Transport capacity data processing method and device

Abstract

The invention discloses a transport capacity data processing method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of obtaining a first waybill number; querying planning trajectory data and first actual trajectory data corresponding to the first waybill number, and performing comparison calculation to obtain a first overlap ratio; in response to determining that the first overlap ratio is within a preset first threshold range, querying an operation data set corresponding to the first waybill number, the operation data set comprising a plurality of operation index data corresponding to a plurality of sampling moments; and combining the operation data set and the first coincidence degree, and inputting the operation data set and the first coincidence degree into a trained anomaly recognition model to obtain an anomaly probability corresponding to each sampling moment so as to correspondingly mark the first waybill number. Therefore, the embodiment of the invention can solve the technical problem that the existing transport capacity data anomaly detection efficiency is low.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for processing transportation capacity data. Background Art

[0002] At present, the transportation and logistics industry is booming, providing users with convenient and fast long-distance transportation services for goods, greatly improving the time efficiency of goods delivery.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] While users enjoy convenient and efficient transportation and logistics services, irregular transportation operations often occur, and there are few technical solutions for multi-dimensional analysis of capacity data and full-process supervision. This is because different capacity management systems correspond to different sampling rules, which leads to variable dimensions of overall capacity data, different sequence lengths, and poor consistency. In the existing technology, only by analyzing the time-series vehicle status data (such as average speed, average driving distance, etc.) and monitoring shooting data, it is determined whether the driving time and rest time of the transport driver are compliant, but there is a lack of in-depth supervision of the driver's abnormal operations during the entire driving process (such as abnormal routes, abnormal destinations, abnormal vehicle speeds, etc.), so it is impossible to optimize and adjust the relevant responsible persons, or transport vehicles, or transport tasks in a more comprehensive manner. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a capacity data processing method and device, which can solve the technical problem of low efficiency of existing capacity data anomaly detection.

[0006] To achieve the above-mentioned purpose, according to one aspect of an embodiment of the present invention, a method for processing capacity data is provided, including obtaining a first waybill number; querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations to obtain a first degree of overlap; in response to determining that the first degree of overlap is within a preset first threshold range, querying an operation data set corresponding to the first waybill number, wherein the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively; merging the operation data set with the first degree of overlap to input into a trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment respectively, so as to mark the first waybill number accordingly.

[0007] Optionally, before obtaining the first coincidence degree, the method includes:

[0008] The starting point overlap, the end point overlap and the trajectory overlap corresponding to the planned trajectory data and the first actual trajectory data are calculated and weighted combined.

[0009] Optionally, after obtaining the first overlap degree, the method further includes:

[0010] In response to determining that the first overlap is not within a preset first threshold range, the first waybill number is marked as abnormal, and the marking reason is recorded as the first overlap is invalid.

[0011] Optionally, before inputting into the trained anomaly recognition model, the following is included:

[0012] Determine the transport vehicle type and transport task type corresponding to the first waybill number as filtering conditions, and filter the waybill database to obtain the second waybill number;

[0013] Querying the second actual trajectory data corresponding to the second waybill number, and reading the first actual trajectory data to perform comparative calculation to obtain a second coincidence degree;

[0014] In response to determining that the second overlap is within a preset second threshold range, the operational data set, the first overlap and the second overlap are combined to serve as input data for a trained anomaly recognition model.

[0015] Optionally, after obtaining the second coincidence degree, the method further comprises:

[0016] In response to determining that the second degree of overlap is not within a preset second threshold range, the first waybill number is marked as abnormal, and the marking reason is recorded as the second degree of overlap corresponding to the second waybill number is invalid.

[0017] Optionally, before inputting into the trained anomaly recognition model, the following is included:

[0018] Use the long short-term memory network model to build an anomaly recognition model;

[0019] Inputting a preset training data set into the anomaly recognition model, training the anomaly recognition model with the goal of making the corresponding output approach the corresponding target value;

[0020] In response to the corresponding loss function reaching a minimum, a trained anomaly recognition model is obtained.

