Logistics order and exception processing method and device, equipment and storage medium

Abstract

The invention relates to the field of logistics transportation, and discloses a logistics order abnormity processing method and device, equipment and a storage medium. The method comprises the steps of obtaining feature data of a logistics order, wherein the feature data of the logistics order comprises order identification information, logistics track information, timestamp information and parcel state information; determining the initial state of the logistics order based on the feature data of the logistics order, and obtaining an initial state judgment result; according to the initial state judgment result, classifying the logistics order into a logistics order in a normal state or a logistics order in an abnormal state; and performing order association analysis on the logistics order in the abnormal state, processing the logistics order in the abnormal state according to an analysis result, and generating a processing result. According to the logistics order exception handling method provided by the invention, a whole-process intelligent closed loop from data acquisition, state judgment, classification management to exception handling is formed, and the logistics operation efficiency and the user experience are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of logistics and transportation technology, and in particular to a logistics order exception processing method, device, equipment and storage medium. Background Art

[0002] As an innovative feature within the e-commerce platform's logistics service system, "Miao Miao Cha" offers a core value beyond simply providing forecasts of logistics track updates. It also establishes an efficient and accurate logistics information exchange platform. By monitoring every step of the logistics network in real time, including sorting, transportation, and delivery, this feature can quickly and accurately predict and notify e-commerce platforms and users of the next logistics track update, significantly improving the timeliness and predictability of logistics information.

[0003] However, faced with growing logistics volumes and a complex and volatile transportation environment, the existing trajectory prediction methods relied upon by Miaomiaocha have gradually exposed limitations due to their inaccurate accuracy. Traditional prediction models often rely on historical data analysis, but their predictive capabilities are limited when faced with unexpected events such as package loss or damage, or inclement weather affecting delivery speeds. These unforeseen events not only disrupt logistics plans but also prevent the platform from providing users with timely and accurate information on logistics updates, impacting user experience and the e-commerce platform's credibility.

[0004] It can be seen that the existing logistics status judgment logic has significant flaws. When determining whether logistics are normal, it fails to fully utilize the latest logistics trajectory information, resulting in a high rate of misjudgment of logistics anomalies and an inability to accurately reflect the true status of logistics. This not only affects users' trust in logistics information, but also increases the cost of ineffective verification for logistics companies. Summary of the Invention

[0005] The main purpose of this invention is to improve the accuracy and reliability of logistics status judgment, reduce misjudgments, and optimize the logistics information feedback mechanism.

[0006] A first aspect of the present invention provides a method for handling logistics order anomalies, comprising: obtaining characteristic data of a logistics order, the characteristic data of the logistics order including order identification information, logistics track information, timestamp information, and package status information; based on the characteristic data of the logistics order, confirming the initial status of the logistics order and obtaining an initial status judgment result; according to the initial status judgment result, classifying the logistics order as a logistics order with normal status or a logistics order with abnormal status; performing order association analysis on the logistics order with abnormal status, processing the logistics order with abnormal status according to the analysis result, and generating a processing result.

[0007] Optionally, in a first implementation method of the first aspect of the present invention, the order identification information of the logistics order is analyzed to determine whether the logistics entity corresponding to the logistics order matches the preset target logistics entity; if not, an initial state judgment result of the abnormal logistics entity is generated; if matched, the logistics track information, order identification information and timestamp information of the logistics order are analyzed to identify the logistics status of the logistics order, and a corresponding initial state judgment result is generated based on the logistics status.

[0008] Optionally, in a second implementation method of the first aspect of the present invention, based on the logistics track information of the logistics order, whether it belongs to the collection status is identified according to the preset stage identification rules, and a corresponding collection logistics status is generated according to the identification result; based on the order identification information of the logistics order, the logistics order is divided into international pieces and non-international pieces according to the preset time classification rules; for the international pieces, the latest scanning track timestamp is obtained based on the timestamp information, and it is identified whether it is within the preset time threshold; for the non-international pieces, it is identified as a normal time status, and a corresponding time logistics status is generated according to the identification result; based on the collection logistics status and the time logistics status, a corresponding initial status judgment result is generated.

[0009] Optionally, in a third implementation of the first aspect of the present invention, if the initial status judgment result is that the logistics entity is abnormal, the logistics order is classified as a logistics order with an abnormal status; if the initial status judgment result is any one of express collection, transportation, transit, delivery and signed for status, the logistics order is classified as a logistics order with a normal status; if the initial status judgment result is any one of collection timeout, transportation delay, delivery stagnation, transit stagnation and no track, the logistics order is classified as a logistics order with an abnormal status.

[0010] Optionally, in a fourth implementation method of the first aspect of the present invention, it is detected whether the package status information corresponding to the logistics order shows that there is damage in appearance, and whether the geographical location information corresponding to the logistics order has a target logistics site within a preset range; if the package status information shows that there is damage in appearance, or the geographical location information does not have a target logistics site within the preset range, the logistics order is determined to be a problem item; if the logistics order is determined to be a problem item, the arrival time of the logistics order is updated, and an abnormal feedback result including a description of the abnormal type is generated; if the logistics order is determined to be a non-problem item, it is identified whether it is in one of the rejected and collected states; if so, a normal feedback result including a description of the status type is generated; if not, the arrival time of the logistics order is updated and abnormal feedback is triggered.

