A vehicle logistics intelligent management method and system based on a TMS system

Abstract

The application provides a whole vehicle logistics intelligent management method and system based on a TMS system, and relates to the technical field of intelligent control.The method comprises the following steps: automatically triggering a prepayment process according to a task state, associating a third-party payment platform to complete the payment; after the task is completed, making the driver submit a reimbursement application through a mobile terminal, and automatically generating a reimbursement detail which is calculated in combination with the comprehensive adjustment parameters and is pushed to the driver for confirmation; after the reimbursement detail is confirmed by the driver, a multi-level audit process of a district manager and finance is performed in combination with the optimized comprehensive adjustment parameters; after the audit is passed, a reimbursement payment process is automatically triggered to complete the cost settlement; and the contract management, reconciliation management, quality loss and violation handling are synchronously updated to realize whole-process closed-loop management.The application can realize whole-process intelligentization of whole vehicle logistics.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent management method and system for vehicle logistics based on a TMS system. Background Technology

[0002] With the development of information technology, Transportation Management Systems (TMS) have been widely used in the field of whole vehicle logistics to achieve digital scheduling and tracking of transportation tasks. However, some existing TMS-based logistics management methods have the following shortcomings.

[0003] Firstly, in terms of monitoring the transportation process, most existing systems can only achieve simple vehicle location reporting and trajectory playback, lacking in-depth spatial analysis and quantitative evaluation of on-the-go transportation behavior. Location data is only processed as discrete point information, which cannot depict the spatial geometric characteristics of the transportation route as a whole, and it is even more difficult to extract effective parameters reflecting dynamic characteristics such as driving behavior, route deviation, or regional traffic difficulty. This makes it impossible for managers to accurately grasp the real situation in the transportation process, and there is a lack of data-supported decision-making basis for abnormal driving and route optimization. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for intelligent management of vehicle logistics based on TMS system, which can realize intelligent management of the entire vehicle logistics process.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] Firstly, a method for intelligent management of vehicle logistics based on a TMS system, the method comprising:

[0007] Step S1: Generate a dispatch order according to the plan and assign the dispatch order to the designated driver's mobile terminal so that the driver can execute the task after receiving the dispatch order and update the task status in real time;

[0008] Step S2: Based on the continuous location data collected from the reported locations along the route, perform spatial geometric modeling on the location data to construct a spatial reference plane for the transportation route; based on the distribution characteristics of the location data, divide the reference plane into several analysis grids and map each location data to the corresponding analysis grid; for each set of location data within the analysis grid, calculate the volume of the smallest enclosing sphere formed by its spatial distribution range and calculate the spatial centroid coordinates of the set to generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process;

[0009] Step S3: The prepayment process is automatically triggered based on the task status, and the payment is completed by linking a third-party payment platform. After the task is completed, the driver submits a reimbursement application through a mobile terminal, and a reimbursement detail calculated in conjunction with the comprehensive adjustment parameters is automatically generated and pushed to the driver for confirmation.

[0010] Step S4: After the reimbursement details are confirmed by the driver, the regional manager and finance department conduct a multi-level review process based on the comprehensive adjustment parameters.

[0011] Step S5: After approval, the reimbursement and payment process is automatically triggered to complete the settlement of expenses; contract management, reconciliation management, and handling of quality damage and violations are updated simultaneously to achieve closed-loop management of the entire process.

[0012] Secondly, an intelligent vehicle logistics management system based on a TMS system includes:

[0013] The update module is used to generate dispatch orders according to the plan and distribute the dispatch orders to the designated driver's mobile terminal so that the driver can execute the task after receiving the dispatch order and update the task status in real time.

