Sorting data analysis method, electronic device, storage medium, and program product

Abstract

Embodiments of the present application provide a sorting data analysis method, an electronic device, a storage medium and a program product, relating to the technical field of logistics sorting and the technical field of artificial intelligence. The method comprises: obtaining sorting data generated by at least two sorting devices in the running process, the sorting data comprising backflow data, the backflow data referring to data corresponding to a waybill that enters a backflow channel in the sorting process due to unsuccessful sorting by the sorting device; classifying and processing the sorting data according to the device type of the sorting device to obtain at least one set of detail data corresponding to the device type; performing backflow positioning analysis on the at least one set of detail data through an artificial intelligence model to generate a backflow positioning result; and generating an analysis report and outputting the analysis report according to the backflow positioning result. The method is used to realize automatic and efficient analysis of sorting data and improve the positioning efficiency and accuracy of abnormal reasons.

Description

Technical Field

[0001] This application relates to the fields of logistics sorting technology and artificial intelligence technology, and in particular to a sorting data analysis method, electronic device, storage medium and program product. Background Technology

[0002] In a logistics sorting system, the sorting equipment in a transfer center generates a large amount of normal sorting data and return data generated during abnormal sorting. It is necessary to determine the cause of abnormal sorting based on the return data.

[0003] In related technologies, sorting data is mainly screened, analyzed, and optimized manually. However, manual analysis is inefficient and cannot quickly locate the cause of anomalies. Therefore, there is an urgent need for an efficient sorting data analysis method to quickly locate the cause of anomalies. Summary of the Invention

[0004] This application provides a sorting data analysis method, electronic device, storage medium, and program product to achieve automated and efficient analysis of sorting data and improve the efficiency and accuracy of anomaly cause location.

[0005] In a first aspect, embodiments of this application provide a sorting data analysis method, including:

[0006] Acquire sorting data generated during the operation of at least two sorting devices. The sorting data includes return data, which refers to the data corresponding to waybills that failed to be sorted successfully during the sorting process and entered the return channel.

[0007] Based on the type of sorting equipment, the sorting data is classified and processed to obtain at least one set of detailed data corresponding to the type of equipment.

[0008] By using an artificial intelligence model to perform backflow location analysis on at least one set of detailed data, a backflow location result is generated. The backflow location result is used to indicate the abnormal sorting equipment that generated the backflow data, as well as the target component in the abnormal sorting equipment. The target component is one or more components contained in the sorting equipment.

[0009] Based on the reflux positioning results, an analysis report is generated and output.

[0010] In one possible implementation, the sorting data is classified according to the equipment type of the sorting equipment to obtain at least one set of detailed data corresponding to the equipment type, including:

[0011] Based on the equipment type, sorting data is classified into multi-shipment sorting data and single-shipment sorting data. Single-shipment sorting data includes unloading sorting data and loading sorting data. Multi-shipment sorting data is generated by sorting multiple goods that are pre-packaged in a bulk bag after unpacking. Single-shipment sorting data is generated by sorting goods transported in single-item form.

[0012] Dimensional aggregation is performed on the multi-shipment sorting data and the single-shipment sorting data respectively to obtain the corresponding multi-shipment detail table and single-shipment detail table. The dimension fields of the detail table include at least one of the following: equipment type, site code, total processing volume, abnormal processing volume, return rate, and return reason.

[0013] In one possible implementation, an artificial intelligence model is used to perform backflow location analysis on at least one set of detailed data to generate backflow location results, including:

[0014] Extract multi-dimensional feature data related to the reasons for return from the detailed data. The multi-dimensional feature data includes scanner dimension, mobile sorting carrier component dimension, and grid dimension.

[0015] By analyzing multi-dimensional features using an artificial intelligence model, the target component that generates the reflux data is identified, and the reflux positioning result is obtained. The target component is at least one of the scanner, the mobile sorting carrier component, and the grid.

[0016] In one possible implementation, an analysis report is generated and output based on the backflow location results, including:

[0017] The results of the return positioning are analyzed by an artificial intelligence model to generate optimization suggestions, which include at least one of the following: suggestions for parameter adjustment of abnormal sorting equipment, maintenance suggestions, and suggestions for feeding operation specifications;

[0018] Based on the reflux positioning results and optimization suggestions, an analysis report is generated and output.

[0019] In one possible implementation, an analysis report is generated based on the backflow location results and optimization suggestions, including:

[0020] Based on the return location results, an analysis text is generated, which includes the device identifier of the abnormal sorting device and the return rate corresponding to the abnormal sorting device.