[0011] Optionally, in a fifth implementation of the first aspect of the present invention, characteristic data of the logistics order with abnormal status is obtained; an association model is constructed, and the association relationship between the logistics order with abnormal status and other orders is analyzed through the association model to determine whether there is an associated order; if there is an associated order, the similarity between the characteristic data of the logistics order with abnormal status and the associated order is calculated, and a merge processing result or a split verification processing result is output according to the similarity result: if there is no associated order, the timestamp information of the logistics order with abnormal status is optimized, the update time is dynamically generated, and the pre-built logistics arrival reminder mechanism is triggered according to the update time, and the processing result including the cause of the abnormality and the update time is output.

[0012] Optionally, in a sixth implementation of the first aspect of the present invention, final distribution feature data of the logistics order is obtained, and the final distribution feature data of the logistics order includes distribution time, target routing information and logistics trajectory after distribution; the distribution time is compared with the preset distribution time limit to determine whether there is a distribution delay anomaly; the matching degree between the logistics trajectory after distribution and the target routing information is analyzed to determine whether there is a misassignment anomaly; if there is a distribution delay or misassignment anomaly, the corresponding anomaly cause and response strategy are generated; if the distribution is normal, the logistics order is updated to the completed final distribution status result, and an estimated delivery time is generated.

[0013] The second aspect of the present invention provides a logistics order exception processing device, including: an acquisition module for acquiring characteristic data of a logistics order; a judgment module for confirming the initial state of the logistics order based on the characteristic data of the logistics order and obtaining an initial state judgment result; a classification module for classifying the logistics order into a normal logistics order or an abnormal logistics order according to the initial state judgment result; a processing module for performing order association analysis on the abnormal logistics order, processing the abnormal logistics order according to the analysis result, and generating a processing result.

[0014] Optionally, in a first implementation method of the second aspect of the present invention, the judgment module includes: a judgment unit, used to analyze the order identification information of the logistics order, and determine whether the logistics entity corresponding to the logistics order matches the preset target logistics entity; an identification unit, used to generate an initial state judgment result of the abnormality of the logistics entity when there is no match; when there is a match, the logistics track information, order identification information and timestamp information of the logistics order are analyzed to identify the logistics state of the logistics order, and generate a corresponding initial state judgment result based on the logistics state.

[0015] Optionally, in a second implementation method of the second aspect of the present invention, the identification unit is specifically used to identify whether it belongs to the collection status based on the logistics trajectory information of the logistics order according to the preset stage identification rules, and generate a corresponding collection logistics status according to the identification result; based on the order identification information of the logistics order, the logistics order is divided into international pieces and non-international pieces according to the preset time classification rules; for the international pieces, the latest scanning trajectory timestamp is obtained based on the timestamp information to identify whether it is within the preset time threshold; for the non-international pieces, it is identified as a normal time status, and a corresponding time logistics status is generated according to the identification result; based on the collection logistics status and the time logistics status, a corresponding initial state judgment result is generated.

[0016] Optionally, in a third implementation of the second aspect of the present invention, the classification module includes: a first classification unit, for classifying the logistics order as a logistics order with an abnormal status if the initial status judgment result is that the logistics entity is abnormal; classifying the logistics order as a logistics order with an abnormal status if the initial status judgment result is any one of collection timeout, transportation delay, stagnation in delivery, stagnation in transit and no track; a second classification unit, for classifying the logistics order as a logistics order with a normal status if the initial status judgment result is any one of express collection, transportation, transit, delivery and signed for status.

[0017] Optionally, in a fourth implementation method of the second aspect of the present invention, the logistics order exception handling device further includes: a detection module for detecting whether the package status information corresponding to the logistics order shows that there is damage in appearance, and detecting whether the geographical location information corresponding to the logistics order has a target logistics site within a preset range; a logistics exception handling module for determining that the logistics order is a problem item when the package status information shows that there is damage in appearance, or the geographical location information does not have a target logistics site within a preset range; when it is determined that the logistics order is a problem item, updating the arrival time of the logistics order and generating an exception feedback result including an exception type description; when it is determined that the logistics order is not a problem item, identifying whether it is one of a rejected item and a collected status; if so, generating a normal feedback result including a status type description; if not, updating the arrival time of the logistics order and triggering exception feedback.

[0018] Optionally, in a fifth implementation of the second aspect of the present invention, the processing module includes: an abnormal data acquisition unit, used to acquire characteristic data of the logistics order with abnormal status; an association unit, used to construct an association model, and analyze the association relationship between the logistics order with abnormal status and other orders through the association model to determine whether there is an associated order; if there is an associated order, calculate the similarity between the characteristic data of the logistics order with abnormal status and the associated order, and output a merged processing result or a split verification processing result based on the similarity result: if there is no associated order, optimize the timestamp information of the logistics order with abnormal status, dynamically generate an update time, and trigger a pre-built logistics arrival reminder mechanism according to the update time, and output a processing result including the cause of the abnormality and the update time.

[0019] Optionally, in a sixth implementation of the second aspect of the present invention, the logistics order exception handling device further includes: a distribution exception handling module for obtaining the final distribution feature data of the logistics order, the final distribution feature data of the logistics order including the distribution time, target routing information and the logistics trajectory after distribution; comparing the distribution time with the preset distribution time limit to determine whether there is a distribution delay exception; analyzing the matching degree between the logistics trajectory after distribution and the target routing information to determine whether there is a misassignment exception; if there is a distribution delay or misassignment exception, generating a corresponding exception cause and response strategy; if the distribution is normal, updating the logistics order to the completed final distribution status result, and generating an estimated delivery time.

