A garment production quality traceability system and method

Abstract

This invention belongs to the field of garment production traceability technology. It provides a garment production quality traceability system and method, comprising: first, performing similarity analysis on traceability data of small batches affiliated with large-volume orders within a factory to identify highly similar affiliated groups; then, integrating similar non-traceability data of small batches and large-volume orders within each group into data pairs, calculating Euclidean distance, and ranking them from low to high based on distance; next, combining historical traceability records of similar small batch orders, counting the number of anomalies in various types of non-traceability data, and ranking traceability anomalies from high to low based on the number of anomalies; finally, determining the traceability data to be added based on the two ranking results. If the top-ranked data is the same, it is directly added; otherwise, the minimum value is selected from the sum of the rankings and added to the small batch order traceability data. This effectively avoids confusion between small batch and large-volume order traceability data, reduces the traceability error rate, and improves the efficiency of tracing quality issues in small batch orders.

Description

Technical Field

[0001] This invention belongs to the field of garment production traceability technology, specifically a garment production quality traceability system and method. Background Technology

[0002] Garment production involves several stages, including raw material procurement (fabric and accessories), cutting, sewing, finished product inspection, warehousing and logistics, forming a complete chain from raw materials to finished garments. Quality traceability records key information at each stage (such as raw material batches, process parameters, and inspection results) to accurately trace the source of quality problems, ensure compliance, and reduce the risk of recalls.

[0003] In the actual process of garment production, there are small-batch garment customization scenarios, specifically garments produced according to users' personalized needs, such as customized shirts in the C2M model (e.g., users specify collar style and sleeve length), customized maternity wear (special size + anti-radiation function), and clothing for people with disabilities (e.g., durable pants and easy-to-wear jackets for wheelchair users).

[0004] However, in order to simplify operations, reduce management costs, and avoid processing traceability data for small batch orders separately (such as generating codes and entering information separately), most small and medium-sized factories or factories that adopt a "batch + customization hybrid production model" will "attach" the production data of small batch orders (such as raw material batches and process records) to the data of large batch orders of the same style or produced at the same time. When the data is highly similar, it will lead to confusion of traceability data, and even break the traceability link of small batch clothing, resulting in problems such as "wanting to check but not being able to" or "the data found does not match the actual situation".

[0005] Therefore, the present invention provides a garment production quality traceability system and method. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: a method for tracing the quality of garment production, comprising the following steps:

[0008] Similarity analysis is performed on traceability data formed by small-batch orders and large-volume orders that are affiliated within the factory, and highly similar affiliation groups are identified in the affiliation combination of small-batch orders and large-volume orders.

[0009] Perform differential analysis on the non-traceable data of small batch orders and large batch orders within the highly similar affiliation group, and rank the non-traceable data of small batch orders by differentiation.

[0010] Based on the historical traceability records of similar small batch orders, various non-traceability data of small batch orders are ranked for traceability anomalies.

[0011] If the top-ranked non-traceable data in the differentiated ranking of various types of non-traceable data and the ranking of traceability anomalies is the same non-traceable data, then the top-ranked non-traceable data will be added as new traceable data to the traceability data of the affiliated small batch order. If they are not the same non-traceable data, then the traceability data to be added will be determined from the various types of non-traceable data based on the differentiated ranking of various types of non-traceable data and the ranking of traceability anomalies, and added to the traceability data of the affiliated small batch order.

[0012] As a further technical solution of the present invention, the traceability data includes multiple traceability data groups, and each traceability data group contains a type of production data.

[0013] As a further technical solution of the present invention, the similarity analysis process is as follows:

[0014] Based on any traceability data group of small batch orders under affiliation, similar traceability groups are obtained through type comparison;

[0015] The proportion of similar traceability groups in the traceability data groups of affiliated small batch orders is statistically analyzed to obtain the data similarity value, which is then summed with the data similarity value to obtain the traceability confusion value.

[0016] If the traceability confusion value is greater than or equal to the traceability confusion threshold, the combination of small batch orders and large batch orders will be marked as a highly similar combination group.

