A method and system for realizing intelligent management of knitted sock production

Abstract

The present invention relates to the technical field of intelligent supplier management, and specifically to a method and system for realizing intelligent management of knitting sock production. First of all, by conducting color evaluation, appearance evaluation and physical property evaluation on sample socks, the present invention can comprehensively detect the quality of raw materials, quantify the color difference defect index, appearance defect index and performance deviation index, and ensure that the raw materials provided by suppliers meet the production requirements; secondly, by calculating the quality index and comparing it with the set threshold, the present invention can objectively screen suppliers meeting the quality standards, improve the supplier screening efficiency and ensure the stability and consistency of the purchased raw materials; finally, by combining the calculation of weighted cost performance, the present invention re-evaluates the default suppliers, takes into account quality and delivery cycle, dynamically optimizes the supplier selection, and ensures that suppliers with high cost performance and reliable performance are always selected in the supply chain, thereby improving production efficiency, reducing risks and enhancing market competitiveness.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent supplier management, and in particular to a method and system for realizing intelligent management of knitted socks production. Background Art

[0002] Intelligent management has gradually become an important means to improve production efficiency, reduce costs and improve product quality in all walks of life. In the knitted socks production industry, intelligent management has gradually achieved efficient and low-cost production goals by integrating advanced equipment, automated production processes and big data analysis. However, the implementation of intelligent management is not limited to the production process, supply chain management is also crucial. The selection and management of suppliers directly affects the quality of raw materials, delivery cycle and production costs. However, supplier management is often easily overlooked in many cases.

[0003] At present, there are still many drawbacks in the supplier management of many enterprises. Enterprises usually manage suppliers too statically. Once a supplier is selected, there is often a lack of dynamic monitoring and continuous optimization mechanisms. This may cause quality fluctuations or delivery delays in the long-term cooperation process, affecting the stability of production and product quality. Therefore, optimizing supplier management and establishing a scientific and effective supplier evaluation system have become important tasks to improve the level of intelligent production management of enterprises.

[0004] Therefore, a method and system for realizing intelligent management of knitted socks production are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for realizing intelligent management of knitted socks production. By evaluating the color, appearance and physical properties of sample socks, the quality of raw materials can be fully monitored to ensure that the finished socks meet the requirements in various quality standards; by calculating the quality index, the quality performance of different suppliers can be quantified and comprehensively evaluated, providing a scientific basis for supplier screening and procurement decisions; by weighted cost-effectiveness calculation, the quality and delivery punctuality of suppliers can be comprehensively considered, and the supplier priority can be dynamically adjusted, thereby optimizing supply chain management.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for realizing intelligent management of knitted socks production, comprising:

[0008] Establishing a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and actual delivery cycle;

[0009] Selecting raw material suppliers to obtain pre-selected suppliers, using the raw materials provided by the pre-selected suppliers to prepare sample socks, and performing color evaluation, appearance evaluation, and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index, and a performance deviation index;

[0010] Calculating a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, uploading the quality index and the supplier name corresponding to the index to the supplier database;

[0011] Calculate the cost performance ratio according to the quality index and the raw material purchase price, and upload the cost performance ratio to the supplier database; prioritize the pre-selected suppliers according to the cost performance ratio, and set the pre-selected supplier with the highest priority as the default supplier;

[0012] Predicting raw material consumption based on historical production data, and generating orders based on the obtained raw material consumption prediction results; dispatching the orders to the default suppliers, recording the actual delivery cycle of the default suppliers and uploading it to the supplier database;

[0013] The finished socks are randomly inspected, the quality index is recalculated, the weighted cost performance of the default supplier is calculated based on the quality index and the actual delivery cycle, the pre-selected suppliers are re-prioritized, and the supplier database is updated.

[0014] Further, obtaining the color difference defect index includes:

[0015] Identify color standards, calibrate measuring equipment, and prepare sample socks;

[0016] Use a spectrophotometer to measure the color values ​​of all sample socks and calculate the average color difference;

[0017] Calculating the color difference defect index according to the average color difference;

[0018] The calculation formula of the color difference defect index is:

[0019] ;

[0020] in represents the color difference defect index, represents the average color difference, Indicates the maximum permissible color difference.

[0021] Further, obtaining the appearance defect index includes:

[0022] The surface flatness and fabric density of all sample socks were tested using a deep learning model, and the surface flatness defect rate and fabric density defect rate were counted.