[0020] The third aspect of the present invention provides a logistics order exception processing device, comprising: a memory and at least one processor, wherein the memory stores computer-readable instructions, and the memory and the at least one processor are interconnected through a line; the at least one processor calls the computer-readable instructions in the memory so that the logistics order exception processing device executes each step of the logistics order exception processing method as described above.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the various steps of the logistics order exception processing method described above.

[0022] Beneficial effects: In the technical solution of the present invention, by obtaining logistics order feature data including order identification information, logistics track information, timestamp information, and package status information, multi-dimensional data support is provided for judging the initial status of the logistics order; based on the feature data, the initial status of the logistics order is confirmed and the initial status judgment result is generated, and potential abnormal situations such as logistics entity abnormalities and collection timeouts are identified in time, which greatly improves its efficiency compared with traditional manual screening; orders are classified as normal or abnormal according to the initial status judgment result, and automatic classification management of logistics orders is realized, so that normal orders can be quickly transferred and abnormal orders can be accurately located; order association analysis is performed on logistics orders with abnormal status, and by building an association model to determine whether there are related orders, and then output processing results such as merging processing, splitting verification or optimizing timestamps, which effectively shortens the processing time of abnormal orders, and at the same time generates processing results including the cause of the abnormality and the update time, thereby improving the transparency of logistics information and enhancing the user's predictability of the logistics status, forming an intelligent closed loop of the entire process from data collection, status judgment, classification management to exception processing, and effectively improving logistics operation efficiency and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A first flow chart of the method for handling logistics order exceptions provided by an embodiment of the present invention;

[0024] Figure 2 A second flow chart of the logistics order exception handling method provided by an embodiment of the present invention;

[0025] Figure 3 A third flow chart of the logistics order exception handling method provided by an embodiment of the present invention;

[0026] Figure 4 A fourth flow chart of the logistics order exception handling method provided by an embodiment of the present invention;

[0027] Figure 5 A fifth flow chart of the logistics order exception handling method provided in an embodiment of the present invention;

[0028] Figure 6 A sixth flow chart of the logistics order exception handling method provided by an embodiment of the present invention;

[0029] Figure 7 A seventh flow chart of the logistics order exception handling method provided in an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of the structure of a logistics order exception processing device provided by an embodiment of the present invention;

[0031] Figure 9 Another structural diagram of the logistics order exception processing device provided by an embodiment of the present invention;

[0032] Figure 10 A schematic diagram of the structure of a logistics order exception processing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The embodiments of the present invention provide a method, device, equipment and storage medium for handling logistics order exceptions, which obtains characteristic data of logistics orders, including order identification information, logistics track information, timestamp information, and package status information; based on the characteristic data of the logistics order, confirms the initial status of the logistics order and obtains an initial status judgment result; according to the initial status judgment result, classifies the logistics order as a logistics order with a normal status or a logistics order with an abnormal status; performs order association analysis on the logistics order with an abnormal status, processes the logistics order with an abnormal status according to the analysis result, and generates a processing result. The present invention solves the problem that the existing logistics orders have a high misjudgment rate of exceptions and cannot accurately feedback the true status of logistics. By forming an intelligent closed loop for the entire process of data collection, status judgment, classification management, and exception handling, it effectively improves logistics operation efficiency and user experience.

[0034] The terms "first," "second," "third," "fourth," and the like (if any) in the description and claims of the present invention and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0035] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 The first embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0036] S100: Acquire characteristic data of a logistics order, where the characteristic data includes order identification information, logistics track information, timestamp information, and package status information;

[0037] In this embodiment, the order identification information includes the waybill number, the text information of the waybill, the logistics entity code, and other data used to uniquely identify the order and the associated logistics service entity; the logistics track information includes the node data of each logistics link such as collection, transportation, transit, and distribution, such as the sorting center name and the logistics transportation route; the timestamp information is the operation time of each logistics node, such as the logistics collection time and the transit scanning time; the package status information includes the package appearance image recorded through security inspection or photography, and the geographic location data collected by the GIS geographic information system.

[0038] For example, when a user places an order through an e-commerce platform, the system automatically generates a logistics order containing a waybill number. When picking up the package, the logistics company uses a scanning device to collect the text information on the waybill, take pictures of the package's appearance, and obtain the collection location coordinates through GPS positioning. At the same time, it records the collection timestamp to form a complete set of feature data.

[0039] S200: confirming the initial state of the logistics order based on the characteristic data of the logistics order and obtaining an initial state judgment result;

[0040] In this embodiment, the natural language processing technology (NLP, Neuro-Linguistic Programming) is first used to analyze the text in the order identification information, and the pre-trained language model (BERT,

[0041] Bidirectional Encoder Representations from Transformers determines whether the logistics entity is the target courier for system integration. If there is no match, such as if the logistics provider is not a partner, an initial status judgment result of "logistics entity abnormality" is generated. If a match is found, and the provider is confirmed to be a partner for system integration, a time series data processing model is further utilized. Specifically, a long short-term memory (LSTM) network or a transformer network can be used to analyze logistics trajectory information and timestamp information to identify the logistics status of the logistics order.

