A textile traceability system and method based on big data

Abstract

The present invention discloses a textile traceability system and method based on big data, and relates to the field of textile traceability technology. The method includes: collecting traceability data from various links of textile production; wherein the various links of textile production include raw material procurement link, production and processing link, quality inspection link and logistics and transportation link; analyzing and mining the traceability data after cleaning and pretreatment, including: dividing textiles into different categories through cluster analysis; analyzing the spatiotemporal variation characteristics of the traceability data after cleaning and pretreatment; mining the association relationship between various links of textile production through FP-Growth association rules; associating textiles with textile data information, and establishing a textile link traceability system, wherein the textile link traceability system includes a QR code logo corresponding to the textile. The present application can improve the efficiency of textile supply chain management.

Description

Technical Field

[0001] The present application relates to the technical field of textile traceability, and in particular, to a textile traceability system and method based on big data. Background Art

[0002] With the development of the global economy and the increasing attention of consumers to product quality and safety, textile traceability has become a hot topic. The traditional production model has problems of information asymmetry and opaque supply chain, which leads to unguaranteed product quality and uncontrollable risks.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a textile traceability system and method based on big data to solve the above technical problems.

[0005] The present application provides a textile traceability system based on big data, including a data collection and integration unit, a data cleaning and preprocessing unit, a data analysis and mining unit, a model building and optimization unit, and a link tracing unit, wherein:

[0006] The data collection and integration unit is used to collect traceability data from various links of textile production; wherein the various links of textile production include raw material procurement, production and processing, quality inspection and logistics and transportation; the traceability data includes supplier information, production batch information, production process parameters and quality inspection reports;

[0007] The data cleaning and preprocessing unit is used to clean and preprocess the traceability data, remove invalid data and abnormal data, and perform mean filling for missing data according to the upper and lower time series; wherein the preprocessing includes normalization, registration and spatialization;

[0008] The data analysis and mining unit is used to analyze and mine the cleaned and pre-processed traceability data, including: dividing the textiles into different categories through cluster analysis; analyzing the spatiotemporal variation characteristics of the cleaned and pre-processed traceability data; mining the association relationship between the various links of the textile production through the FP-Growth association rule; wherein the spatiotemporal variation characteristics of the traceability data are analyzed, including: time series matching for the traceability data of different links; sorting the cleaned and pre-processed traceability data in the order of the links from front to back, smoothing the time intervals between the data of each link included in the cleaned and pre-processed traceability data to obtain a smoothed time interval sequence; symbolizing each data in the smoothed time interval sequence; wherein the links are in the order from front to back: the raw material procurement link, the production and processing link, the quality inspection link, and the logistics and transportation link;

[0009] The model building and optimization unit is used to build a textile prediction model based on the results of analyzing and mining the traceability data, wherein the textile prediction model is used to predict the quality grade information and production efficiency information of the textile;

[0010] The link tracing unit is used to associate the textile with the data information of the textile to establish a link tracing system for the textile, wherein the link tracing system for the textile includes a QR code identifier corresponding to the textile; the data information of the textile includes traceability data, quality grade information and production efficiency information of the textile at each link.

[0011] Furthermore, the textile traceability system based on big data also includes: a visualization and monitoring unit, which is used to visualize the production process of the textile in the form of dashboards, charts and reports based on the established link traceability system of the textile after associating the textile with the data information of the textile and establishing the link traceability system of the textile, so as to monitor the production process of the textile in real time.

[0012] Furthermore, the textile traceability system based on big data also includes: a feedback optimization unit, which is used to improve and optimize the production process of the textile based on the feedback and suggestions of consumers obtained from the link traceability system of the textile.