[0021] Render the return rate as a bar chart according to the device identification dimension, and render the return reason field in the detailed data as a pie chart according to the proportion;

[0022] The bar charts, pie charts, and optimization suggestions are combined to generate an analysis report.

[0023] In one possible implementation, the method also includes:

[0024] Determine the sorting area corresponding to the sorting equipment;

[0025] Set corresponding return thresholds and target groups for the sorting area. The target groups are used to indicate the destination of the analysis report push.

[0026] When the return rate of the sorting equipment exceeds the return threshold, the analysis report will be pushed to the target group; and / or,

[0027] The analysis report will be pushed to the target group at a preset time.

[0028] In one possible implementation, the method also includes:

[0029] Obtain historical reflux sample data and annotation information. The annotation information includes the reasons for the reflux data in the historical reflux sample data, the sorting equipment and components that generated the reflux data, and optimization measures.

[0030] The initial model is trained in a supervised manner using historical backflow samples and labeled information to generate an artificial intelligence model;

[0031] The artificial intelligence model is privately deployed to the site server where the sorting equipment is located.

[0032] Secondly, embodiments of this application provide a sorting data analysis device, comprising:

[0033] The acquisition module is used to acquire sorting data generated by at least two sorting devices during operation. The sorting data includes return data, which refers to the data corresponding to waybills that failed to be sorted successfully during the sorting process and entered the return channel.

[0034] The processing module is used to classify and process the sorting data according to the equipment type of the sorting equipment, and obtain at least one set of detailed data corresponding to the equipment type;

[0035] The analysis module is used to perform backflow location analysis on at least one set of detailed data through an artificial intelligence model, and generate backflow location results. The backflow location results are used to indicate the abnormal sorting equipment that generated the backflow data, as well as the target component in the abnormal sorting equipment. The target component is one or more components contained in the sorting equipment.

[0036] The generation module is used to generate and output an analysis report based on the reflux positioning results.

[0037] In one possible implementation, the processing module is specifically used for:

[0038] Based on the equipment type, sorting data is classified into multi-shipment sorting data and single-shipment sorting data. Single-shipment sorting data includes unloading sorting data and loading sorting data. Multi-shipment sorting data is generated by sorting multiple goods that are pre-packaged in a bulk bag after unpacking. Single-shipment sorting data is generated by sorting goods transported in single-item form.

[0039] Dimensional aggregation is performed on the multi-shipment sorting data and the single-shipment sorting data respectively to obtain the corresponding multi-shipment detail table and single-shipment detail table. The dimension fields of the detail table include at least one of the following: equipment type, site code, total processing volume, abnormal processing volume, return rate, and return reason.

[0040] In one possible implementation, the analysis module is specifically used for:

[0041] Extract multi-dimensional feature data related to the reasons for return from the detailed data. The multi-dimensional feature data includes scanner dimension, mobile sorting carrier component dimension, and grid dimension.

[0042] By analyzing multi-dimensional features using an artificial intelligence model, the target component that generates the reflux data is identified, and the reflux positioning result is obtained. The target component is at least one of the scanner, the mobile sorting carrier component, and the grid.

[0043] In one possible implementation, the generation module is specifically used for:

[0044] The results of the return positioning are analyzed by an artificial intelligence model to generate optimization suggestions, which include at least one of the following: suggestions for parameter adjustment of abnormal sorting equipment, maintenance suggestions, and suggestions for feeding operation specifications;

[0045] Based on the reflux positioning results and optimization suggestions, an analysis report is generated and output.

[0046] In one possible implementation, the generation module is also used for:

[0047] Based on the return location results, an analysis text is generated, which includes the device identifier of the abnormal sorting device and the return rate corresponding to the abnormal sorting device.

[0048] Render the return rate as a bar chart according to the device identification dimension, and render the return reason field in the detailed data as a pie chart according to the proportion;

[0049] The bar charts, pie charts, and optimization suggestions are combined to generate an analysis report.

[0050] In one possible implementation, the sorting data analysis device is also used for:

[0051] Determine the sorting area corresponding to the sorting equipment;

[0052] Set corresponding return thresholds and target groups for the sorting area. The target groups are used to indicate the destination of the analysis report push.

[0053] When the return rate of the sorting equipment exceeds the return threshold, the analysis report will be pushed to the target group; and / or,

[0054] The analysis report will be pushed to the target group at a preset time.

[0055] In one possible implementation, the sorting data analysis device is also used for:

[0056] Obtain historical reflux sample data and annotation information. The annotation information includes the reasons for the reflux data in the historical reflux sample data, the sorting equipment and components that generated the reflux data, and optimization measures.

