Product electronic code shipping document exception registration system

Abstract

The present application relates to logistics information management and abnormality detection technical field, specifically to a kind of product electronic code consigning single abnormal registration system, comprising: data acquisition and alignment step, according to product electronic code collection track, node access, handover and environmental record;Environmental analysis step, to the threshold part of three-axis vibration acceleration composite value and temperature deviation value respectively accumulated and generate damage assessment results;Abnormality identification step, combined determination of integrated data and damage assessment results, output delay, wrong, information error and damage abnormality registration type;Threshold adjustment and hierarchical response step, according to historical false alarm, false alarm correction probability threshold, and according to abnormal score output warning, verification work order or suspend flow instruction;The present application significantly improves the accuracy and processing efficiency of consigning single abnormal registration.

Description

Technical Field

[0001] This invention relates to the field of logistics information management and anomaly detection technology, specifically to a product electronic code consignment note anomaly registration system. Background Technology

[0002] During the logistics and transportation process, products are subject to various risks of abnormal situations such as delays, misdelivery, information errors, and damage under conditions of multi-node transfer, handover, and environmental exposure. In the process of identifying and registering abnormalities corresponding to the consignment note, it is usually necessary to collect data from multiple sources, such as transportation trajectory, node entry and exit, handover records, and transportation environment, around the same product electronic code, and judge the abnormal status of the consignment note in the entire transportation chain based on this data.

[0003] However, existing technologies for judging anomalies in consignment notes typically process trajectory information, handover information, and environmental information from different collection points separately, lacking a mechanism for unified merging and time-series organization around the same consignment note. This makes it difficult to form comprehensive data that can be directly used for judgment among scattered data. Especially when node flow anomalies and environmental damage risks coexist, relying solely on a single data item or simple rules for judgment can lead to confusion of anomaly types and affect the accuracy of anomaly registration results. At the same time, for continuous environmental factors such as temperature deviations and vibration exceeding limits, without technical means to combine quantitative analysis with the transportation process, it is difficult to effectively reflect the degree of damage risk. Furthermore, if a fixed threshold and a single response method are uniformly adopted for different logistics routes and different anomaly types, it will lead to early warning bias and processing imbalance, making it difficult to adapt to the management needs brought about by differences in routes and the severity of anomalies in actual logistics scenarios. Summary of the Invention

[0004] The purpose of this invention is to provide a product electronic code consignment note anomaly registration system, addressing the following technical problems: Existing consignment note anomaly judgment technologies have shortcomings in the unified aggregation and continuous environmental factor quantitative analysis of multi-source data when facing simultaneous node flow anomalies and environmental damage risks, as well as in threshold adjustment and response processing based on different logistics routes and anomaly types. There is an urgent need to propose a product electronic code consignment note anomaly registration system capable of unified identification and registration of anomalies within the same data chain, accurate quantification of environmental damage risks, and dynamic adjustment of early warning thresholds and tiered responses. The purpose of this invention can be achieved through the following technical solutions:

[0005] 1. A product electronic code consignment note anomaly registration system, characterized in that it includes:

[0006] The data acquisition module is used to acquire trajectory data, logistics node time data, environmental data and handover records corresponding to the product electronic code, and to generate comprehensive consignment note data including the environmental data by aligning the trajectory data, the logistics node time data, the environmental data and the handover records according to the product electronic code by time and converting them into a format.

[0007] The environmental analysis module is used to extract environmental data from the comprehensive data of the shipping order, assess the degree of damage caused by environmental factors, and generate damage assessment results.

[0008] An anomaly identification module is used to calculate the probability of occurrence of various preset anomaly types based on the comprehensive data of the consignment note and the damage assessment results, and select the anomaly type with the highest probability as the anomaly registration type of the consignment note corresponding to the electronic code of the product;

[0009] The threshold adjustment module is used to set the probability thresholds corresponding to the various anomaly types, and adjust the probability thresholds based on historical false alarms and false negatives data from historical manual verification results;

[0010] The graded response module is connected to the anomaly identification module and the threshold adjustment module respectively. It is used to compare the maximum probability value corresponding to the anomaly registration type with the corresponding probability threshold to complete the early warning judgment, calculate the anomaly score based on the maximum probability value and the preset severity coefficient, trigger the corresponding level of processing instruction based on the comparison result of the anomaly score and the preset score threshold, and output the processing status information.

[0011] 2. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that the data acquisition module includes a data acquisition unit and a data alignment unit;

[0012] The data acquisition unit collects the trajectory data and logistics node time data corresponding to the product electronic code through an RFID reader / writer, collects environmental data through environmental sensors, including real-time temperature and humidity inside the vehicle and the composite value of triaxial vibration acceleration of the goods, and collects the handover record through a logistics terminal. The handover record includes the identity of the handover person and the handover status code. The handover status code is used to represent at least one handover status among completed handover, rejection, abnormal acceptance, and missing information.

[0013] The data alignment unit uses the product electronic code as the primary key to align the trajectory data, the handover status code, and the environmental data according to time. It generates route deviation data by comparing the preset transportation route with the trajectory data, generates handover comparison data by comparing the handover status code with the preset handover rules, and calculates the node dwell time and transportation time based on the logistics node time data. It then generates comprehensive consignment note data that includes the node dwell time, transportation time, route deviation data, environmental data, and handover comparison data.

[0014] 3. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that the environmental data includes the composite value of triaxial vibration acceleration at various times and real-time temperature; the process by which the environmental analysis module calculates the damage assessment result is as follows:

[0015] The damage assessment result is generated by comprehensively calculating the portion of the triaxial vibration acceleration composite value that exceeds the preset safe vibration upper limit threshold, the preset vibration influence coefficient, the temperature deviation value of the real-time temperature from the preset safe temperature range for goods, and the preset temperature influence coefficient; wherein, the vibration influence coefficient and the temperature influence coefficient are proportional coefficients determined based on the correspondence between the environmental data of historical damaged shipping documents and the actual damage rate.

