Search

CN-122022633-A - Data sampling inspection method and system for coal transportation vehicle in highway logistics

CN122022633ACN 122022633 ACN122022633 ACN 122022633ACN-122022633-A

Abstract

The invention discloses a data sampling inspection method and a system for coal transportation vehicles in highway logistics, which belong to the technical field of transportation supervision, realize comprehensive and accurate collection of transportation vehicle information through arrangement and acquisition of target vehicle transportation data, ensure the accuracy of follow-up sampling inspection calculation, generate a risk weight value conforming to a real-time state through calculation of transportation path space deviation data and target vehicle real-time illegal data, calculate a fairness weight value conforming to a fairness supervision principle through fairness sampling inspection compensation function on the non-inspected time length of the target vehicle and the total inspected proportion of the target vehicle, and form a fairness sampling inspection mechanism combining risk weight value and fairness weight value, thereby ensuring the supervision efficiency of transportation vehicle sampling inspection and avoiding the occurrence of supervision blind areas.

Inventors

  • LI MINGSHUN

Assignees

  • 中豹供应链管理(武汉)有限公司

Dates

Publication Date
20260512
Application Date
20260203

Claims (10)

  1. 1. The data sampling inspection method for the coal transportation vehicle in the road logistics is characterized by comprising the following steps of: acquiring real-time transportation data of a target vehicle and historical business data of the target vehicle, and performing data preprocessing on the real-time transportation data of the target vehicle and the historical business data of the target vehicle to form transportation data of the target vehicle; Acquiring a preset standard transportation path, generating a real-time target vehicle running track and real-time target vehicle violation data based on the target vehicle transportation data, comparing the real-time target vehicle running track with the standard transportation path to calculate transportation path space deviation data, and calculating a risk weight value of the target vehicle according to the real-time target vehicle violation data and the transportation path space deviation data, wherein the risk weight value of the target vehicle is used for representing the false transportation suspicion degree of the target vehicle; acquiring a preset fairness sampling compensation function, generating a target vehicle inspected record based on the target vehicle transportation data, calculating the non-inspected time length of the target vehicle and the total inspected proportion of the target vehicle according to the target vehicle inspected record, and calculating a fairness weight value of the target vehicle by using the fairness sampling compensation function, wherein the fairness weight value of the target vehicle is used for representing sampling attention degree of the target vehicle; Acquiring a preset risk preference adjustment coefficient, and carrying out weight fusion on a risk weight value of the target vehicle and a fairness weight value of the target vehicle according to the risk preference adjustment coefficient to generate real-time comprehensive sampling probability of the target vehicle; and performing sampling inspection judgment on the target vehicle according to the real-time comprehensive sampling inspection probability of the target vehicle to obtain sampling inspection judgment results, and generating sampling inspection decision instructions for the target vehicle based on the sampling inspection judgment results so as to perform sampling inspection on the target vehicle according to the sampling inspection decision instructions.
  2. 2. The method for data spot check of coal transportation vehicles in road logistics according to claim 1, wherein acquiring target vehicle real-time transportation data and target vehicle historical business data, and performing data preprocessing on the target vehicle real-time transportation data and the target vehicle historical business data to form target vehicle transportation data, comprises: acquiring real-time position data of a target vehicle through a vehicle-mounted positioning terminal of the target vehicle, wherein the real-time position data of the target vehicle comprises longitude and latitude coordinates of the target vehicle, instantaneous speed of the target vehicle and position change data of the target vehicle; acquiring real-time load data of a target vehicle through a vehicle load sensing terminal of the target vehicle; Acquiring target vehicle transportation metadata from a vehicle manifest management platform, wherein the target vehicle transportation metadata comprises a target vehicle manifest number, a target vehicle license plate number, target vehicle shipping mine coordinates, target vehicle receiving platform coordinates, a target vehicle loading and unloading area and a target vehicle planned loading capacity; According to the license plate number of the target vehicle, acquiring vehicle basic information, the accumulated transportation trip number, the total driving mileage, the history violation record, the history spot check record and the history detection data of the target vehicle from a vehicle information database; Integrating the target vehicle real-time position data, the target vehicle real-time load data and the target vehicle transportation metadata into target vehicle real-time transportation data, and taking the vehicle basic information, the vehicle accumulated transportation trip number, the vehicle total driving mileage, the vehicle history violation record, the vehicle history spot check record and the vehicle history detection data of a target vehicle as target vehicle history business data; Performing time alignment processing, coordinate screening processing, load data filtering processing and data integrity verification on the target vehicle real-time transportation data to form first pre-target vehicle transportation data, and performing data deduplication processing, outlier rejection processing and data consistency verification on the target vehicle historical service data to form second pre-target vehicle transportation data; Binding the first pre-target vehicle transportation data and the second pre-target vehicle transportation data, and carrying out structural processing on the bound first pre-target vehicle transportation data and second pre-target vehicle transportation data to generate target vehicle transportation data.