[0057] The initial model is trained in a supervised manner using historical backflow samples and labeled information to generate an artificial intelligence model;

[0058] The artificial intelligence model is privately deployed to the site server where the sorting equipment is located.

[0059] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0060] The memory stores the instructions that the computer executes;

[0061] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0062] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0063] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0064] The sorting data analysis method, electronic device, storage medium, and program product provided in this application acquire sorting data from multiple sorting devices and process it according to device type. Combined with an artificial intelligence model, it completes the return location and accurate identification of abnormal devices and target components, and finally automatically generates and outputs an analysis report. It can completely replace manual sorting data processing and anomaly analysis operations, avoid the inefficiency and errors caused by manual processing, realize the automated and efficient analysis of sorting data, and improve the efficiency and accuracy of anomaly cause location. Attached Figure Description

[0065] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0066] Figure 1 Flowchart of the sorting data analysis method provided in this application Figure 1 ;

[0067] Figure 2 Flowchart of the sorting data analysis method provided in this application Figure 2 ;

[0068] Figure 3 A schematic diagram of an analysis report provided for this application;

[0069] Figure 4 A schematic diagram of the sorting data analysis device provided in this application;

[0070] Figure 5 A schematic diagram of the structure of the electronic device provided in this application.

[0071] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0072] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0073] In a logistics sorting system, the sorting equipment in a transfer center generates a large amount of sorting data during operation, including key indicators such as total sorting volume, abnormal handling volume, return rate, and equipment line number.

[0074] In related technologies, sorting systems rely on manual screening, analysis, and optimization suggestions for return data. However, with the rapid growth of logistics volume, the number of sorting devices and the scale of data are increasing exponentially, making it difficult for traditional manual processing methods to meet the demands for real-time performance, accuracy, and efficiency. For example, in large transfer centers, the daily sorting volume can reach tens of millions of pieces, and the return volume may exceed one million. Manual analysis requires a significant amount of time and is prone to missing key anomalies. Furthermore, anomalies in sorting equipment may involve multiple factors such as scanner malfunctions, abnormal cart operation, and improper compartment configuration, requiring comprehensive diagnosis based on business rules and equipment operation data.

[0075] Based on this, this application provides a sorting data analysis method. By acquiring sorting data from multiple sorting devices and classifying and processing it according to device type, and combining it with an artificial intelligence model to complete the return location and accurate identification of abnormal devices and target components, the method automatically generates and outputs an analysis report. This method can completely replace manual sorting data processing and anomaly analysis operations, avoid the inefficiency and errors caused by manual processing, realize the automated and efficient analysis of sorting data, and improve the efficiency and accuracy of anomaly cause location.

[0076] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0077] It should be noted that the execution entity in each embodiment of this application can be a server, processor, microprocessor, etc. The specific execution entity in each embodiment of this application is not limited, and it can be selected and set according to actual needs. In the following embodiments, a server is used as an example of the execution entity, which does not constitute a limitation on the actual execution entity.

[0078] Figure 1 Flowchart of the sorting data analysis method provided in this application Figure 1 ,like Figure 1 As shown, the method includes:

[0079] S101. Obtain sorting data generated during the operation of at least two sorting devices.

[0080] The sorting data includes return data, which refers to the data corresponding to waybills that failed to be sorted successfully during the sorting process and entered the return channel.

[0081] Sorting data refers to the data generated during the operation of sorting equipment. Sorting data also includes normal sorting data, which refers to the data generated when goods are sorted by the sorting equipment, successfully matched to the slots, and delivered normally.

[0082] Sorting equipment refers to devices used for automated sorting of goods in logistics transfer centers. Logistics transfer centers include various types of sorting equipment, and each type may include multiple units. For example, according to equipment type, sorting equipment can be divided into small-item sorting machines, single-item sorting machines, etc. Sorting equipment may include multiple components, such as scanners, mobile sorting carrier components, and sorting slots. The mobile sorting carrier component refers to the physical component in the sorting equipment used to carry and transport goods to be sorted and place them into the sorting slots at designated locations, such as a trolley running along a track. In the following embodiments, the mobile sorting carrier component and the trolley can be used interchangeably.

[0083] Sorting data for a preset time period can be retrieved from the data storage platform at preset times. For example, sorting data generated between 8:00 AM the previous day and 8:00 AM the current day can be retrieved from the data storage platform at 8:00 AM every day.

[0084] S102. Based on the equipment type of the sorting equipment, classify and process the sorting data to obtain at least one set of detailed data corresponding to the equipment type.

