Logistics order abnormity monitoring method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a logistics order abnormity monitoring method and device, electronic equipment and a storage medium, and belongs to the technical field of logistics. The method comprises the following steps: collecting current logistics multi-modal information of each logistics in-transit order; wherein the current logistics multi-modal information is an information flow of a plurality of data modals of the logistics in-transit order in the current logistics link; performing logistics state identification on each logistics in-transit order according to the current logistics multi-mode information to obtain a current logistics state of each logistics in-transit order; if the current logistics state is a logistics abnormal state, acquiring logistics abnormal data of the logistics in-transit order; and generating logistics early warning information according to the logistics abnormal data, and sending the logistics early warning information to a preset logistics monitoring end to enable the logistics monitoring end to perform abnormity processing on the logistics in-transit order. According to the embodiment of the invention, the manpower for monitoring the abnormity of the logistics order can be improved, and the processing efficiency of the abnormity of the logistics order is improved.

Description

Technical Field

[0001] This application relates to the field of logistics technology, and in particular to a method and apparatus, electronic device and storage medium for monitoring abnormal logistics orders. Background Technology

[0002] Logistics is the core artery connecting production and consumption, and its stability and efficiency are crucial. Massive numbers of logistics orders flow through multiple stages, including pickup, transshipment, and delivery, inevitably leading to anomalies due to various reasons such as damaged packages, delivery delays, information errors, or accidental loss. The efficiency and accuracy of handling these abnormal logistics orders directly affect the relationship between users and logistics platforms.

[0003] In related technologies, to quickly resolve abnormal logistics orders, processing is typically only initiated when a user files a complaint or reports an anomaly, and the cause of the anomaly during the entire logistics process is identified. However, the processing of logistics orders is highly dependent on manual labor. Monitoring personnel or after-sales customer service need to manually retrieve and integrate information from multiple independent business systems, including structured order data, unstructured transportation logs, package photos, and multimodal logistics-related information such as delivery receipts. This integration is difficult, leading to low efficiency in handling abnormal logistics orders. Therefore, how to reduce manpower for monitoring abnormal logistics orders and improve the efficiency of handling abnormal logistics orders has become an urgent technical problem to be solved. Summary of the Invention

[0004] The main objective of this application is to provide a method, device, electronic device, and storage medium for monitoring abnormal logistics orders, aiming to reduce manpower required for monitoring abnormal logistics orders and improve the processing efficiency when abnormal logistics orders occur.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for monitoring logistics order anomalies, the method comprising: Collect current logistics multimodal information for each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple data modalities of the logistics order in transit in the current logistics link; Based on the current multimodal logistics information, the logistics status of each logistics order in transit is identified to obtain the current logistics status of each logistics order in transit. If the current logistics status is an abnormal logistics status, collect the logistics abnormality data of the logistics orders in transit; Based on the abnormal logistics data, a logistics early warning message is generated and sent to a preset logistics monitoring terminal so that the logistics monitoring terminal can handle the abnormal logistics orders in transit.

[0006] In some embodiments, the current logistics multimodal information includes: current logistics time-series data and current logistics transportation route; The step of identifying the logistics status of each in-transit order based on the current logistics multimodal information to obtain the current logistics status of each in-transit order includes: Differential time-series data are extracted from the current logistics time-series data based on preset reference logistics time-series data; wherein, the reference logistics time-series data is time-series data referenced for the entire logistics chain; The current logistics transportation route is evaluated for deviation based on a preset reference logistics transportation route to obtain route deviation evaluation data. The logistics status of each in-transit order is identified by using a preset logistics status identification model, the difference time series data, and the route deviation evaluation data, thereby obtaining the current logistics status of each in-transit order.

[0007] In some embodiments, the step of identifying the logistics status of each in-transit order using a preset logistics status identification model, the difference time series data, and the route deviation evaluation data to obtain the current logistics status of each in-transit order includes: The difference time series data is compared with a preset difference threshold to obtain first comparison information; wherein, the difference time series data includes at least one of the following: difference in logistics update frequency, difference in dwell time, and difference in transportation time; The route deviation assessment data is compared with a preset deviation threshold to obtain second comparison information; The first comparison information and the second comparison information are input into the logistics status recognition model to identify the logistics status, thereby obtaining the current logistics status of each logistics order in transit.

[0008] In some embodiments, the method further includes: Receive an order anomaly trigger request; wherein, the order anomaly trigger request includes the order identifier number of the logistics history order and the anomaly feedback content; Based on the abnormal feedback content, the abnormality category is identified to obtain the order abnormality category; The historical logistics multimodal information of the logistics historical order is obtained based on the order identifier; wherein, the historical logistics multimodal information is the information flow of the logistics historical order in multiple modes throughout the entire logistics chain; The cause of the order anomaly is determined based on the order anomaly category and the historical logistics multimodal information.

[0009] In some embodiments, determining the order anomaly cause information based on the order anomaly category and the historical logistics multimodal information includes: Based on the order anomaly category, target modal data and historical logistics time-series data are extracted from the historical logistics multimodal information; wherein, the target modal data is data whose modality matches the order anomaly category; Anomalies are located based on the target modal data and the historical logistics time series data to obtain abnormal logistics nodes. Based on the abnormal logistics nodes, related modal data is extracted from the historical logistics multimodal information; wherein, the related modal data is used to assist in the analysis of the causes of the anomalies; The abnormal order cause information is determined by using a preset abnormal cause analysis model, the target modal data, and the associated modal data.

[0010] In some embodiments, the step of locating abnormal logistics nodes based on the target modal data and the historical logistics time-series data includes: Abnormal logistics features are obtained by extracting abnormal features from the target modal data; The abnormal logistics nodes are obtained by locating the anomalies in the historical logistics time series data based on the abnormal logistics characteristics.

[0011] In some embodiments, after determining the order anomaly cause information through a preset anomaly cause analysis model, the target modal data, and the associated modal data, the method further includes: Input the target modal data, the associated modal data, and the order anomaly reason information into a preset anomaly result feedback template to obtain anomaly result feedback information; The abnormal result feedback information is sent to the logistics monitoring terminal so that the logistics monitoring terminal can verify the cause of the abnormality of the logistics historical orders.