[0023] Calculating the appearance defect index according to the surface flatness defect rate and the fabric density defect rate;

[0024] The calculation formula of the appearance defect index is:

[0025] ;

[0026] in, represents the appearance defect index, represents the surface flatness defect rate, It represents the fabric density defect rate.

[0027] Further, obtaining the performance deviation index includes:

[0028] All sample socks were divided into four parts, and the maximum stretch test, elastic recovery test, moisture absorption test and air permeability test were carried out respectively, and the average stretch deviation, average recovery deviation, average moisture absorption deviation and average air permeability deviation were calculated;

[0029] calculating the performance deviation index based on the average stretch deviation, the average recovery deviation, the average hygroscopicity deviation, and the average air permeability deviation;

[0030] The calculation formula of the performance deviation index is:

[0031] ;

[0032] in, represents the performance deviation index, represents the average tensile deviation, represents the mean recovery deviation, represents the mean hygroscopicity deviation, The average air permeability deviation is represented.

[0033] Furthermore, the formula for calculating the quality index is:

[0034] ;

[0035] in, represents the quality index, represents the color difference defect index, represents the appearance defect index, represents the performance deviation index, represents the color difference weight, represents the appearance weight, Indicates performance weight.

[0036] Furthermore, raw material consumption forecasting based on historical production data includes:

[0037] Collect historical production data and clean and process them to obtain pre-processed data;

[0038] Performing key feature variable selection and standardization processing on the preprocessed data to obtain standardized data;

[0039] The standardized data is divided into a training set and a validation set, the training set is fitted by a regression model, and the regression model is validated by using the validation set to obtain a pre-trained model;

[0040] The pre-trained model is used to predict the raw material consumption to obtain a raw material consumption prediction result.

[0041] Furthermore, the calculation formula for the weighted cost performance is:

[0042] ;

[0043] in, represents the weighted cost performance, Indicates cost-effectiveness, Represents the difference between two consecutive quality indices, Indicates setting the delivery cycle, Indicates the actual lead time.

[0044] The present invention also proposes a system for realizing intelligent management of knitted socks production, comprising:

[0045] A database initialization module is used to establish a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and delivery cycle;

[0046] A sample evaluation module is used to select raw material suppliers to obtain pre-selected suppliers, use the raw materials provided by the pre-selected suppliers to prepare sample socks, and respectively perform color evaluation, appearance evaluation and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index and a performance deviation index;

[0047] a quality evaluation module, configured to calculate a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, upload the quality index and the supplier name corresponding to the index to the supplier database;

[0048] A cost performance evaluation module is used to calculate the cost performance according to the quality index and the raw material purchase price, and upload the cost performance to the supplier database; prioritize the pre-selected suppliers according to the cost performance, and set the pre-selected supplier with the highest priority as the default supplier;

[0049] An order generation module is used to predict raw material consumption based on historical production data, and generate orders based on the obtained raw material consumption prediction results; dispatch the orders to the default supplier, record the delivery cycle of the default supplier and upload it to the supplier database;

[0050] The database update module is used to perform random sampling on finished socks, calculate the quality index, weight the cost performance of the default supplier according to the quality index and the actual delivery cycle, re-prioritize the pre-selected suppliers, and update the supplier database.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. Carrying out color evaluation, appearance evaluation and physical property evaluation on sample socks can comprehensively and systematically test the quality of raw materials. By quantifying the color difference defect index, appearance defect index and performance deviation index, it is possible to accurately evaluate the color consistency, appearance quality and physical properties of the raw materials provided by each supplier, ensuring that the raw materials provided by the supplier meet the production requirements.

[0053] 2. By calculating the quality index and comparing it with the set threshold, suppliers that meet the quality standards can be objectively screened out. Uploading the quality index and suppliers to the supplier database provides companies with a transparent and quantitative supplier evaluation system. This method not only improves the efficiency of supplier screening, but also ensures that companies can choose suppliers with stable quality and excellent performance when purchasing raw materials, thereby reducing production risks and improving product consistency and market competitiveness.