[0021] Optionally, the first waybill number is marked accordingly, including:

[0022] Determine whether there is a sampling time at which the abnormal probability is not within a preset third threshold range;

[0023] If yes, the first waybill number is marked as abnormal, and the sampling time whose abnormal probability is not within the range of the third threshold is screened to be used as the corresponding marking reason;

[0024] If not, the first waybill number is marked normally.

[0025] In addition, the present invention also provides a capacity data processing device, including an acquisition module for acquiring a first waybill number; a calculation module for querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations to obtain a first degree of overlap; an acquisition module for, in response to determining that the first degree of overlap is within a preset first threshold range, querying an operation data set corresponding to the first waybill number, wherein the operation data set includes multiple operation indicator data corresponding to multiple sampling moments respectively; an identification module for merging the operation data set with the first degree of overlap to input into a trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment, so as to mark the first waybill number accordingly.

[0026] An embodiment of the above invention has the following advantages or beneficial effects: first, the present invention achieves the technical purpose of clarifying the object of capacity anomaly detection by obtaining the first waybill number, and the first waybill number is also used as a screening condition for screening the corresponding source of capacity data; and, the present invention obtains the first degree of overlap by querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations, thereby achieving the purpose of determining the rationality and standardization of transportation of the first waybill number based on the trajectory data dimension. The dynamic time warping algorithm adopted by the present invention can accurately compare and calculate trajectory data of different sequence lengths, effectively overcoming the defect in the prior art that it is difficult to analyze capacity data with complex formats; in addition, the present invention queries the first in response to determining that the first degree of overlap is within a preset first threshold range. An operation data set corresponding to the waybill number, wherein the operation data set includes multiple operation indicator data corresponding to multiple sampling moments respectively, and uses a preset first threshold range to quickly mark and filter out the waybill numbers with too low trajectory data standardization, thereby improving the efficiency of the present invention in detecting anomalies in capacity data, and collects operation data sets for waybill numbers with roughly qualified trajectory data standardization, so as to lay a data foundation for abnormal risk determination in subsequent operation indicator dimensions; at the same time, the present invention combines the operation data set with the first overlap degree to input into a trained abnormality recognition model to obtain the abnormal probability corresponding to each sampling moment, so as to mark the first waybill number accordingly, effectively integrating the trajectory data dimension and the operation indicator dimension of the entire transportation process, thereby obtaining a more accurate calculation result of the waybill abnormality probability.

[0027] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0029] Figure 1 is a schematic diagram of the main process of the method for processing transportation capacity data according to the first embodiment of the present invention;

[0030] Figure 2 is a schematic diagram of the main process of the capacity data processing method according to the second embodiment of the present invention;

[0031] Figure 3 is a schematic diagram of the main process of the capacity data processing method according to the third embodiment of the present invention;

[0032] Figure 4 is a schematic diagram of main modules of a capacity data processing device according to a first embodiment of the present invention;

[0033] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0034] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0036] Figure 1 FIG. 1 is a schematic diagram of the main flow of the method for processing transportation data according to the first embodiment of the present invention. Figure 1 As shown, the capacity data processing method includes:

[0037] Step S101, obtaining the first waybill number.

[0038] In an embodiment, the object of capacity analysis is clarified by obtaining the first waybill number, and the corresponding trajectory data and operation data set are obtained by screening the first waybill number, which can serve as data support for capacity analysis, so that the present invention can be used to identify abnormal capacity risks for the vehicle (i.e., the first waybill number), and then according to the abnormal risk identification results, the business rationality of the corresponding driver can be determined, and the corresponding transportation tasks can be optimized and adjusted to promote the healthy development of the transportation industry.

[0039] Step S102, querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations to obtain a first degree of coincidence.

