Method and device for real-time monitoring of logistics in and out of port based on node state machine and multi-dimensional penetration

Abstract

The application relates to the technical field of logistics monitoring, and discloses a logistics in-out port real-time monitoring method based on a node state machine and multidimensional penetration; the method comprises the following steps: generating a standardized logistics event stream through a rule engine by taking scanning track data, order data and configuration data in a distributed message system as input; constructing a node state machine model, automatically attributing a waybill to one of a plurality of preset state nodes, the plurality of state nodes being mutually exclusive and having an ordered flow relationship; based on the attribution result of the node state machine, adopting second-level aggregation for multidimensional hierarchical drilling, and displaying a sortable and first-column-frozen waybill detail list through state penetration; introducing an AI algorithm chain integrating unsupervised anomaly detection, time series pattern clustering and supervised auxiliary classification based on the waybill detail list, establishing a data-driven dynamic abnormal behavior benchmark model, and having the advantages of accurate node state attribution and automatic operation diagnosis.

Description

Technical Field

[0001] This invention relates to the field of logistics monitoring technology, and in particular to a method and device for real-time monitoring of logistics entry and exit from ports based on node state machines and multi-dimensional penetration. Background Technology

[0002] Traditional logistics monitoring systems typically only display simple statuses of packages (such as "in transit" or "on delivery"), failing to accurately and in real-time pinpoint the specific delays within the complex multi-level service network of "distribution centers, branches, salespersons, and delivery stations." Branch managers often have no choice but to passively wait for problems to occur, finding it difficult to proactively and in real-time identify operational bottlenecks (such as distribution delays, undelivered packages at branches, or unsigned packages by salespersons), and even less able to drill down from macro-level indicators to specific problem tracking numbers for intervention. Existing technologies suffer from high data latency, coarse-grained monitoring, and a lack of multi-dimensional (such as three-segment codes, customer data, and salesperson data) penetration analysis capabilities.

[0003] Therefore, a method and device for real-time monitoring of logistics inbound and outbound ports based on node state machines and multi-dimensional penetration are provided to solve the above problems. Summary of the Invention

[0004] The main objective of this invention is to solve the technical problems in the prior art, such as logistics monitoring systems that can only display the simple status of express packages, cannot accurately and in real time locate the specific delay links of express packages in complex multi-level service networks, have high data latency, coarse monitoring granularity, and lack multi-dimensional penetrating analysis capabilities.

[0005] The first aspect of this invention provides a real-time monitoring method for logistics inbound and outbound operations based on node state machines and multi-dimensional penetration, the method comprising: The scanning trajectory data, order data, and configuration data from the distributed messaging system are used to generate standardized logistics event streams through a rules engine; A node state machine model is constructed, which establishes a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically attributed to one of a plurality of preset state nodes. The plurality of state nodes are mutually exclusive and have an ordered flow relationship. The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and the sortable list of waybill details with the first column frozen is displayed through state penetration. Based on the waybill details list, an AI algorithm chain integrating unsupervised anomaly detection, time-series pattern clustering, and supervised assisted classification is introduced to establish a data-driven dynamic abnormal behavior benchmark model.

[0006] Optionally, the rules in the rule engine include, but are not limited to, redirecting and mapping trajectories, and uniformly mapping non-standard nodes to standard network points or distribution nodes.

[0007] Optionally, the node state machine model includes at least an inbound state machine and an outbound state machine, wherein the inbound state machine is used to manage the final circulation status of the waybill after it enters the destination city, and the outbound state machine is used to manage the initial circulation status of the waybill from pickup to shipment to the destination city.