[0012] To achieve the above objectives, a second aspect of this application provides a monitoring device for abnormal logistics orders, the device comprising: The order acquisition module is used to collect the current logistics multimodal information of each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple modes of the logistics order in transit in the current logistics link; The logistics status identification module is used to identify the logistics status of each logistics order in transit based on the current logistics multimodal information, and obtain the current logistics status of each logistics order in transit. The data acquisition module is used to collect logistics abnormality data of the logistics orders in transit if the current logistics status is a logistics abnormality status. The early warning module is used to generate logistics early warning information based on the logistics anomaly data, and send the logistics early warning information to a preset logistics monitoring terminal so that the logistics monitoring terminal can handle the logistics orders in transit abnormally.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The method, apparatus, electronic device, and storage medium for monitoring logistics order anomalies proposed in this application collect current multimodal logistics information for each logistics order in transit at the current moment, determine the current logistics status of each logistics order in transit based on the current multimodal logistics information, and collect logistics anomaly data for the logistics order in transit when the current logistics status is an abnormal logistics status. This abnormal logistics data is then used to generate logistics early warning information, which is reported to the logistics monitoring terminal. Therefore, this application transforms the analysis of logistics orders in transit from post-event analysis to pre-event monitoring of the logistics status of logistics orders in transit. Furthermore, by combining current multimodal logistics information for pre-event monitoring of logistics anomalies, it can accurately detect logistics anomalies and promptly provide feedback to logistics after-sales personnel at the logistics monitoring terminal. This allows logistics after-sales personnel to respond promptly to logistics after-sales service, improving logistics service quality and operational efficiency. Attached Figure Description

[0016] Figure 1 This is a system framework diagram of the logistics order anomaly monitoring method provided in the embodiments of this application; Figure 2 This is a flowchart of the logistics order anomaly monitoring method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the process for proactive anomaly monitoring of logistics orders in the logistics order anomaly monitoring method provided in this application embodiment; Figure 4 yes Figure 2 The flowchart of step S202 in the document; Figure 5 yes Figure 4 The flowchart of step S403 in the process; Figure 6 This is a flowchart illustrating the passive anomaly response of logistics orders in transit in the logistics order anomaly monitoring method provided in this application embodiment; Figure 7 This is a flowchart of a method for monitoring logistics order anomalies provided in another embodiment of this application; Figure 8 yes Figure 7 The flowchart of step S704 in the process; Figure 9 yes Figure 8 The flowchart of step S802 in the process; Figure 10 This is a flowchart of a method for monitoring logistics order anomalies provided in another embodiment of this application; Figure 11 This is a schematic diagram of the structure of the logistics order anomaly monitoring device provided in the embodiments of this application; Figure 12 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] 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.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] First, let's analyze some of the terms used in this application: Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0021] Large Language Models (LLMs) are deep learning models trained on massive amounts of text data, enabling them to generate natural language text or understand the meaning of language text. These models can provide in-depth knowledge and language production on a wide range of topics through training on large datasets. Their core idea is to learn patterns and structures of natural language through large-scale unsupervised training, mimicking human language cognition and generation processes to some extent.

[0022] Computer vision refers to machine vision that uses cameras and computers to identify, track, and measure targets, replacing the human eye. It further processes images to create visually more suitable images for human observation or for instrument detection. As a scientific discipline, computer vision studies related theories and technologies, attempting to build artificial intelligence systems capable of extracting 'information' from images or multidimensional data.

[0023] Multi-source heterogeneous information: This is a technology that integrates and comprehensively analyzes data from multiple data sources, including different devices, sensors, systems, or networks. Its purpose is to leverage multi-faceted information from different data sources to obtain more comprehensive and accurate data, and to provide better business applications and services.

[0024] In the global supply chain system, logistics is the core artery connecting production and consumption, and its stability and efficiency are crucial. Massive numbers of logistics orders flow through multiple stages, including pickup, transshipment, and delivery. Inevitably, anomalies will occur due to various reasons, such as damaged packages, delivery delays, information errors, or accidental loss. The efficiency and accuracy of handling these abnormal logistics orders directly affect user experience and the reputational costs for businesses.

[0025] Traditional monitoring models for abnormal logistics orders suffer from significant technical bottlenecks. First, traditional monitoring is a "passive response" model, typically initiating processing only after a customer complaint or a clear error status (such as "lost package") is recorded in the system. This leads to delays in problem detection and negatively impacts customer satisfaction. Second, the process is highly reliant on manual intervention. Monitoring personnel or customer service staff must manually retrieve and integrate information from multiple independent business systems (such as order management, warehousing systems, and transportation tracking), including structured order data, unstructured transportation logs, and multimodal information such as package photos and delivery receipts. This information integration is difficult and inefficient. Finally, the accuracy of manual judgment heavily relies on individual experience, lacking unified and intelligent analytical standards, making it difficult to discern the underlying causes of anomalies or predict potential risks from massive, fragmented data.

[0026] In related technologies, applying large language models to logistics order management has achieved breakthroughs in text understanding and image recognition, demonstrating a powerful ability to analyze single information sources. However, general-purpose AI models lack deep domain knowledge of complex logistics chains and cannot effectively integrate time-series data (stay time at each station), spatial geographic information (such as transportation trajectories), and external environmental factors (such as weather and traffic) for comprehensive causal reasoning. They might be able to identify a damaged package in a photo, but they struggle to correlate and analyze data such as abnormal stay times or weather conditions prior to the damage, making it difficult to accurately pinpoint the cause of the anomaly.

[0027] Based on this, embodiments of this application provide a method, apparatus, electronic device, and storage medium for monitoring logistics order anomalies. The aim is to monitor the logistics status of in-transit orders in real time by combining current multimodal information of the orders. When an abnormal logistics status is detected, the system automatically collects the corresponding logistics anomaly data, generates a logistics early warning message, and sends it to the logistics monitoring terminal, thus achieving automatic monitoring and reporting of abnormal logistics statuses. Furthermore, by combining multimodal information of in-transit orders with the analysis of abnormal logistics statuses, the system can more accurately and efficiently analyze abnormal states and promptly report logistics anomaly data, facilitating timely processing of abnormal logistics orders and reducing the complaint rate caused by abnormal logistics orders.

[0028] The logistics order anomaly monitoring method, device, electronic equipment, and storage medium provided in this application are specifically described through the following embodiments. First, the logistics order anomaly monitoring method in this application embodiment is described.