[0085] The equipment type is categorized based on the type of goods being handled. During cargo transportation, small items, whether lightweight or bulky, are typically packaged in bulk bags for transport. Upon arrival at the transfer center, these bags are unpacked, and then sorted individually using small-item sorting equipment. Other, heavier or bulkier goods are usually transported individually and sorted using single-item sorting machines upon arrival at the transfer center.

[0086] For example, extract the sorting data corresponding to the small item sorting machine and integrate it to obtain a set of detailed data; extract the sorting data corresponding to the single item sorting machine and integrate it to obtain another set of detailed data.

[0087] Detailed data refers to the information and statistical characteristics of sorting equipment obtained after classifying and processing the sorting data. For example, for the detailed data corresponding to small item sorting machines, for each small item sorting machine, the corresponding site code, equipment code, total sorting data volume, successful sorting data volume, abnormal sorting data volume, and return data volume caused by various return reasons are statistically analyzed.

[0088] The sorting data is categorized to obtain at least one set of detailed data corresponding to each equipment type. This can be achieved as follows: Structured query statements are retrieved from a data processing experience library. These statements indicate the rules for extracting dimension fields from the detailed data. The structured query statements are then used to query, filter, and aggregate the sorting data corresponding to different equipment types to obtain the corresponding detailed data. These structured query statements include, for example, queries indicating how to filter sorting data for small-item machines and extract equipment code fields.

[0089] S103. Perform backflow location analysis on at least one set of detailed data using an artificial intelligence model to generate backflow location results.

[0090] The return location result is used to indicate the abnormal sorting device that generates the return data, as well as the target component in the abnormal sorting device. The target component is one or more components contained in the sorting device.

[0091] Components on the sorting equipment may include scanners, mobile sorting carriers, and grids. The target component refers to the part in the sorting equipment that causes backflow. For example, when backflow data is generated because scanner number 1 failed to read the data, the target component is scanner number 1.

[0092] The categorized detailed data is input into the artificial intelligence model. The artificial intelligence model performs multi-dimensional feature intelligent diagnosis based on the cause of return, scanner, grid, and mobile sorting carrier components, locates the abnormal sorting equipment that caused the return, identifies the target component in the abnormal sorting equipment that caused the return, and generates the return location result.

[0093] Reasons for return flow include, for example, the scanner failing to read the data, the part exceeding the edge of the conveyor belt, the temporary storage area being full and the slot being locked, and the cart sorting failing. For each reason for return flow, the abnormal sorting equipment and target component that generated the return flow data are located to obtain the return flow location result.

[0094] For example, the return rate of sorting equipment LF-1 is 0.62%. Among them, the return reason is that the scanner failed to read the data, and the high-frequency scanners are No. 1 (1006 pieces) and No. 2 (676 pieces); the return reason is that the maximum number of cycles is exceeded, and the frequently locked grids are No. 001 (561 pieces) and No. 002 (127 pieces), etc.

[0095] S104. Based on the reflux positioning results, generate and output an analysis report.

[0096] The return location results can be converted into images, which specifically display key indicators such as the average return rate, total processing volume, and total exception handling volume of the sorting equipment. This allows business developers to intuitively understand the overall operation of the sorting equipment. Furthermore, bar charts, pie charts, and other methods can be used to display the return rate and reasons for return of each sorting equipment, helping business developers to quickly identify sorting equipment with a high return rate.

[0097] The sorting data analysis method provided in this application acquires sorting data from multiple sorting devices and processes it according to device type. It then combines an artificial intelligence model to complete the return location and accurate identification of abnormal devices and target components. Finally, it automatically generates and outputs an analysis report. This method can completely replace manual sorting data processing and anomaly analysis operations, avoiding the inefficiency and errors caused by manual processing. It achieves automated and efficient analysis of sorting data and improves the efficiency and accuracy of anomaly cause location.

[0098] Optionally, the sorting data is classified according to the equipment type of the sorting equipment to obtain at least one set of detailed data corresponding to the equipment type, including: classifying the sorting data into multi-item sorting data and single-item sorting data according to the equipment type, wherein the single-item sorting data includes unloading sorting data and loading sorting data, the multi-item sorting data is the data generated by unpacking and sorting multiple goods pre-packaged in a bag, and the single-item sorting data is the data generated by sorting goods transported in single-item form; and performing dimensional aggregation on the multi-item sorting data and the single-item sorting data respectively to obtain the corresponding multi-item detail table and the single-item detail table, wherein the dimension fields of the detail table include at least one of the following: equipment type, site code, total processing volume, abnormal processing volume, return rate, and return reason.