[0008] Optionally, the arrival state machine includes the following state nodes arranged in flow order: In-transit status node: indicates that the waybill has been sent by the distribution center, but the destination point has not yet performed any scanning operation; the in-transit status node further includes the following sub-status: total amount of goods held up in the distribution center, total amount of goods sent out in the distribution center, amount of goods in transit, and amount of goods sent out within 72 hours. At the branch status node: it indicates that the waybill has arrived at the branch but has not yet been distributed to the branch or salesperson; the branch status node further includes the following sub-status: the number of items received at the branch, the number of items distributed to the salesperson at the branch, and the number of items distributed to the branch at the branch. At the branch status node: it indicates that the waybill has been distributed to the branch, but the branch has not yet taken action or has not yet distributed it further; At the salesperson status node: it indicates that the waybill has been distributed to the salesperson but has not yet been signed for; the salesperson status node further includes the following sub-status: distributed but unsigned quantity, quantity to be delivered to the door, and quantity of missing items. At the station / cabinet status node: it indicates that the waybill has been received into the warehouse or into the cabinet but has not yet been dispatched; the station / cabinet status node further includes the following sub-status: quantity received into the warehouse / cabinet, quantity in the warehouse / cabinet, and quantity not dispatched after more than 72 hours; Abnormal Receipt / Return Status Node: Indicates that the waybill status is abnormal receipt or has been marked as a returned item; The "Received" status node indicates that the waybill has been successfully received. The outbound status machine assigns the waybill to the following status nodes arranged in the flow order based on the trajectory data of order generation, pickup, delivery, transit, and dispatch: customer status node, branch status node, central status node, terminal status node, and receipt status node.

[0009] Optionally, the flow between the status nodes is a unidirectional ordered flow, and the waybill belongs to only one of the multiple status nodes at any given time.

[0010] Optionally, the attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and a sortable list of waybill details with the first column frozen is displayed through state penetration, specifically including: Obtain the current state node and its sub-state attribution results for each waybill output by the node state machine model; Based on the query conditions input by the user, the query conditions include at least one or more of the following: time range, product type, pickup and distribution identifier, and platform type. For waybills that meet the query conditions, second-level counting and statistics are performed on each status node and its sub-indicators to generate real-time aggregated results. In response to user commands, a multi-dimensional drill-down analysis is performed on the real-time aggregation results, including: Drill down by dimension: Drill down from the current summary level to a finer-grained dimension, which includes, but is not limited to, outlets, three-segment codes, and salespersons; Status Penetration: In response to a user's click on any status node or indicator value, generate and display a detailed list of all waybill numbers that constitute that indicator; The detailed list is provided with intelligent sorting and freezing functions. The intelligent sorting function allows users to sort the list in ascending or descending order by any index column, and the freezing function fixes the first column of the list.

[0011] Optionally, the step of introducing an AI algorithm chain that integrates unsupervised anomaly detection, time-series pattern clustering, and supervised assisted classification based on the waybill detail list to establish a data-driven dynamic anomaly behavior benchmark model includes: In response to a user's click on any waybill number in the waybill details list, the system obtains and displays the complete lifecycle trajectory information of that waybill. The lifecycle trajectory information includes at least one or more of the following: the dispatch time of the previous station, the arrival time, the distribution time, the delivery clerk, the signing time, the abnormal signing type and identifier, and the marking of a problem item or a returned item. An anomaly detection module based on unsupervised learning is constructed. The anomaly detection module continuously collects the historical trajectory sequence features of all waybills, establishes a benchmark model of normal operation behavior, and calculates the anomaly score of each waybill in real time. When the anomaly score exceeds a preset dynamic threshold, the waybill is marked as suspected non-standard operation. A pattern discovery module based on temporal clustering is constructed to perform similarity clustering on waybill trajectory sequences with high anomaly scores, automatically grouping similar abnormal trajectories into the same abnormal pattern category and generating feature descriptions for each category; A classification assistance module based on supervised learning is constructed. The classifier is trained based on manually labeled sample data and performs secondary prediction on the suspected waybills marked by the anomaly detection module to help determine whether manual review is required. Based on the combined output of the anomaly detection module, pattern discovery module, and classification assistance module, missed scan diagnostic information is generated, and the order numbers of non-standard operations are automatically identified and marked. The identification results are then fed back to the monitoring dashboard for indicator display.

[0012] A second aspect of the present invention provides a real-time monitoring device for logistics inbound and outbound operations based on node state machines and multi-dimensional penetration, the device comprising: The event stream standard module is used to generate standardized logistics event streams from scan trajectory data, order data, and configuration data in a distributed messaging system through a rules engine; The state machine model construction module is used to construct a node state machine model. The node state machine model constructs a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically attributed to one of a number of preset state nodes. The number of state nodes are mutually exclusive and have an ordered flow relationship. The statistics and analysis module is used to perform multi-dimensional drill-down by aggregating the attribution results based on the node state machine at the second level, and to display a sortable list of waybill details with the first column frozen through state penetration; The decision support module is used to introduce an AI algorithm chain that integrates unsupervised anomaly detection, time-series pattern clustering, and supervised auxiliary classification based on the waybill details list, and to establish a data-driven dynamic abnormal behavior benchmark model.