[0029] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0030] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0031] The logistics order anomaly monitoring method provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the logistics order anomaly monitoring method, but is not limited to the above forms.

[0032] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0033] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics in logistics orders is required, the user's permission or consent will be obtained first. Furthermore, the collection, use, and processing of this data will comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user will be obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent will the necessary user-related data for the proper functioning of these embodiments be acquired.

[0034] like Figure 1 As shown, Figure 1 The system framework diagram for the application of a method for monitoring logistics order anomalies is shown. Figure 1 The system comprises a server, a logistics monitoring terminal, and a logistics information collection terminal. Multiple logistics information collection terminals are configured, each responsible for collecting logistics-related information corresponding to a logistics order, specifically information from order creation to receipt. It's important to note that logistics orders include in-transit orders and historical orders; in-transit orders are those currently in transit, while historical orders are those already signed for and completed. Specifically, a newly created logistics order is defined as an in-transit order. The logistics information collection terminal gathers the current multimodal logistics information from the creation to receipt stage of the in-transit order and uploads this information to the server at preset time intervals. The server performs status analysis and data filtering on this information to determine the current logistics status. Furthermore, if an abnormal logistics status is detected, the abnormal data is sent to the logistics monitoring terminal, where logistics after-sales personnel promptly confirm or handle the anomaly. Once the in-transit order is completed, it is defined as a historical order, and the current multimodal logistics information pre-stored on the server is defined as historical multimodal logistics information. When feedback from historical logistics orders is received, historical multimodal logistics information can be directly retrieved for anomaly analysis, enabling post-event anomaly analysis of logistics orders. Therefore, by constructing a system consisting of a server, at least one logistics information collection terminal, and a logistics monitoring terminal, pre-event anomaly warnings and post-event anomaly analysis of logistics orders can be achieved. This is equivalent to a closed-loop monitoring system with both pre-event warning and post-event analysis modes. It can automatically identify logistics anomalies, perform in-depth attribution analysis after anomaly reporting, and report the anomaly causes to the monitoring terminal for verification and after-sales service, thereby reducing the complaint rate for logistics anomalies.

[0035] Figure 2 This is an optional flowchart of the logistics order anomaly monitoring method provided in the embodiments of this application. Figure 2The method may include, but is not limited to, steps S201 to S204.

[0036] Step S201: Collect the current logistics multimodal information of each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple data modalities of the logistics order in transit in the current logistics link; Step S202: Based on the current multimodal logistics information, identify the logistics status of each logistics order in transit to obtain the current logistics status of each logistics order in transit. Step S203: If the current logistics status is abnormal, collect logistics abnormal data for orders in transit. Step S204: Generate logistics early warning information based on logistics anomaly data, and send the logistics early warning information to the preset logistics monitoring terminal so that the logistics monitoring terminal can handle the anomalies of logistics orders in transit.

[0037] Steps S201 to S205 of this embodiment involve collecting current multimodal logistics information of orders in transit and monitoring their logistics status based on this information. When an abnormal logistics status is detected, abnormal logistics data is collected, a logistics warning is generated based on this data, and the warning is sent to the logistics monitoring terminal. This allows logistics after-sales personnel at the monitoring terminal to promptly take over the orders and handle the anomalies. Therefore, this embodiment monitors the status of orders in transit by combining multimodal logistics information. When an anomaly is detected, it is reported promptly, enabling advance warning of logistics orders, predictive identification and proactive intervention of potential abnormal orders, reducing customer complaint rates, and improving customer experience.

[0038] In step S201 of some embodiments, the method for monitoring logistics order anomalies is applied to the logistics platform of the server. A logistics order created by a customer on the logistics platform and in the logistics transportation state is defined as a logistics order in transit. A signed-for logistics order is defined as a logistics history order. The logistics platform stores the logistics orders in transit and their current multimodal logistics information together in the in-transit order database, and stores the logistics history orders and their historical multimodal logistics information in the history order database. When a logistics order in transit is signed for, it becomes a logistics history order and is transferred from the in-transit order database to the history order database. Therefore, by constructing the in-transit order database and the history order database, a data foundation is provided for subsequent pre-event monitoring and post-event analysis. It should be noted that the current logistics multimodal information is collected periodically according to a preset time period, and the logistics time nodes of logistics orders in transit are updated periodically according to the preset time period. If the logistics time node is the sign-for-delivery node, the logistics orders in transit are defined as logistics historical orders, the current logistics multimodal information is defined as historical logistics multimodal information, and the logistics historical orders and historical logistics multimodal information are transferred to the historical order database for storage, so as to provide a basis for judgment in subsequent anomaly investigation.

[0039] Furthermore, to conserve storage space in the historical order database, the storage duration of the historical order database will be collected periodically according to a preset time period. Logistics historical orders and historical logistics multimodal information stored for longer than the preset duration will be deleted. It should be noted that the preset duration is set by the after-sales personnel in the backend, based on the latest duration among historical feedback durations. For example, the preset duration can be set to six months, one year, or two years, which saves storage space in the historical order database without affecting the subsequent investigation of anomalies in logistics historical orders.

[0040] In some embodiments, the current logistics multimodal information consists of information flows of various data modalities generated by logistics orders in transit along the current logistics link. This current logistics multimodal information is collected and updated according to a preset time period to achieve real-time status monitoring of logistics orders in transit. The logistics link consists of multiple logistics time nodes, which can be any of the following: order creation node, pickup node, transit node, delivery node, and package receipt node. The data modality can be any of the following: text modality, structured modality, image modality, video modality, and time-series modality, etc. This embodiment does not limit the data modality. If the current time is the order creation node, the current logistics multimodal information includes the order creation time, logistics order number, and package information generated during order creation. If the current time is the pickup node, the current logistics multimodal information includes the pickup time, package dimensions, package weight, packaging photos, and express delivery fee invoices generated during the pickup process. Package dimensions and package weight are structured modal data, while package photos are image modal data. If the current time is a transit node, the current multimodal logistics information includes arrival times at various stations, package transport status, weather data, and traffic congestion data generated during the transit process. Package transport status data is in time-series mode, while weather and traffic congestion data are in structured or text-based mode. For a delivery node, the current multimodal logistics information includes arrival times at grid points and delivery times generated during the delivery process, which are in time-series mode. If the current time is a package receipt node, the current multimodal logistics information includes the package receipt time and receipt image generated when the customer signs for the package. The package receipt time is in time-series mode, while the receipt image is in image mode. Therefore, for different logistics time nodes, the current multimodal logistics information consists of data corresponding to different data modes. This information is also key data for monitoring logistics anomalies at the corresponding logistics time nodes, enabling accurate analysis of the logistics status at each time node.