[0042] For example, when the logistics entity is matched with the target express company for system docking, the collection records within the preset time are extracted. If the collection track is not detected and the time exceeds the preset threshold, such as if the collection track is still not detected for more than 48 hours, the initial state judgment result of "collection timeout" is generated; for the collected orders, the logistics track features are extracted through the convolutional neural network (CNN, Convolutional Neural Networks). If the node update frequency during transportation is lower than the preset frequency and the timestamp interval exceeds the preset threshold, such as if the node is still not updated for more than 24 hours, the initial state judgment result of "transportation delay" is generated.

[0043] S300: Classify the logistics order as a normal logistics order or an abnormal logistics order according to the initial status judgment result;

[0044] In this embodiment, if the initial status judgment result is one of the normal states such as "Express Collection," "In Transit," "In Transit," "Delivery," or "Signed for," the order is classified as a normal logistics order. If the initial status judgment result is one of the abnormal states such as "Logistics Entity Abnormal," "Pickup Timeout," "Transport Delay," "Delivery Stalled," or "No Track," the order is classified as an abnormal logistics order. For example, if the initial status judgment result of an order is "Signed for," it is directly marked as a normal order and enters the completion process. If the initial status judgment result is "No Track," the exception handling process is triggered.

[0045] S400: Perform order association analysis on the logistics order with abnormal status, process the logistics order with abnormal status according to the analysis result, and generate a processing result.

[0046] In this example, a graph neural network (GNN) model is first constructed. Using the order ID, delivery information, and logistics trajectory of abnormal logistics orders as node features, the model analyzes whether there are related orders, such as other orders with the same recipient and the same shipping location. If related orders exist, the similarity of features such as address and recipient is calculated using a comparative learning algorithm.

[0047] For example, if the similarity threshold is set to 80%, when the order similarity is greater than or equal to 80%, the "merge processing" result is output, the logistics trajectory of the abnormal order and the related orders are merged and analyzed, and the estimated arrival time is uniformly updated; when the order similarity is less than 80%, the "split verification" result is output, triggering a manual verification of the consistency between the abnormal order's delivery note information and the actual item. In addition, if there are no related orders, the reinforcement learning algorithm (DQN, DeepQ-Network) is used to optimize the timestamp information, and the reward function is trained based on historical processing data, such as user satisfaction weight and processing efficiency weight. The update time is dynamically generated, such as the traceability time plus 1 day. The logistics arrival reminder mechanism is triggered through the rule engine, and the processing result containing the "abnormal type" is output, for example, the processing result output: transportation delay, estimated arrival time: 2025-06-05 18:00.

[0048] This embodiment provides a method for handling logistics order exceptions. By acquiring logistics order feature data including order identification information, logistics trajectory information, timestamp information, and package status information, multi-dimensional data support is provided for determining the initial status of logistics orders. Based on the feature data, the initial status of the logistics order is confirmed and an initial status judgment result is generated, allowing timely identification of potential abnormal situations such as logistics entity abnormalities and pickup timeouts. Compared with traditional manual screening, its efficiency is greatly improved. Orders are classified as normal or abnormal based on the initial status judgment result, realizing automated classification management of logistics orders, ensuring the rapid flow of normal orders and accurate location of abnormal orders. Order association analysis is performed on logistics orders with abnormal status. By building an association model to determine whether there are related orders, processing results such as merging, splitting and verification, or optimizing timestamps are output, effectively shortening the processing time of abnormal orders. At the same time, a processing result containing the cause of the abnormality and the update time is generated, improving the transparency of logistics information and enhancing user predictability of logistics status. This forms an intelligent closed loop for the entire process from data collection, status judgment, classification management to exception handling, effectively improving logistics operation efficiency and user experience.

[0049] Reference Figure 2 The second embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0050] S210: Analyze the order identification information of the logistics order to determine whether the logistics entity corresponding to the logistics order matches the preset target logistics entity;

[0051] S220: If there is no match, generate an initial state judgment result indicating that the logistics entity is abnormal;

[0052] S230. If there is a match, the logistics track information, order identification information and timestamp information of the logistics order are analyzed to identify the logistics status of the logistics order, and a corresponding initial status judgment result is generated based on the logistics status.

[0053] In this embodiment, the first step is to perform a logistics entity matching verification: using natural language processing technology (NLP, Neuro-Linguistic Programming) to input the text in the order identification information, such as the courier company name, into the pre-trained language model (BERT,

[0054] The system then uses a BidirectionalEncoderRepresentationsfromTransformers (BidirectionalEncoderRepresentationsfromTransformers) to perform semantic matching with the system's pre-set target logistics entity library, such as a list of cooperating express delivery companies. For example, if the label text contains "Yunda Express" and the target logistics entity library contains "Yunda Express," it is considered a match. If "JD Express" is entered and there is no record in the target logistics entity library, an initial status judgment of "Logistics Entity Abnormal" is generated, triggering an alert.

[0055] If the logistics entity matches, logistics status recognition is further realized: the logistics trajectory information and timestamp information are analyzed through the long short-term memory network (LSTM). As an example, when the logistics entity matches the target express company connected to the system, the collection records within the preset time are extracted. If the collection trajectory is not detected and the time exceeds the preset threshold, such as if the collection trajectory is still not detected after more than 48 hours, the initial status judgment result of "collection timeout" is generated; for the collected orders, the logistics trajectory features are extracted through the convolutional neural network (CNN). If the node update frequency during transportation is lower than the preset frequency and the timestamp interval exceeds the preset threshold, such as if the node has not been updated for more than 24 hours, the initial status judgment result of "transportation delay" is generated.