[0014] The evaluation module is used to perform spatial geometric modeling on the location data collected from the reported locations along the route, and construct a spatial reference plane for the transportation route; based on the distribution characteristics of the location data, the reference plane is divided into several analysis grids, and each location data is mapped to the corresponding analysis grid; for the location data set within each analysis grid, the volume of the smallest enclosing sphere formed by its spatial distribution range is calculated, and the spatial centroid coordinates of the set are calculated, to generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process;

[0015] The push module is used to automatically trigger the prepayment process based on the task status and complete the payment through a third-party payment platform. After the task is completed, the driver can submit a reimbursement application through a mobile terminal, and the module automatically generates a reimbursement detail calculated based on the comprehensive adjustment parameters and pushes it to the driver for confirmation.

[0016] The processing module is used to optimize the multi-level review process of the regional manager and finance department after the driver confirms the expense details, combined with the comprehensive adjustment parameters. After the review is approved, the expense payment process is automatically triggered to complete the settlement of expenses. Contract management, reconciliation management, quality damage and violation handling are updated in sync to achieve closed-loop management of the whole process.

[0017] The above-described solution of the present invention has at least the following beneficial effects:

[0018] By performing spatial geometric modeling and gridding analysis on continuous vehicle location data, and introducing geometric feature parameters such as the volume of the minimum enclosing sphere and the coordinates of the spatial centroid, the abstract transportation trajectory is transformed into quantifiable comprehensive adjustment parameters. This enables the system to accurately assess the dynamic characteristics of the driver's behavior on the road (such as route deviation and regional driving difficulty). By automatically integrating the comprehensive adjustment parameters into the calculation of reimbursement details, the cost settlement is no longer fixed, static data, but rather dynamic data that reflects the actual transportation process cost, thereby reducing the company's operating costs. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating an intelligent vehicle logistics management method based on a TMS system, as provided in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of an intelligent vehicle logistics management system based on a TMS system provided by an embodiment of the present invention. Detailed Implementation

[0021] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0022] like Figure 1 As shown, an embodiment of the present invention proposes an intelligent management method for vehicle logistics based on a TMS system, the method comprising the following steps:

[0023] Step S1: Generate a dispatch order according to the plan and assign the dispatch order to the designated driver's mobile terminal so that the driver can execute the task after receiving the dispatch order and update the task status in real time;

[0024] Step S2: Based on the continuous location data collected from the reported locations along the route, perform spatial geometric modeling on the location data to construct a spatial reference plane for the transportation route; based on the distribution characteristics of the location data, divide the reference plane into several analysis grids and map each location data to the corresponding analysis grid; for each set of location data within the analysis grid, calculate the volume of the smallest enclosing sphere formed by its spatial distribution range and calculate the spatial centroid coordinates of the set to generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process;

[0025] Step S3: The prepayment process is automatically triggered based on the task status, and the payment is completed by linking a third-party payment platform. After the task is completed, the driver submits a reimbursement application through a mobile terminal, and a reimbursement detail calculated in conjunction with the comprehensive adjustment parameters is automatically generated and pushed to the driver for confirmation.

[0026] Step S4: After the reimbursement details are confirmed by the driver, the regional manager and finance department conduct a multi-level review process based on the comprehensive adjustment parameters.

[0027] Step S5: After approval, the reimbursement and payment process is automatically triggered to complete the settlement of expenses; contract management, reconciliation management, and handling of quality damage and violations are updated simultaneously to achieve closed-loop management of the entire process.

[0028] In this embodiment, task instructions are automatically distributed to the driver's mobile terminal, enabling direct delivery of task commands and real-time updates to task status. This reduces information transmission delays, allowing for more timely driver responses and more precise control of task progress by the management end, thus lowering collaboration costs. Based on location data modeling, grid division, and the generation of comprehensive adjustment parameters (minimum bounding sphere volume, spatial centroid coordinates), the system can accurately assess the dynamic characteristics of the transportation process (such as route deviation and driving stability), and promptly identify transportation anomalies. Automatic prepayment triggering, automatic generation of reimbursement details combined with comprehensive adjustment parameters (ensuring cost calculations accurately reflect actual transportation conditions), and optimized multi-level review processes significantly reduce manual operations, lower financial error rates, and shorten prepayment and reimbursement cycles, improving the driver's financial settlement experience. After reimbursement, contracts, reconciliation, and handling of quality damage and violations are updated simultaneously, achieving full-process data linkage and control from task issuance and in-transit monitoring to cost settlement and subsequent problem handling. This avoids process interruptions and improves the integrity and controllability of vehicle logistics management.