[0016] 4. A product electronic code consignment note anomaly registration system according to claim 1, characterized in that the anomaly identification module includes a first classification unit, a second classification unit, and a comprehensive judgment unit; the first classification unit extracts the node sequence, planned arrival time, and actual arrival time corresponding to the product electronic code based on the trajectory data and the logistics node time data, and combines them with the handover status code in the handover record;

[0017] Based on the extracted node sequence, the handover status code, the planned arrival time, and the actual arrival time, the first classification unit uses an anomaly classification method to output multiple first anomaly sub-probabilities corresponding to delay anomalies, misdelivery anomalies, and information error anomalies.

[0018] The second classification unit extracts the environmental data from the comprehensive data of the consignment note, combines it with the damage assessment results, extracts the ratio of the total duration of environmental exceedances to the transportation time in the entire process of the consignment note, establishes the correlation information between environmental changes and cargo damage status using the time-series correlation method, and outputs a second anomaly probability for identifying damage anomalies.

[0019] The comprehensive judgment unit merges the multiple first anomaly sub-probabilities with the second anomaly probabilities and outputs the final probabilities of the multiple anomaly types. The calculation parameters used by the anomaly classification method and the time-series correlation method are updated according to historical comparison records. The historical comparison records are used to statistically analyze the deviation between the system output probability and the actual anomaly type corresponding to the historical manual verification results.

[0020] 5. A product electronic code consignment note anomaly registration system according to claim 4, characterized in that, when the first classification unit identifies anomalies, it identifies the misdelivery anomaly based on the mismatch between the node sequence corresponding to the product electronic code and the preset target node sequence, identifies the information error anomaly based on the mutual exclusion of the handover status codes at the same node, and identifies the delay anomaly based on the comparison between the planned arrival time and the actual arrival time.

[0021] 6. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that the trajectory data corresponds to multiple logistics routes, the threshold adjustment module sets and updates the probability threshold for each of the logistics routes, and the process of adjusting the probability threshold is as follows:

[0022] Using a preset number of days as an evaluation period, the historical manual verification results of each logistics route within that period are statistically analyzed, and the actual false alarm rate and actual missed alarm rate are calculated.

[0023] The actual false alarm rate is calculated based on the sum of the number of waybills falsely reported by the system and the number of abnormal waybills correctly warned by the system within the current assessment period; the actual missed alarm rate is calculated based on the sum of the number of waybills that were missed by the system and confirmed as genuinely abnormal by historical manual verification and the number of abnormal waybills correctly warned by the system.

[0024] Based on the actual false alarm rate, the actual false negative rate, and the preset step size parameter, the threshold modification amount is determined; and the probability threshold of the current period is adjusted in the corresponding direction according to the threshold modification amount to generate a new probability threshold.

[0025] 7. A product electronic code consignment note anomaly registration system according to claim 1, characterized in that the preset severity coefficients include damage anomaly coefficient, misdelivery anomaly coefficient, information error anomaly coefficient, and delay anomaly coefficient, and the graded response module pre-stores the preset severity coefficients corresponding to the various anomaly types respectively.

[0026] The graded response module determines the anomaly score based on the maximum probability value corresponding to the anomaly registration type and the corresponding preset severity coefficient;

[0027] This module is configured with a scoring threshold based on historical anomaly handling maintenance logs. The scoring threshold includes a first scoring threshold and a second scoring threshold, where the first scoring threshold is less than the second scoring threshold. The module determines the subsequent processing procedure based on the anomaly score.

[0028] When the abnormal score is less than the first score threshold, it is determined to be a minor abnormality, and a logistics status change warning is pushed to the preset receiving and sending terminals as the processing instruction.

[0029] When the abnormal score is greater than or equal to the first score threshold and less than the second score threshold, it is determined to be a moderate abnormality, a verification work order is generated and dispatched to the current node manager for on-site review as the processing instruction;

[0030] When the anomaly score is greater than or equal to the second score threshold, it is determined to be a severe anomaly. A pause flow control command is sent to the preset logistics terminal or scheduling system of the current logistics node to stop the subsequent logistics flow of the consignment note. It is then determined whether the current node is included in the preset target node sequence: if it is included, the index position number of the current node in the preset target node sequence is extracted; if it is not included, the legal node that was correctly registered in the preset target node sequence before the anomaly occurred is taken as the base node, and the index position number of the base node in the preset target node sequence is extracted; the index position number is compared with the total number of nodes in the preset target node sequence; if the index position number is less than half of the total number of nodes, a return command is triggered as the processing command; if the index position number is greater than or equal to half of the total number of nodes, a transfer and reissue command is triggered as the processing command.

[0031] 8. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that the system further includes a record management module;

[0032] The record management module is used to record the anomaly registration type output by the anomaly identification module, the processing instructions output by the hierarchical response module, the historical manual verification results, and the processing status information.

[0033] The record management module has a built-in report template, generates carrier assessment data based on the recorded data, and feeds back the processing status information to the anomaly identification module to correct the calculation parameters in the anomaly type probability calculation.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. This system uses a data acquisition module to align and uniformly convert the trajectory, time, environment, and handover data obtained based on the product's electronic code according to time. This design solves the problems of scattered processing of multi-source data and lack of time-series organization in existing technologies. It integrates scattered data into complete consignment note data, avoiding confusion caused by relying on single data for judgment, and effectively improving the accuracy of consignment note anomaly type identification and registration.

[0036] 2. This system extracts environmental data from comprehensive data through the environmental analysis module to assess the degree of damage, and the anomaly identification module calculates the probability of various anomalies based on the damage assessment results. This design solves the problem that continuous environmental factors are difficult to combine with the quantitative analysis of the transportation process, can effectively reflect the actual degree of damage risk during transportation, accurately distinguish between node flow anomalies and environmental damage risks, and improve the judgment ability in complex scenarios.

[0037] 3. This system adjusts the probability threshold based on historical false alarms and missed alarms using a threshold adjustment module, and triggers corresponding level processing instructions by calculating anomaly scores using a graded response module combined with severity coefficients. This setting overcomes the shortcomings of fixed thresholds and single response methods, adapts to differences in routes and severity in logistics scenarios, reduces early warning deviations, and achieves precise graded control and dynamic handling of abnormal status of consignment notes. Attached Figure Description

[0038] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0039] Figure 1 This is a schematic diagram of the structure of an electronic code consignment note exception registration system according to the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0041] like Figure 1 As shown, a product electronic code consignment note anomaly registration system includes:

[0042] The data acquisition module is used to acquire trajectory data, logistics node time data, environmental data and handover records corresponding to the product electronic code, and to generate comprehensive consignment note data including environmental data by aligning the trajectory data, logistics node time data, environmental data and handover records with time and converting the format based on the product electronic code.

