Method, device and computer equipment for identifying abnormal state of logistics package

Abstract

The application relates to a logistics package abnormal state identification method and device, computer equipment and a storage medium, which comprises the following steps: obtaining the logistics characteristics of a logistics package at a business node; obtaining the respective logistics characteristics and historical flow states of a plurality of historical packages; comparing the logistics characteristics of the logistics package and the logistics characteristics of the historical packages to obtain a feature similarity, and obtaining a target historical package from the plurality of historical packages according to the feature similarity; and determining whether the current flow state of the logistics package is abnormal according to the historical flow state corresponding to the target historical package, so that the logistics characteristics of the business node and the historical flow states of the past logistics packages can be comprehensively considered, the timely monitoring and early warning of the abnormal flow state of the logistics package can be realized, the post-facto passive discovery of the flow abnormality can be changed into the timely and active discovery in the process or the advance prediction before the process, and the early warning efficiency is effectively improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer equipment, and storage medium for identifying abnormal states of logistics parcels. Background Technology

[0002] In the logistics and transportation process, the flow of logistics parcels from pickup to delivery involves multiple business stages, such as last-mile delivery, parcel transportation, and site transfer. While existing technologies can collect data from each business stage through business systems, abnormal situations in logistics parcels are only passively discovered some time after they occur, making it difficult to provide early warnings or timely detection of anomalies. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, device, computer equipment, and storage medium for identifying abnormal states of logistics parcels to address the aforementioned technical problems.

[0004] A method for identifying abnormal status of logistics parcels, the method comprising:

[0005] Obtain the logistics characteristics of the package at the business node;

[0006] Obtain the logistics characteristics and historical transit status of each of the multiple historical packages;

[0007] The logistics characteristics of the logistics package and the logistics characteristics of the historical package are compared to obtain the feature similarity, and the target historical package is obtained from multiple historical packages based on the feature similarity.

[0008] Based on the historical circulation status of the target historical package, determine whether the current circulation status of the logistics package is abnormal.

[0009] Optionally, obtaining the logistics characteristics of the logistics package at the business node includes:

[0010] The system receives data streams sent by the business system and extracts raw logistics characteristics from the data streams; the business system is used to collect logistics characteristics of logistics packages at multiple business nodes.

[0011] Obtain the first feature vector corresponding to the original logistics features, and use it as the logistics features of the logistics package.

[0012] Optionally, the historical circulation status includes abnormal status and normal status. Determining whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package includes:

[0013] Identify target historical packages whose historical circulation status is abnormal, and obtain the ratio of target historical packages in abnormal status to the total number of target historical packages;

[0014] When the ratio exceeds the ratio threshold, the current circulation status of the logistics package is determined to be abnormal.

[0015] Optionally, before obtaining the logistics characteristics and historical transit status of each of the multiple historical packages, the method further includes:

[0016] Obtain the package type corresponding to the logistics package, and obtain the package type corresponding to each of the multiple candidate packages;

[0017] Multiple candidate packages whose corresponding package type matches the package type of the logistics package are identified and used as the multiple historical packages.

[0018] Optionally, obtaining the package type corresponding to each of the multiple candidate packages includes:

[0019] Obtain key-value pairs corresponding to multiple candidate packages; the key-value pairs store the package type of the candidate packages;

[0020] For each candidate package, determine the package type based on the key-value pair corresponding to that candidate package.

[0021] Optionally, it also includes:

[0022] Obtain the package identifier and circulation status corresponding to the candidate package, and obtain the logistics characteristics of the candidate package at multiple business nodes;

[0023] Based on the logistics characteristics of the candidate packages, the package type corresponding to the candidate packages is determined;

[0024] Using the package identifier and package type corresponding to the candidate package, a key corresponding to the candidate package is generated; and using the logistics characteristics and circulation status corresponding to the candidate package, a value corresponding to the candidate package is generated.

[0025] Based on the key and value corresponding to the candidate package, determine the key-value pair corresponding to the candidate package.

[0026] Optionally, obtaining the logistics characteristics of candidate packages at multiple business nodes includes:

[0027] For each business node, obtain logistics features of the same type for multiple candidate packages, and obtain the vectors corresponding to each logistics feature of the same type; wherein, the multiple logistics features of the same type are discrete logistics features.

[0028] Multiple vectors are processed into a continuous form, and each vector after continuous processing is standardized. The standardized result is then used as the second feature vector of the corresponding candidate package.

[0029] Optionally, determining the parcel type corresponding to the candidate parcel based on its logistics characteristics includes:

[0030] Obtain a preset identification feature type; the identification feature type is a feature type used to identify package type;

[0031] From the logistics features corresponding to the candidate packages, obtain the logistics features that match the identification feature type, and use them as the package type corresponding to the candidate packages.