[0013] A third aspect of the present invention provides an electronic device, the electronic device comprising a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the electronic device to execute the various steps of the real-time monitoring method for logistics entry and exit based on node state machine and multi-dimensional penetration as described above.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the various steps of the real-time monitoring method for logistics entry and exit based on node state machine and multi-dimensional penetration as described above.

[0015] The technical solution of this invention solves the problem of unclear status of express parcels in multi-level structures by constructing a complex node state machine model. Through the dual penetration analysis capability of state and dimension, users can not only penetrate from macro indicators to view details, but also dynamically switch analysis dimensions during the penetration process, which greatly improves the efficiency of problem location. The system's built-in functions such as "missed scan diagnosis" and "timeout warning" upgrade passive monitoring to proactive problem discovery and process auditing, providing data support for refined management of outlets. Through a configurable rule engine, it effectively handles complex business scenarios such as grid warehouse mapping, directly operated outlets, village components, misdelivered parcels, and dispatching on behalf of others, ensuring data accuracy and possessing strong industry versatility and technological advancement. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the structure of a real-time monitoring device for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] This invention provides a real-time monitoring method for logistics inbound and outbound shipments based on node state machines and multi-dimensional penetration. The method includes generating a standardized logistics event flow from scan trajectory data, order data, and configuration data in a distributed message system using a rule engine; constructing a node state machine model, where each waybill has a lifecycle; automatically attributing the waybill to one of a set of pre-defined state nodes based on the scan site and operation type of the waybill's latest trajectory, where these state nodes are mutually exclusive and have an ordered flow relationship; using second-level aggregation to perform multi-dimensional drill-down based on the attribution results from the node state machine, and displaying a sortable waybill detail list with the first column frozen through state penetration; and introducing an AI algorithm chain that integrates unsupervised anomaly detection, temporal pattern clustering, and supervised assisted classification based on the waybill detail list to establish a data-driven dynamic anomaly behavior benchmark model. This invention solves the technical problems in existing technologies, such as logistics monitoring systems typically only displaying the simple status of packages, failing to accurately and in real-time locate the specific delay points of packages in complex multi-level service networks, high data latency, coarse monitoring granularity, and lack of multi-dimensional penetration analysis capabilities.

[0018] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention 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 described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a 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.

[0019] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 The first embodiment of the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in this invention includes: The system generates standardized logistics event flows from scan trajectory data, order data, and configuration data in a distributed messaging system using a rule engine. The rules in the rule engine include, but are not limited to, redirecting and mapping trajectories, and uniformly mapping non-standard nodes to standard outlets or distribution nodes. Specifically, the scan trajectory data includes scan_type, scan_site, nxt_scan_site, etc.; order data includes system time, customer code, and address; and configuration data includes frequency, grid warehouse mapping, and directly operated outlets. Trajectories are redirected and mapped, such as for tracks with scan_type=18 / 97 / 81 / 30 / 24. Non-standard nodes include grid warehouses, trunk grid warehouses, and directly operated outlets. A node state machine model is constructed, which establishes a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically assigned to one of a set of pre-defined state nodes. These state nodes are mutually exclusive and have an ordered flow relationship. The node state machine model includes at least an inbound state machine and an outbound state machine. The inbound state machine is used to manage the final flow status of the waybill after it enters the destination city, and the outbound state machine is used to manage the initial flow status of the waybill from pickup to shipment to the destination city. The arrival state machine contains the following state nodes arranged in flow order: In-transit status node: indicates that the waybill has been sent by the distribution center, but the destination point has not yet performed any scanning operation; the in-transit status node further includes the following sub-status: total amount of goods held up in the distribution center, total amount of goods sent out in the distribution center, amount of goods in transit, and amount of goods sent out within 72 hours. At the branch status node: it indicates that the waybill has arrived at the branch but has not yet been distributed to the branch or salesperson; the branch status node further includes the following sub-status: the number of items received at the branch, the number of items distributed to the salesperson at the branch, and the number of items distributed to the branch at the branch. At the branch status node: it indicates that the waybill has been distributed to the branch, but the branch has not yet taken action or has not yet distributed it further; At the salesperson status node: it indicates that the waybill has been distributed to the salesperson but has not yet been signed for; the salesperson status node further includes the following sub-status: distributed but unsigned quantity, quantity to be delivered to the door, and quantity of missing items. At the station / cabinet status node: it indicates that the waybill has been received into the warehouse or into the cabinet but has not yet been dispatched; the station / cabinet status node further includes the following sub-status: quantity received into the warehouse / cabinet, quantity in the warehouse / cabinet, and quantity not dispatched after more than 72 hours; Abnormal Receipt / Return Status Node: Indicates that the waybill status is abnormal receipt or has been marked as a returned item; The "Received" status node indicates that the waybill has been successfully received. The outbound state machine, based on trajectory data such as order generation, pickup, delivery, transit, and dispatch, assigns waybills to the following state nodes arranged in the flow order: customer state node, branch state node, central state node, terminal state node, and receipt state node. This complex node state machine model solves the problem of unclear states in multi-level structures for express shipments.