[0043] The environmental analysis module is used to extract environmental data from the comprehensive data of the shipping order, assess the degree of damage caused by environmental factors, and generate damage assessment results.

[0044] The anomaly identification module is used to calculate the probability of occurrence of various preset anomaly types based on the comprehensive data of the consignment note and the damage assessment results, and select the anomaly type with the highest probability as the anomaly registration type of the consignment note corresponding to the product electronic code;

[0045] The threshold adjustment module is used to set probability thresholds for various anomaly types and adjust the probability thresholds based on historical false alarm and false negative data from historical manual verification results.

[0046] The graded response module is connected to the anomaly identification module and the threshold adjustment module respectively. It is used to compare the maximum probability value corresponding to the anomaly registration type with the corresponding probability threshold to complete the early warning judgment, calculate the anomaly score based on the maximum probability value and the preset severity coefficient, trigger the corresponding level of processing instruction based on the comparison result of the anomaly score and the preset score threshold, and output the processing status information.

[0047] The data acquisition module includes a data acquisition unit and a data alignment unit;

[0048] The data acquisition unit collects trajectory data and logistics node time data corresponding to the product's electronic code through an RFID reader / writer, collects environmental data through environmental sensors, including real-time temperature and humidity inside the vehicle compartment and the composite value of triaxial vibration acceleration of the goods, and collects handover records through the logistics terminal. The handover records include the identity of the handover personnel and the handover status code. The handover status code is used to represent at least one handover status among completed handover, rejection, abnormal acceptance, and missing information.

[0049] The data alignment unit uses the product electronic code as the primary key, aligns the trajectory data, handover status code, and environmental data according to time, generates route deviation data by comparing the preset transportation route with the trajectory data, generates handover comparison data by comparing the handover status code with the preset handover rules, and calculates the node dwell time and transportation time based on the logistics node time data, generating comprehensive consignment note data that includes node dwell time, transportation time, route deviation data, environmental data, and handover comparison data.

[0050] Environmental data includes the composite values ​​of triaxial vibration acceleration at various times and real-time temperature; the process by which the environmental analysis module calculates the damage assessment results is as follows:

[0051] The damage assessment result is generated by comprehensively calculating the portion of the triaxial vibration acceleration composite value that exceeds the preset safe vibration upper limit threshold, the preset vibration influence coefficient, the temperature deviation value of the real-time temperature from the preset safe temperature range for goods, and the preset temperature influence coefficient. Among them, the vibration influence coefficient and the temperature influence coefficient are proportional coefficients determined based on the correspondence between the environmental data of historical damaged shipping documents and the actual damage rate.

[0052] This implementation method focuses on the abnormal registration of electronic code consignment notes during the multi-node logistics flow of products. It gathers the trajectory records, node entry and exit records, handover records, and transportation environment records of the consignment note during transportation, based on the same electronic code of the product, and completes the abnormal identification and registration within the data chain of the same consignment note. It organizes the data scattered at different collection terminals into unified data, quantifies the damage risk based on environmental factors, and outputs the abnormal registration type.

[0053] In this embodiment, the data acquisition module includes a data acquisition unit and a data alignment unit; the data acquisition unit maintains a data connection with the RFID reader, environmental sensors and logistics terminal; the data acquisition unit acquires trajectory data and logistics node time data corresponding to the product electronic code through the RFID reader, acquires environmental data through the environmental sensor, including real-time temperature and humidity inside the vehicle and the composite value of triaxial vibration acceleration of the goods, and acquires handover records through the logistics terminal;

[0054] The handover record must include at least the identity of the person handing over the goods and the handover status code. The handover status code is used to indicate at least one of the following states: handover completed, rejection, abnormal receipt, or missing information. After the data from each source enters the system, it is not directly used as the basis for anomaly judgment. Instead, it is first merged into the corresponding consignment note name according to the product electronic code, and then sorted in order according to the time information attached to each item.

[0055] During the data processing, the data alignment unit uses the product electronic code as the primary key and writes the trajectory data, handover status code and environmental data into the data record of the same consignment according to the time correspondence. It also calculates the dwell time of the consignment at each node and the transportation time between adjacent nodes by combining the logistics node time data. For trajectory data, the system compares the actual sequence of nodes passed with the target node sequence in the preset transportation route one by one, counts the number of mismatched nodes and generates route deviation data accordingly.

[0056] For handover records, the system compares the handover status code with the preset handover rules to generate handover comparison data. The preset handover rules are a set of legal handover status codes for the current logistics node, and the handover comparison data is a Boolean matching identifier that indicates whether the actual handover status code belongs to the set of legal handover status codes.

[0057] After the above processing, comprehensive data of the consignment note is generated. This comprehensive data includes at least node dwell time, transportation time, route deviation data, environmental data, and handover comparison data. It is written into the data storage area corresponding to the consignment note for direct access by the subsequent environmental analysis module and anomaly identification module, thereby avoiding the need to backtrack to the original data collection points in the subsequent judgment stage.

[0058] The environmental analysis module extracts environmental data from the comprehensive data of the consignment note and calculates the probability of damage caused by environmental factors; the environmental sensors perform isochronous sampling at fixed standard time intervals during each node of the flow and during transportation; for the composite value of triaxial vibration acceleration at each moment, the system compares it with the preset safe vibration upper limit threshold one by one, and when the composite value is higher than the threshold, the excess part is retained;

[0059] When the synthesized value is not higher than the threshold, it is treated as zero; the results obtained at each time are directly accumulated and then multiplied by the preset vibration influence coefficient to obtain the first result; for the real-time temperature at each time, the system compares it with the safe temperature range of the goods. When the real-time temperature is within the range, the corresponding temperature deviation value is treated as zero; when the real-time temperature is lower than the lower limit of the range or higher than the upper limit of the range, the temperature deviation value is taken as the absolute value of the difference between the real-time temperature and the corresponding boundary value; the temperature deviation values ​​at each time are accumulated and then multiplied by the preset temperature influence coefficient to obtain the second result. The temperature influence coefficient also internalizes the above-mentioned fixed time interval factor.