  3. 3. The method for data sampling inspection of coal transportation vehicles in road logistics according to claim 1, wherein obtaining a preset standard transportation path, generating a real-time travel track of a target vehicle and real-time violation data of the target vehicle based on the target vehicle transportation data, comparing the real-time travel track of the target vehicle with the standard transportation path to calculate transportation path space deviation data, and calculating a risk weight value of the target vehicle according to the real-time violation data of the target vehicle and the transportation path space deviation data, comprising: Based on the target vehicle transportation data, extracting each historical coordinate position and each real-time coordinate position of the target vehicle, sequencing each historical coordinate position and each real-time coordinate position according to a time sequence to form a track coordinate position sequence of the target vehicle, sequentially connecting the track coordinate position sequences of the target vehicle, and performing smoothing treatment to form a real-time running track of the target vehicle; acquiring a preset standard transportation path, projecting the real-time running track of the target vehicle onto the standard transportation path, and taking each historical coordinate position and each real-time coordinate position as a projection marking point; For each projection marking point, calculating the normal offset distance of each projection marking point to the standard transportation path; Acquiring a preset deviation threshold, counting the average deviation distance and the maximum deviation distance of a target vehicle according to the normal deviation distances corresponding to the projection marking points, judging the normal deviation distances corresponding to the projection marking points according to the deviation threshold, and calculating the duty ratio of the projection marking points with the normal deviation distances exceeding the deviation threshold as an abnormal deviation duty ratio; Integrating the average offset distance, the maximum offset distance and the abnormal offset duty ratio of the target vehicle to obtain transportation path space offset data; Extracting parking point coordinates of the target vehicle based on the target vehicle transportation data; acquiring a preset loading and unloading area, performing abnormality judgment on the parking point coordinates of a target vehicle by utilizing the loading and unloading area, and marking the parking point coordinates with coordinates exceeding the loading and unloading area as abnormal parking positions; For the abnormal parking position, extracting parking duration corresponding to the abnormal parking position and target vehicle parking front and rear load variation corresponding to the abnormal parking position from the target vehicle transportation data; Acquiring a preset abnormal parking time length threshold value and a normal load change rate, judging the parking time length corresponding to the abnormal parking position by utilizing the abnormal parking time length threshold value to generate an abnormal parking event when the parking time length exceeds the abnormal parking time length threshold value, calculating the load change rate corresponding to the abnormal parking position according to the load change rate before and after parking of the target vehicle corresponding to the abnormal parking position, and judging the load change rate corresponding to the abnormal parking position by utilizing the normal load change rate to generate an abnormal loading and unloading event when the load change rate is higher than the normal load change rate; counting the abnormal stay events and the abnormal loading and unloading events of the target vehicle to obtain real-time violation data of the target vehicle; quantizing the transportation path space deviation data into a target vehicle path deviation degree coefficient, quantizing the target vehicle real-time violation data into a target vehicle violation degree coefficient, and calculating a basic risk weight value of the target vehicle according to the following formula (1): (1) Wherein, the Representing the target vehicle path offset degree coefficient, Representing the target vehicle violation degree coefficient, Is a preset path risk weight factor, which is a preset path risk weight factor, For a preset risk of violation weighting factor, Representing a base risk weight value for the target vehicle; and adjusting the basic risk weight value of the target vehicle based on the target vehicle transportation data to obtain a risk weight value of the target vehicle.
  4. 4. The method for data spot check of coal transportation vehicles in road logistics according to claim 3, wherein adjusting the base risk weight value of the target vehicle based on the target vehicle transportation data to obtain the risk weight value of the target vehicle comprises: Extracting a vehicle history violation record of a target vehicle based on the target vehicle transportation data, counting the history violation number of the target vehicle according to the vehicle history violation record of the target vehicle, and generating a history risk weight adjustment factor according to the history violation number of the target vehicle, wherein if the history violation number of the target vehicle is 0, the generated history risk weight adjustment factor is 1, and if the history violation number of the target vehicle is not 0, the value of the history risk weight adjustment factor is adjusted upwards according to the history violation number of the target vehicle; Based on the historical risk weight adjustment factor, adjusting a base risk weight value of the target vehicle by the following formula (2): (2) Wherein, the Representing the historical risk weight adjustment factor, A risk weight value representing the target vehicle.