[0056] Reference Figure 3 The third embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0057] S231. Based on the logistics track information of the logistics order, identify whether it is in the pickup state according to a preset stage identification rule, and generate a corresponding pickup logistics state according to the identification result;

[0058] S232. Based on the order identification information of the logistics order, classify the logistics order into international shipments and non-international shipments according to a preset time classification rule;

[0059] S233. For the international shipment, obtain the latest scanning trajectory timestamp based on the timestamp information and identify whether it is within a preset time threshold; for the non-international shipment, identify it as being in a normal timeliness status and generate a corresponding timeliness logistics status based on the identification result;

[0060] S234. Generate a corresponding initial status judgment result based on the collection logistics status and the time-sensitive logistics status.

[0061] In this embodiment, the first step is to identify the pickup status. Based on the node types in the logistics track information, such as "Picked Up" and "Pickup Failed," preset stage identification rules, such as whether the node name contains the keyword "Pickup," are used to determine whether the item is in the pickup state. If no pickup node is detected and the order generation exceeds a preset time, such as 48 hours, the system determines that the item has "Pickup Timed Out."

[0062] Then, international and non-international parcels are classified: by the waybill number prefix in the order identification information, such as the Arabic numerals starting with "779" to identify international parcels, or the keywords such as "international" and "import" in the waybill text to identify international parcels, the order types are divided according to the preset time classification rules. As an example, for international parcels, the timestamp of the latest scan track is extracted and compared with the preset time threshold. For example, if the preset time threshold is set to 72 hours, the latest scan track timestamp exceeds 72 hours, and an "abnormal international parcel time" judgment result is generated. It should be understood that the timestamp refers to the electronic data recorded by the scanning equipment during the logistics process for the specific time point of the logistics node operation, such as collection, sorting, transit, and delivery. The latest scan track timestamp specifically refers to the time point when the logistics order was last scanned and recorded in the current state. For non-international parcels, the default mark is "normal time" unless the track data shows that it has been stagnant for more than a preset number of days, such as if the track data shows that it has been stagnant for more than 3 days.

[0063] Finally, the initial status judgment result is generated by combining the collection logistics status and the delivery logistics status. For example, if an order is identified as "Collected" but the delivery time of the international shipment is abnormal, the initial status judgment result is "Abnormal delivery time (international shipment)". If an order is identified as "Collected" and the delivery time of the international shipment is normal, the initial status judgment result is "Normal delivery time (international shipment)".

[0064] Reference Figure 4 The fourth embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0065] S310: If the initial status judgment result is that the logistics entity is abnormal, classify the logistics order as a logistics order with abnormal status;

[0066] S320: If the initial status judgment result is any one of the following states: collected by courier, in transit, in transit, in delivery, and received, the logistics order is classified as a normal logistics order;

[0067] S330. If the initial status judgment result is any one of pickup timeout, transportation delay, delivery stagnation, transit stagnation and no track, the logistics order is classified as an abnormal status logistics order.

[0068] In this embodiment, if the initial status judgment result is a normal status such as "Collected by Express Delivery," "In Transit," or "Signed for," the order is classified as a normal logistics order and enters the regular tracking process. If the order shows "In Transit" and the delivery time is normal, it is classified as a normal logistics order.

[0069] If the initial status judgment results are "logistics entity abnormality," "collection timeout," or "delivery stalled," the order is classified as abnormal. For example, if the text on the shipping label identifies an order with a non-partnered logistics provider, the order will be marked abnormal and tracking will be terminated. Furthermore, multiple levels of alerts will be triggered for abnormal orders, such as a yellow alert for "collection timeout" and a red alert for "no track."

[0070] Reference Figure 5 The fifth embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0071] S340: Detecting whether the package status information corresponding to the logistics order shows any external damage, and detecting whether the geographic location information corresponding to the logistics order has a target logistics site within a preset range;

[0072] S350: If the package status information indicates that the package is externally damaged, or the geographic location information does not have a target logistics station within a preset range, the logistics order is determined to be a problem item;

[0073] S360: If the logistics order is determined to be a problem item, the arrival time of the logistics order is updated, and an exception feedback result including a description of the exception type is generated;

[0074] S370. If it is determined that the logistics order is not a problem item, identify whether it is in one of the rejected and collected states; if so, generate a normal feedback result including a description of the status type; if not, update the arrival time of the logistics order and trigger abnormal feedback.

[0075] In this embodiment, a problem package is first identified. Computer vision techniques, such as the highly efficient object detection model (YOLOv5, YouOnlyLookOnceversion5), are used to analyze the security inspection image in the package status information to detect features such as damage or deformation. A GIS (Geographic Information System) is then used to determine whether there are valid logistics sites within a preset range of the package's most recent destination, such as within 50 kilometers of the most recent destination. If the image indicates damage to the package's outer packaging, or if the GIS system indicates "no nearby site," the package is identified as a problem package.