[0099] Single shipments can be sorted separately on unloading and loading sorting equipment. The single shipment sorting data can be further divided into unloading sorting data and loading sorting data.

[0100] Optionally, before classifying the sorting data, the process may include: cleaning the sorting data to remove invalid and duplicate data; and standardizing the cleaned sorting data to convert it into a uniform structured format.

[0101] Optionally, the dimensions of the detailed table can also include equipment number, designed capacity, overall equipment efficiency, average overall equipment efficiency, etc. Reasons for return can include incorrect collection and distribution, sorting plan disabled, scanner not successfully read, returned dispatched waybill, parts exceeding the edge of the trolley belt, temporary storage area full and lockable slots, exceeding the maximum cycle number, trolley sorting failure, equipment action but not actually sorted successfully, sorting timeout, lost object, illegal waybill number, multiple returned items, multiple tags identified, unknown address, etc.

[0102] The dimensions of the single-send detail table are the same as those of the multi-send detail table. For example, the multi-send detail table can be shown in Table 1 below:

[0103]

[0104] In the above method, sorting data is divided into multi-item sorting data and single-item sorting data according to equipment type, and corresponding detailed tables are generated by aggregation according to dimensions. This enables structured and hierarchical processing of sorting data in multiple scenarios, avoids analysis interference caused by data mixing of different business forms, and provides data support for more efficient location of return flow anomalies in the future.

[0105] Optionally, an artificial intelligence model is used to perform backflow location analysis on at least one set of detailed data to generate backflow location results, including: extracting multi-dimensional feature data related to the cause of backflow from the detailed data, the multi-dimensional feature data including scanner dimension, mobile sorting carrier component dimension and grid dimension; analyzing the multi-dimensional features through the artificial intelligence model to determine the target component that generated the backflow data, and obtaining backflow location results, the target component being at least one of scanner, mobile sorting carrier component and grid.

[0106] Different reasons for reflow correspond to different feature data. For example, if the scanner fails to read the data, the feature data is at the scanner dimension; if the temporary storage area is full and the slot is locked, the feature data is at the slot dimension; if the trolley sorting fails, the feature data is at the mobile sorting carrier component dimension, etc. By analyzing these multi-dimensional features using an artificial intelligence model, the target component generating the reflow data is identified, resulting in the reflow location result.

[0107] The above method improves the accuracy and precision of anomaly location by extracting multi-dimensional features of the scanner, moving sorting carrier components, and grid, and then using an artificial intelligence model to analyze and locate the target components.

[0108] Optional, see Figure 2 , Figure 2 Flowchart of the sorting data analysis method provided in this application Figure 2 The method includes:

[0109] S201. Obtain sorting data generated during the operation of at least two sorting devices.

[0110] S202. Based on the equipment type of the sorting equipment, classify and process the sorting data to obtain at least one set of detailed data corresponding to the equipment type.

[0111] S203. Perform backflow location analysis on at least one set of detailed data using an artificial intelligence model to generate backflow location results.

[0112] S204. Analyze the reflux positioning results using an artificial intelligence model and generate optimization suggestions.

[0113] The optimization suggestions include at least one of the following: suggestions for adjusting parameters of abnormal sorting equipment, maintenance suggestions, and suggestions for feeding operation specifications.

[0114] For example, regarding the issue of parts exceeding the edge of the conveyor belt, optimization suggestions could include providing specialized training for parts-feeding personnel at the parts-feeding station, and optimizing the oversized settings and loading parameters of the parts-feeding station. Regarding the issue of scanners failing to read parts, optimization suggestions could include checking the scanner lens / under-scan dust cleaning, checking scanner focus, analyzing unread photos via File Transfer Protocol (FTP), and checking for issues with waybills (broken needles, incorrect format, improper pasting, lamination, etc.). Regarding the issue of exceeding the maximum cycle count, optimization suggestions could include coordinating with packing personnel, ensuring timely cleaning of the chute by packing personnel, providing timely on-site intervention for corresponding area management, adjusting the balance of parts in the sorting plan for compartments with high locking frequency, and providing operational training or increasing the number of packing personnel for compartments with long locking times. Regarding the issue of multiple items being returned, optimization suggestions could include searching for multiple item images locally.

[0115] S205. Based on the reflux positioning results and optimization suggestions, generate and output an analysis report.