  5. 5. The method for data sampling inspection of coal transportation vehicles in road logistics according to claim 1, wherein obtaining a preset fairness sampling inspection compensation function, generating a target vehicle inspected record based on the target vehicle transportation data, calculating a target vehicle non-inspected duration and a target vehicle total inspected proportion according to the target vehicle inspected record, and calculating a fairness weight value of the target vehicle by using the fairness sampling inspection compensation function, comprising: extracting a vehicle history spot check record of the target vehicle based on the target vehicle transportation data, and analyzing the vehicle history spot check record of the target vehicle to select the spot check record which completes the spot check closed loop verification as a target vehicle detected record; Sequentially arranging the target vehicle detected records according to the time sequence to form a target vehicle detected record sequence, and selecting the last target vehicle detected record from the target vehicle detected record sequence as a target detected record; obtaining the detected completion time of the target detected record, and obtaining the non-detected duration of the target vehicle by utilizing the difference between the current time and the detected completion time of the target detected record; based on the target vehicle transportation data, extracting the accumulated transportation trip number of the target vehicle, and counting the detected records of the target vehicle to obtain the effective detected times of the target vehicle; calculating the total inspected proportion of the target vehicle according to the effective inspected times of the target vehicle and the accumulated transportation trip number of the target vehicle; Acquiring a preset fairness sampling compensation function, calculating a time compensation term according to the non-inspected time length of the target vehicle by using the fairness sampling compensation function, calculating a frequency compensation term according to the total inspected proportion of the target vehicle, and bringing the time compensation term and the frequency compensation term into the fairness sampling compensation function to generate a fairness sampling compensation coefficient of the target vehicle; And obtaining fairness weight boundary constraint, and performing numerical scaling processing on the fairness sampling inspection compensation coefficient by utilizing the fairness weight boundary constraint to obtain a fairness weight value of the target vehicle.
  6. 6. The method for data spot inspection of coal transportation vehicles in road logistics according to claim 5, wherein calculating a time compensation term from the non-inspected duration of the target vehicle using the fair spot inspection compensation function comprises: Acquiring a standard sampling detection duration preset in the fairness sampling detection compensation function, and judging the non-detected duration of the target vehicle by using the standard sampling detection duration; If the non-inspected time length of the target vehicle does not exceed the standard sampling time length, calculating a corresponding time compensation term by using the fairness sampling compensation function according to the non-inspected time length of the target vehicle by the following formula (3): (3) Wherein, the A preset time compensation upper limit in the fairness sampling detection compensation function is set, Compensating the increment coefficient for the time preset in the fairness sampling inspection compensation function, Indicating the period of time that the target vehicle is not examined, A time compensation term is represented and is used to represent, A base number that is a natural logarithm; If the non-inspected time length of the target vehicle exceeds the standard sampling time length, calculating a corresponding time compensation term by using the fairness sampling compensation function according to the non-inspected time length of the target vehicle by the following formula (4): (4) Wherein, the A blind-spot detection and compensation base number preset in the fairness sampling-spot detection compensation function is obtained; Correspondingly, calculating a frequency compensation term according to the total detected proportion of the target vehicle, and bringing the time compensation term and the frequency compensation term into the fairness sampling compensation function to generate a fairness sampling compensation coefficient of the target vehicle, wherein the method comprises the following steps: based on the fairness sampling compensation function, calculating a corresponding frequency compensation term according to the total inspected proportion of the target vehicle through the following formula (5): (5) Wherein, the A frequency compensation growth coefficient preset in the fairness sampling inspection compensation function is used for obtaining the frequency compensation growth coefficient, Representing a preset expected inspected proportion in the fairness sampling compensation function, Representing the total inspected proportion of the target vehicle, Representing the term of frequency compensation, Obtaining a function for the maximum value; Based on the fairness sampling compensation function, calculating a fairness sampling compensation coefficient of the target vehicle according to the time compensation term and the frequency compensation term through the following formula (6): (6) Wherein, the A time compensation weight factor preset in the fairness sampling inspection compensation function is used for the fairness sampling inspection, A frequency compensation weight factor preset in the fairness sampling detection compensation function, Representing a fair spot compensation coefficient of the target vehicle, and the time compensation weight factor Greater than the frequency compensation weight factor 。