[0060] The first result is added to the second result to generate the damage assessment result. The vibration influence coefficient and temperature influence coefficient can be predetermined based on the correspondence between the environmental data of historical damaged shipments and the actual damage rate, and are updated during the system maintenance cycle. For example: assuming the safe temperature range of the goods is 2℃ to 8℃, the preset upper limit threshold of safe vibration is 5g, the preset vibration influence coefficient is 0.05, and the temperature influence coefficient is 0.1; at three consecutive collection moments in a certain transportation, the recorded composite values ​​of triaxial vibration acceleration are 4g, 6g, and 7g, and the real-time temperatures are 8℃, 10℃, and 11℃, respectively; for the vibration data, the excess at the three moments are 0g, 1g, and 2g, respectively, with a cumulative value of 3g, which is multiplied by the vibration influence coefficient of 0.05 to obtain the first result of 0.15; for the temperature data, the temperature deviation values ​​corresponding to the upper limit of 8℃ at the three moments are 0℃, 2℃, and 3℃, respectively, with a cumulative value of 5℃, which is multiplied by the temperature influence coefficient of 0.1 to obtain the second result of 0.5.

[0061] The final damage assessment result is The vibration impact coefficient and temperature impact coefficient are set as follows: The system extracts shipping orders confirmed to be environmentally damaged from historical data, and calculates the cumulative average of the vibration exceeding the limit and the cumulative average of the temperature deviation during the transportation of this batch of shipping orders.

[0062] The actual damage rate of the batch of goods is allocated to vibration and temperature according to the preset environmental attribution weights. The preset environmental attribution weights are set by the system based on the physical characteristics of different types of goods. The specific calibration method is as follows: extract the seismic vulnerability rating and temperature sensitivity rating of the corresponding goods from the packaging engineering standard library, calculate the percentage of each in the total rating, and use these percentages as the attribution weights of vibration and temperature, respectively.

[0063] Among them, the seismic vulnerability rating and the temperature sensitivity rating are both pre-set dimensionless quantitative values. The total rating is the arithmetic sum of the seismic vulnerability rating value and the temperature sensitivity rating value of the corresponding goods. Then, the damage rate obtained by allocation is divided by the corresponding cumulative average value, thereby calculating the corresponding proportional coefficient for each unit of vibration exceeding the limit or temperature deviation.

[0064] Based on the above, the anomaly identification module uses the comprehensive data of the consignment note and the damage assessment results as the basis for judgment, and calculates the probability of occurrence of various preset anomaly types. The anomaly types include at least delay anomaly, misdelivery anomaly, information error anomaly, and damage anomaly. The system selects the maximum value from the probabilities corresponding to each anomaly type and determines the anomaly registration type corresponding to the maximum value as the anomaly registration type of the electronic code consignment note for that product. This result is sent to the hierarchical response module, while retaining the maximum probability value for comparison with the corresponding probability threshold.

[0065] The threshold adjustment module pre-sets probability thresholds for various anomaly types. These thresholds are not fixed but are adjusted based on historical false alarms and missed alarms obtained from historical manual verification results. The adjusted thresholds are saved as the current valid thresholds for the graded response module to read. After receiving the anomaly registration type, the maximum probability value corresponding to that type, and the corresponding probability threshold, the graded response module first performs an early warning judgment.

[0066] When the maximum probability value reaches the corresponding threshold, the system enters the processing instruction generation process. This module further multiplies the maximum probability value by a preset severity coefficient to obtain an anomaly score. The anomaly score is then compared with a preset score threshold to determine the processing level and output the corresponding processing instruction, while simultaneously generating processing status information. The processing status information includes one of the following: pending, dispatched, processing, or completed status, and is associated with and saved with the corresponding consignment note.

[0067] If individual environmental sensors fail to return valid data within a certain data collection period, the system can read a valid environmental record from the previous valid period of the consignment as a temporary value and add a missing identifier to that record, enabling subsequent modules to recognize the data and continue calculations according to predetermined rules. When a sensor fails to return valid data, and the current period is the first data collection period for the consignment, resulting in the absence of a previous valid period, the system will extract the arithmetic mean of the upper and lower limits of the cargo's safe temperature range as a temporary temperature value and extract a value of... The combined triaxial vibration acceleration value is used as a temporary vibration value and written into the record of this acquisition cycle;

[0068] If there is a time sequence conflict in the trajectory data or handover records, the data alignment unit will prioritize retaining records with complete sources and continuous time identifiers, and retain conflicting records separately to avoid directly overwriting the already formed consignment note composite data; for scattered records that cannot be aligned, the system can continue to save them under the corresponding product electronic code name, and then merge them into the composite data after the relevant time records are supplemented later, thereby avoiding the interruption of the abnormal registration process of the entire consignment note due to the lack of single data items;

[0069] In this logistics flow scenario, from the moment a consignment note enters the transportation chain, it continuously generates a comprehensive record corresponding to the product's electronic code. The damage assessment results provided by the environmental analysis module, together with the node time consumption, route deviation, and handover comparison results, constitute the basis for subsequent anomaly identification. This enables the system to identify delays or misdeliveries in the processing of the same consignment note and to register the damage risks caused by temperature deviations or excessive vibration based on environmental data, thereby completing the unified registration of consignment note anomaly types.

[0070] In this embodiment, the anomaly identification module includes a first classification unit, a second classification unit, and a comprehensive judgment unit; the first classification unit extracts the node sequence, planned arrival time, and actual arrival time corresponding to the product electronic code based on trajectory data and logistics node time data, and combines them with the handover status code in the handover record;

[0071] The first classification unit, based on the extracted node sequence, handover status code, planned arrival time and actual arrival time, uses an anomaly classification method to output multiple first anomaly subprobabilities corresponding to delay anomalies, misdelivery anomalies and information error anomalies.

[0072] The second classification unit extracts environmental data from the comprehensive data of the consignment note. Combined with the damage assessment results, it extracts the ratio of the total duration of environmental exceedances to the transportation time throughout the entire consignment note process. It then uses a time-series correlation method to establish the correlation information between environmental changes and cargo damage status, and outputs a second anomaly probability for identifying damage anomalies.