[0020] The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and the sortable list of waybill details with the first column frozen is displayed through state penetration. Based on the waybill details list, an AI algorithm chain integrating unsupervised anomaly detection, time-series pattern clustering, and supervised assisted classification is introduced to establish a data-driven dynamic abnormal behavior benchmark model.

[0021] Please see Figure 1 The second embodiment of the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in this invention includes: The system generates standardized logistics event flows from scan trajectory data, order data, and configuration data in a distributed messaging system using a rule engine. The rules in the rule engine include, but are not limited to, redirecting and mapping trajectories, and uniformly mapping non-standard nodes to standard outlets or distribution nodes. Specifically, the scan trajectory data includes scan_type, scan_site, nxt_scan_site, etc.; order data includes system time, customer code, and address; and configuration data includes frequency, grid warehouse mapping, and directly operated outlets. Trajectories are redirected and mapped, such as for tracks with scan_type=18 / 97 / 81 / 30 / 24. Non-standard nodes include grid warehouses, trunk grid warehouses, and directly operated outlets. A node state machine model is constructed, which establishes a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically assigned to one of a set of pre-defined state nodes. These state nodes are mutually exclusive and have an ordered flow relationship. The node state machine model includes at least an inbound state machine and an outbound state machine. The inbound state machine is used to manage the final flow status of the waybill after it enters the destination city, and the outbound state machine is used to manage the initial flow status of the waybill from pickup to shipment to the destination city. The arrival state machine contains the following state nodes arranged in flow order: In-transit status node: indicates that the waybill has been sent by the distribution center, but the destination point has not yet performed any scanning operation; the in-transit status node further includes the following sub-status: total amount of goods held up in the distribution center, total amount of goods sent out in the distribution center, amount of goods in transit, and amount of goods sent out within 72 hours. At the branch status node: it indicates that the waybill has arrived at the branch but has not yet been distributed to the branch or salesperson; the branch status node further includes the following sub-status: the number of items received at the branch, the number of items distributed to the salesperson at the branch, and the number of items distributed to the branch at the branch. At the branch status node: it indicates that the waybill has been distributed to the branch, but the branch has not yet taken action or has not yet distributed it further; At the salesperson status node: it indicates that the waybill has been distributed to the salesperson but has not yet been signed for; the salesperson status node further includes the following sub-status: distributed but unsigned quantity, quantity to be delivered to the door, and quantity of missing items. At the station / cabinet status node: it indicates that the waybill has been received into the warehouse or into the cabinet but has not yet been dispatched; the station / cabinet status node further includes the following sub-status: quantity received into the warehouse / cabinet, quantity in the warehouse / cabinet, and quantity not dispatched after more than 72 hours; Abnormal Receipt / Return Status Node: Indicates that the waybill status is abnormal receipt or has been marked as a returned item; The "Received" status node indicates that the waybill has been successfully received. The outbound state machine, based on trajectory data such as order generation, pickup, delivery, transit, and dispatch, assigns waybills to the following state nodes arranged in the flow order: customer state node, branch state node, central state node, terminal state node, and receipt state node. This complex node state machine model solves the problem of unclear states in multi-level structures for express shipments.

[0022] The flow between the status nodes is a unidirectional and orderly flow, and the waybill belongs to only one of the multiple status nodes at any given time.