[0041] like Figure 3 As shown, Figure 3 This demonstrates the current multimodal logistics information collected at each logistics time point throughout the entire logistics chain. Specifically, it collects data from various data modes generated at the order creation node, receiving node, transit transportation node, delivery node, and package signing node. This data is stored in the current logistics database as reference data for monitoring anomalies in logistics orders throughout the entire logistics chain, and can also serve as reference data for subsequent order feedback analysis.

[0042] In step S202 of some embodiments, logistics status identification mainly involves identifying the current logistics status of orders in transit. This requires identifying the logistics status of each order in transit to achieve proactive anomaly scanning and pre-emptive anomaly monitoring. As previously disclosed, the current multimodal logistics information collected at different logistics time points contains different data modalities. Therefore, the criteria for judging whether the logistics status is abnormal differ at different time points based on data analysis of different data modalities.

[0043] In some embodiments, taking a logistics order in transit at a transit node as an example, the current logistics multimodal information includes current logistics time-series data and current logistics transportation route. The current logistics time-series data represents the time sequence of the logistics order in transit, specifically the current logistics update frequency, current logistics dwell time, and current transportation time. The current logistics update frequency is the update frequency of the logistics status, and the current logistics dwell time is the dwell time of the logistics order at the transit station. The current logistics transportation route is the route that the logistics order in transit has already taken at the current moment.

[0044] Please refer to Figure 4 In some embodiments, step S202 may include, but is not limited to, steps S401 to S403: Step S401: Extract the difference time series data from the current logistics time series data based on the preset reference logistics time series data; wherein, the reference logistics time series data is the time series data referenced for the entire logistics chain; Step S402: Evaluate the route deviation of the current logistics transportation route based on the preset reference logistics transportation route to obtain route deviation evaluation data; Step S403: The logistics status of each in-transit order is identified by using a preset logistics status identification model, difference time series data, and route deviation evaluation data to obtain the current logistics status of each in-transit order.

[0045] In step S401 of some embodiments, the reference logistics time-series data is the time-series data of express parcels of the same category as the logistics in-transit orders throughout the entire logistics chain, representing the reference logistics update frequency, reference logistics dwell time, and reference transportation time at each logistics time node. It should be noted that the difference time-series data is the time-series data where there are differences between the reference logistics time-series data and the current logistics time-series data, including differences in logistics update frequency, dwell time, and transportation time. Specifically, the difference between the current logistics update frequency and the reference logistics update frequency is used as the logistics update frequency difference; the difference between the current logistics dwell time and the reference logistics dwell time is used as the dwell time difference; and the difference between the current transportation time and the reference transportation time is used as the transportation time difference. Therefore, the current logistics status of the logistics in-transit orders can be comprehensively analyzed at transit transportation nodes through the update frequency difference, dwell time difference, and transportation time difference.

[0046] In step S402 of some embodiments, in addition to analyzing the current logistics status of logistics orders in transit from the difference data in logistics time sequence, the degree of deviation between the reference logistics transportation route and the current logistics transportation route can also be analyzed. Therefore, the difference route between the reference logistics transportation route and the current logistics transportation route is calculated, and the route deviation assessment data is determined according to the proportion of the difference route on the reference logistics transportation route.

[0047] In step S403 of some embodiments, the logistics status recognition model is a pre-trained multimodal large model applied to the field of logistics status recognition. As the core of monitoring logistics orders in transit, the multimodal large model analyzes real-time collected differential time-series data and route deviation assessment data to discover potential anomalies and determine the current logistics status. It should be noted that the multimodal large model is capable of simultaneously processing text modality, image modality, audio modality, video modality, and time-series modality data. Therefore, differential time-series data and differential structured route deviation data can be directly fed into the multimodal large model for current logistics status recognition, or current multimodal logistics information can be directly input into the multimodal large model for current logistics status recognition, thereby achieving anomaly monitoring of logistics orders in transit.

[0048] In steps S401 to S403 of this embodiment, real-time logistics status identification is performed on logistics orders in transit. First, difference time series data and route deviation assessment data are calculated. Then, the current logistics status of logistics orders in transit is analyzed together with the multimodal large model, difference time series data and route deviation assessment data, making the logistics status analysis of logistics orders in transit more accurate and faster.

[0049] In one embodiment, the current logistics status of transit nodes is primarily based on differential time-series data and route deviation assessment data. In addition, the current logistics status of order creation nodes is analyzed based on fast-moving item information to determine the presence of abnormal item information. For receiving nodes, the current logistics status is accurately determined by analyzing multiple modal data during the pickup process, including pickup time, dimensions, package weight, package photos, and courier fee invoices. Similarly, data from corresponding data modalities is collected for each logistics time point and comprehensively analyzed to determine the current logistics status, enabling anomaly monitoring at different logistics time points.

[0050] Please refer to Figure 5 In some embodiments, step S403 may include, but is not limited to, steps S501 to S503: Step S501: Compare the difference time series data with a preset difference threshold to obtain first comparison information; wherein, the difference time series data includes at least one of the following: difference in logistics update frequency, difference in dwell time, and difference in transportation time. Step S502: Compare the route deviation assessment data with the preset deviation threshold to obtain the second comparison information; Step S503: Input the first comparison information and the second comparison information into the logistics status recognition model to identify the logistics status and obtain the current logistics status of each logistics order in transit.

[0051] In step S501 of some embodiments, the difference threshold is used to measure the degree of difference between the current logistics time-series data and the reference logistics time-series data, and the difference threshold includes an update frequency difference threshold, a dwell time difference threshold, and a transportation difference threshold. Comparing the difference time-series data with the difference threshold specifically involves at least one of the following: comparing the logistics update frequency difference with the update frequency difference threshold, comparing the dwell time difference with the dwell time difference threshold, and comparing the transportation time difference with the transportation difference threshold to determine first comparison information.