[0076] Processing is then performed based on the type of problem and non-problem items. For example, for problem items, the arrival time is updated to "traceability time + 5 days," and exception feedback is generated through the rule engine, including image evidence and geographic analysis results. For example, an exception feedback message such as "Exception type: Damage; Exception description: The upper left corner of the package is damaged, and there is no sorting center within 50 kilometers" is generated. For non-problem items, if the image appears normal and a valid logistics station exists, further determination is made as to whether the item is "Rejected" or "Collected." If so, it is marked as normal; if not, the arrival time is updated and an exception feedback message is triggered, such as "Delivery stalled, expected delay 24 hours."

[0077] Reference Figure 6 The sixth embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0078] S410: Acquire characteristic data of the logistics order with abnormal status;

[0079] S420: Build a correlation model, analyze the correlation relationship between the abnormal logistics order and other orders through the correlation model, and determine whether there are related orders;

[0080] S430: If there are related orders, calculate the similarity between the feature data of the abnormal logistics order and the related order, and output a merge processing result or a split verification processing result based on the similarity result:

[0081] S440. If there is no associated order, optimize the timestamp information of the logistics order with abnormal status, dynamically generate the update time, and trigger the pre-built logistics arrival reminder mechanism according to the update time, and output the processing result including the abnormal reason and update time.

[0082] In this embodiment, the order ID, delivery information, and logistics trajectory of the abnormal logistics order are first obtained. Then, an association model is constructed. Using a graph neural network (GNN) model, the order ID, delivery information, and logistics trajectory of the abnormal logistics order are used as graph nodes to construct an order association graph to form the association model. For example, multiple orders placed by the same recipient within three days are automatically associated as a "same-user order group." Similarity calculation and processing are performed using the association model. If there are associated orders, the feature similarity between the abnormal logistics order and the associated orders is calculated, such as address matching and logistics trajectory overlap. For example, a similarity threshold is set to 80%. If the order similarity is greater than or equal to 80%, a "merge processing" result is output, and the estimated arrival time is uniformly adjusted, such as to a preset average delivery time. If the order similarity is less than 80%, a "split verification" result is output, prompting manual verification of the shipping order information consistency. If there is no associated order, the timestamp information is optimized using a reinforcement learning algorithm (DQN, DeepQ-Network). The reward function, such as user satisfaction weight and processing efficiency weight, is trained in combination with historical processing data. The update time is dynamically generated, such as the shortest update time is 30 minutes and the longest update time is 1 day. The logistics arrival reminder is triggered, and the processing result containing the "abnormal type" is output, for example, the output is: transportation delay, estimated arrival time: 2025-06-05-18:00.

[0083] Reference Figure 7 The seventh embodiment of the logistics order exception processing method in the embodiment of the present invention includes:

[0084] S500: Acquire final distribution characteristic data of a logistics order, wherein the final distribution characteristic data of the logistics order includes distribution time, target routing information, and logistics track after distribution;

[0085] S600: Compare the distribution time with the preset distribution time limit to determine whether there is a distribution delay anomaly; analyze the matching degree between the logistics trajectory after distribution and the target routing information to determine whether there is a misdistribution anomaly;

[0086] S700: If there is a distribution delay or misdistribution exception, generate the corresponding exception cause and response strategy; if the distribution is normal, update the logistics order to the final distribution status result as completed, and generate an estimated delivery time.

[0087] In this embodiment, a distribution delay detection is first performed: the distribution time from the final distribution feature data is extracted and compared with a preset distribution time limit, such as a 4-hour time limit. If the actual distribution time exceeds 8 hours, it is determined to be a "distribution delay anomaly," and a cause analysis is generated, such as "sorting center equipment failure." Next, a mismatch anomaly detection is performed: a path matching algorithm is used to compare the degree of match between the post-distribution logistics trajectory and the target routing information. For example, if the order's target route is "Shanghai Pudong," but the post-distribution trajectory indicates a destination of "Beijing Chaoyang," and the match is less than 50%, a "mismatch anomaly" is determined, triggering a path adjustment mechanism. Furthermore, if the distribution time and trajectory are both correct, the order status is updated to "Final-level distribution completed," and an estimated delivery time (e.g., "Expected delivery at 2025-06-05-15:00") is generated based on real-time traffic data.

[0088] The above describes the logistics order exception processing method in the embodiment of the present invention. The following describes the logistics order exception processing device in the embodiment of the present invention. Please refer to Figure 8 In one embodiment of the present invention, a device for handling logistics order exceptions includes:

[0089] Acquisition module 10, used to obtain characteristic data of logistics orders;

[0090] A judgment module 20 is configured to determine the initial state of the logistics order based on the characteristic data of the logistics order and obtain an initial state judgment result;

[0091] A classification module 30 is configured to classify the logistics order into a normal logistics order or an abnormal logistics order according to the initial status judgment result;

[0092] The processing module 40 is used to perform order association analysis on the logistics order with abnormal status, process the logistics order with abnormal status according to the analysis result, and generate a processing result.

[0093] In this embodiment, by obtaining the characteristic data of the logistics order, confirming the initial status of the logistics order and generating the initial status judgment result, the order is classified as normal or abnormal according to the initial status judgment result, and order association analysis is performed on the logistics orders with abnormal status. By building an association model, it is determined whether there are related orders, and then the processing results such as merging processing, splitting verification or optimizing timestamps are output, which effectively shortens the processing time of abnormal orders and forms an intelligent closed loop of the entire process from data collection, status judgment, classification management to exception processing, effectively improving logistics operation efficiency and user experience.