[0040] In an embodiment, the type of transport vehicle and the type of transport task corresponding to the first waybill number can be first determined, and then the planned trajectory data corresponding to the type of transport task and the type of transport vehicle can be matched in the database, and the corresponding first actual trajectory data can be queried according to the first waybill number. The trajectory data includes multiple coordinate data corresponding to multiple sampling moments, for example, it can be in the form of [sampling moment 1, longitude 1, dimension 1; sampling moment 2, longitude 2, dimension 2; sampling moment 3, ...]. According to the first overlap calculated by comparing the planned trajectory data and the first actual trajectory data, a comprehensive judgment is made on the standardization of the transportation route and driving time of the corresponding driver, which is also equivalent to measuring the abnormal risk of the corresponding vehicle (i.e., the first waybill number) in the space-time trajectory dimension; therefore, the value of the first overlap provides an important basis for judging the rationality of the space-time trajectory of the first waybill number; obviously, the higher the value of the first overlap, the more reasonable and standardized the transportation space-time trajectory of the corresponding first waybill number, and the lower the value of the first overlap, the greater the space-time trajectory deviation of the corresponding first waybill number. However, considering that even in the case of similar transportation tasks and similar transportation vehicles, the possibility of different road conditions is also very high, which leads to a decrease in the corresponding space-time trajectory overlap. For example, a traffic accident occurs on a certain section of road, so the driver chooses to take a detour in this transportation task. In this case, it cannot be judged that the driver's transportation trajectory is unreasonable solely for this reason.

[0041] In a further embodiment, the corresponding first threshold range can be pre-set based on practical experience (i.e., the probability of abnormal transportation road conditions obtained based on massive transportation capacity data), so as to judge whether the calculated first overlap is within the preset first threshold range, and then roughly judge the abnormal capacity risk of the corresponding transport vehicle (i.e., the first waybill number) in the dimension of the space-time trajectory. In a further embodiment, the set first threshold range can be a larger numerical range, so as to roughly screen the driving space-time trajectory of each waybill number, and filter out a small number of waybill numbers with extremely irregular and unreasonable driving routes and driving durations. That is, after obtaining the first overlap, in response to determining that the first overlap is not within the preset first threshold range, the first waybill number is abnormally marked, and the reason for the mark is recorded as invalid first overlap. On the other hand, for most waybill numbers whose corresponding overlap is within the first threshold range, the following abnormal identification model is called to perform another abnormal determination on the order number in the dimension of comprehensive operation indicators to obtain a more comprehensive and accurate abnormal risk determination result. After processing in this step, by establishing the first threshold range, some waybill numbers with obvious driving trajectory abnormalities can be quickly filtered out, thereby narrowing the judgment range of the following abnormal recognition model and improving the efficiency of the present invention in judging the abnormal risk of waybill numbers.

[0042] In some embodiments, the first degree of overlap may include corresponding transport starting point similarity, transport end point similarity and transport trajectory similarity, or the first degree of overlap is obtained by weighted calculation of the three similarities, so as to make a more accurate comprehensive judgment on the capacity anomaly of the first waybill number in more data dimensions.

[0043] In some embodiments, in order to expand the comparison calculation range of trajectory data similarity and improve the accuracy of comparison calculation of trajectory data similarity, a dynamic time warping algorithm can be used to calculate the overlap corresponding to multiple trajectory data. According to the dynamic time warping algorithm, firstly, a sequence corresponding to the planned trajectory data and the first actual trajectory data is generated, and each point of the sequence can be a corresponding trajectory coordinate; the distance between each point in the above two sequences is calculated to obtain a distance matrix; in the distance matrix, a path from the upper left corner to the lower right corner of the matrix is ​​found so that the sum of elements on the path is minimized, and the sum of elements can be used to characterize the similarity of the two trajectory sequences. The use of the dynamic time warping algorithm can accurately compare trajectory data of different dimensions (i.e., each trajectory data includes a different number of sampling moments), thereby overcoming the shortcomings of the prior art means: when comparing and calculating the Euclidean distance of two time series, it is greatly affected by the different lengths of the two time series, resulting in inaccurate calculation results and difficulty in judging whether the transportation trajectory is abnormal or not.