[0017] As a further technical solution of the present invention, the method for obtaining the similar traceability group is as follows:

[0018] If a traceability data group of the same type as the traceability data group of the small batch orders exists in the traceability data generated by the large batch orders that are being affiliated, then the traceability data group of the affiliated small batch orders will be marked as the same type of traceability group.

[0019] As a further technical solution of the present invention, the method for obtaining the data similarity value is as follows:

[0020] Mark the same type of traceability group in the traceability data group of the affiliated small batch orders as the first type of traceability group. In the traceability data of the affiliated large batch orders, mark the traceability data group of the same type as the first type of traceability group as the second type of traceability group.

[0021] Calculate the Euclidean distance between the first and second similar tracing groups. If the Euclidean distance is less than or equal to the Euclidean distance threshold, then mark the first similar tracing group as a highly similar tracing group.

[0022] The proportion of highly similar traceability groups within the same category is counted to obtain the data similarity value.

[0023] As a further technical solution of the present invention, the non-traceable data refers to other production data that is not currently used as traceable data in the production data of small batch orders or large batch orders. The non-traceable data includes multiple non-traceable data groups, and each non-traceable data group contains one type of production data.

[0024] A further technical solution of the present invention is as follows: the process of performing difference analysis and ranking the various production data of small batch orders by difference is as follows:

[0025] Consolidate similar non-traceable data groups between small batch orders and large batch orders into non-traceable data group pairs;

[0026] Calculate the Euclidean distance between each non-traceable data group and the non-traceable data groups it contains, and sort the various types of non-traceable data of small batch orders in order of Euclidean distance from low to high.

[0027] As a further technical solution of the present invention, the process of tracing and ranking anomalies is as follows:

[0028] For various non-traceable data groups of small batch orders, the number of anomalies for each type of non-traceable data group is determined based on the historical traceability records of similar small batch orders, and the various types of non-traceable data of small batch orders are sorted in descending order of the number of anomalies.

[0029] As a further technical solution of the present invention, the process of determining traceability supplementary data from various types of non-traceability data is as follows:

[0030] Based on any non-traceable data, sum the non-traceable data in the differentiated ranking and the traceability anomaly ranking, and select the non-traceable data with the smallest sum of rankings as the traceability supplement data.

[0031] A garment production quality traceability system, the system comprising:

[0032] Order Affiliation Analysis Module: Performs similarity analysis on traceability data formed by small batch orders and large batch orders affiliated within the factory, and identifies highly similar affiliation groups in the affiliation combination of small batch orders and large batch orders;

[0033] Data Difference Analysis Module: Performs corresponding difference analysis on the non-traceable data of small batch orders and large batch orders within highly similar affiliation groups, and ranks the various types of non-traceable data of small batch orders by difference;

[0034] Traceability Anomaly Analysis Module: Based on historical traceability records of similar small batch orders, this module ranks various non-traceability data of small batch orders for traceability anomalies.

[0035] Traceability Data Supplement Module: If the top-ranked non-traceability data in the differentiated ranking and traceability anomaly ranking of various types of non-traceability data is the same non-traceability data, then the top-ranked non-traceability data will be added as new traceability data to the traceability data of the affiliated small batch order. If they are not the same non-traceability data, then the traceability supplement data will be determined from various types of non-traceability data based on the differentiated ranking and traceability anomaly ranking, and added to the traceability data of the affiliated small batch order.

[0036] The beneficial effects of this invention are as follows: First, similarity analysis is performed on the traceability data of small batches affiliated with large-volume orders within the factory. By statistically analyzing the proportion of similar traceability groups, the data similarity value is obtained. The Euclidean distance between similar traceability groups is calculated to screen highly similar groups and obtain data similarity values. The traceability confusion value is summed to identify highly similar affiliated groups. Next, the non-traceability data of the same type of small batches and large-volume orders within the group are integrated into data pairs. After calculating the Euclidean distance, the data is ranked from low to high based on the distance. Then, combined with the historical traceability records of similar small batches of orders, the number of anomalies of various types of non-traceability data is statistically analyzed and the traceability anomalies are ranked from high to low based on the number of anomalies. Finally, the traceability data to be added is determined based on the two ranking results. If the top-ranked data is the same, it is directly added. Otherwise, the data with the minimum summation ranking is added to the traceability data of the small batch orders. This technical solution effectively avoids confusion between traceability data and traceability data of small batches and large batches by accurately identifying highly similar groups that are easily confused in traceability data, mining differentiated non-traceability data, and determining supplementary data by combining historical anomalies. This reduces the traceability error rate and improves the traceability efficiency for quality issues of small batch orders. It realizes the upgrade of traceability data from basic coverage to precision and efficiency, perfectly adapts to the quality control needs of multi-order affiliation scenarios in garment production, and provides a scientific and efficient solution for quality traceability in garment production. Attached Figure Description