[0094] Reference Figure 9 In this embodiment, the judgment module 20 includes:

[0095] The judgment unit 21 is used to analyze the order identification information of the logistics order and determine whether the logistics entity corresponding to the logistics order matches the preset target logistics entity;

[0096] The identification unit 22 is used to generate an initial state judgment result of an abnormal logistics entity when there is no match; when there is a match, the logistics track information, order identification information and timestamp information of the logistics order are analyzed to identify the logistics state of the logistics order, and generate a corresponding initial state judgment result based on the logistics state.

[0097] Reference Figure 9 In this embodiment, the identification unit 22 is specifically used to identify whether it belongs to the collection status based on the logistics trajectory information of the logistics order according to the preset stage identification rules, and generate the corresponding collection logistics status according to the identification result; based on the order identification information of the logistics order, the logistics order is divided into international pieces and non-international pieces according to the preset time classification rules; for the international pieces, the latest scanning trajectory timestamp is obtained based on the timestamp information to identify whether it is within the preset time threshold; for the non-international pieces, it is identified as a normal time status, and the corresponding time logistics status is generated according to the identification result; based on the collection logistics status and the time logistics status, a corresponding initial state judgment result is generated.

[0098] Reference Figure 9 In this embodiment, the classification module 30 includes:

[0099] The first classification unit 31 is configured to classify the logistics order as an abnormal logistics order if the initial status judgment result is that the logistics entity is abnormal; and classify the logistics order as an abnormal logistics order if the initial status judgment result is any one of: pickup timeout, transportation delay, delivery stagnation, transit stagnation, and no track.

[0100] The second classification unit 32 is used to classify the logistics order as a normal logistics order if the initial status judgment result is any one of the express collection, transportation, transit, delivery and signed for status.

[0101] Reference Figure 9 In this embodiment, the logistics order exception processing device further includes:

[0102] A detection module 50 is configured to detect whether the status information of the package corresponding to the logistics order indicates any external damage, and to detect whether the geographical location information corresponding to the logistics order includes a target logistics site within a preset range;

[0103] The logistics exception processing module 60 is used to determine that the logistics order is a problem item when the package status information shows that there is external damage, or the geographical location information does not have a target logistics site within a preset range; when the logistics order is determined to be a problem item, the arrival time of the logistics order is updated, and an exception feedback result including a description of the exception type is generated; when the logistics order is determined to be a non-problem item, whether it is in one of the rejected and collected states is identified; if so, a normal feedback result including a description of the status type is generated; if not, the arrival time of the logistics order is updated and an exception feedback is triggered.

[0104] Reference Figure 9 In this embodiment, the processing module 40 includes:

[0105] The abnormal data acquisition unit 41 is used to acquire characteristic data of the logistics order with abnormal status;

[0106] The association unit 42 is used to build an association model, analyze the association relationship between the logistics order with abnormal status and other orders through the association model, and determine whether there is an associated order; if there is an associated order, calculate the similarity between the feature data of the logistics order with abnormal status and the associated order, and output the merge processing result or the split verification processing result according to the similarity result: if there is no associated order, optimize the timestamp information of the logistics order with abnormal status, dynamically generate the update time, and trigger the pre-built logistics arrival reminder mechanism according to the update time, and output the processing result including the cause of the abnormality and the update time.

[0107] Reference Figure 9 In this embodiment, the logistics order exception processing device further includes:

[0108] The distribution exception processing module 70 is used to obtain the final distribution feature data of the logistics order, which includes the distribution time, target routing information and the logistics trajectory after distribution; compare the distribution time with the preset distribution time limit to determine whether there is a distribution delay exception; analyze the matching degree between the logistics trajectory after distribution and the target routing information to determine whether there is a misdistribution exception; if there is a distribution delay or misdistribution exception, generate the corresponding exception cause and response strategy; if the distribution is normal, update the logistics order to the completed final distribution status result and generate an estimated delivery time.

[0109] above Figure 8 and Figure 9 The logistics order exception processing device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The logistics order exception processing device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0110] Figure 10It is a structural diagram of a logistics order exception processing device provided by an embodiment of the present invention. The logistics order exception processing device 1000 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 1100 (for example, one or more processors) and a memory 1200, and one or more storage media 1300 (for example, one or more massive storage devices) storing application programs 1310 or data 1320. Among them, the memory 1200 and the storage medium 1300 can be short-term storage or persistent storage. The program stored in the storage medium 1300 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the logistics order exception processing device 1000. Furthermore, the processor 1100 can be configured to communicate with the storage medium 1300 to execute a series of instruction operations in the storage medium 1300 on the logistics order exception processing device 1000.

[0111] The logistics order exception processing device 1000 may also include one or more power supplies 1400, one or more wired or wireless network interfaces 1500, one or more input and output interfaces 1600, and / or one or more operating systems 1330 such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 10 The device structure shown does not constitute a limitation on the logistics order exception processing device 1000, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0112] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are run on a computer, the computer executes the steps of the logistics order exception handling method.