[0029] In a preferred embodiment of the present invention, step S1, generating a dispatch order according to the plan and allocating the dispatch order to a designated driver's mobile terminal, so that the driver can execute the task after receiving the dispatch order and update the task status in real time, includes:

[0030] Obtain the basic transportation plan required to generate the dispatch order. The plan source needs to be combined with the core demand scenarios of full truckload logistics, mainly including two types:

[0031] First, there is the order-driven planning: triggered by upstream sales orders, dealer vehicle pickup needs, or OEM vehicle delivery needs, the TMS system automatically synchronizes order information (such as vehicle model, quantity, recipient address, and delivery time limit), and combines it with the vehicle storage location in the inventory management module (such as OEM warehouse and transit warehouse) to initially determine the basic parameters of the transportation task.

[0032] Second, there is the scheduling optimization plan: This is manually created by logistics managers based on the regional capacity balance needs. For example, for idle capacity with unmatched orders, short-distance transfers, vehicle allocation, and other tasks can be planned, or existing temporary transportation needs can be adjusted (such as expedited delivery or route changes) to ensure that the plan matches the actual capacity resources.

[0033] After the plan is generated, the system will first perform compliance verification, such as confirming whether the transportation route is within the permitted scope, whether the delivery time limit is reasonable, and whether the required vehicle type (such as flatbed truck or enclosed truck) is consistent with the capacity information. After the verification is passed, a list of plans to be generated for dispatch orders will be generated, and the dispatch order construction stage will begin.

[0034] Based on the confirmed transportation plan, the TMS system automatically generates a dispatch order containing comprehensive task information. The dispatch order must cover the core needs of the driver in performing the task, specifically including:

[0035] Basic task information: Task number (unique identifier for subsequent tracking), task type (e.g., order transportation, transshipment, dispatch), origin and destination (accurate to the door address, synchronously linked to map coordinates, supporting mobile terminal navigation), planned departure time / arrival time, and specific information of the vehicle to be transported (vehicle identification number, vehicle model, vehicle status, such as whether inspection has been completed).

[0036] Execution requirements information: contact person and contact information for loading and unloading (to avoid gaps in driver communication), cargo handover requirements (such as the need for paper receipts and electronic receipts), and special transportation specifications (such as requirements for securing fragile vehicles and temperature control standards for low-temperature transportation).

[0037] Related support information: recommended route (generated based on the best historical route for driver reference), information on service points along the way (such as the location of gas stations and repair stations), and emergency contact person (logistics dispatcher's phone number for communication in case of emergencies).

[0038] After the transfer order is generated, the system automatically associates it with the corresponding plan number, forming a plan-transfer order binding relationship, which facilitates subsequent tracing of the task source.

[0039] To ensure that dispatch orders are accurately assigned to suitable drivers, the system uses multi-dimensional screening logic to identify specific drivers, avoiding wasted capacity or unexecuted tasks due to blind allocation.

[0040] Step 1: Basic condition screening. Select candidates from the driver database who meet the task requirements, such as holding a driver's license for the corresponding vehicle type (e.g., an A2 license for large trucks), having no outstanding traffic violations / accident records, and having a fleet that covers the task's start and end areas (to avoid empty driving costs caused by cross-regional dispatch).

[0041] Step 2: Dynamic status matching, prioritizing drivers who are currently idle or whose tasks are about to be completed. For example, the TMS system can obtain the driver's task status in real time (such as having completed the previous task or returning), calculate the estimated time for them to reach the task starting point, and ensure that they can execute the new task on time.