[0037] The invention will now be further described with reference to the accompanying drawings.

[0038] Figure 1 This is a flowchart of the steps in Embodiment 1 of the present invention;

[0039] Figure 2 This is the logic judgment diagram of Embodiment 1 of the present invention;

[0040] Figure 3 This is a flowchart of Embodiment 2 of the present invention. Detailed Implementation

[0041] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0042] Example 1

[0043] Please see Figure 1-2 As shown in the figure, a method for tracing the quality of garment production according to an embodiment of the present invention includes the following steps:

[0044] Step 1: Perform similarity analysis on the traceability data formed by small batch orders and large batch orders that are affiliated within the factory, and identify highly similar affiliation groups in the affiliation combination of small batch orders and large batch orders.

[0045] In step one, the traceability data formed by the affiliated small batch orders and the affiliated large batch orders is obtained by technicians through the compilation and summarization of the production data of the orders. The traceability data includes multiple traceability data groups, and each traceability data group contains one type of production data.

[0046] It should be noted that traceability data is obtained by sorting, extracting and summarizing production data. Its data coverage is smaller than that of production data. Production data includes, but is not limited to, material data of garment orders (such as parameters such as cotton content and polyester content), processing data (such as parameters such as cutting width and area), and process data (such as parameters such as cutting times and dyeing area).

[0047] In step one, the similarity analysis process is as follows:

[0048] Any traceability data group based on affiliated small batch orders;

[0049] If there is a traceability data group of the same type as the traceability data group of the small batch orders in the traceability data generated by the large batch orders that are affiliated, then the traceability data group of the affiliated small batch orders will be marked as the same type of traceability group.

[0050] If there is no traceability data group of the same type as the traceability data group of the small batch orders in the traceability data generated by the large batch orders that are attached, then the traceability data group of the attached small batch orders will be marked as a different type of traceability group.

[0051] The proportion of similar traceability groups in the traceability data groups of affiliated small batch orders is counted to obtain the data similarity value;

[0052] Mark the same type of traceability group in the traceability data group of the affiliated small batch orders as the first type of traceability group. In the traceability data formed by the affiliated large batch orders, mark the traceability data group of the same type as the first type of traceability group as the second type of traceability group.

[0053] Calculate the Euclidean distance between the first and second similar tracing groups. If the Euclidean distance is less than or equal to a threshold, mark the first similar tracing group as a highly similar tracing group. If the Euclidean distance is greater than the threshold, mark the second similar tracing group as a highly similar tracing group.

[0054] It should be noted that when calculating the Euclidean distance between the first and second similar tracing groups, if a parameter that exists in one similar tracing group does not exist in the other similar tracing group, the parameter in the other similar tracing group is defaulted to 0. For example, if the first similar tracing group contains the cutting area, but the second similar tracing group does not contain the cutting area, the cutting area in the second similar tracing group is defaulted to 0.

[0055] The proportion of highly similar traceability groups within the first similar traceability group is used to obtain the data similarity value.

[0056] The summation of similar values ​​and identical values ​​in the data yields the traceability confusion value.