[0052] In step S502 of some embodiments, a higher transportation deviation assessment data indicates a greater degree of deviation from the current logistics transportation route, and a higher potential risk of anomalies in logistics orders in transit. It should be noted that the deviation threshold is used to measure whether the route deviation assessment data is too high.

[0053] In step S503 of some embodiments, the current logistics status is identified by using both the first comparison information and the second comparison information together, which can more accurately identify the abnormal logistics status of orders in transit. Specifically, if the first comparison information indicates that the difference in logistics update frequency is greater than or equal to the update frequency difference threshold, and / or the difference in dwell time is greater than or equal to the dwell time difference threshold, and / or the difference in transportation time is greater than or equal to the transportation difference threshold, or if the second comparison information indicates that the transportation deviation is greater than or equal to the deviation threshold, the current logistics status is determined to be an abnormal logistics status. Conversely, the current logistics status is determined to be a normal logistics status.

[0054] In steps S501 to S503 of this embodiment, the current logistics status of transit nodes is identified by analyzing any one of the following data: logistics update frequency difference, dwell time difference, transportation time difference, and route deviation assessment data. This can accurately identify the abnormal logistics status of logistics orders in transit and realize pre-emptive abnormal monitoring of logistics orders in transit.

[0055] In some embodiments, the current logistics multimodal information may further include: current logistics text data, current logistics product data, current logistics image data, current logistics weather data, and current logistics traffic data. As disclosed above, the current logistics text data includes the order creation time, courier item information, pickup time, and arrival time at each station for logistics orders in transit. The current logistics product data includes the package dimensions and weight corresponding to the logistics orders in transit, representing the volume, size, and weight of the packages corresponding to the logistics orders in transit. The current logistics image data includes packaging photos and package receipt photos of logistics orders in transit at the current time. The current logistics weather data is the weather data for the logistics orders in transit at the current moment, which can be used to analyze the reasons for delays in pickup, transportation, and receipt. The current logistics traffic data is used to analyze traffic congestion for logistics orders in transit at the current moment, and can also be used to analyze the reasons for delays in pickup, transportation, and receipt.

[0056] Specifically, if current logistics text data is collected, a large language model can be used to perform content understanding on the current logistics text data to output abnormal text content, and the current logistics status of the logistics order in transit can be determined based on the abnormal text content. For example, if the large language model identifies abnormal text content such as "package damaged," "transit station closed due to typhoon," or "pickup failed due to inability to contact the user" from the current logistics text data, it can determine that the current logistics status of the logistics order in transit is an abnormal logistics status.

[0057] If current logistics product data is collected, multimodal large-scale modeling can be used to analyze the data for structural anomalies to determine the current logistics status. For example, if the multimodal large-scale model identifies abnormal structural data in the current logistics product data, the current logistics status can be determined as an abnormal logistics state based on this abnormal structural data.

[0058] If current logistics image data is collected, abnormal image features are directly identified using a computer vision model, and the current logistics status of orders in transit is determined based on these abnormal image features. For example, if the computer vision model identifies the abnormal image feature as "package damage," the current logistics status is determined to be an abnormal logistics status.

[0059] If current logistics weather and traffic data are collected, an obstacle assessment is performed on these data. Specifically, the current weather data is matched with preset abnormal weather data to determine a weather anomaly score. Traffic congestion is determined based on the current traffic data, and a traffic anomaly score is then determined based on the traffic congestion score. Finally, the obstacle assessment data is determined based on both the weather and traffic anomaly scores. The obstacle assessment data is represented by a score; a higher score indicates a higher degree of transportation obstruction for the logistics order throughout its lifecycle, and vice versa. If the obstacle assessment data exceeds the preset obstacle level, the current logistics status of the logistics order in transit is determined to be an abnormal logistics status.

[0060] As disclosed above, in addition to identifying the current logistics status at transit nodes, it can also identify the current logistics status based on data collected at different logistics time points, thereby achieving comprehensive anomaly monitoring of logistics orders in transit.

[0061] In step S203 of some embodiments, if the current logistics status is a logistics abnormal status, it indicates that there is an abnormal risk in the logistics orders in transit. It is necessary to collect the logistics abnormal data at the current moment, and the logistics abnormal data belongs to the data in the current logistics multimodal information that causes the current logistics status to be a logistics abnormal status.

[0062] For example, at transit points, abnormal logistics data is extracted from the current multimodal logistics information based on the first and second comparison information. Therefore, abnormal logistics data can be current logistics time-series data and current logistics transportation routes. At other logistics time points, abnormal logistics data can be current logistics image data showing "package damage characteristics," current logistics weather data, and current logistics traffic data. Therefore, collecting abnormal logistics data corresponding to abnormal logistics states can provide it to logistics after-sales personnel at the logistics monitoring end for logistics analysis and verification. This allows after-sales personnel to handle abnormalities promptly before customer complaints arise, improving the accuracy of abnormality handling and reducing the number of logistics complaints.

[0063] In some embodiments, if the current logistics status is normal, the current logistics multimodal information of the logistics orders in transit is collected according to a preset time period, and the current logistics status of the logistics orders in transit is analyzed through the current logistics multimodal information until the logistics orders in transit reach the logistics receipt node.

[0064] In step S204 of some embodiments, logistics early warning information is generated by logistics anomaly data. Specifically, logistics prompt information, current logistics time nodes and logistics anomaly data are integrated into logistics early warning information, and the logistics early warning information is displayed in a pop-up window at the logistics monitoring terminal. This can clearly show at which logistics time node the logistics order in transit has an anomaly, and perform anomaly analysis and anomaly handling based on the logistics anomaly data, thereby improving the accuracy of anomaly handling and reducing the logistics complaint rate.

[0065] like Figure 6 As shown, Figure 6 As shown, by extracting logistics orders in transit from the in-transit order database and collecting current multimodal logistics information for each order, this information is input into a multimodal large-scale model for analysis to determine the current logistics status of the in-transit order at the current logistics time node. Specifically, if the in-transit order is at a transit node, the current logistics status can be analyzed from the following aspects: determining the degree of deviation of the current logistics transportation route from the reference route, the difference between the current dwell time and the reference dwell time at a certain transit station, and the difference between the current logistics update frequency and the reference logistics update frequency. If the current logistics status is determined to be an abnormal logistics status, corresponding logistics abnormality data is collected from the current logistics multimodal information, and logistics early warning information is generated based on this data. This early warning information is then reported to the logistics monitoring terminal, where logistics after-sales personnel can intervene before customer complaints, confirming and handling the abnormality, thereby reducing the customer complaint rate and improving customer satisfaction with logistics. In addition, if the current logistics status of an order in transit is detected as normal, continue to monitor the order in transit at subsequent logistics time points until the corresponding package is signed for.