[0013] The present application also provides a textile traceability method based on big data, comprising: collecting traceability data from various links of textile production; wherein the various links of textile production include raw material procurement, production and processing, quality inspection and logistics and transportation; the traceability data includes supplier information, production batch information, production process parameters and quality inspection reports;

[0014] Clean and preprocess the traceability data, remove invalid data and abnormal data, and fill in the missing data with mean value according to the upper and lower time series; wherein the preprocessing includes normalization, registration and spatialization;

[0015] Analyzing and mining the traceability data after cleaning and pretreatment, including: classifying the textiles into different categories through cluster analysis; analyzing the temporal and spatial variation characteristics of the traceability data after cleaning and pretreatment; mining the association relationship between various links of the textile production through FP-Growth association rules;

[0016] The temporal and spatial variation characteristics of the traceability data are analyzed, including: time series matching of the traceability data of different links; sorting the cleaned and preprocessed traceability data in order from front to back, smoothing the time intervals between the data of each link included in the cleaned and preprocessed traceability data to obtain a smoothed time interval sequence; symbolizing each data in the smoothed time interval sequence; wherein the links are in order from front to back: the raw material procurement link, the production and processing link, the quality inspection link, and the logistics and transportation link;

[0017] Based on the results of analyzing and mining the traceability data, a textile prediction model is established, wherein the textile prediction model is used to predict the quality grade information and production efficiency information of the textile;

[0018] The textile is associated with the data information of the textile to establish a link traceability system for the textile, wherein the link traceability system for the textile includes a QR code identification corresponding to the textile; the data information of the textile includes traceability data, quality grade information and production efficiency information of the textile at each link.

[0019] Furthermore, mining the association relationship between the various links of the textile production through the FP-Growth association rule includes: constructing a frequent pattern tree according to the result of symbolic processing, and using each computing node of the big data platform to count the occurrence frequency of data items in each link; based on the occurrence frequency of the data items in each link, using the FP-Growth association rule to filter out frequent item sets from the frequent pattern tree; constructing link association rules according to the filtered frequent item sets; and mining the association relationship between the various links of the textile production according to the link association rules.

[0020] Furthermore, the textile prediction model includes a production efficiency identification network and a quality grade identification model. Based on the results of analyzing and mining the traceability data, the production efficiency identification network is established, including: based on the association relationship between the various links of the textile production mined, the traceability data after cleaning and pretreatment is captured with extreme features, and the time division result is determined according to the captured difference; the time division result is subjected to a change optimization search between the upper and lower time division nodes, and the division time node is determined according to the search result;

[0021] Building a production efficiency identification sub-network based on the segmented time nodes;

[0022] The cleaned and preprocessed traceability data is mapped to a preset coordinate system according to the time interval sequence to construct a plurality of linear regression equations; wherein the coordinate distance density p of the preset coordinate system is calculated;

[0023] The multiple linear regression equations are merged into the production efficiency identification sub-network to obtain the production efficiency identification network.

[0024] Furthermore, the quality level identification model performs the following steps:

[0025] Calculate the preliminary steady-state coefficient of the cleaned and preprocessed traceability data according to the preset sample size constraint;

[0026] Using textile quality standard information as a preset steady-state evaluation result, the traceability data after cleaning and pretreatment is evaluated;

[0027] If the evaluation result does not meet the preset condition, adjust the preset sample size constraint until the evaluation result meets the preset condition.

[0028] Furthermore, after associating the textile with the data information of the textile and establishing a link traceability system for the textile, the method also includes: based on the established link traceability system for the textile, visually displaying the production process of the textile in the form of dashboards, charts and reports to monitor the production process of the textile in real time.

[0029] Furthermore, consumers can obtain traceability data information, quality grade information and production efficiency information of the textiles at each link by scanning the QR code corresponding to the textiles, and feed back feedback and suggestions on the textiles to the link traceability system of the textiles.

[0030] Furthermore, based on the feedback and suggestions from consumers obtained from the link traceability system of the textile, the production process of the textile is improved and optimized.

