group has a cotton textile division and an apparel division, both of which plan to relocate their operations. The dynamic evaluation module 300 automatically loads different indicator weight thresholds and evaluation benchmarks based on the differences in the sub-sectors.
[0068] In the cotton textile sector: the weighting of the capacity migration adaptability index is 0.45 (focusing on the compatibility of spinning equipment and large-scale production capacity), the weighting of the supply chain restructuring cost index is 0.35 (focusing on cotton raw material coordination and transportation costs), and the weighting of the order response timeliness index is 0.20 (cotton textile orders have relatively long cycles and moderate timeliness sensitivity). In the apparel and garment industry: the production capacity migration adaptability index has a weight of 0.30 (focusing on the compatibility of sewing equipment and the availability of skilled sewing workers), the supply chain restructuring cost index has a weight of 0.30 (focusing on the matching of fabrics and accessories and rapid response collaboration), and the order response timeliness index has a weight of 0.40 (apparel and garment orders are mostly rapid response orders and are highly time-sensitive). When a user selects the apparel sector, the system automatically loads the corresponding weight thresholds to make the evaluation results more in line with the characteristics of the sub-sector.
[0069] Figure 4 This is a schematic diagram of the electronic device 600 provided in an embodiment of this application. Figure 4 As shown, the electronic device 600 of this embodiment includes a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, it implements the steps of the above-described model executing the corresponding method.
[0070] Electronic device 600 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 600 may include, but is not limited to, a processor 601 and a memory 602. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 600 and does not constitute a limitation on electronic device 600. It may include more or fewer components than shown, or different components.
[0071] The processor 601 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
[0072] The memory 602 can be an internal storage unit of the electronic device 600, such as a hard disk or RAM of the electronic device 600. The memory 602 can also be an external storage device of the electronic device 600, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the electronic device 600. The memory 602 can also include both internal and external storage units of the electronic device 600. The memory 602 is used to store computer programs and other programs and data required by the electronic device.
[0073] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0074] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium (e.g., a computer-readable storage medium). Based on this understanding, all or part of the processes executed by the methods of each module / unit in this application can also be implemented by a computer program instructing related hardware. The computer program can be stored in a readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
[0075] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A digital assessment and decision-making model for the transfer of the textile and apparel industry, characterized in that, include: The data access module is configured to acquire multi-source data related to the transfer of the textile and apparel industry. The indicator system construction module is connected to the data access module and is configured to construct an evaluation indicator system containing multiple evaluation dimensions, wherein the evaluation dimensions correspond to the business areas covered by the multi-source data. The dynamic evaluation module is connected to the data access module and the indicator system construction module, respectively, and is configured to dynamically assign weights to each indicator in the evaluation indicator system based on the multi-source data, and calculate the evaluation score of each candidate region. The decision-making simulation module is connected to the dynamic evaluation module and is configured to build a digital simulation model for candidate areas whose evaluation scores meet a preset threshold, simulate the operational effects under different transfer schemes, and generate optimized transfer decision information based on the simulation results, which is then fed back to the dynamic evaluation module to adjust the evaluation parameters.
2. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The data access module includes an edge computing gateway deployed at the textile enterprise, and the acquired multi-source data includes at least: Manufacturing execution system production data, enterprise resource planning system resource data, textile equipment operating condition data collected by IoT devices, and logistics data in supply chain management system; The edge computing gateway is configured to preprocess and extract features from real-time vibration data, temperature and humidity data, and production data of textile equipment, and upload the compressed feature vector to the dynamic evaluation module.
3. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The evaluation indicator system constructed by the indicator system construction module includes at least the following: Capacity migration adaptability indicators, supply chain restructuring cost indicators, and order response time indicators; The business areas corresponding to the capacity migration adaptability index are textile equipment and human resources; the business areas corresponding to the supply chain restructuring cost index are raw materials and supporting collaboration; and the business areas corresponding to the order response timeliness index are production cycle and logistics clearance.
4. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 3, characterized in that, The capacity migration adaptability index includes the textile equipment compatibility index. The dynamic evaluation module uses graph neural network algorithms to calculate the process connection similarity and control system protocol matching degree between the equipment to be transferred and the existing equipment in the target area, based on the textile equipment operating condition data and equipment model data in the multi-source data, in order to determine the textile equipment compatibility index.
5. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 3, characterized in that, The cost indicators for supply chain restructuring include the completeness of outsourcing support; The dynamic evaluation module, based on the target region enterprise distribution data in the multi-source data, determines the completeness of the external cooperation by analyzing the enterprise distribution density and network connectivity in the industrial chain links. The aforementioned industrial chain links include at least spinning, weaving, dyeing and finishing, printing, embroidery and finishing.
6. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 3, characterized in that, The order response timeliness indicators include at least the potential for reducing the cycle from order acceptance to first shipment, the flexible production adjustment capability for fast-response orders, and the efficiency of cross-border logistics and customs clearance. The dynamic evaluation module calculates various sub-indicators of the order response timeliness index based on the production history data, order fulfillment data, and logistics timeliness data from the multi-source data.
7. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The dynamic evaluation module uses an improved entropy-weighted analytic hierarchy process to dynamically assign weights to the evaluation index system. The improved entropy-weighted analytic hierarchy process includes: using the analytic hierarchy process to determine subjective weights, using the entropy weight method to calculate objective weights based on the degree of variation of each indicator in the multi-source data, and solving for the optimal combined weights by constructing a combined weight optimization model.
8. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The digital inference model constructed by the decision inference module is a digital twin model; The digital twin model maps the textile equipment layout, logistics network nodes, and supporting enterprise distribution of the target area based on the target area's geographic information, equipment layout data, and supporting enterprise distribution data from the multi-source data. The decision-making simulation module is also configured to introduce a supply chain resilience assessment submodule into the digital twin model, which simulates the supply chain recovery time and capacity loss range when a sudden event occurs in the target area based on historical supply chain disruption data from the multi-source data, and uses the supply chain resilience score as a correction factor for the decision output.
9. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The model also includes a feedback optimization module, which is connected to the data access module and the decision inference module, respectively. It is configured to continuously collect actual operating data of the transferred area after the enterprise completes part of the capacity transfer, compare and analyze the data with the inference results of the decision inference module, calculate the prediction deviation, and adjust the evaluation parameters of the dynamic evaluation module and the inference parameters of the decision inference module.
10. The digital assessment and decision-making model for the transfer of the textile and apparel industry according to claim 1, characterized in that, The dynamic evaluation module is also configured to preset corresponding indicator weight thresholds and evaluation benchmarks according to different sub-sectors of the textile and apparel industry; the sub-sectors include at least cotton textiles, wool textiles, garments and industrial textiles.