[0032] A device for identifying abnormal status of logistics packages, the device comprising:

[0033] The logistics feature acquisition module is used to acquire the logistics features of logistics packages at business nodes;

[0034] The historical tracking status acquisition module is used to acquire the logistics characteristics and historical tracking status of multiple historical packages.

[0035] The comparison module is used to compare the logistics characteristics of the logistics package with the logistics characteristics of the historical package to obtain the feature similarity, and to obtain the target historical package from multiple historical packages based on the feature similarity.

[0036] The circulation status identification module is used to determine whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package.

[0037] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method as described in any of the preceding claims.

[0038] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as described in any of the preceding claims.

[0039] The aforementioned method, apparatus, computer equipment, and storage medium for identifying abnormal states of logistics parcels acquire logistics characteristics of the parcels at business nodes, obtain the corresponding logistics characteristics and historical circulation status of multiple historical parcels, compare the logistics characteristics of the current parcels with those of the historical parcels to obtain feature similarity, and based on the feature similarity, select a target historical parcel from the multiple historical parcels. Based on the historical circulation status corresponding to the target historical parcel, determine whether the current circulation status of the logistics parcel is abnormal. This method can comprehensively consider the logistics characteristics of business nodes and the historical circulation status of past logistics parcels, enabling timely monitoring and early warning of abnormal circulation status of logistics parcels. It transforms the passive discovery of circulation anomalies after the fact into timely and proactive discovery during the event or early prediction before the event, effectively improving the efficiency of early warning. Attached Figure Description

[0040] Figure 1 This is an application environment diagram of a method for identifying abnormal states of logistics parcels in one embodiment;

[0041] Figure 2 This is a flowchart illustrating a method for identifying abnormal states of logistics parcels in one embodiment;

[0042] Figure 3 This is a flowchart illustrating the steps for generating key-value pairs in one embodiment;

[0043] Figure 4 This is a schematic diagram of the architecture of a method for identifying abnormal states of logistics parcels in one embodiment;

[0044] Figure 5 This is a flowchart illustrating a method for identifying abnormal states of logistics parcels in another embodiment;

[0045] Figure 6 This is a structural block diagram of a logistics package abnormality identification device in one embodiment;

[0046] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] This application provides a method for identifying abnormal states of logistics parcels, which can be applied to, for example... Figure 1In the application environment shown, server 102 can communicate with business system 104, which can record the logistics characteristics of packages. Server 102 and / or business system 104 can be implemented using independent servers or a server cluster consisting of multiple servers.

[0049] In one embodiment, such as Figure 2 As shown, a method for identifying abnormal states of logistics packages is provided, which can be applied to... Figure 1 Taking server 102 as an example, the following steps may be included:

[0050] Step 201: Obtain the logistics characteristics of the logistics package at the business node.

[0051] As an example, a business node can correspond to a logistics business link. The logistics characteristics of a business node can characterize the logistics status and / or logistics business attributes of a logistics package in the corresponding logistics business link. There can be one or more business nodes. When a logistics package has been transferred through multiple logistics business links, the logistics characteristics of a business node can include logistics characteristics of different dimensions of the logistics package, such as the logistics characteristics of the logistics package in various logistics business links such as pickup and transfer.

[0052] In practical applications, servers can monitor logistics packages. During monitoring, they can obtain the logistics characteristics of the packages at business nodes. The monitored packages may be those that have not yet arrived at their destination or have not been signed for by the user. Business nodes may include at least one of historical business nodes and current business nodes. Historical business nodes may be the business nodes corresponding to the logistics business links that the packages have already been transferred or completed, while current business nodes may be the business nodes corresponding to the logistics business links that the packages are currently in.

[0053] Step 202: Obtain the logistics characteristics and historical transfer status of each of the multiple historical packages.

[0054] As an example, a historical package can be a package for which the logistics task has been completed.

[0055] In practical implementation, since the logistics task of the historical package has been completed, the circulation status of the historical package in the logistics task can be obtained, that is, the historical circulation status. The historical circulation status can be information that represents whether the historical package has circulated normally in the logistics business process.

[0056] After obtaining the historical circulation status, the logistics characteristics and historical circulation status of the historical packages can be stored in a preset storage module, such as a file system. Then, when the server monitors the logistics packages, it can obtain the logistics characteristics and historical circulation status of each of the multiple historical packages from the storage module.

[0057] Step 203: Compare the logistics characteristics of the logistics package with the logistics characteristics of the historical package to obtain the feature similarity, and obtain the target historical package from multiple historical packages based on the feature similarity.

[0058] After obtaining the logistics characteristics of the historical packages, the logistics characteristics of the current packages can be compared with those of the historical packages to determine the degree of similarity between the logistics characteristics of the current packages and the logistics characteristics of the historical packages. This yields the feature similarity score, and the target historical packages can be selected from multiple historical packages based on the feature similarity score.