[0044] In some embodiments, in order to determine the abnormal risk of the first waybill number in the dimension of the authenticity of the transport task, the transport vehicle type and transport task type corresponding to the first waybill number can be determined as a filtering condition, and the second waybill number is filtered in the waybill library; the second actual trajectory data corresponding to the second waybill number is queried, and the first actual trajectory data is read to perform a comparative calculation to obtain a second overlap; in response to determining that the second overlap is within the preset second threshold range, the operation data, the first overlap and the second overlap are combined as input data for the trained abnormal recognition model. After the processing of this step, the second waybill number whose transport vehicle type and transport task type are different from the first waybill number is screened, and the second overlap of the actual trajectory routes corresponding to the two waybill numbers is compared and analyzed; if the first waybill number and the second waybill number both correspond to real transport tasks, since their transport task types and transport vehicle types are different, the corresponding second overlap should be smaller; if the first waybill number or the second waybill number corresponds to the edited (non-actual) transport trajectory data, then the second overlap obtained by comparison is likely to have a higher value.

[0045] In a further embodiment, the authenticity of the transport track of the first waybill number can be roughly judged by a preset second threshold range and a randomly selected second waybill number corresponding to different transport tasks and different transport vehicle types. That is, after obtaining the second overlap, in response to determining that the second overlap is not within the preset second threshold range, the first waybill number is marked as abnormal, and the reason for the mark is recorded as the invalid second overlap corresponding to the second waybill number. In a further embodiment, the second threshold range can also be set to a larger numerical range, so as to achieve the technical purpose of quickly filtering out the first waybill number with lower authenticity, and calling the following abnormal recognition model for the first waybill number with roughly reasonable authenticity, so as to further calculate the corresponding abnormal capacity risk.

[0046] Step S103, in response to determining that the first overlap degree is within a preset first threshold range, querying an operation data set corresponding to the first waybill number, wherein the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively.

[0047] In an embodiment, in response to determining that the first degree of overlap is within a preset reasonable range (i.e., a first threshold range), an operational data set corresponding to the first waybill number is collected to further judge the rationality of the capacity of the first waybill number in the dimension of operational indicators. For example, the operational data set may be in the form of {[sampling time 1, operational indicator 1, operational indicator 2, operational indicator 3, operational indicator 4, operational indicator 5, …], [sampling time 2, operational indicator 1, operational indicator 2, operational indicator 3, operational indicator 4, operational indicator 5, …], [sampling time 3, operational indicator 1, operational indicator 2, operational indicator 3, operational indicator 4, operational indicator 5, …], …}. In the actual calculation and processing process, the multiple operational indicator data may include order quantity, loading weight, loading volume, mileage, vehicle rated load, carrier, and the like.

[0048] In some embodiments, the rationality of the first overlap and the second overlap corresponding to the first waybill number can also be judged. In response to the first overlap and the second overlap being respectively within the preset first threshold range and the second threshold range, that is, determining that the first overlap and the second overlap are both roughly reasonable, the corresponding operation data set is collected to call the following anomaly recognition model, thereby integrating the transport trajectory standardization dimension (corresponding to the first overlap), the transport trajectory authenticity dimension (corresponding to the second overlap) and the rationality dimension of the entire operation indicator data (corresponding to the output result of the anomaly recognition model described below) to obtain a more accurate and convincing abnormal risk judgment result of the first waybill number.

[0049] Step S104, merging the operation data set and the first overlap degree to input into the trained anomaly recognition model, obtaining the anomaly probability corresponding to each sampling moment, and marking the first waybill number accordingly.