[0057] Understandably, the physical meaning of the traceability confusion value lies in the fact that it is calculated using data similarity and data homonym values. The data homonym value reflects the degree of similarity in data type between the traceability data set of the affiliated small batch order and the traceability data set of the affiliated large batch order. The higher the degree of similarity, the more susceptible the traceability data of the large batch order is to be affected by the traceability data of the large batch order when tracing subsequent small batch orders, and the higher the risk of traceability data confusion. Similarly, the data homonym value reflects the degree of similarity in data parameters between the traceability data of the affiliated small batch order and the traceability data of the affiliated large batch order. The higher the degree of similarity, the more susceptible the traceability data of the large batch order is to be affected by the traceability data of the large batch order when tracing subsequent small batch orders, and the higher the risk of traceability data confusion.

[0058] In some embodiments, the trace obfuscation value is compared with the trace obfuscation threshold;

[0059] If the traceability confusion value is greater than or equal to the traceability confusion threshold, the combination of small batch orders and large batch orders will be marked as a highly similar combination group.

[0060] If the trace obfuscation value is less than the trace obfuscation threshold, no action is taken.

[0061] Understandably, the significance of step one lies in accurately identifying highly similar order groups that are easily confused in terms of traceability data from the order groupings. This provides a clear target for subsequent targeted optimization of traceability data for small batch orders, avoiding interference between batch order data and small batch orders during traceability due to similar data types and parameters, and identifying the traceability groups that need improvement from the source.

[0062] Step 2: Conduct a difference analysis on the non-traceable data of small batch orders and large batch orders within the highly similar affiliation group, and rank the various types of non-traceable data of small batch orders by difference;

[0063] In step two, the non-traceable data refers to other production data that is not currently used as traceable data in the production data of small batch orders or large batch orders. The non-traceable data includes multiple non-traceable data groups, and each non-traceable data group contains one type of production data.

[0064] It should be noted that the reason for conducting the corresponding difference analysis on the non-traceable data of the same type for small batch orders and large batch orders within the highly similar affiliation group is that the non-traceable data of the same type for both small batch orders and large batch orders have not been used as traceable data. Therefore, finding the non-traceable data with the greatest difference to supplement the traceable data of small batch orders can improve the difference between the traceable data of small batch orders and the traceable data of large batch orders, which is beneficial to the traceability of subsequent small batch orders.

[0065] In step two, a difference analysis is performed, and the process of ranking the various production data of small batch orders by difference is as follows:

[0066] Consolidate similar non-traceable data groups between small batch orders and large batch orders into non-traceable data group pairs;

[0067] Calculate the Euclidean distance between each non-traceable data group and the non-traceable data groups it contains, and sort the various types of non-traceable data for small batch orders in order of Euclidean distance from low to high. The smaller the Euclidean distance, the higher the non-traceable data group ranks for the non-traceable data of small and medium batch orders.

[0068] Understandably, the significance of step two lies in: by mining the differences between non-traceable data of the same type, filtering out non-traceable data that differs significantly from large-volume orders, providing a "candidate pool" to supplement traceability data for small-batch orders, laying the foundation for improving the differentiation between the two types of order traceability data, and solving the problem of traceability data homogenization.

[0069] Step 3: Based on the historical traceability records of similar small batch orders, rank the various non-traceability data of small batch orders for traceability anomalies;

[0070] In step three, the same type of small batch order refers to the small batch order with the same type as the small batch order in the highly similar affiliation group. The historical traceability record contains the abnormal production data types (such as material data, processing data, and process data) determined after tracing the same type of small batch order multiple times.

[0071] The process of tracing and ranking anomalies is as follows:

[0072] For various non-traceable data groups of small batch orders, the number of anomalies for each type of non-traceable data group is determined based on the historical traceability records of similar small batch orders, and the various types of non-traceable data of small batch orders are sorted in descending order of the number of anomalies.

[0073] For example, the various non-traceable data groups for small batch orders include material data, processing data, and process data. Based on historical traceability records, the number of times the abnormal production data is identified as material data is 5, processing data is 8, and process data is 6. Note that the same number of times is ranked equally.

[0074] Understandably, the significance of step three lies in: combining historical traceability anomaly records to prioritize the screening of non-traceability data types that frequently exhibit anomalies in small batch orders during traceability, ensuring that the supplementary traceability data directly addresses historical traceability pain points and improves the coverage and early warning capabilities of traceability data for anomalies.