  7. 7. The method for data sampling inspection of coal transportation vehicles in road logistics according to claim 1, wherein obtaining a preset risk preference adjustment coefficient, performing weight fusion on a risk weight value of the target vehicle and a fairness weight value of the target vehicle according to the risk preference adjustment coefficient, and generating real-time comprehensive sampling inspection probability of the target vehicle, comprises: acquiring a preset risk preference adjustment coefficient, carrying out weight fusion on a risk weight value of the target vehicle and a fairness weight value of the target vehicle according to the risk preference adjustment coefficient, and calculating real-time comprehensive sampling probability of the target vehicle according to the following formula (7): (7) Wherein, the For the risk preference adjustment factor, A risk weight value representing the target vehicle, A fair spot compensation coefficient representing the target vehicle, And representing the real-time comprehensive sampling probability of the target vehicle.
  8. 8. The method for data spot inspection of coal transportation vehicles in highway logistics according to claim 1, wherein performing spot inspection judgment on the target vehicle according to the real-time comprehensive spot inspection probability of the target vehicle to obtain a spot inspection judgment result, generating a spot inspection decision instruction on the target vehicle based on the spot inspection judgment result to perform spot inspection on the target vehicle according to the spot inspection decision instruction, comprising: generating a random decision threshold by using a safe random number generator, wherein the random decision threshold is a floating point number uniformly distributed in a [0,1 ] interval; performing sampling inspection judgment on the real-time comprehensive sampling inspection probability of the target vehicle by utilizing the random decision threshold value to obtain sampling inspection judgment results; If the sampling detection judging result is that the real-time comprehensive sampling detection probability of the target vehicle is not higher than the random decision threshold, a sampling detection wheel space instruction is generated for the target vehicle; if the sampling detection judging result is that the real-time comprehensive sampling detection probability of the target vehicle is higher than the random decision threshold, a sampling detection decision instruction is generated for the target vehicle, so that sampling detection is carried out on the target vehicle according to the sampling detection decision instruction; after generating a spot check decision instruction for the target vehicle or a spot check wheel space instruction for the target vehicle, the method further comprises: Integrating the target vehicle transportation data, the risk weight value of the target vehicle, the fairness weight value of the target vehicle and the spot check judging result to form a target vehicle real-time spot check record, and sending the target vehicle real-time spot check record to a vehicle information database for data storage.
  9. 9. The data sampling inspection system for the coal transportation vehicle in the road logistics is characterized by being applied to the data sampling inspection method for the coal transportation vehicle in the road logistics according to any one of claims 1-8, and comprising the following steps: The data acquisition unit is used for acquiring real-time transportation data of the target vehicle and historical business data of the target vehicle, and carrying out data preprocessing on the real-time transportation data of the target vehicle and the historical business data of the target vehicle so as to form transportation data of the target vehicle; the risk weight calculation unit is used for acquiring a preset standard transportation path, generating a real-time target vehicle running track and real-time target vehicle violation data based on the target vehicle transportation data, comparing the real-time target vehicle running track with the standard transportation path to calculate transportation path space deviation data, and calculating a risk weight value of the target vehicle according to the real-time target vehicle violation data and the transportation path space deviation data, wherein the risk weight value of the target vehicle is used for representing the false transportation suspicion degree of the target vehicle; The fairness weight calculation unit is used for obtaining a preset fairness sampling compensation function, generating a target vehicle inspected record based on the target vehicle transportation data, calculating the non-inspected time length of the target vehicle and the total inspected proportion of the target vehicle according to the target vehicle inspected record, and calculating a fairness weight value of the target vehicle by utilizing the fairness sampling compensation function, wherein the fairness weight value of the target vehicle is used for representing sampling attention degree of the target vehicle; The comprehensive sampling probability calculation unit is used for acquiring a preset risk preference adjustment coefficient, carrying out weight fusion on the risk weight value of the target vehicle and the fairness weight value of the target vehicle according to the risk preference adjustment coefficient, and generating real-time comprehensive sampling probability of the target vehicle; The sampling inspection decision unit is used for carrying out sampling inspection decision on the target vehicle according to the real-time comprehensive sampling inspection probability of the target vehicle to obtain sampling inspection decision results, and generating sampling inspection decision instructions for the target vehicle based on the sampling inspection decision results so as to carry out sampling inspection on the target vehicle according to the sampling inspection decision instructions.
  10. 10. An electronic device comprising a memory, a processor and a transceiver in communication with each other in sequence, wherein the memory is configured to store a computer program, the transceiver is configured to send and receive a message, and the processor is configured to read the computer program and perform the method for sampling data of a coal transportation vehicle in a highway logistics according to any one of claims 1 to 8.