[0116] Based on the return flow location results and optimization suggestions, an analysis report is generated, including: generating analysis text based on the return flow location results, which includes the equipment identifier of the abnormal sorting equipment and the return flow rate corresponding to the abnormal sorting equipment; rendering the return flow rate as a bar chart according to the equipment identifier dimension, and rendering the return flow reason field in the detailed data as a pie chart according to the proportion; and integrating the bar chart, pie chart and optimization suggestions to generate an analysis report.

[0117] Equipment identifiers are used to uniquely identify a specific sorting device. Examples of equipment identifiers include equipment number LF-1. See also... Figure 3 , Figure 3This is a schematic diagram of an analysis report provided in this application. The report shows the analysis report for the small-item sorting machine. The top displays the average return rate, total processing volume, and total abnormal processing volume. The left side shows a bar chart of the return rate, indicating the three sorting devices with the highest return rates: LF-9, LF-2, and LF-7, with return rates of 0.62%, 0.56%, and 0.54%, respectively. The right side shows a pie chart of the return cause distribution, with the following results: parts exceeding the edge of the trolley belt (50.75%), scanner failure to read (29.73%), exceeding the maximum cycle time (8.99%), multiple item returns (8.84%), illegal waybill numbers (1.44%), and others (0.25%). The bottom shows the return location results and optimization suggestions for each return cause (only a portion is shown).

[0118] The above method automatically generates optimization suggestions such as parameter adjustment, equipment maintenance, and operation specifications based on the return flow positioning results. It can directly transform the anomaly analysis conclusions into executable improvement plans, eliminating the need for manual summarization and experience-based judgment, and improving the efficiency of sorting data analysis. By presenting abnormal equipment identification and return flow rate in bar charts and return flow reasons in pie charts, it can visually and intuitively display the distribution of anomalies and the proportion of causes, reducing the cost of data understanding and helping business personnel quickly identify high-anomaly equipment and the main reasons for return flow.

[0119] Optionally, after generating the analysis report, the method further includes pushing the analysis report to a target group, specifically including: determining the sorting area corresponding to the sorting equipment; setting a corresponding return threshold and target group for the sorting area, the target group being used to indicate the destination of the analysis report push; pushing the analysis report to the target group when the return rate of the sorting equipment is greater than the return threshold; and / or, pushing the analysis report to the target group at a preset time.

[0120] Different sorting areas correspond to different return thresholds and target groups, which are pre-set. Return thresholds are, for example, 1% or 1.5%, and target groups are the work groups of business personnel. Push notifications can be flexibly configured, including scheduled pushes and pushes based on preset conditions. Scheduled pushes send the analysis report to the target group at a fixed time each day, such as sending the previous day's analysis report and a download link for detailed data to the target group at 8:00 AM daily. Preset conditions send the analysis report to the target group when the return rate of the sorting equipment exceeds the return threshold.

[0121] A sample push notification template reads: "Hello, the analysis report for the small item sorting machine on January 1st has been generated. Please check it out! 1. Detailed data download link: https: / / 123, 2. The analysis report is shown in the image below."

[0122] The above method, by configuring return thresholds and target groups according to sorting sites and supporting threshold triggering and timed push, can achieve accurate, proactive, and on-demand distribution of abnormal reports, avoid interference from invalid information, ensure that abnormal data reaches the responsible person in a timely manner, and further improve the overall efficiency of sorting data analysis and abnormal handling.

[0123] Optionally, before implementing the sorting data analysis method, a configuration phase is also included, specifically including: acquiring historical return sample data and annotation information, the annotation information including the reasons for the return data in the historical return sample data, the sorting equipment and components that generated the return data, and optimization measures; using the historical return samples and annotation information to perform supervised fine-tuning training on the initial model to generate an artificial intelligence model; and privately deploying the artificial intelligence model to the site server where the sorting equipment is located.

[0124] Historical reflux sample data consists of detailed reflux data generated by each sorting device over a period of time (e.g., the last 90 days). The initial model can be a general model in the domain that has the ability to understand text, classify data, reason about causes, and generate suggestions.

[0125] Historical return sample data can be obtained from the database. Business personnel can annotate each sample data. The historical return samples and description information are used to build a training dataset. The historical return sample data is used as input, and the reasons for return, the sorting equipment and components that generate the return data, and optimization measures are used as output labels to fine-tune the initial model to obtain an artificial intelligence model. Then, the artificial intelligence model is deployed to the sorting site, so that the artificial intelligence model can be called through the application programming interface (API).

[0126] In the above method, the initial model is fine-tuned by using real return samples and labeled information, so that the resulting artificial intelligence model can accurately identify the return characteristics of different sorting equipment and different components, avoiding the problems of inaccuracy and misjudgment of the general model. The artificial intelligence model can directly output positioning results and optimization suggestions without manual analysis, thus improving analysis efficiency.