[0059] In practice, similarity conditions can be preset. These conditions can be either a preset number of historical packages with the highest feature similarity or historical packages whose feature similarity meets a similarity threshold. After determining the feature similarity, the target historical package whose feature similarity meets the similarity condition can be obtained from multiple historical packages.

[0060] Step 204: Determine whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package.

[0061] Once the target historical package is identified, its current status can be determined based on its historical transit status. Specifically, since the target historical package and the current package meet a similarity condition, the current transit status of the current package can be predicted based on the historical transit status of the target historical package, thus determining whether the current transit of the current package is normal or abnormal.

[0062] In this embodiment, by acquiring the logistics characteristics of a logistics package at a business node, acquiring the logistics characteristics and historical circulation status of multiple historical packages, comparing the logistics characteristics of the logistics package with those of the historical packages to obtain feature similarity, and based on the feature similarity, obtaining a target historical package from multiple historical packages, and determining whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package, this method can comprehensively consider the logistics characteristics of the business node and the historical circulation status of past logistics packages to achieve timely monitoring and early warning of abnormal circulation status of logistics packages. This transforms the passive discovery of circulation anomalies after the fact into timely proactive discovery during the event or early prediction before the event, effectively improving the efficiency of early warning.

[0063] In one embodiment, obtaining the logistics characteristics of a logistics package at a business node may include the following steps:

[0064] Receive the data stream sent by the business system and obtain the original logistics characteristics from the data stream; obtain the first feature vector corresponding to the original logistics characteristics as the logistics characteristics of the logistics package.

[0065] As an example, a business system can be used to collect logistics characteristics of logistics packages at multiple business nodes. There can be multiple business systems, each of which can monitor one or more logistics business links of the logistics package and obtain the logistics characteristics of the logistics package at the corresponding business node.

[0066] Specifically, the flow of a logistics parcel from pickup to delivery involves multiple logistics business stages. These stages can be monitored by multiple business systems, such as a last-mile delivery system, a parcel transportation system, and a site transfer system, which monitor the pickup, transportation, and transfer stages respectively. However, anomalies in logistics parcels can occur at any stage of the logistics process. Once an anomaly occurs, the difficulty in inter-system communication makes it challenging to promptly identify the cause.

[0067] Based on this, in this embodiment, the business system can send the original logistics characteristics of the logistics package to the server via data stream, enabling the server to receive the logistics characteristics of the logistics package at multiple business nodes. Specifically, a stream processing platform can be pre-configured, which can be a high-throughput distributed publish-subscribe messaging system, such as the Kafka stream processing platform. After each business system collects the original logistics characteristics of the logistics package from different dimensions, it can send them to the stream processing platform in the form of data streams, and the stream processing platform will forward the data streams to the server.

[0068] After receiving the data stream, the server can extract the original logistics characteristics of the logistics package from the data stream, and based on the original logistics characteristics, obtain the corresponding first feature vector as the logistics characteristics of the logistics package.

[0069] In this embodiment, by receiving the data stream sent by the business system and obtaining the original logistics characteristics from the data stream, the first feature vector corresponding to the original logistics characteristics is obtained as the logistics characteristics of the logistics package. This can summarize the logistics characteristics of each logistics business link, thereby improving the efficiency of problem location, timely detection of abnormal logistics package circulation status, and, based on the data stream processing method, monitoring the circulation status of massive logistics packages, greatly improving data processing efficiency and providing a computational basis for quickly and timely detection of abnormal circulation status.

[0070] In one embodiment, the historical circulation status may include abnormal status and normal status. Determining whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package may include the following steps:

[0071] Identify target historical packages whose historical circulation status is abnormal, and obtain the ratio of target historical packages in abnormal status to the total number of target historical packages; when the ratio exceeds the ratio threshold, determine that the current circulation status of the logistics package is abnormal.

[0072] The target historical parcel total is the total number of target historical parcels.

[0073] In practical applications, for each target historical package, the historical circulation status of each package can be used to identify those with abnormal circulation status. Then, the total number of target historical packages can be obtained, along with the ratio between the abnormal packages and the total number of target historical packages. After determining this ratio, it can be checked whether it exceeds a preset threshold, such as 0.5. If the ratio exceeds the threshold, the current circulation status of the package is determined to be abnormal; if the ratio does not exceed the threshold, the current circulation status of the package is determined to be normal.

[0074] In this embodiment, by obtaining the ratio of the target historical package in an abnormal state to the total number of target historical packages, when the ratio exceeds the ratio threshold, it is determined that the current circulation status of the logistics package is abnormal, which can timely determine or predict the circulation status of the logistics package.