[0050] In an embodiment, by merging the operational data set with the first overlap and inputting them into a trained anomaly recognition model, the previously calculated trajectory data overlap of the first waybill number is effectively included in the judgment range of the anomaly recognition model, thereby effectively ensuring the comprehensiveness and richness of the data source used for calculation by the anomaly recognition model, so as to ensure that the output anomaly probability at each sampling moment is based on a comprehensive judgment result of multiple dimensions.

[0051] In a further embodiment, the operational data set, the first overlap degree and the second overlap degree may be combined as input data for a trained anomaly recognition model to further increase the breadth of data sources for the anomaly recognition model, thereby improving the accuracy of the corresponding calculation results.

[0052] In some embodiments, in order to make the calculation and analysis of the abnormal recognition model closer to the input data format of the present invention, thereby improving the accuracy of the abnormal risk judgment of the abnormal recognition model for the present invention, the abnormal recognition model can be constructed using a long short-term memory network model before inputting into the trained abnormal recognition model; the preset training data set is input into the abnormal recognition model, and the abnormal recognition model is trained with the corresponding output approaching the corresponding target value as the goal; in response to the corresponding loss function reaching the minimum, a trained abnormal recognition model is obtained. In this step, by using the long short-term memory network model (LSTM) to construct the abnormal recognition model, the problems of gradient vanishing and long-term dependence that occur when the deep neural network model calculates the input data with a long span (i.e., the operation data set) can be effectively avoided, so that the present invention can more accurately calculate the abnormal risk probability corresponding to each sampling moment.

[0053] In a further embodiment, in order to clearly and concisely mark the first waybill number according to the output result of the abnormal identification model to improve the processing efficiency of the downstream processing end, when marking the first waybill number accordingly, it can be determined whether there is an abnormal probability corresponding to a certain sampling moment that is not within the preset third threshold range; if so, the first waybill number is marked abnormally, and the sampling moment with the abnormal probability not within the third threshold range is screened as the corresponding marking reason, and the screened sampling moment is used as the abnormal sampling moment; if not, the first waybill number is marked normally. After processing in this step, the waybill number with abnormal sampling moments is marked abnormally, and the waybill number without abnormal sampling moments is marked as normal, thereby clarifying the standard for the abnormal identification model to judge the waybill number as positive or abnormal, and then all the waybill numbers to be processed can be quickly distinguished. On the other hand, by recording the abnormal sampling moments, it helps the downstream processing end to more smoothly locate the abnormal problem of the first waybill number and formulate the corresponding processing plan, thereby promoting the optimized management and active adjustment of the transportation and logistics industry.

[0054] In a further embodiment, in order to more efficiently divert the waybill numbers that are detected to have a high risk of abnormality using the present technical solution, the first waybill number marked as abnormal can be sent to a designated processing port, so that the person responsible for the first waybill number can be adjusted and optimized in a targeted manner through the designated downstream processing end. In actual processing, the abnormal waybill numbers screened out by the first threshold range, the second threshold range and the abnormality recognition model may have the following transportation problems: the transportation task is not true (corresponding to abnormal actual trajectory data, or abnormal part of the operating data, such as abnormal vehicle load), the transportation model is not true (corresponding to abnormal actual trajectory data, or abnormal part of the operating data, such as abnormal vehicle load or abnormal vehicle rated load), unreasonable vehicle load (corresponding to abnormal vehicle load or abnormal vehicle rated load), irregular transportation operation (corresponding to abnormal actual trajectory data, such as abnormal transportation departure and arrival time), and so on. Through the designated downstream processing end, judgment is made based on the abnormal mark of the abnormal waybill number, and corresponding rectification conclusions are drawn, such as re-training the driver or adjusting a certain planned trajectory data, thereby achieving the effect of monitoring, inspecting and optimizing the transportation industry to promote the healthy development of the transportation industry.

[0055] Figure 2 is a schematic diagram of the main process of a method for processing capacity data according to a second embodiment of the present invention, the method for processing capacity data comprising:

[0056] Step S201, obtaining the first waybill number.