[0066] This embodiment sets up real-time anomaly monitoring for logistics orders in transit, and also sets up post-event anomaly analysis for logistics historical orders. The anomaly analysis for logistics historical orders is explained in detail below.

[0067] Please refer to Figure 7 In some embodiments, the method for monitoring abnormal logistics orders may also include, but is not limited to, steps S701 to S704: Step S701: Receive an order exception trigger request; wherein, the order exception trigger request includes the order identifier number of the logistics history order and the exception feedback content; Step S702: Identify the abnormality category based on the abnormality feedback content to obtain the order abnormality category; Step S703: Obtain historical logistics multimodal information of historical logistics orders based on the order identifier; wherein, historical logistics multimodal information is the information flow of historical logistics orders in multiple modes throughout the entire logistics chain; Step S704: Determine the reason for the order abnormality based on the order abnormality category and historical logistics multimodal information.

[0068] In step S701 of some embodiments, the order anomaly trigger request is an anomaly trigger request initiated by the customer on the logistics platform for a logistics historical order, specifically a trigger request when "logistics order complaint / anomaly" occurs. The logistics anomaly trigger request includes an order identifier and anomaly feedback content. The order identifier serves as a unique identifier for the logistics historical order, and may specifically be the logistics order number. The anomaly feedback content is the complaint / suggestion filled out by the customer on the logistics platform for the logistics historical order. The anomaly category of the logistics historical order can be determined through the anomaly feedback content.

[0069] In step S702 of some embodiments, the order anomaly categories include: damaged parcel category, failed pickup category, delayed transportation category, and delayed delivery category, etc. These order anomaly categories can be used to analyze historical logistics order anomalies and assist in identifying the causes of the anomalies. Specifically, natural language processing technology is used to understand the anomaly feedback content to determine the core semantics, and the order anomaly category is determined based on the core semantics.

[0070] For example, natural language processing (NLP) technology can be used to understand the core semantics of abnormal customer feedback (such as "hole" or "thing is broken") to determine the order abnormality category as the package damage category.

[0071] In step S703 of some embodiments, as disclosed above, historical logistics multimodal information is stored in the historical order database. Therefore, historical logistics multimodal information of historical logistics orders is extracted from the historical order database according to the order identifier. The historical logistics multimodal information is composed of data from various data modes generated throughout the entire logistics chain. Therefore, the historical logistics multimodal information is specifically composed of data from various data modes generated by the order creation node → receiving node → transit transportation node → dispatch node → express delivery node.

[0072] In step S704 of some embodiments, different order anomaly categories need to be analyzed for anomaly causes using data from different data modalities. Therefore, by combining the order anomaly category and historical logistics multimodal information to investigate the causes of anomalies in historical logistics orders, the order anomaly cause information can be accurately identified.

[0073] In steps S701 to S704 of this embodiment, when a customer actively reports a logistics anomaly, the abnormal feedback content is first identified using natural language processing technology to determine the order anomaly category. Then, the order anomaly category and historical logistics multimodal information are combined to find the cause of the anomaly. This accurately analyzes the cause of the logistics order anomaly, so that logistics after-sales personnel can quickly respond to customers based on the cause of the anomaly, thereby improving the efficiency and accuracy of problem handling.

[0074] Please refer to Figure 8 In some embodiments, step S704 includes, but is not limited to, steps S801 to S804: Step S801: Extract target modal data and historical logistics time-series data from historical logistics multimodal information according to the order anomaly category; wherein, the target modal data is data that matches the modality with the order anomaly category; Step S802: Based on the target modal data and historical logistics time series data, anomaly location is performed to obtain the abnormal logistics node; Step S803: Extract associated modal data from historical logistics multimodal information based on abnormal logistics nodes; wherein, the associated modal data is used to assist in the analysis of abnormal causes; Step S804: Determine the order anomaly cause information through the preset anomaly cause analysis model, target modal data, and associated modal data.

[0075] In step S801 of some embodiments, this embodiment sets up a mapping relationship table between each order anomaly category and data modality. The target modality is determined in the mapping relationship table based on the order anomaly category, and target modality data is extracted from historical logistics multimodal information based on the target modality. Target modality data is key data for analyzing the causes of order anomalies. Historical logistics time-series data records the time-series information of historical logistics orders throughout the entire logistics chain, used to analyze the logistics time nodes where anomalies occurred. For example, if the order anomaly category is package damage, historical logistics image data for each logistics time node needs to be collected as target modality data. The cause of package damage can be found through historical logistics image data. If the order anomaly category is transportation delay, historical logistics weather data, historical logistics traffic data, and historical logistics transportation data during the transportation process of historical logistics orders are collected as target modality data to analyze the causes of transportation delays.

[0076] In step S802 of some embodiments, an anomaly can be located through target modal data, and the logistics time node at which the anomaly occurred can be located through historical logistics time sequence data, so as to determine the abnormal logistics node.

[0077] In step S803 of some embodiments, the associated modal data is data related to abnormal logistics nodes and abnormal order categories, which can assist in investigating the causes of anomalies. As disclosed above, in the category of damaged parcels, after determining that a parcel is damaged, the associated modal data collected can include historical logistics text data, historical logistics structured data, and historical logistics time-series data, etc., to perform cross-validation and logical reasoning by combining text modal, image modal, time-series modal, and structured modal data to find the causes of abnormal logistics historical orders.

[0078] In step S804 of some embodiments, the anomaly cause analysis model is also a model trained by a multimodal large model, which can accurately identify the causes of anomalies and output accurate order anomaly cause information. It should be noted that the anomaly cause analysis model can process data from multiple data modalities, and can combine text modalities, image modalities, time-series modalities, and structured modalities for cross-validation and logical reasoning to output accurate order anomaly causes.