[0075] In one embodiment, prior to the step of obtaining the logistics characteristics and historical transit status of each of the multiple historical packages, the method further includes:

[0076] Obtain the package type corresponding to the logistics package, and obtain the package type corresponding to each of the multiple candidate packages; determine the multiple candidate packages whose corresponding package type matches the package type of the logistics package, and use them as the multiple historical packages.

[0077] As an example, a candidate package could be a logistics package for which the logistics task has been completed.

[0078] In practical applications, packages can be pre-categorized to obtain multiple package types. When monitoring logistics packages, the package type corresponding to the logistics package can be obtained, as well as the package type corresponding to multiple candidate packages. After obtaining the package type corresponding to each candidate package, the multiple candidate packages that match the package type of the logistics package can be treated as historical packages.

[0079] In one example, since the number of historical packages increases with the size of the historical sample, to improve server processing efficiency, a preset number of candidate packages matching the package type can be included as historical packages. This preset number can be determined based on data processing speed requirements or the server's data processing capabilities. For instance, when the server processes logistics features from various business systems using a data stream approach, and has high data processing speed requirements, the preset number could be 20,000.

[0080] In this embodiment, by identifying multiple candidate packages whose corresponding package type matches the package type of the logistics package, and using them as multiple historical packages, the scope of historical packages can be effectively narrowed, data processing efficiency can be improved, and the circulation status of logistics packages can be determined in a timely manner.

[0081] In one embodiment, such as Figure 3 As shown, the method may further include the following steps:

[0082] Step 301: Obtain the package identifier and circulation status corresponding to the candidate package, and obtain the logistics characteristics of the candidate package at multiple business nodes in the system.

[0083] As an example, the logistics features corresponding to a candidate package may include logistics features describing the static attributes of the candidate package, and logistics features describing the dynamic attributes of the candidate package.

[0084] Among them, logistics features describing static attributes can also be called static attribute data. These are the initial characteristics of logistics packages, such as the origin, destination, product type, timeliness type, weight, and logistics cost of the package.

[0085] Logistics characteristics describing dynamic attributes can also be called dynamic behavioral data. This can be information describing which business stage (also known as the parcel lifecycle) a package is in and the business characteristics of the current stage. For example, the departure time of the current business stage, the duration of the current business stage, the number of sites the current business stage passes through, whether the business stage follows the planned route, whether it departs according to the planned schedule, and whether it arrives according to the planned schedule. The business stages can include the receiving stage, the transit stage, and the delivery stage.

[0086] In practice, candidate packages can have a package identifier that uniquely identifies the package. During the processing of the logistics task, the business system can record the corresponding logistics characteristics of the candidate package. When the logistics task is completed, the flow status of the candidate package in the logistics task can be recorded.

[0087] The server can then obtain the package identifier and circulation status of the candidate package within a preset time period, and obtain the logistics characteristics of the candidate package at multiple business nodes during the circulation process from the business system.

[0088] Step 302: Determine the package type corresponding to the candidate package based on the logistics characteristics corresponding to the candidate package.

[0089] After obtaining the logistics characteristics of the candidate packages, the package type can be determined based on one or more of these logistics characteristics.

[0090] Step 303: Using the package identifier and package type corresponding to the candidate package, generate the key corresponding to the candidate package; and using the logistics characteristics and circulation status corresponding to the candidate package, generate the value corresponding to the candidate package.

[0091] Step 304: Determine the key-value pair corresponding to the candidate package based on the key and value corresponding to the candidate package.

[0092] In this embodiment, the logistics characteristics of candidate packages can be set to key-value pair format, i.e., "key / value" format.

[0093] In practical applications, the package identifier and package type corresponding to the candidate package can be used to generate the key corresponding to the package. For example, the package type can be used as a prefix, and the package identifier UUID corresponding to the candidate package can be concatenated with it. The concatenation result "prefix|UUID" can be used as the key corresponding to the candidate package.

[0094] For the value in the key-value pair, the logistics features and circulation status of the candidate package can be used to generate it. Specifically, the circulation status can be a circulation status label, where abnormal circulation can be represented by "0" and normal circulation by "1". Then, the logistics feature "X" corresponding to the candidate package is concatenated with the circulation status "Label" to obtain the corresponding value, i.e., value = X|Label.

[0095] After obtaining the key and value corresponding to the candidate package, key-value pairs can be generated using the key and value, and then the key-value pairs can be saved in memory.

[0096] In this embodiment, key-value pairs can be used to store the logistics features and other information related to candidate packages, which can effectively save storage space and provide a data foundation for quickly matching historical packages in the future.

[0097] In one embodiment, obtaining the package type corresponding to each of the multiple candidate packages may include the following steps:

[0098] Obtain key-value pairs corresponding to multiple candidate packages; the key-value pairs store the package type of the candidate package; for each candidate package, determine the package type based on the key-value pair corresponding to the candidate package.