[0054] 3. Re-evaluating default suppliers through weighted cost-effectiveness calculations and adjusting them based on quality index and actual delivery cycle can help dynamically optimize supplier selection. This method can comprehensively consider the supplier's quality performance and delivery punctuality, and promptly reflect changes in the supplier's performance in actual production. In this way, companies can re-rank pre-selected suppliers based on the latest supplier data, thereby ensuring that the supply chain always selects cost-effective, stable and reliable suppliers. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of a method for realizing intelligent management of knitted socks production provided by an embodiment of the present invention;

[0056] Figure 2 A raw material consumption prediction flow chart provided for an embodiment of the present invention;

[0057] Figure 3 A system structure diagram for realizing intelligent management of knitted socks production provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] The first embodiment provided by the present invention is as follows:

[0060] A method for realizing intelligent management of knitted socks production, such as Figure 1 As shown, including:

[0061] S100: Establishing a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and actual delivery cycle;

[0062] S200: selecting raw material suppliers to obtain pre-selected suppliers, using the raw materials provided by the pre-selected suppliers to prepare sample socks, and respectively performing color evaluation, appearance evaluation, and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index, and a performance deviation index;

[0063] Further, obtaining the color difference defect index includes:

[0064] Identify color standards, calibrate measuring equipment, and prepare sample socks;

[0065] Use a spectrophotometer to measure the color values ​​of all sample socks and calculate the average color difference;

[0066] Calculating the color difference defect index according to the average color difference;

[0067] The calculation formula of the color difference defect index is:

[0068] ;

[0069] in represents the color difference defect index, represents the average color difference, Indicates the maximum permissible color difference.

[0070] Specifically, the color value is passed Color space representation, where Indicates the brightness value (0 is black, 100 is white), Represents the red and green axis values ​​(positive values ​​are red, negative values ​​are green), Indicates the yellow-blue axis value (positive value is yellow, negative value is blue); the measuring instrument is a spectrophotometer, and a black reference plate is required for calibration before using the spectrophotometer; the color difference calculation formula is:

[0071] ;

[0072] in, , and are the brightness value, red-green axis value, and yellow-blue axis value of the sample respectively. , and They are the brightness value, red-green axis value and yellow-blue axis value of the standard Pantone color card. The color difference data of some sample socks are shown in Table 1. The average color difference can be obtained according to the average color difference calculation formula. The average color difference calculation formula is:

[0073] ;

[0074] in, represents the number of samples, represents the color difference of sample i; finally, the color difference defect index is calculated according to the average color difference, wherein the maximum allowable color difference can be set manually, and the value of the maximum allowable color difference in this embodiment is 2.

[0075] Table 1 Color difference data of some sample socks

[0076]

[0077] By providing suppliers with scientific quantitative indicators for their raw materials through the color difference defect index, companies can promptly detect possible color deviations in raw materials, avoid quality problems caused by color inconsistencies, and improve product qualification rates and market competitiveness.

[0078] Further, obtaining the appearance defect index includes:

[0079] The surface flatness and fabric density of all sample socks were tested using a deep learning model, and the surface flatness defect rate and fabric density defect rate were counted.

[0080] Calculating the appearance defect index according to the surface flatness defect rate and the fabric density defect rate;

[0081] The calculation formula of the appearance defect index is:

[0082] ;

[0083] in, represents the appearance defect index, represents the surface flatness defect rate, It represents the fabric density defect rate.

[0084] Specifically, the deep learning model is not limited. The deep learning model used in this embodiment is YOLO v5. The surface flatness detection includes loop protrusion, hairiness, knots and loop undocking. The fabric density detection includes uneven thickness and needle path deviation. The statistical data of the surface flatness defect rate and fabric density defect rate of the sample socks are shown in Table 2. Finally, the appearance defect index is calculated by the surface flatness defect rate and the fabric density defect rate.

[0085] Table 2 Statistics of surface flatness defect rate and fabric density defect rate of sample socks

[0086]

[0087] Testing the surface flatness and fabric density of sample socks can accurately identify quality problems in the production process of raw materials provided by suppliers. The surface flatness defect rate and fabric density defect rate provide a quantitative basis for evaluating the quality of raw materials, helping companies to promptly discover and resolve potential quality risks of raw materials.

[0088] Further, obtaining the performance deviation index includes:

[0089] All sample socks were divided into four parts, and the maximum stretch test, elastic recovery test, moisture absorption test and air permeability test were carried out respectively, and the average stretch deviation, average recovery deviation, average moisture absorption deviation and average air permeability deviation were calculated;

[0090] calculating the performance deviation index based on the average stretch deviation, the average recovery deviation, the average hygroscopicity deviation, and the average air permeability deviation;

[0091] The calculation formula of the performance deviation index is:

[0092] ;

[0093] in, represents the performance deviation index, represents the average tensile deviation, represents the mean recovery deviation, represents the mean hygroscopicity deviation, The average air permeability deviation is represented.