[0113] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0114] If the integrated unit is implemented in the form of 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 the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0115] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for handling logistics order exceptions, characterized in that: The logistics order exception processing method includes: Obtaining characteristic data of a logistics order, wherein the characteristic data of the logistics order includes order identification information, logistics track information, timestamp information, and package status information; Based on the characteristic data of the logistics order, confirm the initial state of the logistics order and obtain an initial state judgment result; Classifying the logistics order as a normal logistics order or an abnormal logistics order according to the initial status judgment result; An order association analysis is performed on the logistics order with abnormal status, the logistics order with abnormal status is processed according to the analysis result, and a processing result is generated.

2. The logistics order exception handling method according to claim 1 is characterized in that: Based on the characteristic data of the logistics order, the initial state of the logistics order is confirmed to obtain an initial state judgment result, which specifically includes: Analyze the order identification information of the logistics order to determine whether the logistics entity corresponding to the logistics order matches the preset target logistics entity; If there is no match, an initial state judgment result indicating abnormality of the logistics entity is generated; If there is a match, the logistics track information, order identification information and timestamp information of the logistics order are analyzed to identify the logistics status of the logistics order, and a corresponding initial status judgment result is generated based on the logistics status.

3. The method for handling logistics order exceptions according to claim 2, characterized in that: Analyze the logistics track information, order identification information, and timestamp information of the logistics order to identify the logistics status of the logistics order, and generate a corresponding initial status judgment result based on the logistics status, specifically including: Based on the logistics track information of the logistics order, identify whether it belongs to the pickup state according to the preset stage identification rules, and generate the corresponding pickup logistics state according to the identification result; Based on the order identification information of the logistics order, classify the logistics order into international shipments and non-international shipments according to preset time classification rules; For international shipments, the latest scanning trajectory timestamp is obtained based on the timestamp information to identify whether it is within the preset time threshold; for non-international shipments, it is identified as being in a normal timeliness status, and the corresponding timeliness logistics status is generated based on the identification result; A corresponding initial status judgment result is generated based on the collection logistics status and time-sensitive logistics status.

4. The logistics order exception handling method according to claim 1 is characterized in that: According to the initial status judgment result, the logistics order is classified as a normal logistics order or an abnormal logistics order, specifically including: If the initial status judgment result is that the logistics entity is abnormal, the logistics order is classified as a logistics order with abnormal status; If the initial status judgment result is any one of the following states: collected by courier, in transit, in transit, in delivery, and received, the logistics order is classified as a normal logistics order; If the initial status judgment result is any one of the following: pickup timeout, transportation delay, delivery stagnation, transit stagnation and no track, the logistics order will be classified as an abnormal status logistics order.

5. The method for handling logistics order exceptions according to claim 4, characterized in that: If the status of the logistics order is classified as an abnormal logistics order, the method further includes: Detecting whether the package status information corresponding to the logistics order shows any external damage, and detecting whether the geographical location information corresponding to the logistics order has a target logistics site within a preset range; If the package status information shows that the package is externally damaged, or the geographic location information does not have a target logistics station within a preset range, the logistics order is determined to be a problem item; If the logistics order is determined to be a problem item, the arrival time of the logistics order is updated, and an exception feedback result including a description of the exception type is generated; If it is determined that the logistics order is not a problem item, identify whether it is in one of the rejected and collected states; if so, generate a normal feedback result including a description of the status type; if not, update the arrival time of the logistics order and trigger abnormal feedback.

6. The logistics order exception handling method according to claim 1 is characterized in that: Perform order association analysis on the logistics order with abnormal status, process the logistics order with abnormal status according to the analysis result, and generate a processing result, specifically including: Obtaining characteristic data of the logistics order with abnormal status; Constructing a correlation model, analyzing the correlation relationship between the abnormal logistics order and other orders through the correlation model, and determining whether there are related orders; If there are related orders, calculate the similarity between the feature data of the abnormal logistics order and the related order, and output the merge processing result or split verification processing result based on the similarity result: If there is no associated order, optimize the timestamp information of the logistics order with abnormal status, dynamically generate the update time, and trigger the pre-built logistics arrival reminder mechanism according to the update time, and output the processing result including the abnormal reason and update time.

7. The logistics order exception processing method according to any one of claims 1 to 6, characterized in that: The method further comprises: Acquire final distribution feature data of the logistics order, wherein the final distribution feature data of the logistics order includes distribution time, target routing information, and logistics track after distribution; Compare the distribution time with the preset distribution time to determine whether there is a distribution delay anomaly; analyze the matching degree between the logistics trajectory after distribution and the target routing information to determine whether there is a misdistribution anomaly; If there is a distribution delay or misassignment exception, the corresponding exception cause and response strategy are generated; if the distribution is normal, the logistics order is updated to the final distribution status result as completed, and an estimated delivery time is generated.

8. A logistics order exception processing device, characterized in that: include: Acquisition module, used to obtain characteristic data of logistics orders; a judgment module, configured to determine the initial state of the logistics order based on the characteristic data of the logistics order and obtain an initial state judgment result; A classification module, configured to classify the logistics order into a normal logistics order or an abnormal logistics order according to the initial status judgment result; The processing module is used to perform order association analysis on the logistics order with abnormal status, process the logistics order with abnormal status according to the analysis result, and generate a processing result.

9. A logistics order exception processing device, characterized in that: comprising a memory and at least one processor, wherein the memory has computer-readable instructions stored therein; The at least one processor calls the computer-readable instructions in the memory to execute the various steps of the logistics order exception processing method as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the various steps of the logistics order exception handling method as described in any one of claims 1 to 7 are implemented.

Images