[0099] In the specific implementation, the key-value pairs corresponding to each candidate package can be pre-stored in memory. When the server filters historical packages, it can obtain the key-value pairs corresponding to each candidate package. The key-value pairs store the package type corresponding to the candidate package. Then, by obtaining the key-value pairs corresponding to the candidate packages, the package type corresponding to the candidate packages can be determined.

[0100] In one example, when multiple candidate packages whose corresponding package types match the package types of logistics packages are identified as multiple historical packages, the package types corresponding to the logistics packages and candidate packages can be used as prefixes of keys and stored in key-value pairs. Key-value pairs with the same prefix can be matched using the prefix corresponding to the logistics packages, thereby identifying multiple historical packages. The key-value pairs corresponding to multiple historical packages can be stored in a list.

[0101] In this embodiment, the package type corresponding to the candidate package can be quickly obtained, which effectively improves the data processing speed and provides a basis for quickly matching historical packages.

[0102] In one embodiment, obtaining the logistics characteristics of candidate packages at multiple business nodes includes:

[0103] For each business node, obtain logistics features of the same type for multiple candidate packages, and obtain the vectors corresponding to each logistics feature of the same type; perform continuous processing on multiple vectors, and perform data standardization on each vector after continuous processing, and determine the data standardization result as the second feature vector of the corresponding candidate package.

[0104] Among them, multiple logistics features of the same type are discrete logistics features.

[0105] In practical applications, for multiple candidate packages corresponding to the same business node, logistics features with the same characteristic type can be obtained, and their corresponding vectors can be determined. After obtaining the vectors corresponding to the logistics features with the same characteristic type, the multiple vectors can be processed into a continuous vector. Simultaneously, since the logistics features have different dimensions, the multiple vectors after continuous processing can be standardized. The standardized data result of each vector can then be determined as the second feature vector of the corresponding candidate package. Data standardization can be performed using the following formula:

[0106]

[0107] Where, x iLet x be a component in the second eigenvector, and u be multiple components x. i The corresponding mean, σ, represents multiple components x. i The corresponding variance. Data scaling can be achieved through data standardization, for example, scaling logistics features of the same type to have a mean of 0 and a variance of 1.

[0108] In one example, the process of obtaining the first feature vector corresponding to the original logistics feature is the same as that of obtaining logistics features of the same feature type as candidate packages. Specifically, for the original logistics feature X, and the logistics feature of the same feature type as the candidate package is logistics feature Y, then when obtaining the first feature vector, we can obtain the vector x corresponding to logistics feature X, where the length of vector x is the same as the length of vector y corresponding to logistics feature Y. After obtaining vector x, we can perform continuous processing and data standardization on vector x to obtain the first feature vector.

[0109] In this embodiment, for each business node, logistics features of the same type for multiple candidate packages can be obtained, and vectors corresponding to the logistics features of the same type can be obtained. The multiple vectors are processed into a continuous vector, and each vector after continuous processing is standardized. The standardized data result is determined as the second feature vector of the corresponding candidate package. This can effectively improve the accuracy, reliability and measurability of the second feature vector, and provide a basis for the accurate calculation of feature similarity.

[0110] In one embodiment, comparing the logistics characteristics of the logistics package with the logistics characteristics of the historical packages to obtain feature similarity may include the following steps:

[0111] Obtain the Euclidean distance between the first feature vector and the second feature vector, and use the Euclidean distance as the feature similarity.

[0112] After obtaining the first feature vector, the Euclidean distance between the first and second feature vectors can be calculated, and this Euclidean distance can be used as the feature similarity. The calculation can be performed using the formula shown below:

[0113]

[0114] Where Y is the first eigenvector, X is the second eigenvector, and x... i y is a component in the second eigenvector. i These are the components in the first eigenvector.

[0115] In this embodiment, by obtaining the Euclidean distance between the first feature vector and the second feature vector as feature similarity, the similarity between the logistics package and the historical package can be quantified, providing a basis for subsequently determining the circulation status of the logistics package.

[0116] In one embodiment, determining the package type corresponding to the candidate package based on the logistics characteristics of the candidate package may include the following steps:

[0117] Obtain a preset identification feature type; from the logistics features corresponding to the candidate package, obtain the logistics features that match the identification feature type, and use them as the package type corresponding to the candidate package.

[0118] Among them, the identification feature type can be a logistics feature type used to identify the type of package.

[0119] In practical implementation, a candidate package can have multiple logistics features of different types. From these multiple feature types, a feature type can be pre-determined as the identification feature type. After obtaining the logistics features corresponding to the candidate package, the server can obtain the pre-defined identification feature type, and then extract the logistics features from the candidate package's logistics features that match the identification feature type, which will be used as the package type of the candidate package.