[0073] The comprehensive judgment unit merges multiple first anomaly sub-probabilities and second anomaly probabilities to output the final probabilities of various anomaly types. The calculation parameters used by the anomaly classification method and the time-series correlation method are updated according to historical comparison records. The historical comparison records are used to statistically analyze the deviation between the system output probability and the actual anomaly type corresponding to the historical manual verification results.

[0074] When identifying anomalies, the first classification unit identifies misdelivery anomalies based on the mismatch between the node sequence corresponding to the product's electronic code and the preset target node sequence; identifies information error anomalies based on the occurrence of mutually exclusive handover status codes at the same node; and identifies delay anomalies based on the comparison between the planned arrival time and the actual arrival time.

[0075] The anomaly identification module is further divided into a first classification unit, a second classification unit, and a comprehensive judgment unit, which respectively handle node flow anomalies and environmental damage anomalies, and merge their respective results into the final anomaly probability.

[0076] In this embodiment, the comprehensive data of the consignment note, in addition to the aforementioned node dwell time, transportation time, route deviation data, environmental data, and handover comparison data, also includes node sequences, handover status codes in the handover records, planned arrival time, and actual arrival time. The first classification unit extracts the node sequences, handover status codes, planned arrival time, and actual arrival time from the comprehensive data, and uses an anomaly classification method to calculate and output the corresponding first anomaly sub-probabilities for delay anomalies, misdelivery anomalies, and information error anomalies. For misdelivery anomalies, the system compares the node sequence formed by the product's electronic code in actual circulation with a preset target node sequence.

[0077] When the two do not match, the first classification unit multiplies the proportion of the number of mismatched nodes to the total number of nodes in the preset target node sequence by a preset base probability weight for mis-transmission, thereby quantifying and calculating the first sub-probability of the mis-transmission anomaly. For information error anomalies, the system checks whether mutually exclusive handover status codes appear under the same node, such as the simultaneous occurrence of completed handover and missing information, or completed handover and rejection, which cannot coexist, in the same node record. If such a conflict exists, the first sub-probability of the information error anomaly is directly assigned to a preset conflict extreme probability, such as 0.95. If there is no conflict, it is assigned to a minimum value.

[0078] For delay anomalies, the system compares the planned arrival time with the actual arrival time. When the actual arrival time deviates from the planned arrival time, the system multiplies the actual time exceeding the planned time by a preset unit delay penalty coefficient to obtain the first anomaly sub-probability corresponding to the delay anomaly. The system sets an absolute upper limit truncation mechanism for the calculation result of this sub-probability: if the product of the time measure and the unit delay penalty coefficient is greater than 1, the first anomaly sub-probability corresponding to the delay anomaly is forcibly truncated and set to 1.

[0079] Similarly, the calculation of the error transmission anomaly sub-probability also applies a hard truncation of no more than 1 to avoid overflow of the sub-probability value; for example: assuming the preset target node sequence contains 5 nodes, 2 nodes in the actual flow are mismatched, and the preset error transmission base probability weight is 0.5, then the calculated error transmission anomaly sub-probability is... ;

[0080] If the actual arrival time exceeds the planned time by 10 hours, and the preset unit delay penalty coefficient is 0.03 / hour, then the product is... If no overflow occurs, the probability of delay anomaly is 0.3; if the delay exceeds 40 hours, the product is 1.2, triggering the upper limit truncation mechanism, and the actual probability of delay anomaly is 1; after multiple first anomaly probabilities are formed, they are written into the anomaly probability record area of ​​the current consignment note, waiting to be called by the comprehensive judgment unit.

[0081] The second classification unit specifically processes environmental data. This unit extracts environmental data from the comprehensive data of the shipping order and incorporates the damage assessment results output by the previous environmental analysis module into the judgment. Since damage anomalies are often related to the continuous process of environmental changes rather than determined by isolated records at a single moment, the second classification unit uses a time-series correlation method to extract the corresponding information between the environmental change process and the damage status of the goods, and outputs a second anomaly probability for identifying damage anomalies.

[0082] Temporal correlation is based on the environmental records of the same consignment in the continuous transportation process. It determines the correspondence between the cumulative situation of vibration exceeding the limit and temperature deviation in the entire process of the consignment and the damage assessment result. The specific quantification steps are as follows: the system traverses the environmental data in chronological order and extracts the total duration of environmental exceeding the limit in the entire process of the consignment. That is, for all sampling times that meet any of the exceeding conditions, such as the combined triaxial vibration acceleration value being greater than the safety threshold or the real-time temperature deviating from the safety range, the differences between the corresponding adjacent sampling timestamps are globally discretely accumulated.

[0083] The duration is divided by the transportation time of the shipment to obtain the environmental degradation time percentage. The damage assessment result generated by the previous environmental analysis module is multiplied by the environmental degradation time percentage to obtain the comprehensive damage factor. This comprehensive damage factor is then input into a preset probability mapping transformation logic. Specifically, the comprehensive damage factor is directly divided by a preset maximum limit factor value. The maximum limit factor value is a constant set based on the maximum comprehensive damage factor obtained from the statistics of historical damaged shipments. When the calculated result is greater than 1, it is truncated to 1, thus converting it into a value between 0 and 1, which serves as the second anomaly probability for identifying damage anomalies. After the second anomaly probability is generated, it is also written into the anomaly probability record area and is read by the comprehensive judgment unit along with multiple sub-probabilities of the first classification unit.

[0084] The comprehensive judgment unit summarizes multiple first abnormal sub-probabilities and second abnormal probabilities under the same consignment note, and merges them to output the final probabilities of multiple abnormal types. This output result does not simply retain the original values ​​of each branch, but forms a comparable abnormal probability set according to the preset merging rules, and then selects the abnormal type with the highest probability as the abnormal registration type of the consignment note corresponding to the electronic code of the product.

[0085] The preset merging rules are as follows: multiple first anomaly sub-probabilities are directly mapped to the corresponding anomaly type fields in the set, and the second anomaly probabilities are mapped to the damage anomaly field. If an anomaly sub-probability takes a minimum value because it does not meet the triggering condition, such as the information error anomaly being set to a fixed basic minimum probability of 0.01 when there is no state conflict, then the set fields are filled with this basic minimum probability. In this way, the unified construction of the probability set is completed through non-interference parallel assembly logic, providing a consistent data format for subsequent peer comparisons.