[0094] Specifically, the average tensile deviation is calculated as:

[0095] ;

[0096] in, Indicates the number of sample socks participating in the maximum tensile test, Indicates The maximum tensile force that a sample sock can withstand, Indicates the standard tensile force provided by the industry standard; the elastic recovery test stretches the sample socks to a specified length, holds it for a period of time, then releases it, and measures the length after recovery. The calculation formula for the average recovery deviation is:

[0097] ;

[0098] in, Indicates the number of sample socks participating in the elastic recovery test, Indicates Elastic recovery rate of sample socks, Indicates the standard elastic recovery rate provided by the industry standard, Indicates the length of the sample socks after recovery, Indicates the length of the sample socks after stretching, Indicates the initial length of the sample socks; the hygroscopicity test is to immerse the sock sample in water for a certain period of time and then take it out, and measure the difference between the mass after absorbing water and the dry mass. The calculation formula for the average hygroscopicity deviation is:

[0099] ;

[0100] in, Indicates the number of sample socks participating in the moisture absorption test, Indicates The moisture absorption ratio of the sample socks is Indicates the standard moisture absorption ratio provided by the industry standard. Indicates the weight of the sample socks after absorbing water, Indicates the initial weight of the sample socks; the air permeability test is to measure the pressure difference required for air to pass through a certain area of ​​the socks on the air permeability tester. The calculation formula for the average air permeability deviation is:

[0101] ;

[0102] Where, s represents the number of sample socks participating in the air permeability test, Indicates Sample socks tested pressure difference, Indicates the standard pressure difference provided by the industry standard; the test deviation data of the sample socks are shown in Table 3.

[0103] Table 3 Table of deviation data of various tests of sample socks

[0104]

[0105] By subjecting sample socks to maximum stretch test, elastic recovery test, moisture absorption test and air permeability test, we can comprehensively evaluate the performance of the raw materials provided by the supplier. By calculating the average deviation value of each test, we can quantify the degree of performance fluctuation of the raw materials and comprehensively derive the performance deviation index. The performance deviation index can provide a scientific basis for supplier quality assessment and ensure that the selected supplier can provide stable and high-quality raw materials.

[0106] S300: Calculating a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, uploading the quality index and the supplier name corresponding to the index to the supplier database;

[0107] Furthermore, the formula for calculating the quality index is:

[0108] ;

[0109] in, represents the quality index, represents the color difference defect index, represents the appearance defect index, represents the performance deviation index, represents the color difference weight, represents the appearance weight, Indicates performance weight.

[0110] Specifically, in this embodiment, the first threshold is set to 85%.

[0111] By combining the color difference defect index, appearance defect index and performance deviation index to calculate the comprehensive quality index, the quality of raw materials provided by suppliers can be comprehensively evaluated. This comprehensive evaluation method not only covers multiple dimensions such as color consistency, appearance quality and physical properties of raw materials, but also reflects the quality fluctuation of each dimension through quantitative indicators, ensuring that companies can accurately identify the quality performance of suppliers in all aspects, thereby achieving more scientific screening and evaluation of suppliers.

[0112] S400: Calculate the cost performance ratio according to the quality index and the raw material purchase price, and upload the cost performance ratio to the supplier database; prioritize the pre-selected suppliers according to the cost performance ratio, and set the pre-selected supplier with the highest priority as the default supplier;

[0113] Specifically, the calculation formula for cost performance is:

[0114] ;

[0115] in, Indicates cost-effectiveness, Indicates the purchase price of raw materials.

[0116] S500: forecasting raw material consumption according to historical production data, and generating an order according to the obtained raw material consumption forecast result; dispatching the order to the default supplier, recording the actual delivery cycle of the default supplier and uploading it to the supplier database;

[0117] Furthermore, raw material consumption is predicted based on historical production data. Figure 2 As shown, including:

[0118] Collect historical production data and clean and process them to obtain pre-processed data;

[0119] Performing key feature variable selection and standardization processing on the preprocessed data to obtain standardized data;

[0120] The standardized data is divided into a training set and a validation set, the training set is fitted by a regression model, and the regression model is validated by using the validation set to obtain a pre-trained model;

[0121] The pre-trained model is used to predict the raw material consumption to obtain a raw material consumption prediction result.