[0120] For example, if the identification feature type can be the package product type and the package timeliness type, then the specific logistics features corresponding to the package product type and the package timeliness type can be obtained as the package type.

[0121] In this embodiment, logistics features that match the identification feature type can be obtained from the logistics features corresponding to the candidate package, and these can be used as the package type corresponding to the candidate package. This allows for the use of preset standards to distinguish the types of candidate packages.

[0122] In one embodiment, the method may further include the following steps:

[0123] When it is determined that the circulation status of the logistics package is abnormal, an alarm message is generated; the alarm message is sent to the monitoring terminal associated with the logistics package.

[0124] As an example, the monitoring terminal could be a handheld terminal device used by operators performing logistics tasks.

[0125] In practical applications, when an abnormality is detected in the flow of a logistics package, an alarm message can be generated and sent to the monitoring terminal associated with the package, prompting relevant operators to conduct an anomaly check. When operators confirm the anomaly, they can resolve it promptly and report the cause of the anomaly to the server. If the check determines that the anomaly does not exist, a false alarm reason can be uploaded to the server. The server can persist the received false alarm reasons and subsequently remove data based on these reasons.

[0126] In this embodiment, when an abnormal flow status of a logistics package is determined, an alarm message is generated and sent to the monitoring terminal associated with the logistics package. This allows relevant personnel to be notified in a timely manner to resolve the abnormality, thereby ensuring the timeliness of logistics transportation and reducing transportation costs.

[0127] To enable those skilled in the art to better understand the above steps, the following example illustrates the embodiments of this application, but it should be understood that the embodiments of this application are not limited thereto.

[0128] Taking a server as a stream processing system as an example, this embodiment can adopt the following... Figure 4 The framework shown includes a data source module, a stream processing system, and a data service module.

[0129] Specifically, the data source module is a replayable Kafka stream processing platform. The stream processing platform can receive streaming data from various business systems, classify the streaming data according to the topic (type), and send the streaming data to the stream processing system.

[0130] After receiving streaming data, the stream processing system can first store the data in a data warehouse tool (such as Hive) for batch processing, define the computational logic, and then use the dataset in the data warehouse tool to create a model. This model can provide real-time anomaly monitoring and alarm services and pre-event anomaly identification services for logistics packages. The real-time anomaly monitoring and alarm services and pre-event anomaly identification services can be monitored and identified using the abnormal status identification methods for logistics packages described above.

[0131] In actual monitoring, the received data stream can be vectorized, processed continuously, and standardized. The logistics packages can be monitored through a pre-built model. At the same time, the processed data can be stored in the storage module.

[0132] For stream processing systems, such as Figure 5 As shown, after receiving a data stream, the stream processing system can save the data stream in the file system and train an anomaly detection model offline. When monitoring logistics packages, the pre-event anomaly detection model can be loaded and combined with the in-event business anomaly calculation logic to calculate the received data stream, determine in-event or pre-event anomalies, generate alarm information, and push it to relevant systems and operator terminals. After receiving the alarm information, operators can check whether there are any anomalies in the logistics packages. If so, they can resolve the anomaly and upload the cause of the anomaly to the stream processing system; if not, they can determine the cause of the false alarm and upload it to the stream processing system. The stream processing system can store the cause of the anomaly and the cause of the false alarm in the file system.

[0133] It should be understood that, although Figure 1-5The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1-5 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0134] In one embodiment, such as Figure 6 As shown, a device for identifying abnormal status of logistics packages is provided, the device may include:

[0135] The logistics feature acquisition module 601 is used to acquire the logistics features of logistics packages at business nodes;

[0136] The historical circulation status acquisition module 602 is used to acquire the logistics characteristics and historical circulation status of each of the multiple historical packages.

[0137] The comparison module 603 is used to compare the logistics characteristics of the logistics package and the logistics characteristics of the historical package to obtain the feature similarity, and to obtain the target historical package from multiple historical packages based on the feature similarity.

[0138] The circulation status identification module 604 is used to determine whether the current circulation status of the logistics package is abnormal based on the historical circulation status corresponding to the target historical package.

[0139] In one embodiment, the logistics feature acquisition module 601 includes:

[0140] The data stream acquisition submodule is used to receive the data stream sent by the business system and obtain the original logistics characteristics from the data stream; the business system is used to collect the logistics characteristics of logistics packages at multiple business nodes;

[0141] The first feature vector acquisition submodule is used to acquire the first feature vector corresponding to the original logistics features, as the logistics features of the logistics package.

[0142] In one embodiment, the historical flow status includes abnormal status and normal status, and the flow status identification module 604 includes:

[0143] The ratio calculation submodule is used to identify target historical packages with an abnormal historical circulation status and obtain the ratio of the target historical packages with the abnormal status to the total number of target historical packages.