[0075] Step 4: If the top-ranked non-traceable data in the differentiated ranking and traceability anomaly ranking of various types of non-traceable data is the same non-traceable data, then the top-ranked non-traceable data will be added as new traceable data to the traceability data of the affiliated small batch order. If they are not the same non-traceable data, then the traceability data to be added will be determined from various types of non-traceable data by combining the differentiated ranking and traceability anomaly ranking of various types of non-traceable data, and added to the traceability data of the affiliated small batch order.

[0076] In step four, if the top-ranked non-traceable data in the differentiated ranking and traceability anomaly ranking of various non-traceable data is the same non-traceable data, then the top-ranked non-traceable data will be added as new traceable data to the traceability data of the affiliated small batch order. For example, if material data is non-traceable data and ranks first in both the differentiated ranking and traceability anomaly ranking, then the material data will also be used as the traceability data of the affiliated small batch order.

[0077] In step four, the process of determining the traceability supplementary data from various types of non-traceability data is as follows:

[0078] Based on any non-traceable data, sum the non-traceable data in the differentiated ranking and the traceability anomaly ranking, select the non-traceable data with the smallest sum of rankings as the traceability supplement data, and supplement it into the traceability data of the affiliated small batch orders;

[0079] Understandably, the significance of step four lies in: through the collaborative decision-making of differentiation and anomaly ranking, scientifically determining the optimal traceability supplementary data, ensuring that the supplementary data can effectively distinguish large-volume orders, and specifically addressing historical traceability anomaly issues, thereby achieving precise optimization of traceability data for small-batch orders.

[0080] The technical solution of this invention is as follows: by identifying highly similar groups and supplementing differentiated and highly abnormal correlation data, the confusion between traceability data of small batches and large batches of orders is avoided, and the traceability error rate is reduced; by combining historical abnormal records, the supplementary data focuses on weak links in traceability, and the traceability efficiency for quality problems of small batch orders is improved; the dimensions of traceability data for small batch orders are dynamically improved, and the traceability data is upgraded from "basic coverage" to "precise and efficient", which is adapted to the quality control needs of multiple order affiliation scenarios in garment production.

[0081] Example 2

[0082] Based on the same inventive concept as the garment production quality traceability method in the foregoing embodiments, such as Figure 3 As shown, this application provides a garment production quality traceability system, wherein the system specifically includes:

[0083] Order Affiliation Analysis Module: Performs similarity analysis on traceability data formed by small batch orders and large batch orders affiliated within the factory, and identifies highly similar affiliation groups in the affiliation combination of small batch orders and large batch orders;

[0084] Data Difference Analysis Module: Performs corresponding difference analysis on the non-traceable data of small batch orders and large batch orders within highly similar affiliation groups, and ranks the various types of non-traceable data of small batch orders by difference;

[0085] Traceability Anomaly Analysis Module: Based on historical traceability records of similar small batch orders, this module ranks various non-traceability data of small batch orders for traceability anomalies.

[0086] Traceability Data Supplement Module: If the top-ranked non-traceability data in the differentiated ranking and traceability anomaly ranking of various types of non-traceability data is the same non-traceability data, then the top-ranked non-traceability data will be added as new traceability data to the traceability data of the affiliated small batch order. If they are not the same non-traceability data, then the traceability supplement data will be determined from various types of non-traceability data based on the differentiated ranking and traceability anomaly ranking, and added to the traceability data of the affiliated small batch order.