[0031] Based on the embodiments provided in this application, product quality and safety can be guaranteed. Specifically, textiles are products closely related to people's daily lives, and their quality and safety are directly related to the health and rights of consumers. Establishing a textile traceability system based on big data can help companies promptly discover and solve quality problems in the production process and ensure that products meet standards and regulations; it can improve the efficiency of supply chain management. That is, traditional supply chain management has problems of information asymmetry and opaque processes, resulting in low production efficiency and increased costs. Through big data technology, real-time monitoring and data analysis of the supply chain can be achieved, production processes and resource allocation can be optimized, and the efficiency and flexibility of supply chain management can be improved; it can meet consumer needs. Specifically, consumers are increasingly concerned about product quality and safety, and they hope to understand the production process and source of products. Establishing a textile traceability system based on big data can meet consumer needs, improve product transparency and trust, and enhance brand competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the embodiments of the present invention and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0033] Figure 1 is a structural diagram of an optional textile traceability system based on big data according to an embodiment of the present application;

[0034] Figure 2 The present invention is a flowchart of an optional textile traceability method based on big data according to an embodiment of the present application.

[0035] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0036] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0037] In this application, since big data technology has the ability to process massive data, extract related information, and realize data mining, it can provide strong technical support for building a textile traceability system. By collecting, integrating and analyzing a large amount of data, it is possible to monitor and track the entire textile production process, from raw material procurement to production and processing to product sales, and achieve full visualization and transparent management.

[0038] Alternatively, if Figure 1 As shown, the present application provides a textile traceability system based on big data, including a data collection and integration unit 101, a data cleaning and preprocessing unit 102, a data analysis and mining unit 103, a model building and optimization unit 104 and a link tracing unit 105, wherein:

[0039] The data collection and integration unit 101 is used to collect traceability data from various links of textile production; wherein the various links of textile production include raw material procurement, production and processing, quality inspection and logistics and transportation; the traceability data includes supplier information, production batch information, production process parameters and quality inspection reports;

[0040] The data cleaning and preprocessing unit 102 is used to clean and preprocess the traceability data, remove invalid data and abnormal data, and fill the missing data with the mean value according to the upper and lower time series; wherein the preprocessing includes normalization, registration and spatialization;

[0041] The data analysis and mining unit 103 is used to analyze and mine the traceability data after cleaning and pretreatment, including: classifying textiles into different categories through cluster analysis; analyzing the spatiotemporal variation characteristics of the traceability data after cleaning and pretreatment; mining the association relationship between various links of textile production through FP-Growth association rules; wherein, the spatiotemporal variation characteristics of the traceability data are analyzed, including: time series matching for traceability data of different links; sorting the traceability data after cleaning and pretreatment in the order of links from front to back, smoothing the time intervals between the data of each link included in the traceability data after cleaning and pretreatment, and obtaining a smoothed time interval sequence; symbolizing each data in the smoothed time interval sequence; wherein, the links are in the order from front to back: raw material procurement link, production and processing link, quality inspection link and logistics and transportation link;

[0042] A model building and optimization unit 104 is used to build a textile prediction model based on the results of analyzing and mining the traceability data, wherein the textile prediction model is used to predict the quality grade information and production efficiency information of the textile;

[0043] The link tracing unit 105 is used to associate the textile with the data information of the textile to establish a link tracing system for the textile, wherein the link tracing system for the textile includes a QR code identification corresponding to the textile; the data information of the textile includes traceability data, quality grade information and production efficiency information of the textile at each link.