[0023] The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and a sortable list of waybill details with the first column frozen is displayed through state penetration; specifically including: Obtain the current state node and its sub-state attribution results for each waybill output by the node state machine model; Based on the query conditions input by the user, the query conditions include at least one or more of the following: time range, product type, pickup and distribution identifier, and platform type. For waybills that meet the query conditions, second-level counting and statistics are performed on each status node and its sub-indicators to generate real-time aggregated results. In response to user commands, a multi-dimensional drill-down analysis is performed on the real-time aggregation results, including: Drill down by dimension: Drill down from the current summary level to a finer-grained dimension, which includes, but is not limited to, outlets, three-segment codes, and salespersons; Status Penetration: In response to a user's click on any status node or indicator value, generate and display a detailed list of all waybill numbers that constitute that indicator; The detailed list is provided with intelligent sorting and freezing functions. The intelligent sorting function allows users to sort the list in ascending or descending order by any index column, and the freezing function fixes the first column of the list to improve data readability.

[0024] Based on the waybill details list, an AI algorithm chain integrating unsupervised anomaly detection, time-series pattern clustering, and supervised assisted classification is introduced to establish a data-driven dynamic abnormal behavior benchmark model.

[0025] Please see Figure 1 The third embodiment of the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in this invention includes: The system generates standardized logistics event flows from scan trajectory data, order data, and configuration data in a distributed messaging system using a rule engine. The rules in the rule engine include, but are not limited to, redirecting and mapping trajectories, and uniformly mapping non-standard nodes to standard outlets or distribution nodes. Specifically, the scan trajectory data includes scan_type, scan_site, nxt_scan_site, etc.; order data includes system time, customer code, and address; and configuration data includes frequency, grid warehouse mapping, and directly operated outlets. Trajectories are redirected and mapped, such as for tracks with scan_type=18 / 97 / 81 / 30 / 24. Non-standard nodes include grid warehouses, trunk grid warehouses, and directly operated outlets. A node state machine model is constructed, which establishes a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically assigned to one of a set of pre-defined state nodes. These state nodes are mutually exclusive and have an ordered flow relationship. The node state machine model includes at least an inbound state machine and an outbound state machine. The inbound state machine is used to manage the final flow status of the waybill after it enters the destination city, and the outbound state machine is used to manage the initial flow status of the waybill from pickup to shipment to the destination city. The arrival state machine contains the following state nodes arranged in flow order: In-transit status node: indicates that the waybill has been sent by the distribution center, but the destination point has not yet performed any scanning operation; the in-transit status node further includes the following sub-status: total amount of goods held up in the distribution center, total amount of goods sent out in the distribution center, amount of goods in transit, and amount of goods sent out within 72 hours. At the branch status node: it indicates that the waybill has arrived at the branch but has not yet been distributed to the branch or salesperson; the branch status node further includes the following sub-status: the number of items received at the branch, the number of items distributed to the salesperson at the branch, and the number of items distributed to the branch at the branch. At the branch status node: it indicates that the waybill has been distributed to the branch, but the branch has not yet taken action or has not yet distributed it further; At the salesperson status node: it indicates that the waybill has been distributed to the salesperson but has not yet been signed for; the salesperson status node further includes the following sub-status: distributed but unsigned quantity, quantity to be delivered to the door, and quantity of missing items. At the station / cabinet status node: it indicates that the waybill has been received into the warehouse or into the cabinet but has not yet been dispatched; the station / cabinet status node further includes the following sub-status: quantity received into the warehouse / cabinet, quantity in the warehouse / cabinet, and quantity not dispatched after more than 72 hours; Abnormal Receipt / Return Status Node: Indicates that the waybill status is abnormal receipt or has been marked as a returned item; The "Received" status node indicates that the waybill has been successfully received. The outbound state machine, based on trajectory data such as order generation, pickup, delivery, transit, and dispatch, assigns waybills to the following state nodes arranged in the flow order: customer state node, branch state node, central state node, terminal state node, and receipt state node. This complex node state machine model solves the problem of unclear states in multi-level structures for express shipments.

[0026] The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and a sortable list of waybill details with the first column frozen is displayed through state penetration; specifically including: Obtain the current state node and its sub-state attribution results for each waybill output by the node state machine model; Based on the query conditions input by the user, the query conditions include at least one or more of the following: time range, product type, pickup and distribution identifier, and platform type. For waybills that meet the query conditions, second-level counting and statistics are performed on each status node and its sub-indicators to generate real-time aggregated results. In response to user commands, a multi-dimensional drill-down analysis is performed on the real-time aggregation results, including: Drill down by dimension: Drill down from the current summary level to a finer-grained dimension, which includes, but is not limited to, outlets, three-segment codes, and salespersons; Status Penetration: In response to a user's click on any status node or indicator value, generate and display a detailed list of all waybill numbers that constitute that indicator; The detailed list is provided with intelligent sorting and freezing functions. The intelligent sorting function allows users to sort the list in ascending or descending order by any index column, and the freezing function fixes the first column of the list to improve data readability.