[0057] Step S202, querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number.

[0058] Step S203, using a dynamic time warping algorithm, calculate the corresponding starting point coincidence, end point coincidence and trajectory coincidence, and perform weighted merging calculation to obtain a first coincidence.

[0059] Step S204, determining whether the first overlap degree is within a preset first threshold range, if so, executing step S205, if not, executing step S206.

[0060] Step S205, determining the transport vehicle type and transport task type corresponding to the first waybill number, and using them as filtering conditions to filter in the waybill database to obtain the second waybill number.

[0061] Step S206: Mark the first waybill number as abnormal, and record the marking reason as invalid first overlap.

[0062] Step S207, query the second actual track data corresponding to the second waybill number.

[0063] Step S208: using a dynamic time warping algorithm, comparing and calculating a second degree of overlap between the first actual trajectory data and the second actual trajectory data.

[0064] Preferably, the corresponding starting point coincidence, end point coincidence and track coincidence can be calculated first, and then weighted combination can be performed to obtain the second coincidence.

[0065] Step S209, determining whether the second overlap degree is within a preset second threshold range, if so, executing step S210, if not, executing step S211.

[0066] Step S210, querying the operation data set corresponding to the first waybill number, merging the operation data set, the first overlap degree and the second overlap degree to obtain model input data.

[0067] Preferably, the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively.

[0068] Step S211, marking the first waybill number as abnormal, and recording the marking reason as the second overlap degree corresponding to the second waybill number is invalid.

[0069] Step S212: construct an abnormality recognition model using the long short-term memory network model, and perform training to obtain a trained abnormality recognition model.

[0070] Preferably, a preset training data set is input into the anomaly recognition model, and the anomaly recognition model is trained with the corresponding output approaching the corresponding target value as the goal; in response to the corresponding loss function reaching the minimum, a trained anomaly recognition model is obtained.

[0071] Step S213, inputting the model input data into the trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment.

[0072] Step S214, determining whether there is an abnormal probability corresponding to a certain sampling moment that is not within a preset third threshold range, if so, executing step S215, if not, executing step S216.

[0073] Step S215, marking the first waybill number as abnormal, and screening out abnormal sampling moments whose abnormal probability is not within a third threshold range as corresponding marking reasons.

[0074] Step S216, marking the first waybill number normally.

[0075] Step S217: Send the first waybill number marked as abnormal to a designated processing port.

[0076] Figure 3is a schematic diagram of the main process of a method for processing capacity data according to a third embodiment of the present invention, the method for processing capacity data comprising:

[0077] Step S301, obtaining the first waybill number.

[0078] Step S302, querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number.

[0079] Step S303, using a dynamic time warping algorithm, calculate the corresponding starting point coincidence, end point coincidence and trajectory coincidence, and perform weighted merging calculations to obtain a first coincidence.

[0080] Step S304, determining whether the first overlap degree is within a preset first threshold range, if so, executing step S305, if not, executing step S306.

[0081] Step S305, querying the operation data set corresponding to the first waybill number, merging the operation data set with the first overlap degree, and obtaining model input data.

[0082] Preferably, the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively.

[0083] Step S306: Mark the first waybill number as abnormal, and record the marking reason as invalid first overlap.

[0084] Step S307: construct an abnormality recognition model using the long short-term memory network model, and perform training to obtain a trained abnormality recognition model.

[0085] Preferably, a preset training data set is input into the anomaly recognition model, and the anomaly recognition model is trained with the corresponding output approaching the corresponding target value as the goal; in response to the corresponding loss function reaching the minimum, a trained anomaly recognition model is obtained.

[0086] Step S308: input the model input data into the trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment.

[0087] Step S309, determining whether there is an abnormal probability corresponding to a certain sampling moment that is not within a preset third threshold range, if so, executing step S310, otherwise executing step S311.