[0044] In some embodiments of the present application, the collected data may have problems such as missing values, abnormal values ​​and duplicate values, and data cleaning and preprocessing are required. Through data cleaning, invalid data and abnormal data are removed; through data preprocessing, data normalization, standardization and feature extraction are performed for subsequent analysis and modeling. The cleaned and preprocessed data are analyzed and mined using big data analysis technology. Cluster analysis, association rule mining, time series analysis and other methods can be used to find the correlation and regularity between data. For example, products can be divided into different categories through cluster analysis, and the association relationship between different links can be found through association rule mining. Based on the results of data analysis, a model of textile production is established. Machine learning algorithms such as decision trees, random forests, neural networks, etc. can be used for modeling to predict product quality and production efficiency. At the same time, the model is continuously optimized to improve the accuracy and stability of the prediction. Finally, a complete link traceability system is established to associate the production process and data information of the product with the product itself. Through the logo or QR code on the product, consumers can scan and obtain detailed information about the product, including production location, production time, production process, etc. At the same time, feedback and suggestions from consumers are collected to continuously improve and optimize the production process.

[0045] Based on the embodiments provided in this application, product quality and safety can be guaranteed. Specifically, textiles are products closely related to people's daily lives, and their quality and safety are directly related to the health and rights of consumers. Establishing a textile traceability system based on big data can help companies promptly discover and solve quality problems in the production process and ensure that products meet standards and regulations; it can improve the efficiency of supply chain management. That is, traditional supply chain management has problems of information asymmetry and opaque processes, resulting in low production efficiency and increased costs. Through big data technology, real-time monitoring and data analysis of the supply chain can be achieved, production processes and resource allocation can be optimized, and the efficiency and flexibility of supply chain management can be improved; it can meet consumer needs. Specifically, consumers are increasingly concerned about product quality and safety, and they hope to understand the production process and source of products. Establishing a textile traceability system based on big data can meet consumer needs, improve product transparency and trust, and enhance brand competitiveness.

[0046] Furthermore, the textile traceability system based on big data also includes:

[0047] The visualization and monitoring unit is used to associate the data information of textiles with textiles and establish a link traceability system for textiles. Based on the established link traceability system for textiles, the production process of textiles is visualized in the form of dashboards, charts and reports to monitor the production process of textiles in real time.

[0048] Furthermore, the textile traceability system based on big data also includes: a feedback optimization unit, which is used to improve and optimize the production process of textiles based on consumers' feedback and suggestions obtained from the textile link traceability system.

[0049] Alternatively, if Figure 2 As shown, the present application provides a textile traceability method based on big data, comprising:

[0050] S201, collect traceability data from all links of textile production; where each link of textile production includes raw material procurement, production and processing, quality inspection and logistics and transportation; traceability data includes supplier information, production batch information, production process parameters and quality inspection reports;

[0051] S202, cleaning and preprocessing the traceability data, removing invalid data and abnormal data, and filling the missing data with mean values ​​according to the upper and lower time series; wherein the preprocessing includes normalization, registration and spatialization;

[0052] S203, establishing a textile prediction model based on the results of analyzing and mining the traceability data, wherein the textile prediction model is used to predict quality grade information and production efficiency information of the textile;

[0053] S204, associating the textile with the data information of the textile, and establishing a link traceability system for the textile, wherein the link traceability system for the textile includes a QR code identification corresponding to the textile; the data information of the textile includes traceability data, quality grade information and production efficiency information of the textile at each link.

[0054] In some embodiments of the present application, the traceability data after cleaning and pretreatment are analyzed and mined, including: classifying textiles into different categories through cluster analysis; analyzing the spatiotemporal variation characteristics of the traceability data after cleaning and pretreatment; mining the association relationship between various links of textile production through FP-Growth association rules;

[0055] Among them, the temporal and spatial variation characteristics of the traceability data are analyzed, including: time series matching of traceability data in different links; sorting the cleaned and pre-processed traceability data in order from front to back, smoothing the time intervals between the data of each link included in the cleaned and pre-processed traceability data to obtain a smoothed time interval sequence; symbolizing each data in the smoothed time interval sequence; wherein the links are in the following order from front to back: raw material procurement link, production and processing link, quality inspection link, and logistics and transportation link.