[0122] Specifically, historical production data are collected, which usually include production date, production volume, raw material usage, equipment operation data, environmental factors (such as temperature and humidity), production efficiency, etc. The historical production data are cleaned and processed, including missing value processing, outlier detection and duplicate data deletion, to obtain preprocessed data; key feature variables can be selected through correlation analysis (such as Pearson correlation coefficient) or other algorithms (such as LASSO regression, decision tree, etc.), and the standardization process is Min-Max standardization, that is, compressing the data to the range of [0,1]; the regression model can be a linear regression model, or it can be a ridge regression, LASSO regression or other regression, which is not limited here; use the pre-trained model to predict the raw material consumption of the current production data to obtain the raw material consumption prediction result. The current production data can be this week, this month or this quarter, which is not limited here.

[0123] By forecasting raw material consumption based on historical production data, companies can accurately predict future raw material demand, optimize procurement and production plans, and thus rationally arrange production resources and improve communication efficiency with suppliers.

[0124] S600: Perform random inspection on finished socks, recalculate the quality index, perform weighted cost performance calculation on the default supplier based on the quality index and the actual delivery cycle, re-prioritize the pre-selected suppliers, and update the supplier database.

[0125] Furthermore, the calculation formula for the weighted cost performance is:

[0126] ;

[0127] in, represents the weighted cost performance, Indicates cost-effectiveness, Represents the difference between two consecutive quality indices, Indicates setting the delivery cycle, Indicates the actual lead time.

[0128] Specifically, the finished socks are made of raw materials provided by the default supplier, and recalculating the quality index includes recalculating the color difference defect index, the appearance defect index and the performance deviation index. The calculation process is the same as step S200 and will not be repeated here.

[0129] By randomly sampling finished socks and recalculating the quality index, and calculating the weighted cost-effectiveness of default suppliers in combination with the actual delivery cycle, the actual performance of suppliers can be evaluated in real time. This dynamic evaluation method helps to promptly detect changes in supplier quality or delivery efficiency, and prioritize pre-selected suppliers based on the latest quality data, thereby ensuring that the information in the supplier database always reflects the actual capabilities and market performance of the current suppliers, thereby ensuring that the company can continuously and efficiently provide high-quality knitted products.

[0130] The second embodiment provided by the present invention is as follows:

[0131] In order to realize comprehensive intelligent management, a knitted socks manufacturer has upgraded its supplier management system. The upgraded supplier management system is a system for realizing intelligent management of knitted socks production proposed by the present invention. Figure 3 As shown, including:

[0132] A database initialization module is used to establish a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and delivery cycle;

[0133] A sample evaluation module is used to select raw material suppliers to obtain pre-selected suppliers, use the raw materials provided by the pre-selected suppliers to prepare sample socks, and respectively perform color evaluation, appearance evaluation and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index and a performance deviation index;

[0134] a quality evaluation module, configured to calculate a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, upload the quality index and the supplier name corresponding to the index to the supplier database;

[0135] A cost performance evaluation module is used to calculate the cost performance according to the quality index and the raw material purchase price, and upload the cost performance to the supplier database; prioritize the pre-selected suppliers according to the cost performance, and set the pre-selected supplier with the highest priority as the default supplier;

[0136] An order generation module is used to predict raw material consumption based on historical production data, and generate orders based on the obtained raw material consumption prediction results; dispatch the orders to the default supplier, record the delivery cycle of the default supplier and upload it to the supplier database;

[0137] The database update module is used to perform random sampling on finished socks, calculate the quality index, weight the cost performance of the default supplier according to the quality index and the actual delivery cycle, re-prioritize the pre-selected suppliers, and update the supplier database.

[0138] Further, obtaining the color difference defect index includes:

[0139] Identify color standards, calibrate measuring equipment, and prepare sample socks;

[0140] Use a spectrophotometer to measure the color values ​​of all sample socks and calculate the average color difference;

[0141] Calculating the color difference defect index according to the average color difference;

[0142] The calculation formula of the color difference defect index is:

[0143] ;

[0144] in represents the color difference defect index, represents the average color difference, Indicates the maximum permissible color difference.