[0144] The abnormal status identification submodule is used to determine that the current circulation status of the logistics package is abnormal when the ratio exceeds the ratio threshold.

[0145] In one embodiment, the apparatus further includes:

[0146] The package type acquisition module is used to acquire the package type corresponding to the logistics package, and to acquire the package type corresponding to each of the multiple candidate packages;

[0147] The historical package determination module is used to determine multiple candidate packages whose corresponding package type matches the package type of the logistics package, and to identify them as the multiple historical packages.

[0148] In one embodiment, the package type acquisition module includes:

[0149] The key-value pair retrieval submodule is used to retrieve the key-value pairs corresponding to multiple candidate packages; the key-value pairs store the package type of the candidate packages;

[0150] The key-value pair identification submodule is used to determine the package type for each candidate package based on the key-value pair corresponding to that candidate package.

[0151] In one embodiment, the apparatus further includes:

[0152] The package identifier acquisition module is used to acquire the package identifier and circulation status corresponding to the candidate package, and to acquire the logistics characteristics of the candidate package at multiple business nodes;

[0153] The package type determination module is used to determine the package type corresponding to the candidate package based on the logistics characteristics corresponding to the candidate package;

[0154] The key-value acquisition module is used to generate a key corresponding to the candidate package using the package identifier and package type corresponding to the candidate package, and to generate a value corresponding to the candidate package using the logistics characteristics and circulation status corresponding to the candidate package.

[0155] The key-value pair generation module is used to determine the key-value pair corresponding to the candidate package based on the key and value corresponding to the candidate package.

[0156] In one embodiment, the package identifier acquisition module includes:

[0157] The vector acquisition submodule is used to acquire logistics features of the same type for multiple candidate packages for each business node, and to acquire the vectors corresponding to each logistics feature of the same type; wherein, the multiple logistics features of the same type are discrete logistics features.

[0158] The second feature vector acquisition submodule is used to perform continuous processing on multiple vectors, and to standardize the data of each vector after continuous processing. The data standardization result is determined as the second feature vector of the corresponding candidate package.

[0159] In one embodiment, the root package type determination module includes:

[0160] The feature type acquisition submodule is used to acquire a preset feature type; the feature type is a feature type used to identify package type.

[0161] The matching submodule is used to obtain logistics features that match the identification feature type from the logistics features corresponding to the candidate package, and use them as the package type corresponding to the candidate package.

[0162] In one embodiment, the apparatus further includes:

[0163] The alarm information generation module is used to generate alarm information when it is determined that the circulation status of the logistics package is abnormal;

[0164] The alarm information sending module is used to send the alarm information to the monitoring terminal associated with the logistics package.

[0165] For specific limitations regarding a device for identifying abnormal states of logistics packages, please refer to the limitations of a method for identifying abnormal states of logistics packages described above, which will not be repeated here. Each module in the aforementioned device for identifying abnormal states of logistics packages can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0166] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data streams obtained from business systems. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a method for identifying abnormal states of logistics packages.

[0167] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0168] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0169] Obtain logistics characteristics of logistics packages in multiple business systems; the multiple business systems are used to collect logistics characteristics of the logistics packages in different dimensions;

[0170] Obtain the logistics characteristics and historical transit status of each of the multiple historical packages;

[0171] The logistics characteristics of the logistics package and the logistics characteristics of the historical package are compared to obtain the feature similarity, and the target historical package whose feature similarity meets the preset similarity condition is obtained from multiple historical packages.

[0172] Based on the historical circulation status of the target historical package, determine whether the current circulation status of the logistics package is abnormal.

[0173] In one embodiment, the processor also performs the steps described in the other embodiments when executing the computer program.

[0174] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0175] Obtain logistics characteristics of logistics packages in multiple business systems; the multiple business systems are used to collect logistics characteristics of the logistics packages in different dimensions;

[0176] Obtain the logistics characteristics and historical transit status of each of the multiple historical packages;

[0177] The logistics characteristics of the logistics package and the logistics characteristics of the historical package are compared to obtain the feature similarity, and the target historical package whose feature similarity meets the preset similarity condition is obtained from multiple historical packages.

[0178] Based on the historical circulation status of the target historical package, determine whether the current circulation status of the logistics package is abnormal.

[0179] In one embodiment, the computer program, when executed by a processor, also implements the steps described in the other embodiments above.