[0056] Based on the embodiments provided in this application, product quality and safety can be guaranteed. Specifically, textiles are products closely related to people's daily lives, and their quality and safety are directly related to the health and rights of consumers. Establishing a textile traceability system based on big data can help companies promptly discover and solve quality problems in the production process and ensure that products meet standards and regulations; it can improve the efficiency of supply chain management. That is, traditional supply chain management has problems of information asymmetry and opaque processes, resulting in low production efficiency and increased costs. Through big data technology, real-time monitoring and data analysis of the supply chain can be achieved, production processes and resource allocation can be optimized, and the efficiency and flexibility of supply chain management can be improved; it can meet consumer needs. Specifically, consumers are increasingly concerned about product quality and safety, and they hope to understand the production process and source of products. Establishing a textile traceability system based on big data can meet consumer needs, improve product transparency and trust, and enhance brand competitiveness.

[0057] Furthermore, the association relationship between various links of textile production is mined through FP-Growth association rules, including:

[0058] A frequent pattern tree is constructed based on the results of symbolic processing, and the frequency of occurrence of data items in each link is counted using the computing nodes of the big data platform; based on the frequency of occurrence of data items in each link, the FP-Growth association rule is used to filter out frequent item sets from the frequent pattern tree; link association rules are constructed based on the filtered frequent item sets; and the association relationships between the various links of textile production are mined based on the link association rules.

[0059] Furthermore, the textile prediction model includes a production efficiency identification network and a quality grade identification model. Based on the results of analyzing and mining the traceability data, a production efficiency identification network is established, including:

[0060] According to the association between each link of textile production, the traceability data after cleaning and pretreatment is captured with extreme features, and the time division result is determined according to the captured difference; the time division result is searched for the change between the upper and lower time division nodes, and the division time node is determined according to the search result;

[0061] Build a production efficiency identification sub-network based on segmented time nodes;

[0062] The cleaned and preprocessed traceability data is mapped to a preset coordinate system according to the time interval sequence to construct multiple linear regression equations; wherein the coordinate distance density p of the preset coordinate system is calculated based on the following formula;

[0063]

[0064] Where n is the number of traceability data that have been cleaned and preprocessed; d c Indicates the time interval between the data of each link; d i Indicates the distance between the ith traceability data and the origin of the preset coordinate system; d i-1 Indicates the distance between the (i-1)th traceability data and the origin of the preset coordinate system;

[0065] Multiple linear regression equations are merged into the production efficiency identification sub-network to obtain the production efficiency identification network.

[0066] Furthermore, the quality level identification model performs the following steps:

[0067] Calculate preliminary steady-state coefficients of cleaned and preprocessed traceability data according to preset sample size constraints;

[0068] Using textile quality standard information as the preset steady-state evaluation result, the traceability data after cleaning and pretreatment is evaluated;

[0069] If the evaluation result does not meet the preset conditions, adjust the preset sample size constraint based on the following formula until the evaluation result meets the preset conditions;

[0070]

[0071] Where k is the preset sample size constraint after adjustment; c is the correction factor of the steady-state coefficient; e is the natural constant; y actual is the initial steady-state coefficient; y actual It is the standard steady-state coefficient corresponding to the textile quality standard information.

[0072] Furthermore, after associating the textile with the data information of the textile and establishing the link traceability system of the textile, the method further includes:

[0073] Based on the established textile link traceability system, the textile production process is visualized in the form of dashboards, charts and reports to monitor the textile production process in real time.

[0074] Furthermore, consumers can obtain traceability data information, quality grade information and production efficiency information of textiles at each link by scanning the QR code logo corresponding to the textiles, and feed back their feedback and suggestions on the textiles to the textile link traceability system.

[0075] Furthermore, based on the feedback and suggestions from consumers obtained from the textile link traceability system, the production process of textiles is improved and optimized.