[0079] In steps S801 to S804 of this embodiment, during the investigation of logistics order anomalies, it is necessary to first determine the logistics time node that caused the anomaly as the anomaly time node, and then extract the associated modal data from the historical logistics multimodal information through the anomaly time node. The target modal data and associated modal data can be combined to cross-validate and logically deduce the order anomaly cause information of the logistics historical order, so as to achieve accurate anomaly cause investigation.

[0080] Please see Figure 9 In some embodiments, step S802 includes, but is not limited to, steps S901 to S902: Step S901: Extract abnormal features from the target modal data to obtain abnormal logistics features; Step S902: Based on the abnormal logistics characteristics, perform anomaly location on historical logistics time series data to obtain abnormal logistics nodes.

[0081] In step S901 of some embodiments, in order to more accurately determine the cause of the anomaly, the target modal data is subjected to anomaly feature extraction in advance to obtain abnormal logistics features. Abnormal logistics features can be damaged image features, abnormal logistics weather features, abnormal logistics traffic features, abnormal logistics time sequence features, and abnormal logistics structured features, etc.

[0082] In step S902 of some embodiments, anomaly localization is performed in historical logistics time-series data based on abnormal logistics features. Anomaly localization involves obtaining the logistics time node where the abnormal logistics feature appears as the abnormal logistics node. For example, the target modal data is historical logistics image data. Computer vision technology is used to analyze the historical logistics image data and extract key images. For example, by comparing "packaging photos" and "package receipt images," if the abnormal logistics feature is identified as "package damage feature," the abnormal logistics feature of "package damage feature" is used to determine from the historical logistics time-series data that it occurred at a transit node, and then the transit node is determined as the abnormal logistics node.

[0083] In steps S901 to S902 of this embodiment, abnormal logistics features are first extracted, and then abnormal logistics nodes are located in historical logistics time-series data using these features, ensuring accurate location of abnormal logistics nodes. Therefore, by identifying abnormal logistics nodes, data related to the data modality can be found, thus accurately identifying the cause of order anomalies.

[0084] For example, if the abnormal logistics node is determined to be a transit node, the associated modal data includes the historical logistics traffic data, historical logistics routes, historical logistics weather data, and historical logistics time series data of the transit node. By combining the data from each data modal to conduct anomaly investigation, it can be determined that the cause of the package damage was that the package got wet in the rain, thus achieving accurate investigation of the cause of the order anomaly.

[0085] Please see Figure 10 In some embodiments, please refer to Figure 10 In some embodiments, after step S804, the method for monitoring abnormal logistics orders may also include, but is not limited to, steps S1001 to S1002: Step S1001: Input the target modal data, associated modal data and order anomaly reason information into the preset anomaly result feedback template to obtain anomaly result feedback information; Step S1002: Send the abnormal result feedback information to the logistics monitoring terminal so that the logistics monitoring terminal can verify the cause of the abnormality of the logistics historical orders.

[0086] In step S1001 of some embodiments, the anomaly feedback template is pre-built and may include any of the following: a mini-program template, a PPT template, an EXCEL template, and a WORD template, etc., to facilitate logistics after-sales personnel in selecting the corresponding anomaly feedback template according to their reading habits. Specifically, inputting the target modal data, associated modal data, and order anomaly cause information into the preset anomaly feedback template can not only clarify the root cause of the anomaly but also further verify the accuracy of the order anomaly cause analysis, thereby improving the accuracy of the order anomaly analysis.

[0087] Specifically, if the order anomaly category is "damaged parcel," and the cause of the anomaly is determined to be damage during handling at transit station X, then images of the package during transit and historical logistics time-series data are entered into a preset anomaly result feedback template to obtain anomaly result feedback information. If the order anomaly category is "delayed transport," and the cause of the anomaly is determined to be "delayed transport due to stormy weather," then the order cause information is entered into the anomaly result feedback template, followed by historical logistics time-series data and historical weather data, to obtain anomaly result feedback information.

[0088] In step S1002 of some embodiments, by sending the abnormal result feedback information with target modal data, associated modal data and order abnormality reason information to the logistics monitoring terminal, it can be used as the basis for logistics after-sales personnel to judge abnormality handling and further verify the abnormality. This allows for a quick and accurate response to customers, greatly improving the efficiency and accuracy of handling logistics abnormality issues.

[0089] In steps S1001 to S1002 of this embodiment, an abnormal result feedback information with target modal data, associated modal data and order abnormality reason information is constructed, and the abnormal result feedback information is sent to logistics after-sales personnel so that logistics after-sales personnel can accurately and quickly handle logistics abnormality issues and reduce customer complaint rates.

[0090] This application embodiment constructs a monitoring system for logistics order anomalies that integrates and collects multimodal information across the entire logistics chain. It aggregates structured data, text data, image data, weather data, and traffic data generated throughout the entire lifecycle of a logistics order, providing a data foundation for pre-event anomaly monitoring and post-event anomaly cause analysis. For orders in transit, the system collects current multimodal logistics information in real time to monitor logistics status. If an anomaly is detected, it is promptly reported to logistics after-sales personnel for timely handling, reducing the rate of logistics anomaly complaints. Simultaneously, for historical logistics orders, the system identifies the causes of anomalies based on customer complaints or anomaly feedback from historical multimodal logistics information. This eliminates the need for manual analysis by logistics after-sales personnel, saving manpower. Furthermore, during anomaly cause investigation, cross-validation of different data modalities using target modal data and related modal data allows for accurate analysis of the root causes of logistics order anomalies.

[0091] Please see Figure 11 This application embodiment also provides a monitoring device for abnormal logistics orders, which can implement the above-mentioned method for monitoring abnormal logistics orders. The device includes: The order acquisition module 1101 is used to collect the current logistics multimodal information of each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple modes of the logistics order in transit in the current logistics link; The logistics status identification module 1102 is used to identify the logistics status of each logistics order in transit based on the current logistics multimodal information, and obtain the current logistics status of each logistics order in transit. Data acquisition module 1103 is used to collect logistics abnormality data of orders in transit if the current logistics status is abnormal. The early warning module 1104 is used to generate logistics early warning information based on logistics anomaly data and send the logistics early warning information to the preset logistics monitoring terminal so that the logistics monitoring terminal can handle the abnormality of logistics orders in transit.

[0092] The specific implementation of the logistics order anomaly monitoring device is basically the same as the specific implementation of the logistics order anomaly monitoring method described above, and will not be repeated here.