[0087] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A garment production quality traceability method, characterized by: The method comprises the following steps: Similarity analysis is performed on the traceability data of the small batch orders based on the attachment in the factory and the attached large batch orders, and high similarity attachment groups are identified in the attachment combination of the small batch orders attached to the large batch orders; Difference analysis is performed on the same type of non-traceability data of the small batch orders and the large batch orders in the high similarity attachment group, and the difference ranking of the non-traceability data of the small batch orders is performed; The non-traceability data represents the production data that is not currently used as traceability data in the production data of the small batch orders or the large batch orders, wherein the non-traceability data comprises a plurality of non-traceability data groups, and each non-traceability data group contains one type of production data; The difference ranking process is as follows: The same type of non-traceability data groups between the small batch orders and the large batch orders are integrated into non-traceability data group pairs; The Euclidean distance between the non-traceability data groups contained in each non-traceability data group pair is calculated, and the non-traceability data of the small batch orders is sequentially sorted from low to high according to the Euclidean distance; According to the historical traceability records of the same type of small batch orders, the non-traceability data of the small batch orders is ranked according to the traceability exception; The traceability exception ranking process is as follows: For each type of non-traceability data group of the small batch orders, the number of exceptions of each type of non-traceability data group is determined according to the historical traceability records of the same type of small batch orders, and the non-traceability data of the small batch orders is sequentially sorted from high to low according to the number of exceptions; If the top-ranked non-traceability data in the difference ranking and the traceability exception ranking of the non-traceability data is the same non-traceability data, the top-ranked non-traceability data is supplemented as new traceability data into the traceability data of the attached small batch orders, and if it is not the same non-traceability data, the difference ranking and the traceability exception ranking of the non-traceability data are combined to determine the traceability supplement data in the non-traceability data, and the traceability supplement data is supplemented into the traceability data of the attached small batch orders; The process of determining the traceability supplement data in the non-traceability data is as follows: Based on any non-traceability data, the sum of the difference ranking and the traceability exception ranking of the non-traceability data is calculated, and the non-traceability data with the smallest sum is selected as the traceability supplement data.

2. The garment production quality traceability method according to claim 1, wherein: The traceability data comprises a plurality of traceability data groups, and each traceability data group contains one type of production data.

3. The garment production quality traceability method according to claim 2, wherein: The similarity analysis process is as follows: Based on any traceability data group of the small batch orders attached, the same type of traceability group is obtained through type comparison; The number proportion of the same type of traceability group in the traceability data group of the attached small batch orders is counted to obtain the data similarity value, and the sum of the data similarity value and the data similarity threshold value is obtained to obtain the traceability confusion value; If the traceability confusion value is greater than or equal to the traceability confusion threshold value, the attachment combination of the small batch orders attached to the large batch orders is marked as a high similarity attachment group.

4. The garment production quality traceability method according to claim 3, wherein: The same type of traceability group is obtained in the following manner: If there is a same type of trace data group in the trace data formed by the large batch orders to which the small batch orders are attached, the trace data group of the small batch orders attached is marked as a same type trace group.

5. The garment production quality trace method of claim 4, wherein: The data similarity value is obtained in the following manner: The same type trace groups in the trace data group of the small batch orders attached are marked as first same type trace groups, and in the trace data of the large batch orders to which the small batch orders are attached, the trace data groups of the same type as the first same type trace groups are marked as second same type trace groups; The Euclidean distance between the first same type trace groups and the second same type trace groups is calculated, and if the Euclidean distance is less than or equal to the Euclidean distance threshold, the first same type trace groups are marked as same type high similarity trace groups; The proportion of the number of the same type high similarity trace groups in the first same type trace groups is counted to obtain the data similarity value.

6. A garment production quality traceability system characterized by, The system is used to perform the method of any one of claims 1-5, and the system comprises: An order attachment analysis module: performing similarity analysis on the trace data formed by the small batch orders and the large batch orders to which the small batch orders are attached in the factory, and identifying high similarity attachment groups in the attachment combination of the small batch orders and the large batch orders; A data difference analysis module: performing corresponding difference analysis on each same type non-trace data of the small batch orders and the large batch orders in the high similarity attachment groups, and differentially ranking each type of non-trace data of the small batch orders; A trace abnormality analysis module: according to the historical trace records of the same type small batch orders, differentially ranking each type of non-trace data of the small batch orders; A trace data supplement module: if the first ranked non-trace data in the differentially ranked and trace abnormality ranked non-trace data is the same, the first ranked non-trace data is supplemented as new trace data to the trace data of the small batch orders attached, and if it is not the same, the trace supplement data is determined from the differentially ranked and trace abnormality ranked non-trace data, and is supplemented to the trace data of the small batch orders attached.

Images