[0086] Because logistics routes, handover habits, and environmental changes can vary in different statistical periods, the calculation parameters used by the anomaly classification method and the time-series correlation method are updated based on historical comparison records. The historical comparison records store the deviation between the anomaly probability previously output by the system and the actual anomaly type in the historical manual verification results. The system reads this deviation record during the maintenance cycle and adjusts the subsequent calculation parameters to make the calculation results closer to the subsequent manual verification conclusions.

[0087] If the consignment note's aggregate data is missing either the planned arrival time or the actual arrival time, the first classification unit will not directly output a deterministic delay anomaly judgment. Instead, it will directly assign the first anomaly sub-probability corresponding to the delay anomaly to zero and add a missing item marker to the anomaly probability record area of ​​the consignment note. The delay anomaly sub-probability will be recalculated after subsequent node data is entered.

[0088] If there is an interruption in the node sequence, the system does not immediately regard the interruption as a mistransmission. Instead, it first performs a local comparison based on the recorded node order. Only when a mismatch is found in the local comparison will the sub-probability corresponding to the mistransmission anomaly be increased. In the case of handover status code conflict, if the conflict record comes from repeated upload and the content is completely identical, the system can retain it as a single valid record to avoid the information error anomaly probability being artificially high due to repeated submissions by the terminal.

[0089] In this logistics flow scenario, the first classification unit focuses on handling the sequence, handover, and time limits of the consignment note during the node advancement process, while the second classification unit focuses on handling the correspondence between the transportation environment and the risk of damage. The comprehensive judgment unit unifies the two into the same set of abnormal probabilities, enabling the system to distinguish whether the multiple types of concurrent abnormalities of the same consignment note are due to deviations in node flow or environmental factors that have created a risk of damage, thus providing a consistent data foundation for subsequent early warning judgments and abnormal registration.

[0090] In this embodiment, the trajectory data corresponds to multiple logistics routes. The threshold adjustment module sets and updates the probability threshold for each logistics route. The process of adjusting the probability threshold is as follows:

[0091] Using a preset number of days as an evaluation period, the historical manual verification results for each logistics route within that period are statistically analyzed, and the actual false alarm rate and actual missed alarm rate are calculated.

[0092] The actual false alarm rate is calculated based on the sum of the number of waybills falsely reported by the system and the number of abnormal waybills correctly warned by the system during the current assessment period; the actual missed alarm rate is calculated based on the sum of the number of waybills that were missed by the system and confirmed as genuine abnormalities by historical manual verification and the number of abnormal waybills correctly warned by the system.

[0093] Based on the actual false alarm rate, the actual false negative rate, and the preset step size parameter, the threshold modification amount is determined; and the probability threshold of the current period is adjusted in the corresponding direction according to the threshold modification amount to generate a new probability threshold.

[0094] The preset severity coefficients include damage anomaly coefficient, misdelivery anomaly coefficient, information error anomaly coefficient, and delay anomaly coefficient. The graded response module has preset severity coefficients corresponding to various anomaly types.

[0095] The graded response module determines the anomaly score based on the maximum probability value corresponding to the anomaly registration type and the corresponding preset severity coefficient;

[0096] This module is configured with scoring thresholds based on historical anomaly handling maintenance logs. The scoring thresholds include a first scoring threshold and a second scoring threshold, with the first scoring threshold being lower than the second scoring threshold. The module determines the subsequent processing procedure based on the anomaly score.

[0097] When the anomaly score is less than the first score threshold, it is judged as a minor anomaly, and a logistics status change warning is pushed to the preset receiving and sending terminals as a processing instruction.

[0098] When the abnormal score is greater than or equal to the first score threshold and less than the second score threshold, it is judged as a moderate abnormality, a verification work order is generated and dispatched to the current node manager for on-site review as a processing instruction;

[0099] When the abnormal score is greater than or equal to the second score threshold, it is judged as a severe abnormality. A pause flow control command is sent to the preset logistics terminal or scheduling system of the current logistics node to stop the subsequent logistics flow of the consignment. It is also determined whether the current node is included in the preset target node sequence. If it is included, the index position number of the current node in the preset target node sequence is extracted.

[0100] If not included, the valid node that was correctly registered in the preset target node sequence before the anomaly occurred will be used as the base node, and the index position number of the base node in the preset target node sequence will be extracted; the index position number will be compared with the total number of nodes in the preset target node sequence; if the index position number is less than half of the total number of nodes, a rollback instruction will be triggered as a processing instruction; if the index position number is greater than or equal to half of the total number of nodes, a transfer and reissue instruction will be triggered as a processing instruction.

[0101] The system also includes a record management module;

[0102] The record management module is used to record the anomaly registration types output by the anomaly identification module, as well as the processing instructions, historical manual verification results, and processing status information output by the hierarchical response module.

[0103] The record management module has built-in report templates, which generate carrier assessment data based on the recorded data and feed back the processing status information to the anomaly identification module to correct the calculation parameters in the anomaly type probability calculation.

[0104] This implementation method adjusts the probability threshold according to the logistics route, calculates the anomaly score according to the anomaly type, and records and provides feedback on the processing results.

[0105] In this embodiment, the threshold adjustment module stores a set of probability thresholds for each logistics route corresponding to the trajectory data. The system uses a preset number of days as an evaluation period. After each evaluation period, it reads the historical manual verification results corresponding to each logistics route within that period and calculates the actual false alarm rate and the actual missed alarm rate accordingly. The system subtracts the actual missed alarm rate from the actual false alarm rate and multiplies the difference by a preset step size parameter to obtain the threshold modification amount. The specific value of the preset step size parameter is based on the system's pre-calibrated requirement for limiting the magnitude of a single threshold adjustment. An exemplary value is 0.05, which limits the boundary of the threshold adjustment range to be raised or lowered.

[0106] The threshold modification is superimposed on the probability threshold of the corresponding line in the current period to form a preliminary new threshold value; the threshold adjustment module further performs over-limit verification: setting a lower limit and an upper limit for the absolute probability threshold; when the preliminary new threshold value is lower than the lower limit, the generated new probability threshold is set to the lower limit; when the preliminary new threshold value is higher than the upper limit, the new probability threshold is set to the upper limit.