[0180] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0181] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0182] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for identifying abnormal status of logistics parcels, characterized in that, The method includes: Obtain the logistics characteristics of the logistics package itself at the business node; the logistics characteristics include features describing the static attributes of the package and features describing the dynamic attributes of the package; the features describing the static attributes of the package include the product type of the package, the timeliness type of the package, the weight of the package, and the logistics cost of the package; the features describing the dynamic attributes of the package are used to describe the current business stage of the logistics package and the business characteristics of the logistics package in the current business stage. Based on the keys in the key-value pairs corresponding to the candidate packages, the package type of the candidate packages is determined. The number of target packages N is determined according to the data processing speed requirements or data processing capabilities. The N candidate packages whose corresponding package types match the package type of the logistics package are taken as multiple historical packages. Obtain the logistics characteristics and historical circulation status of each of the multiple historical packages; the historical circulation status of each historical package includes a normal status or an abnormal status; The feature similarity is obtained based on the Euclidean distance between the first feature vector corresponding to the logistics characteristics of the logistics package and the second feature vector corresponding to the logistics characteristics of the historical package, and the target historical package is obtained from multiple historical packages based on the feature similarity. Identify the target historical package whose historical circulation status is in the abnormal state, and obtain the ratio of the target historical package in the abnormal state to the total number of target historical packages. Based on the ratio, determine whether the current circulation status of the logistics package is abnormal.

2. The method according to claim 1, characterized in that, The acquisition of logistics characteristics of logistics packages at business nodes includes: The system receives data streams sent by the business system and extracts raw logistics characteristics from the data streams; the business system is used to collect logistics characteristics of logistics packages at multiple business nodes. Obtain the first feature vector corresponding to the original logistics features, and use it as the logistics features of the logistics package.

3. The method according to claim 1, characterized in that, Determining whether the current circulation status of the logistics package is abnormal based on the ratio includes: When the ratio exceeds the ratio threshold, the current circulation status of the logistics package is determined to be abnormal.

4. The method according to claim 1, characterized in that, Before determining the package type of the candidate packages based on the keys in the key-value pairs corresponding to each of the multiple candidate packages, the method further includes: Retrieve the key-value pairs corresponding to each of the multiple candidate packages.

5. The method according to claim 4, characterized in that, Also includes: Obtain the package identifier and circulation status corresponding to the candidate package, and obtain the logistics characteristics of the candidate package at multiple business nodes; Based on the logistics characteristics of the candidate packages, the package type corresponding to the candidate packages is determined; Using the package identifier and package type corresponding to the candidate package, a key corresponding to the candidate package is generated; and using the logistics characteristics and circulation status corresponding to the candidate package, a value corresponding to the candidate package is generated. Based on the key and value corresponding to the candidate package, determine the key-value pair corresponding to the candidate package.

6. The method according to claim 5, characterized in that, The step of obtaining the logistics characteristics of the candidate package at multiple business nodes includes: For each business node, obtain logistics features of the same type for multiple candidate packages, and obtain the vectors corresponding to each logistics feature of the same type; wherein, the multiple logistics features of the same type are discrete logistics features. Multiple vectors are processed into a continuous form, and each vector after continuous processing is standardized. The standardized result is then used as the second feature vector of the corresponding candidate package.

7. The method according to claim 5, characterized in that, The step of determining the package type corresponding to the candidate package based on the logistics characteristics of the candidate package includes: Obtain a preset identification feature type; the identification feature type is a feature type used to identify package type; From the logistics features corresponding to the candidate packages, obtain the logistics features that match the identification feature type, and use them as the package type corresponding to the candidate packages.

8. A device for identifying abnormal status of logistics parcels, characterized in that, The device includes: The logistics feature acquisition module is used to acquire the logistics features of the logistics package itself at the business node; the logistics features include features describing the static attributes of the package and features describing the dynamic attributes of the package; the features describing the static attributes of the package include the product type of the package, the timeliness type of the package, the weight of the package, and the logistics cost of the package; the features describing the dynamic attributes of the package are used to describe the current business stage of the logistics package and the business characteristics of the logistics package in the current business stage. The historical flow status acquisition module is used to determine the package type of the candidate packages based on the keys in the key-value pairs corresponding to each candidate package, determine the target package quantity N according to the data processing speed requirements or data processing capabilities, and take the N candidate packages whose corresponding package types match the package type of the logistics package as multiple historical packages; acquire the logistics characteristics and historical flow status of each of the multiple historical packages; the historical flow status of each historical package includes a normal status or an abnormal status; The comparison module is used to obtain feature similarity based on the Euclidean distance between the first feature vector corresponding to the logistics features of the logistics package and the second feature vector corresponding to the logistics features of the historical package, and to obtain the target historical package from multiple historical packages based on the feature similarity. The circulation status identification module is used to identify target historical packages whose historical circulation status is the abnormal status, and to obtain the ratio of the target historical packages in the abnormal status to the total number of target historical packages, and to determine whether the current circulation status of the logistics package is abnormal based on the ratio.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images