[0093] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for monitoring abnormal logistics orders. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0094] Please see Figure 12 , Figure 12 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1201 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1202 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1202 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1202 and is called and executed by the processor 1201 to execute the logistics order anomaly monitoring method of the embodiments of this application. The input / output interface 1203 is used to implement information input and output; The communication interface 1204 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1205 transmits information between various components of the device (e.g., processor 1201, memory 1202, input / output interface 1203, and communication interface 1204); The processor 1201, memory 1202, input / output interface 1203 and communication interface 1204 are connected to each other within the device via bus 1205.

[0095] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for monitoring abnormal logistics orders.

[0096] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0097] The logistics order anomaly monitoring method, apparatus, electronic device, and storage medium provided in this application collect multimodal logistics information of logistics orders at each logistics time node, and determine the current logistics status of each logistics time node based on reference logistics information and multimodal logistics information. When the current logistics status is an abnormal logistics status, abnormal logistics data is collected for each logistics time node and reported to the logistics monitoring terminal. Therefore, this application transforms the analysis of logistics orders from a post-event analysis to the monitoring of the logistics status throughout the entire lifecycle of logistics orders. By combining multimodal information for monitoring, abnormal logistics status can be accurately detected, and timely feedback can be provided to logistics after-sales personnel at the logistics monitoring terminal, enabling them to respond promptly to logistics after-sales service and improve logistics service quality and operational efficiency.

[0098] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0099] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0102] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0103] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0104] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0105] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, 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. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0108] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for monitoring abnormal logistics orders, characterized in that, The method includes: Collect current logistics multimodal information for each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple data modalities of the logistics order in transit in the current logistics link; Based on the current multimodal logistics information, the logistics status of each logistics order in transit is identified to obtain the current logistics status of each logistics order in transit. If the current logistics status is an abnormal logistics status, collect the logistics abnormality data of the logistics orders in transit; Based on the abnormal logistics data, a logistics early warning message is generated and sent to a preset logistics monitoring terminal so that the logistics monitoring terminal can handle the abnormal logistics orders in transit.

2. The method according to claim 1, characterized in that, The current logistics multimodal information includes: current logistics time-series data and current logistics transportation routes; The step of identifying the logistics status of each in-transit order based on the current logistics multimodal information to obtain the current logistics status of each in-transit order includes: Differential time-series data are extracted from the current logistics time-series data based on preset reference logistics time-series data; wherein, the reference logistics time-series data is time-series data referenced for the entire logistics chain; The current logistics transportation route is evaluated for deviation based on a preset reference logistics transportation route to obtain route deviation evaluation data. The logistics status of each in-transit order is identified by using a preset logistics status identification model, the difference time series data, and the route deviation evaluation data, thereby obtaining the current logistics status of each in-transit order.

3. The method according to claim 2, characterized in that, The process of identifying the logistics status of each in-transit order using a preset logistics status identification model, the difference time series data, and the route deviation evaluation data, to obtain the current logistics status of each in-transit order, includes: The difference time series data is compared with a preset difference threshold to obtain first comparison information; wherein, the difference time series data includes at least one of the following: difference in logistics update frequency, difference in dwell time, and difference in transportation time; The route deviation assessment data is compared with a preset deviation threshold to obtain second comparison information; The first comparison information and the second comparison information are input into the logistics status recognition model to identify the logistics status, thereby obtaining the current logistics status of each logistics order in transit.

4. The method according to claim 1, characterized in that, The method further includes: Receive an order anomaly trigger request; wherein, the order anomaly trigger request includes the order identifier number of the logistics history order and the anomaly feedback content; Based on the abnormal feedback content, the abnormality category is identified to obtain the order abnormality category; The historical logistics multimodal information of the logistics historical order is obtained based on the order identifier; wherein, the historical logistics multimodal information is the information flow of the logistics historical order in multiple modes throughout the entire logistics chain; The cause of the order anomaly is determined based on the order anomaly category and the historical logistics multimodal information.

5. The method according to claim 4, characterized in that, The step of determining the order anomaly cause information based on the order anomaly category and the historical logistics multimodal information includes: Based on the order anomaly category, target modal data and historical logistics time-series data are extracted from the historical logistics multimodal information; wherein, the target modal data is data whose modality matches the order anomaly category; Anomalies are located based on the target modal data and the historical logistics time series data to obtain abnormal logistics nodes. Based on the abnormal logistics nodes, related modal data is extracted from the historical logistics multimodal information; wherein, the related modal data is used to assist in the analysis of the causes of the anomalies; The abnormal order cause information is determined by using a preset abnormal cause analysis model, the target modal data, and the associated modal data.

6. The method according to claim 5, characterized in that, The step of locating abnormal logistics nodes based on the target modal data and the historical logistics time-series data includes: Abnormal logistics features are obtained by extracting abnormal features from the target modal data; The abnormal logistics nodes are obtained by locating the anomalies in the historical logistics time series data based on the abnormal logistics characteristics.

7. The method according to claim 5, characterized in that, After determining the order anomaly cause information through the preset anomaly cause analysis model, the target modal data, and the associated modal data, the method further includes: Input the target modal data, the associated modal data, and the order anomaly reason information into a preset anomaly result feedback template to obtain anomaly result feedback information; The abnormal result feedback information is sent to the logistics monitoring terminal so that the logistics monitoring terminal can verify the cause of the abnormality of the logistics historical orders.

8. A monitoring device for abnormal logistics orders, characterized in that, The device includes: The order acquisition module is used to collect the current logistics multimodal information of each logistics order in transit; wherein, the current logistics multimodal information is the information flow of multiple modes of the logistics order in transit in the current logistics link; The logistics status identification module is used to identify the logistics status of each logistics order in transit based on the current logistics multimodal information, and obtain the current logistics status of each logistics order in transit. The data acquisition module is used to collect logistics abnormality data of the logistics orders in transit if the current logistics status is a logistics abnormality status. The early warning module is used to generate logistics early warning information based on the logistics anomaly data, and send the logistics early warning information to a preset logistics monitoring terminal so that the logistics monitoring terminal can handle the logistics orders in transit abnormally.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the logistics order anomaly monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for monitoring logistics order anomalies as described in any one of claims 1 to 7.

Images