[0088] Step S310, marking the first waybill number as abnormal, and screening out abnormal sampling moments whose abnormal probability is not within a third threshold range as corresponding marking reasons.

[0089] Step S311, marking the first waybill number normally.

[0090] Step S312: Send the first waybill number marked as abnormal to a designated processing port.

[0091] Figure 4 is a schematic diagram of the main modules of the capacity data processing device according to an embodiment of the present invention, such as Figure 4 As shown, the transport data processing device 400 includes an acquisition module 401, a calculation module 402, a collection module 403 and an identification module 404. The acquisition module 401 acquires the first waybill number; the calculation module 402 queries the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performs comparative calculation to obtain a first overlap; the collection module 403, in response to determining that the first overlap is within a preset first threshold range, queries the operation data set corresponding to the first waybill number, wherein the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively; the identification module 404 merges the operation data set with the first overlap, and inputs them into the trained abnormality recognition model, and obtains the abnormal probability corresponding to each sampling moment, so as to mark the first waybill number accordingly.

[0092] In some embodiments, the calculation module 402 is further used to: before obtaining the first overlap, calculate the start point overlap, end point overlap and trajectory overlap corresponding to the planned trajectory data and the first actual trajectory data, and perform weighted merging calculation.

[0093] In some embodiments, the identification module 404 is further used to: after obtaining the first degree of overlap, in response to determining that the first degree of overlap is not within a preset first threshold range, mark the first waybill number as abnormal and record the marking reason as invalid first degree of overlap.

[0094] In some embodiments, the calculation module 402 is also used to: determine the type of transport vehicle and the type of transport task corresponding to the first waybill number as filtering conditions before inputting into the trained abnormality recognition model, and filter the second waybill number in the waybill database; query the second actual trajectory data corresponding to the second waybill number, and read the first actual trajectory data to perform comparative calculations to obtain a second degree of overlap; in response to determining that the second degree of overlap is within a preset second threshold range, merge the operating data, the first degree of overlap and the second degree of overlap as input data for the trained abnormality recognition model.

[0095] In some embodiments, the identification module 404 is also used to: after obtaining the second degree of overlap, in response to determining that the second degree of overlap is not within a preset second threshold range, mark the first waybill number as abnormal, and record the marking reason as the second degree of overlap corresponding to the second waybill number is invalid.

[0096] In some embodiments, the identification module 404 is also used to: construct an anomaly recognition model using a long short-term memory network model before inputting it into a trained anomaly recognition model; input a preset training data set into the anomaly recognition model, and train the anomaly recognition model with the corresponding output approaching the corresponding target value as the goal; in response to the corresponding loss function being minimized, obtain a trained anomaly recognition model.

[0097] In some embodiments, the identification module 404 is also used to: when marking the first waybill number accordingly, determine whether there is a sampling moment whose abnormal probability is not within a preset third threshold range; if so, mark the first waybill number as abnormal, and screen out the sampling moment whose abnormal probability is not within the third threshold range as the corresponding marking reason; if not, mark the first waybill number normally.

[0098] It should be noted that the capacity data processing method and the capacity data processing device described in the present invention have a corresponding relationship in terms of specific implementation contents, so the repeated contents will not be described again.

[0099] Figure 5 An exemplary system architecture 500 is shown to which the method or device for processing capacity data according to the embodiment of the present invention can be applied.

[0100] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0101] The user can use the terminal devices 501, 502, 503 to interact with the server 505 through the network 504 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 501, 502, 503.

[0102] The terminal devices 501 , 502 , and 503 may be various electronic devices having a page display processing screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0103] The server 505 may be a server that provides various services, such as a background management server (only an example) that provides support for users to use the terminal devices 501, 502, and 503. The background management server may analyze and process the received data such as the product information query request, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0104] It should be noted that the capacity data processing method provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the computing device is generally set in the server 505 .