[0127] Figure 4 A schematic diagram of the sorting data analysis device provided in this application is shown below. Figure 4 As shown, the sorting data analysis device 40 provided in this embodiment includes:

[0128] The acquisition module 41 is used to acquire sorting data generated by at least two sorting devices during operation. The sorting data includes return data, which refers to the data corresponding to waybills that failed to be sorted successfully during the sorting process and entered the return channel.

[0129] Processing module 42 is used to classify and process sorting data according to the equipment type of the sorting equipment to obtain at least one set of detailed data corresponding to the equipment type;

[0130] Analysis module 43 is used to perform backflow location analysis on at least one set of detailed data through an artificial intelligence model, generate backflow location results, and use the backflow location results to indicate the abnormal sorting equipment that generates backflow data, as well as the target component in the abnormal sorting equipment. The target component is one or more components contained in the sorting equipment.

[0131] The generation module 44 is used to generate and output an analysis report based on the reflux positioning results.

[0132] In one possible implementation, processing module 42 is specifically used for:

[0133] Based on the equipment type, sorting data is classified into multi-shipment sorting data and single-shipment sorting data. Single-shipment sorting data includes unloading sorting data and loading sorting data. Multi-shipment sorting data is generated by sorting multiple goods that are pre-packaged in a bulk bag after unpacking. Single-shipment sorting data is generated by sorting goods transported in single-item form.

[0134] Dimensional aggregation is performed on the multi-shipment sorting data and the single-shipment sorting data respectively to obtain the corresponding multi-shipment detail table and single-shipment detail table. The dimension fields of the detail table include at least one of the following: equipment type, site code, total processing volume, abnormal processing volume, return rate, and return reason.

[0135] In one possible implementation, the analysis module 43 is specifically used for:

[0136] Extract multi-dimensional feature data related to the reasons for return from the detailed data. The multi-dimensional feature data includes scanner dimension, mobile sorting carrier component dimension, and grid dimension.

[0137] By analyzing multi-dimensional features using an artificial intelligence model, the target component that generates the reflux data is identified, and the reflux positioning result is obtained. The target component is at least one of the scanner, the mobile sorting carrier component, and the grid.

[0138] In one possible implementation, the generation module 44 is specifically used for:

[0139] The results of the return positioning are analyzed by an artificial intelligence model to generate optimization suggestions, which include at least one of the following: suggestions for parameter adjustment of abnormal sorting equipment, maintenance suggestions, and suggestions for feeding operation specifications;

[0140] Based on the reflux positioning results and optimization suggestions, an analysis report is generated and output.

[0141] In one possible implementation, the generation module 44 is also used for:

[0142] Based on the return location results, an analysis text is generated, which includes the device identifier of the abnormal sorting device and the return rate corresponding to the abnormal sorting device.

[0143] Render the return rate as a bar chart according to the device identification dimension, and render the return reason field in the detailed data as a pie chart according to the proportion;

[0144] The bar charts, pie charts, and optimization suggestions are combined to generate an analysis report.

[0145] In one possible implementation, the sorting data analysis device 40 is also used for:

[0146] Determine the sorting area corresponding to the sorting equipment;

[0147] Set corresponding return thresholds and target groups for the sorting area. The target groups are used to indicate the destination of the analysis report push.

[0148] When the return rate of the sorting equipment exceeds the return threshold, the analysis report will be pushed to the target group; and / or,

[0149] The analysis report will be pushed to the target group at a preset time.

[0150] In one possible implementation, the sorting data analysis device 40 is also used for:

[0151] Obtain historical reflux sample data and annotation information. The annotation information includes the reasons for the reflux data in the historical reflux sample data, the sorting equipment and components that generated the reflux data, and optimization measures.

[0152] The initial model is trained in a supervised manner using historical backflow samples and labeled information to generate an artificial intelligence model;

[0153] The artificial intelligence model is privately deployed to the site server where the sorting equipment is located.

[0154] The sorting data analysis device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0155] Figure 5 A schematic diagram of the structure of the electronic device provided in this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 51 and a memory 52. ​​Optionally, the device 50 further includes a communication component 53. The processor 51, memory 52, and communication component 53 are connected via a bus.

[0156] In a specific implementation, at least one processor 51 executes computer execution instructions stored in memory 52, causing at least one processor 51 to perform the above-described method.