[0027] Based on the waybill details list, an AI algorithm chain integrating unsupervised anomaly detection, time-series pattern clustering, and supervised-assisted classification is introduced to establish a data-driven dynamic anomaly behavior benchmark model, which specifically includes: In response to a user's click on any waybill number in the waybill details list, the system obtains and displays the complete lifecycle trajectory information of that waybill. The lifecycle trajectory information includes at least one or more of the following: the dispatch time of the previous station, the arrival time, the distribution time, the delivery clerk, the signing time, the abnormal signing type and identifier, and the marking of a problem item or a returned item. An anomaly detection module based on unsupervised learning is constructed. This module continuously collects historical trajectory sequence features of all waybills, establishes a baseline model of normal operation behavior, and calculates the anomaly score for each waybill in real time. When the anomaly score exceeds a preset dynamic threshold, the waybill is marked as potentially non-compliant. In one specific implementation, the system continuously collects trajectory sequence features of all waybills in the background (such as the time difference of operation at each node, operation site type, operator, scanning equipment, etc.), and uses an isolated forest or autoencoder algorithm to establish a baseline model of "normal operation behavior." The anomaly score for each waybill is calculated in real time, and when the score exceeds the dynamic threshold, the waybill number is automatically marked as "potentially non-compliant," without relying on any preset rules. This algorithm can detect anomalies outside of manually defined rules, such as: frequent short-term repeated scanning, mismatch between scanning site and GPS location, and a large number of operations outside of working hours. A pattern discovery module based on temporal clustering is constructed to perform similarity clustering on waybill trajectory sequences with high anomaly scores. This automatically groups similar anomaly trajectories into the same anomaly pattern category and generates feature descriptions for each category. In one specific implementation, dynamic time warping and DBSCAN clustering are performed on the trajectory sequences of waybills with high anomaly scores, automatically grouping similar anomaly patterns into the same category. The system automatically generates feature descriptions for each newly generated anomaly category (e.g., "delivery signed for immediately after arrival with a very short interval and no delivery trajectory") and prompts operations personnel to confirm whether it belongs to a new missed scan type. After confirmation, this type of anomaly can be named and included in a new indicator on the monitoring dashboard.

[0028] A classification assistance module based on supervised learning is constructed. A classifier is trained based on manually labeled sample data and performs secondary prediction on suspected waybills marked by the anomaly detection module to help determine whether manual review is required. In a specific implementation, after the operators manually label some abnormal waybill numbers (whether they are genuine or not non-standard operations), the system uses a gradient boosting tree model to train a classifier, which can then directly perform high-precision prediction on new waybills to help determine whether manual review is required.

[0029] Based on the combined output of the anomaly detection module, pattern discovery module, and classification assistance module, missed scan diagnostic information is generated, and the order numbers of non-standard operations are automatically identified and marked. The identification results are then fed back to the monitoring dashboard for indicator display.

[0030] The above describes the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in embodiments of the present invention. The following describes the real-time monitoring device for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in embodiments of the present invention. Please refer to [link to relevant documentation]. Figure 2 The real-time monitoring device for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration in this embodiment of the invention includes, for the above embodiments: Event Stream Standard Module 201 is used to generate standardized logistics event streams from scan trajectory data, order data, and configuration data in a distributed messaging system through a rules engine. The state machine model construction module 202 is used to construct a node state machine model. The node state machine model constructs a life cycle for each waybill. Based on the scanning station and operation type of the latest trajectory of the waybill, the waybill is automatically attributed to one of a number of preset state nodes. The number of state nodes are mutually exclusive and have an ordered flow relationship. The statistics and analysis module 203 is used to perform multi-dimensional hierarchical drill-down by aggregating the attribution results based on the node state machine at the second level, and to display a sortable list of waybill details with the first column frozen through state penetration; The decision support module 204 is used to introduce an AI algorithm chain that integrates unsupervised anomaly detection, time-series pattern clustering and supervised auxiliary classification based on the waybill details list, and establish a data-driven dynamic abnormal behavior benchmark model.