[0105] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0106] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0107] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the computer system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0108] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal page display processor (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0109] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0110] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0111] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0112] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an acquisition module, a calculation module, a collection module and an identification module. The names of these modules do not constitute limitations on the modules themselves in some cases.

[0113] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by a device, the device includes obtaining a first waybill number; querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations to obtain a first degree of overlap; in response to determining that the first degree of overlap is within a preset first threshold range, querying the operation data set corresponding to the first waybill number, wherein the operation data set includes multiple operation indicator data corresponding to multiple sampling moments respectively; merging the operation data set and the first degree of overlap to input into the trained anomaly recognition model, and obtaining the anomaly probability corresponding to each sampling moment, so as to mark the first waybill number accordingly.

[0114] The technical solution according to the embodiment of the present invention can solve the technical problem of low efficiency in abnormal detection of existing transportation data.

[0115] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing transport capacity data, characterized in that: include: Get the first waybill number; Querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculations to obtain a first degree of coincidence; In response to determining that the first overlap degree is within a preset first threshold range, querying an operation data set corresponding to the first waybill number, wherein the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively; The operation data set and the first overlap degree are combined to be input into the trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment, so as to mark the first waybill number accordingly.

2. The method according to claim 1, characterized in that Before getting the first coincidence, including: The starting point overlap, the end point overlap and the trajectory overlap corresponding to the planned trajectory data and the first actual trajectory data are calculated and weighted combined.

3. The method according to claim 1, characterized in that After obtaining the first degree of coincidence, including: In response to determining that the first overlap is not within a preset first threshold range, the first waybill number is marked as abnormal, and the marking reason is recorded as the first overlap is invalid.

4. The method according to claim 1, characterized in that: Before inputting into the trained anomaly recognition model, it includes: Determine the transport vehicle type and transport task type corresponding to the first waybill number as filtering conditions, and filter the waybill database to obtain the second waybill number; Querying the second actual trajectory data corresponding to the second waybill number, and reading the first actual trajectory data to perform comparative calculation to obtain a second coincidence degree; In response to determining that the second overlap is within a preset second threshold range, the operational data set, the first overlap and the second overlap are combined to serve as input data for a trained anomaly recognition model.

5. The method according to claim 4, characterized in that After obtaining the second degree of coincidence, it includes: In response to determining that the second degree of overlap is not within a preset second threshold range, the first waybill number is marked as abnormal, and the marking reason is recorded as the second degree of overlap corresponding to the second waybill number is invalid.

6. The method according to claim 1, characterized in that Before inputting into the trained anomaly recognition model, it includes: Use the long short-term memory network model to build an anomaly recognition model; Inputting a preset training data set into the anomaly recognition model, and training the anomaly recognition model with the goal of the corresponding output approaching the corresponding target value; In response to the corresponding loss function reaching a minimum, a trained anomaly recognition model is obtained.

7. The method according to claim 1, characterized in that The first waybill number is marked accordingly, including: Determine whether there is a sampling time at which the abnormal probability is not within a preset third threshold range; If yes, the first waybill number is marked as abnormal, and the sampling time whose abnormal probability is not within the range of the third threshold is screened to be used as the corresponding marking reason; If not, the first waybill number is marked normally.

8. A capacity data processing device, characterized in that: include: An acquisition module, used to obtain the first waybill number; A calculation module, used for querying the planned trajectory data and the first actual trajectory data corresponding to the first waybill number, and performing comparative calculation to obtain a first coincidence degree; a collection module, configured to query an operation data set corresponding to the first waybill number in response to determining that the first overlap degree is within a preset first threshold range, wherein the operation data set includes a plurality of operation indicator data corresponding to a plurality of sampling moments respectively; An identification module is used to merge the operation data set and the first overlap degree to input into a trained anomaly recognition model to obtain the anomaly probability corresponding to each sampling moment to mark the first waybill number accordingly.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

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

Images