[0107] In the graded response module, the system pre-stores the damage anomaly coefficient, the misdelivery anomaly coefficient, the information error anomaly coefficient, and the delay anomaly coefficient; after the anomaly identification module outputs the anomaly registration type and its maximum probability value, the graded response module first reads the severity coefficient corresponding to the anomaly type and multiplies the two to obtain the anomaly score; the module further configures a first score threshold and a second score threshold, and the first score threshold is less than the second score threshold.

[0108] The specific criteria for determining the first and second scoring thresholds are derived from historical anomaly handling and maintenance logs: the system sets the lowest historical anomaly score corresponding to the medium-severity anomaly work order that is confirmed to require on-site review in the historical records as the first scoring threshold, and sets the lowest historical anomaly score corresponding to the severe anomaly control instruction that is confirmed to require stopping the flow in the historical records as the second scoring threshold, thereby ensuring that the threshold division is directly linked to the severity of actual business handling.

[0109] If the anomaly score is less than the first score threshold, the system determines it as a minor anomaly and pushes a logistics status change warning to the preset receiving and shipping terminals; if the anomaly score is greater than or equal to the first score threshold and less than the second score threshold, the system determines it as a moderate anomaly, generates a verification work order and dispatches it to the current node manager for on-site review; if the anomaly score is greater than or equal to the second score threshold, the system determines it as a severe anomaly, sends a pause flow control command to the preset logistics terminal or scheduling system of the current logistics node to stop the subsequent logistics flow of the consignment, and triggers a return or reassignment command; the processing results at each level form corresponding processing status information and are associated with and saved with the consignment identifier;

[0110] The record management module maintains data interaction with the anomaly identification module and the hierarchical response module. This module records the anomaly registration types output by the anomaly identification module, as well as the processing instructions, historical manual verification results, and processing status information output by the hierarchical response module. Based on the above records, the record management module can generate carrier assessment data according to the built-in report template. This assessment data can be organized around routes, anomaly types, processing timeliness, and consistency of verification results for subsequent statistical analysis.

[0111] Meanwhile, the processing status information is fed back to the anomaly identification module to correct the calculation parameters in the anomaly type probability calculation. The specific correction logic is as follows: the system uses a preset number of processed consignment notes to form an evaluation window, and compares the occurrence probability of each anomaly type output by the system in this window with the final historical manual verification results. If the historical manual verification result of a certain type of anomaly, such as a mis-issuance anomaly, is that it actually occurred, but the first anomaly sub-probability previously calculated by the system is lower than the corresponding probability threshold in three consecutive judgments, resulting in a missed report, then the system automatically adjusts the calculation parameter corresponding to the anomaly type, such as the mis-issuance basic probability weight, by a preset parameter fine-tuning step size, for example, by 0.02.

[0112] Conversely, if the verification result is that the event did not occur, but the probability calculated by the system is higher than the corresponding probability threshold, resulting in a false alarm, then the calculated parameter corresponding to the anomaly type will be subtracted by a value equal to the preset parameter fine-tuning step size; if the parameter value after subtraction is less than or equal to the preset lower limit value of the parameter limit... Then the calculated parameter is forcibly truncated and taken as a value. This is to prevent negative probabilities from causing subsequent data format distortion;

[0113] A closed-loop feedback control is formed between the anomaly identification result, the subsequent manual verification result, and the final processing status. The system updates the anomaly probability of the subsequent output by fine-tuning the step size through preset parameters.

[0114] If the historical manual verification results of a logistics route do not reach the preset statistical sample threshold in the current evaluation period, the threshold adjustment module will keep the probability threshold of the route unchanged in the previous period and add a statistical insufficiency mark to the threshold record of the route. The update will be performed after the data of the subsequent evaluation period is supplemented.

[0115] If the same consignment note falls within the switching boundary of multiple route records at the same time, the system uses the current route corresponding to the actual trajectory data of the consignment note as the threshold reading basis to avoid the same consignment note being repeatedly subject to multiple sets of thresholds in one warning judgment; for the hierarchical response module, if the receiving and dispatching terminal, logistics terminal or scheduling system does not return the receiving result within the current processing cycle, the system will still retain the generated processing instruction and mark the processing status information as pending confirmation so that it can be reissued or manually followed up later, without affecting the abnormal registration of the consignment note that has been completed;

[0116] The system uses threshold updates based on the route dimension to make early warning judgments on consignment notes, classifies different anomaly types based on anomaly scores, and records anomaly registration, manual verification, and processing results through the record management module for carrier assessment and anomaly identification parameter correction.

[0117] It should be noted that the above 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A product electronic code consignment note anomaly registration system, characterized in that, include: The data acquisition module is used to acquire trajectory data, logistics node time data, environmental data and handover records corresponding to the product electronic code, and to generate comprehensive consignment note data including the environmental data by aligning the trajectory data, the logistics node time data, the environmental data and the handover records according to the product electronic code by time and converting them into a format. The environmental analysis module is used to extract environmental data from the comprehensive data of the shipping order, assess the degree of damage caused by environmental factors, and generate damage assessment results. An anomaly identification module is used to calculate the probability of occurrence of various preset anomaly types based on the comprehensive data of the consignment note and the damage assessment results, and select the anomaly type with the highest probability as the anomaly registration type of the consignment note corresponding to the electronic code of the product; The threshold adjustment module is used to set the probability thresholds corresponding to the various anomaly types, and adjust the probability thresholds based on historical false alarms and false negatives data from historical manual verification results; The graded response module is connected to the anomaly identification module and the threshold adjustment module respectively. It is used to compare the maximum probability value corresponding to the anomaly registration type with the corresponding probability threshold to complete the early warning judgment, calculate the anomaly score based on the maximum probability value and the preset severity coefficient, trigger the corresponding level of processing instruction based on the comparison result of the anomaly score and the preset score threshold, and output the processing status information.

2. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that, The data acquisition module includes a data acquisition unit and a data alignment unit; The data acquisition unit collects the trajectory data and logistics node time data corresponding to the product electronic code through an RFID reader / writer, collects environmental data through environmental sensors, including real-time temperature and humidity inside the vehicle and the composite value of triaxial vibration acceleration of the goods, and collects the handover record through a logistics terminal. The handover record includes the identity of the handover person and the handover status code. The handover status code is used to represent at least one handover status among completed handover, rejection, abnormal acceptance, and missing information. The data alignment unit uses the product electronic code as the primary key to align the trajectory data, the handover status code, and the environmental data according to time. It generates route deviation data by comparing the preset transportation route with the trajectory data, generates handover comparison data by comparing the handover status code with the preset handover rules, and calculates the node dwell time and transportation time based on the logistics node time data. It then generates comprehensive consignment note data that includes the node dwell time, transportation time, route deviation data, environmental data, and handover comparison data.

3. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that, The environmental data includes the composite values ​​of triaxial vibration acceleration at various times and real-time temperature; the process by which the environmental analysis module calculates the damage assessment results is as follows: The damage assessment result is generated by comprehensively calculating the portion of the triaxial vibration acceleration composite value that exceeds the preset safe vibration upper limit threshold, the preset vibration influence coefficient, the temperature deviation value of the real-time temperature from the preset safe temperature range for goods, and the preset temperature influence coefficient; wherein, the vibration influence coefficient and the temperature influence coefficient are proportional coefficients determined based on the correspondence between the environmental data of historical damaged shipping documents and the actual damage rate.

4. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that, The anomaly identification module includes a first classification unit, a second classification unit, and a comprehensive judgment unit; the first classification unit extracts the node sequence, planned arrival time, and actual arrival time corresponding to the product electronic code based on the trajectory data and the logistics node time data, and combines them with the handover status code in the handover record. Based on the extracted node sequence, the handover status code, the planned arrival time, and the actual arrival time, the first classification unit uses an anomaly classification method to output multiple first anomaly sub-probabilities corresponding to delay anomalies, misdelivery anomalies, and information error anomalies. The second classification unit extracts the environmental data from the comprehensive data of the consignment note, combines it with the damage assessment results, extracts the ratio of the total duration of environmental exceedances to the transportation time in the entire process of the consignment note, establishes the correlation information between environmental changes and cargo damage status using the time-series correlation method, and outputs a second anomaly probability for identifying damage anomalies. The comprehensive judgment unit merges the multiple first anomaly sub-probabilities with the second anomaly probabilities and outputs the final probabilities of the multiple anomaly types. The calculation parameters used by the anomaly classification method and the time-series correlation method are updated according to historical comparison records. The historical comparison records are used to statistically analyze the deviation between the system output probability and the actual anomaly type corresponding to the historical manual verification results.

5. The product electronic code consignment note anomaly registration system according to claim 4, characterized in that, When identifying anomalies, the first classification unit identifies the misdelivery anomaly based on the mismatch between the node sequence corresponding to the product electronic code and the preset target node sequence, identifies the information error anomaly based on the mutual exclusion of the handover status codes at the same node, and identifies the delay anomaly based on the comparison between the planned arrival time and the actual arrival time.

6. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that, The trajectory data corresponds to multiple logistics routes. The threshold adjustment module sets and updates the probability threshold for each logistics route. The process of adjusting the probability threshold is as follows: Using a preset number of days as an evaluation period, the historical manual verification results of each logistics route within that period are statistically analyzed, and the actual false alarm rate and actual missed alarm rate are calculated. The actual false alarm rate is calculated based on the sum of the number of waybills falsely reported by the system and the number of abnormal waybills correctly warned by the system during the current assessment period. The actual underreporting rate is calculated based on the sum of the number of waybills that were missed by the system but confirmed as genuine anomalies through historical manual verification and the number of abnormal waybills that were correctly alerted by the system. Based on the actual false alarm rate, the actual false negative rate, and the preset step size parameter, the threshold modification amount is determined; and the probability threshold of the current period is adjusted in the corresponding direction according to the threshold modification amount to generate a new probability threshold.

7. The product electronic code consignment note anomaly registration system according to claim 5, characterized in that, The preset severity coefficients include damage anomaly coefficients, misdelivery anomaly coefficients, information error anomaly coefficients, and delay anomaly coefficients. The graded response module stores the preset severity coefficients corresponding to each of the various anomaly types. The graded response module determines the anomaly score based on the maximum probability value corresponding to the anomaly registration type and the corresponding preset severity coefficient; This module is configured with a scoring threshold set based on historical anomaly handling maintenance logs. The scoring threshold includes a first scoring threshold and a second scoring threshold, and the first scoring threshold is smaller than the second scoring threshold. This module determines the subsequent processing procedure based on the anomaly score: When the abnormal score is less than the first score threshold, it is determined to be a minor abnormality, and a logistics status change warning is pushed to the preset receiving and sending terminals as the processing instruction. When the abnormal score is greater than or equal to the first score threshold and less than the second score threshold, it is determined to be a moderate abnormality, a verification work order is generated and dispatched to the current node manager for on-site review as the processing instruction; When the abnormal score is greater than or equal to the second score threshold, it is determined to be a severe abnormality. A pause flow control command is sent to the preset logistics terminal or scheduling system of the current logistics node to stop the subsequent logistics flow of the consignment. It is also determined whether the current node is included in the preset target node sequence. If it is included, the index position number of the current node in the preset target node sequence is extracted. If not included, the valid node that was correctly registered in the preset target node sequence before the exception occurred will be used as the reference node, and the index position number of the reference node in the preset target node sequence will be extracted. The index position number is compared with the total number of nodes in the preset target node sequence; if the index position number is less than half of the total number of nodes, a rollback instruction is triggered as the processing instruction. If the index position number is greater than or equal to half of the total number of nodes, a transfer and reissue instruction is triggered as the processing instruction.

8. The product electronic code consignment note anomaly registration system according to claim 1, characterized in that, The system also includes a record management module; The record management module is used to record the anomaly registration type output by the anomaly identification module, the processing instructions output by the hierarchical response module, the historical manual verification results, and the processing status information. The record management module has a built-in report template, generates carrier assessment data based on the recorded data, and feeds back the processing status information to the anomaly identification module to correct the calculation parameters in the anomaly type probability calculation.

Images