[0145] Further, obtaining the appearance defect index includes:

[0146] The surface flatness and fabric density of all sample socks were tested using a deep learning model, and the surface flatness defect rate and fabric density defect rate were counted.

[0147] Calculating the appearance defect index according to the surface flatness defect rate and the fabric density defect rate;

[0148] The calculation formula of the appearance defect index is:

[0149] ;

[0150] in, represents the appearance defect index, represents the surface flatness defect rate, It represents the fabric density defect rate.

[0151] Further, obtaining the performance deviation index includes:

[0152] All sample socks were divided into four parts, and the maximum stretch test, elastic recovery test, moisture absorption test and air permeability test were carried out respectively, and the average stretch deviation, average recovery deviation, average moisture absorption deviation and average air permeability deviation were calculated;

[0153] calculating the performance deviation index based on the average stretch deviation, the average recovery deviation, the average hygroscopicity deviation, and the average air permeability deviation;

[0154] The calculation formula of the performance deviation index is:

[0155] ;

[0156] in, represents the performance deviation index, represents the average tensile deviation, represents the mean recovery deviation, represents the mean hygroscopicity deviation, The average air permeability deviation is represented.

[0157] Furthermore, the formula for calculating the quality index is:

[0158] ;

[0159] in, represents the quality index, represents the color difference defect index, represents the appearance defect index, represents the performance deviation index, represents the color difference weight, represents the appearance weight, Indicates performance weight.

[0160] Furthermore, raw material consumption forecasting based on historical production data includes:

[0161] Collect historical production data and clean and process them to obtain pre-processed data;

[0162] Performing key feature variable selection and standardization processing on the preprocessed data to obtain standardized data;

[0163] The standardized data is divided into a training set and a validation set, the training set is fitted by a regression model, and the regression model is validated by using the validation set to obtain a pre-trained model;

[0164] The pre-trained model is used to predict the raw material consumption to obtain a raw material consumption prediction result.

[0165] Furthermore, the calculation formula for the weighted cost performance is:

[0166] ;

[0167] in, represents the weighted cost performance, Cost-effectiveness, Represents the difference between two consecutive quality indices, Indicates setting the delivery cycle, Indicates the actual lead time.

[0168] By establishing a supplier database and conducting multi-dimensional evaluations of suppliers, including color, appearance, and physical properties, companies can comprehensively and objectively screen high-quality raw material suppliers. The quality index is calculated by combining the color difference defect index, appearance defect index, and performance deviation index, and the cost performance is calculated based on the raw material purchase price, which helps to determine the actual capabilities and cost performance of suppliers and optimize supplier priority ranking. By predicting raw material consumption through historical production data, companies can generate orders based on accurate demand forecasts and distribute orders to the default supplier with the highest priority to ensure timely delivery and stable supply. During the production process, by spot-checking finished socks and recalculating the quality index, the performance of suppliers can be dynamically evaluated, and the cost performance can be weighted according to the actual delivery cycle. The supplier priority can be adjusted in real time to ensure the flexibility and stability of the supply chain, improve production efficiency, reduce procurement costs, and ensure the stability of the final product quality.

[0169] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for realizing intelligent management of knitted socks production, characterized in that: include: Establishing a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and actual delivery cycle; Selecting raw material suppliers to obtain pre-selected suppliers, using the raw materials provided by the pre-selected suppliers to prepare sample socks, and performing color evaluation, appearance evaluation, and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index, and a performance deviation index; Calculating a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, uploading the quality index and the supplier name corresponding to the index to the supplier database; The formula for calculating the quality index is: ； in, represents the quality index, represents the color difference defect index, represents the appearance defect index, represents the performance deviation index, represents the color difference weight, represents the appearance weight, represents the performance weight; Calculate the cost performance ratio according to the quality index and the raw material purchase price, and upload the cost performance ratio to the supplier database; prioritize the pre-selected suppliers according to the cost performance ratio, and set the pre-selected supplier with the highest priority as the default supplier; The formula for calculating cost performance is: ； in, Indicates cost-effectiveness. Indicates the purchase price of raw materials; Predicting raw material consumption based on historical production data, and generating orders based on the obtained raw material consumption prediction results; dispatching the orders to the default suppliers, recording the actual delivery cycle of the default suppliers and uploading it to the supplier database; The finished socks are randomly inspected, the quality index is recalculated, the weighted cost performance of the default supplier is calculated based on the quality index and the actual delivery cycle, the pre-selected suppliers are re-prioritized, and the supplier database is updated.