Description

Data sampling inspection method and system for coal transportation vehicle in highway logistics Technical Field The invention belongs to the technical field of transportation supervision, and particularly relates to a data sampling inspection method and system for coal transportation vehicles in highway logistics. Background In the field of highway coal logistics transportation, false transportation behaviors (such as empty car counterfeiting of a waybill, illegal unloading on the way, fake license transportation and the like) become outstanding treatment problems of corroding industry profits, disturbing market order and causing tax loss due to the characteristics of large transportation capacity, relatively fixed routes, high transportation frequency and the like. The existing transportation supervision and spot check technology mainly depends on two modes, namely, on-site interception and check based on manual experience, the efficiency is low, the coverage area is limited, the method is severely limited by subjective judgment of check staff, and is difficult to deal with massive transportation flow, and a preliminary informatization rule spot check system generally adopts a fixed-proportion random extraction or simple periodic shaking mode. Firstly, the prior art adopts a random spot inspection mode for the spot inspection of transport vehicles, but cannot accurately guide limited supervision resources to vehicles with high risk suspicion, so that the supervision efficiency is low, the actual detection rate of false transport behaviors is low for a long time, secondly, because the prior spot inspection system adopts a relatively simple, transparent and fixed spot inspection calculation mode, a transport party or an internal person can actively evade an inspection period through analysis rules (such as spot inspection batch interval and vehicle numbering rule) so that the supervision mechanism is similar to a dummy value, and in addition, in the prior spot inspection systems, the risk assessment model trained based on historical data is excessively relied on, although the hit rate of high risk targets can be improved, the compliance vehicles with frequent transport tasks and good historical records can be repeatedly checked, the operation cost and time delay of the compliance vehicles are increased, the overall logistics efficiency is influenced, and part of vehicles with low risk or unobvious data characteristics can be dissociated outside the supervision vision for a long time, so as to form a supervision blind area. From the foregoing, how to improve the supervision efficiency, and to balance the risk with the accurate identification and fair supervision of the data sampling inspection method and system of the coal transportation vehicles in the highway logistics has become an urgent topic to be studied in the field. Disclosure of Invention The invention aims to provide a data sampling inspection method and system for coal transportation vehicles in highway logistics, which are used for solving the problems in the prior art. In order to achieve the above purpose, the present invention adopts the following technical scheme: In a first aspect, the invention provides a data sampling inspection method for coal transportation vehicles in road logistics, which comprises the following steps: acquiring real-time transportation data of a target vehicle and historical business data of the target vehicle, and performing data preprocessing on the real-time transportation data of the target vehicle and the historical business data of the target vehicle to form transportation data of the target vehicle; Acquiring a preset standard transportation path, generating a real-time target vehicle running track and real-time target vehicle violation data based on the target vehicle transportation data, comparing the real-time target vehicle running track with the standard transportation path to calculate transportation path space deviation data, and calculating a risk weight value of the target vehicle according to the real-time target vehicle violation data and the transportation path space deviation data, wherein the risk weight value of the target vehicle is used for representing the false transportation suspicion degree of the target vehicle; acquiring a preset fairness sampling compensation function, generating a target vehicle inspected record based on the target vehicle transportation data, calculating the non-inspected time length of the target vehicle and the total inspected proportion of the target vehicle according to the target vehicle inspected record, and calculating a fairness weight value of the target vehicle by using the fairness sampling compensation function, wherein the fairness weight value of the target vehicle is used for representing sampling attention degree of the target vehicle; Acquiring a preset risk preference adjustment coefficient, and carrying out weight fusion on a risk weight value of the