[0076] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A textile traceability method based on big data, It is characterized in that include: Collect traceability data from all links of textile production; wherein, the various links of textile production include raw material procurement, production and processing, quality inspection and logistics and transportation; the traceability data includes supplier information, production batch information, production process parameters and quality inspection reports; Clean and preprocess the traceability data, remove invalid data and abnormal data, and fill in the missing data with mean value according to the upper and lower time series; wherein the preprocessing includes normalization, registration and spatialization; Analyzing and mining the traceability data after cleaning and pretreatment, including: classifying the textiles into different categories through cluster analysis; analyzing the temporal and spatial variation characteristics of the traceability data after cleaning and pretreatment; mining the association relationship between various links of the textile production through FP-Growth association rules; The temporal and spatial variation characteristics of the traceability data are analyzed, including: time series matching of the traceability data of different links; sorting the cleaned and preprocessed traceability data in order from front to back, smoothing the time intervals between the data of each link included in the cleaned and preprocessed traceability data to obtain a smoothed time interval sequence; symbolizing each data in the smoothed time interval sequence; wherein the links are in order from front to back: the raw material procurement link, the production and processing link, the quality inspection link, and the logistics and transportation link; Based on the results of analyzing and mining the traceability data, a textile prediction model is established, wherein the textile prediction model is used to predict the quality grade information and production efficiency information of the textile; Associating the textile with the data information of the textile to establish a link traceability system for the textile, wherein the link traceability system for the textile includes a QR code identifier corresponding to the textile; the data information of the textile includes traceability data, quality grade information and production efficiency information of the textile at each link; The mining of the association relationship between various links of the textile production by using the FP-Growth association rule includes: A frequent pattern tree is constructed based on the result of symbolic processing, and the occurrence frequency of data items in each link is counted using each computing node of the big data platform; frequent item sets are filtered out from the frequent pattern tree based on the occurrence frequency of data items in each link using the FP-Growth association rule; link association rules are constructed based on the filtered frequent item sets; and the association relationship between the various links of the textile production is mined based on the link association rules; The textile prediction model includes a production efficiency identification network and a quality grade identification model. The production efficiency identification network is established based on the results of analyzing and mining the traceability data, including: According to the association relationship between the various links of the textile production mined, the traceability data after cleaning and preprocessing is captured with extreme features, and the time division result is determined according to the captured difference; the time division result is subjected to the change optimization search between the upper and lower time division nodes, and the division time node is determined according to the search result; Building a production efficiency identification sub-network based on the segmented time nodes; The cleaned and preprocessed traceability data is mapped to a preset coordinate system according to the time interval sequence to construct multiple linear regression equations; wherein the coordinate distance density of the preset coordinate system is calculated. ; The multiple linear regression equations are merged into the production efficiency identification sub-network to obtain the production efficiency identification network.

2. The textile traceability method based on big data according to claim 1, It is characterized in that The quality level identification model performs the following steps: Calculate the preliminary steady-state coefficient of the cleaned and preprocessed traceability data according to the preset sample size constraint; Using textile quality standard information as a preset steady-state evaluation result, the traceability data after cleaning and pretreatment is evaluated; If the evaluation result does not meet the preset condition, adjust the preset sample size constraint until the evaluation result meets the preset condition.

3. The textile tracing method based on big data according to claim 2, It is characterized in that After associating the textile with the data information of the textile and establishing a link traceability system for the textile, the method further includes: Based on the established link traceability system of the textiles, the production process of the textiles is visualized in the form of dashboards, charts and reports, so as to monitor the production process of the textiles in real time.

4. The textile traceability method based on big data according to claim 3, It is characterized in that By scanning the QR code corresponding to the textile, consumers can obtain the traceability data information, quality grade information and production efficiency information of the textile in each link, and feed back their feedback and suggestions on the textile to the link traceability system of the textile.

5. The textile traceability method based on big data according to claim 4, It is characterized in that Based on the feedback and suggestions from consumers obtained from the link traceability system of the textile, the production process of the textile is improved and optimized.

Images