2. A method for realizing intelligent management of knitted socks production according to claim 1, characterized in that: Obtaining the color difference defect index includes: Identify color standards, calibrate measuring equipment, and prepare sample socks; Use a spectrophotometer to measure the color values ​​of all sample socks and calculate the average color difference; Calculating the color difference defect index according to the average color difference; The calculation formula of the color difference defect index is: ； in represents the color difference defect index, represents the average color difference, Indicates the maximum permissible color difference.

3. A method for realizing intelligent management of knitted socks production according to claim 1, characterized in that: Obtaining the appearance defect index includes: The surface flatness and fabric density of all sample socks were tested using a deep learning model, and the surface flatness defect rate and fabric density defect rate were counted. Calculating the appearance defect index according to the surface flatness defect rate and the fabric density defect rate; The calculation formula of the appearance defect index is: ； in, represents the appearance defect index, represents the surface flatness defect rate, It represents the fabric density defect rate.

4. A method for realizing intelligent management of knitted socks production according to claim 1, characterized in that: Obtaining the performance deviation index includes: All sample socks were divided into four parts, and the maximum stretch test, elastic recovery test, moisture absorption test and air permeability test were carried out respectively, and the average stretch deviation, average recovery deviation, average moisture absorption deviation and average air permeability deviation were calculated; calculating the performance deviation index based on the average stretch deviation, the average recovery deviation, the average hygroscopicity deviation, and the average air permeability deviation; The calculation formula of the performance deviation index is: ； in, represents the performance deviation index, represents the average tensile deviation, represents the mean recovery deviation, represents the mean hygroscopicity deviation, represents the average air permeability deviation.

5. A method for realizing intelligent management of knitted socks production according to claim 1, characterized in that: Raw material consumption forecast based on historical production data includes: Collect historical production data and clean and process them to obtain pre-processed data; Performing key feature variable selection and standardization processing on the preprocessed data to obtain standardized data; The standardized data is divided into a training set and a validation set, the training set is fitted by a regression model, and the regression model is validated by using the validation set to obtain a pre-trained model; The pre-trained model is used to predict the raw material consumption to obtain a raw material consumption prediction result.

6. A method for realizing intelligent management of knitted socks production according to claim 1, characterized in that: The calculation formula for the weighted cost performance ratio is: ； in, represents the weighted cost performance, Indicates cost-effectiveness. Represents the difference between two consecutive quality indices, Indicates setting the delivery cycle, Indicates the actual lead time.

7. A system for realizing intelligent management of knitted socks production, characterized in that: include: A database initialization module is used to establish a supplier database; the index items of the supplier database include supplier name, quality index, raw material purchase price, cost performance and delivery cycle; A sample evaluation module is used to select raw material suppliers to obtain pre-selected suppliers, use the raw materials provided by the pre-selected suppliers to prepare sample socks, and respectively perform color evaluation, appearance evaluation and physical property evaluation on the sample socks to obtain a color difference defect index, an appearance defect index and a performance deviation index; a quality assessment module, configured to calculate a quality index according to the color difference defect index, the appearance defect index, and the performance deviation index, and if the quality index is greater than a first threshold, upload the quality index and the supplier name corresponding to the index to the supplier database; The formula for calculating the quality index is: ； in, represents the quality index, represents the color difference defect index, represents the appearance defect index, represents the performance deviation index, represents the color difference weight, represents the appearance weight, represents the performance weight; A cost performance evaluation module is used to calculate the cost performance according to the quality index and the raw material purchase price, and upload the cost performance to the supplier database; prioritize the pre-selected suppliers according to the cost performance, and set the pre-selected supplier with the highest priority as the default supplier; The formula for calculating cost performance is: ； in, Indicates cost-effectiveness. Indicates the purchase price of raw materials; An order generation module is used to predict raw material consumption based on historical production data, and generate orders based on the obtained raw material consumption prediction results; dispatch the orders to the default supplier, record the delivery cycle of the default supplier and upload it to the supplier database; The database update module is used to perform random sampling on finished socks, calculate the quality index, weight the cost performance of the default supplier according to the quality index and the actual delivery cycle, re-prioritize the pre-selected suppliers, and update the supplier database.

Images