[0031] above Figure 2 The real-time monitoring device for logistics entry and exit based on node state machine and multi-dimensional penetration in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The electronic equipment in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0032] Figure 3 This is a schematic diagram of the structure of an electronic device 700 provided in an embodiment of the present invention. The electronic device 700 can vary significantly due to differences in configuration or performance. It may include one or more processors 710 (e.g., one or more processors) and a memory 720, and one or more storage media 730 (e.g., one or more storage devices, including RAM, FLASH, etc.) for storing application programs 733 or data 732. The memory 720 and storage media 730 can be temporary or persistent storage. The program stored in the storage media 730 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the electronic device 700. Furthermore, the processor 710 may be configured to communicate with the storage media 730 and execute the series of instruction operations in the storage media 730 on the electronic device 700.

[0033] The electronic device 700 may also include one or more power supplies 740, one or more input / output interfaces 750, and / or one or more operating systems 731, such as FreeRTOS, Android, etc. Those skilled in the art will understand that... Figure 3 The illustrated electronic device structure does not constitute a limitation on electronic devices and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0034] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a real-time monitoring method for logistics inbound and outbound ports based on a node state machine and multi-dimensional penetration.

[0035] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0036] 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 the present invention, 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 several instructions to cause a computer device (which may be a personal computer, mobile device, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0037] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration, characterized in that, The real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration includes: The scanning trajectory data, order data, and configuration data from the distributed messaging system are used to generate standardized logistics event streams through a rules engine; A node state machine model is constructed, which establishes a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically attributed to one of a plurality of preset state nodes. The plurality of state nodes are mutually exclusive and have an ordered flow relationship. The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and the sortable list of waybill details with the first column frozen is displayed through state penetration. Based on the waybill details list, an AI algorithm chain integrating unsupervised anomaly detection, time-series pattern clustering, and supervised assisted classification is introduced to establish a data-driven dynamic abnormal behavior benchmark model.

2. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration according to claim 1, characterized in that, The rules in the rule engine include, but are not limited to, redirecting and mapping trajectories, and uniformly mapping non-standard nodes to standard network points or distribution nodes.

3. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration as described in claim 1, characterized in that, The node state machine model includes at least an inbound state machine and an outbound state machine. The inbound state machine is used to manage the final circulation status of the waybill after it enters the destination city, and the outbound state machine is used to manage the initial circulation status of the waybill from pickup to shipment to the destination city.

4. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration according to claim 3, characterized in that, The port entry state machine includes the following state nodes arranged in flow order: In-transit status node: indicates that the waybill has been sent by the distribution center, but the destination point has not yet performed any scanning operation; the in-transit status node further includes the following sub-status: total amount of goods held up in the distribution center, total amount of goods sent out in the distribution center, amount of goods in transit, and amount of goods sent out within 72 hours. At the branch status node: it indicates that the waybill has arrived at the branch but has not yet been distributed to the branch or salesperson; the branch status node further includes the following sub-status: the number of items received at the branch, the number of items distributed to the salesperson at the branch, and the number of items distributed to the branch at the branch. At the branch status node: it indicates that the waybill has been distributed to the branch, but the branch has not yet taken action or has not yet distributed it further; At the salesperson status node: it indicates that the waybill has been distributed to the salesperson but has not yet been signed for; the salesperson status node further includes the following sub-status: distributed but unsigned quantity, quantity to be delivered to the door, and quantity of missing items. At the station / cabinet status node: it indicates that the waybill has been received into the warehouse or into the cabinet but has not yet been dispatched; the station / cabinet status node further includes the following sub-status: quantity received into the warehouse / cabinet, quantity in the warehouse / cabinet, and quantity not dispatched after more than 72 hours; Abnormal Receipt / Return Status Node: Indicates that the waybill status is abnormal receipt or has been marked as a returned item; The "Received" status node indicates that the waybill has been successfully received. The outbound status machine assigns the waybill to the following status nodes arranged in the flow order based on the trajectory data of order generation, pickup, delivery, transit, and dispatch: customer status node, branch status node, central status node, terminal status node, and receipt status node.

5. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration according to claim 4, characterized in that, The flow between the status nodes is a unidirectional and orderly flow, and the waybill belongs to only one of the multiple status nodes at any given time.

6. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration according to claim 1, characterized in that, The attribution results based on the node state machine are aggregated at the second level for multi-dimensional drill-down, and a sortable list of waybill details with the first column frozen is displayed through state penetration. Specifically, this includes: Obtain the current state node and its sub-state attribution results for each waybill output by the node state machine model; Based on the query conditions input by the user, the query conditions include at least one or more of the following: time range, product type, pickup and distribution identifier, and platform type. For waybills that meet the query conditions, second-level counting and statistics are performed on each status node and its sub-indicators to generate real-time aggregated results. In response to user commands, a multi-dimensional drill-down analysis is performed on the real-time aggregation results, including: Drill down by dimension: Drill down from the current summary level to a finer-grained dimension, which includes, but is not limited to, outlets, three-segment codes, and salespersons; Status Penetration: In response to a user's click on any status node or indicator value, generate and display a detailed list of all waybill numbers that constitute that indicator; The detailed list is provided with intelligent sorting and freezing functions. The intelligent sorting function allows users to sort the list in ascending or descending order by any index column, and the freezing function fixes the first column of the list.

7. The real-time monitoring method for logistics inbound and outbound ports based on node state machine and multi-dimensional penetration according to claim 6, characterized in that, The aforementioned AI algorithm chain, which integrates unsupervised anomaly detection, temporal pattern clustering, and supervised assisted classification, is introduced based on the waybill detail list to establish a data-driven dynamic anomaly behavior benchmark model, including: In response to a user's click on any waybill number in the waybill details list, the system obtains and displays the complete lifecycle trajectory information of that waybill. The lifecycle trajectory information includes at least one or more of the following: the dispatch time of the previous station, the arrival time, the distribution time, the delivery clerk, the signing time, the abnormal signing type and identifier, and the marking of a problem item or a returned item. An anomaly detection module based on unsupervised learning is constructed. The anomaly detection module continuously collects the historical trajectory sequence features of all waybills, establishes a benchmark model of normal operation behavior, and calculates the anomaly score of each waybill in real time. When the anomaly score exceeds a preset dynamic threshold, the waybill is marked as suspected non-standard operation. A pattern discovery module based on temporal clustering is constructed to perform similarity clustering on waybill trajectory sequences with high anomaly scores, automatically grouping similar abnormal trajectories into the same abnormal pattern category and generating feature descriptions for each category; A classification assistance module based on supervised learning is constructed. The classifier is trained based on manually labeled sample data and performs secondary prediction on the suspected waybills marked by the anomaly detection module to help determine whether manual review is required. Based on the combined output of the anomaly detection module, pattern discovery module, and classification assistance module, missed scan diagnostic information is generated, and the order numbers of non-standard operations are automatically identified and marked. The identification results are then fed back to the monitoring dashboard for indicator display.

8. A real-time monitoring device for logistics inbound and outbound operations based on node state machines and multi-dimensional penetration, used in the real-time monitoring method for logistics inbound and outbound operations based on node state machines and multi-dimensional penetration as described in any one of claims 1-7, characterized in that, include: The event stream standard module is used to generate standardized logistics event streams from scan trajectory data, order data, and configuration data in a distributed messaging system through a rules engine; The state machine model construction module is used to construct a node state machine model. The node state machine model constructs a lifecycle for each waybill. Based on the scanning station and operation type of the waybill's latest trajectory, the waybill is automatically attributed to one of a number of preset state nodes. The number of state nodes are mutually exclusive and have an ordered flow relationship. The statistics and analysis module is used to perform multi-dimensional drill-down by aggregating the attribution results based on the node state machine at the second level, and to display a sortable list of waybill details with the first column frozen through state penetration; The decision support module is used to introduce an AI algorithm chain that integrates unsupervised anomaly detection, time-series pattern clustering, and supervised auxiliary classification based on the waybill details list, and to establish a data-driven dynamic abnormal behavior benchmark model.

9. An electronic device comprising a memory and at least one processor, wherein the memory stores instructions; characterized in that, The at least one processor invokes the instructions in the memory to cause the electronic device to perform the steps of the real-time monitoring method for logistics inbound and outbound ports based on node state machines and multi-dimensional penetration as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement each step of the real-time monitoring method for logistics entry and exit based on node state machine and multi-dimensional penetration as described in any one of claims 1-7.

Images