[0157] The specific implementation process of processor 51 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0158] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0159] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0160] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0161] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0162] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0163] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0164] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0165] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0166] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0168] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0169] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0170] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A sorting data analysis method, characterized in that, include: Acquire sorting data generated by at least two sorting devices during operation, including return data, which refers to the data corresponding to waybills that failed to be sorted successfully by the sorting devices during the sorting process and entered the return channel; According to the equipment type of the sorting equipment, the sorting data is classified and processed to obtain at least one set of detailed data corresponding to the equipment type; The at least one set of detailed data is analyzed by an artificial intelligence model to generate a return location result. The return location result is used to indicate the abnormal sorting device that generated the return data and the target component in the abnormal sorting device. The target component is one or more components included in the sorting device. Based on the reflux positioning results, an analysis report is generated and output.

2. The method according to claim 1, characterized in that, The step of classifying the sorting data according to the equipment type of the sorting equipment to obtain at least one set of detailed data corresponding to the equipment type includes: According to the equipment type, the sorting data is classified into multi-shipment sorting data and single-shipment sorting data. The single-shipment sorting data includes unloading sorting data and loading sorting data. The multi-shipment sorting data is the data generated by unpacking and sorting multiple goods that are pre-packaged in a bulk bag. The single-shipment sorting data is the data generated by sorting goods transported in single-item form. The multi-shipment sorting data and the single-shipment sorting data are aggregated by dimensions to obtain the corresponding multi-shipment detail table and single-shipment detail table. The dimension fields of the detail table include at least one of the following: equipment type, site code, total processing volume, abnormal processing volume, return rate, and return reason.

3. The method according to claim 1, characterized in that, The step of performing backflow location analysis on the at least one set of detailed data using an artificial intelligence model to generate backflow location results includes: Extract multi-dimensional feature data related to the reasons for return from the detailed data. The multi-dimensional feature data includes scanner dimension, mobile sorting carrier component dimension, and grid dimension. The multi-dimensional features are analyzed by the artificial intelligence model to determine the target component that generates the reflux data, and the reflux positioning result is obtained. The target component is at least one of a scanner, a mobile sorting carrier component, and a grid.

4. The method according to claim 1, characterized in that, The step of generating and outputting an analysis report based on the reflux positioning results includes: The artificial intelligence model is used to analyze the return positioning results and generate optimization suggestions, which include at least one of the following: parameter adjustment suggestions, maintenance suggestions, and component supply operation specifications suggestions for the abnormal sorting equipment. Based on the reflux positioning results and the optimization suggestions, the analysis report is generated and output.

5. The method according to claim 4, characterized in that, The step of generating the analysis report based on the reflux positioning results and the optimization suggestions includes: Based on the return location results, an analysis text is generated, which includes the device identifier of the abnormal sorting device and the return rate corresponding to the abnormal sorting device. The return rate is rendered as a bar chart according to the device identification dimension, and the return reason field in the detailed data is rendered as a pie chart proportionally. The bar chart, the pie chart, and the optimization suggestions are combined to generate the analysis report.

6. The method according to claim 1, characterized in that, The method further includes: Determine the sorting area corresponding to the sorting equipment; Set corresponding return thresholds and target groups for the sorting area, whereby the target groups are used to indicate the destination of the analysis report push; When the return rate of the sorting equipment is greater than the return threshold, the analysis report is pushed to the target group; and / or, The analysis report will be pushed to the target group at a preset time.

7. The method according to claim 1, characterized in that, The method further includes: Obtain historical reflux sample data and annotation information, wherein the annotation information includes the reasons for the reflux data in the historical reflux sample data, the sorting equipment and components that generated the reflux data, and optimization measures; The initial model is trained in a supervised manner using the historical reflux samples and the labeled information to generate the artificial intelligence model. The artificial intelligence model is privately deployed to the site server where the sorting equipment is located.

8. A sorting data analysis device, characterized in that, include: The acquisition module is used to acquire sorting data generated by at least two sorting devices during operation. The sorting data includes return data, which refers to the data corresponding to waybills that failed to be sorted successfully by the sorting devices during the sorting process and entered the return channel. The processing module is used to classify the sorting data according to the equipment type of the sorting equipment to obtain at least one set of detailed data corresponding to the equipment type; An analysis module is used to perform backflow location analysis on the at least one set of detailed data using an artificial intelligence model, and generate backflow location results. The backflow location results are used to indicate the abnormal sorting device that generated the backflow data, and the target component in the abnormal sorting device. The target component is one or more components included in the sorting device. The generation module is used to generate and output an analysis report based on the reflux positioning results.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images