[0042] Step 3: Weighting based on historical performance. Prioritize drivers based on their historical task completion data (such as on-time arrival rate, cargo integrity rate, and customer reviews). For example, give higher weight to drivers with excellent historical performance and assign them tasks first, while avoiding assigning high-difficulty tasks (such as long-distance express transportation) to novice drivers.

[0043] After the screening is completed, the system will automatically select one designated driver. If there are multiple qualified drivers, the logistics dispatcher can also manually select one (supports viewing the driver's real-time location and current status to assist in decision-making). After selection, the driver ID will be bound to the dispatch order, and the dispatch order distribution process will begin.

[0044] The TMS system uses a multi-channel push + receipt confirmation mechanism to ensure that dispatch orders accurately reach the designated driver's mobile terminal (usually a logistics-specific APP, which needs to be bound to the TMS system account):

[0045] Distribution channels: Two push methods are triggered simultaneously: one is the in-app pop-up notification (highest priority, displayed directly when the driver opens the app to avoid omission), and the other is the SMS reminder (as a backup channel, for situations where the app is not logged in or the network is poor, the SMS contains the dispatch order number and a link to view details, which will redirect to the dispatch order page in the app after clicking).

[0046] Receipt Confirmation: After viewing the dispatch order, the driver needs to confirm receipt on the mobile terminal (click the confirm execution button). The system receives confirmation feedback in real time. If no confirmation is received within 15 minutes, a second reminder will be automatically sent to the driver. If no confirmation is received within 30 minutes, the system will return the dispatch order to the pending allocation status and notify the logistics dispatcher. The dispatcher will then manually communicate the reason (such as the driver taking temporary leave) and re-match other drivers to avoid gaps where dispatch orders have been issued but not received.

[0047] After the driver confirms receipt of the dispatch order, the task execution phase begins. The system uses a combination of manual triggering and automatic assistance to dynamically synchronize the task status, ensuring that the manager can monitor the progress in real time.

[0048] Manual updates of status nodes: On the task center page of the mobile terminal, the driver clicks the corresponding status button according to the actual execution progress. The core status nodes include: ① Departed (clicked after the driver arrives at the starting point and completes the vehicle handover, the system records the departure time); ② On the way (the driver can update in real time. If there is a traffic jam, maintenance, or other situation, the driver can add an abnormal note on the way, which will be pushed to the dispatcher simultaneously); ③ Arrived at the loading / unloading point (clicked after arriving at the starting point to load or the destination to unload, the system records the arrival time and triggers a loading / unloading reminder); ④ Task completed (clicked after completing the handover of goods and uploading the signed receipt, the task is marked as completed).

[0049] Automatic status update assistance: In conjunction with the on-the-way location reporting function to be used later in step S2, the system verifies and supplements the status manually updated by the driver. For example, if the driver does not click "arrived", but the location data shows that he has entered the destination area (within a 1-kilometer range defined by the map coordinates in the dispatch order), the system will automatically send a reminder to confirm whether the destination has been reached. After the driver confirms, the status is updated. If the location data shows that the driver has deviated from the recommended route by more than 5 kilometers without noting the reason, the system will automatically mark the status as "on the road abnormal" and notify the dispatcher to verify, so as to avoid status delays caused by driver misoperation.

[0050] All status update records are synchronized to the TMS system's management backend in real time. Administrators can view the progress of all tasks through the task status dashboard (such as the number and details of tasks that are pending departure, en route, or completed). Clicking on a single task allows you to view the time point of the status update and the driver's feedback notes, enabling visual tracking of the entire task process.

[0051] Through the above steps, step S1 achieves a closed-loop system encompassing planning, dispatching orders, drivers, and status. This not only solves the problems of inaccurate task instruction transmission and delayed status updates in the existing system, but also provides basic task data support for the subsequent in-transit analysis in step S2 and the triggering of financial processes in step S3.

[0052] In a preferred embodiment of the present invention, step S2, based on the continuous location data collected from the on-the-go location reports, performs spatial geometric modeling on the location data to construct a spatial reference plane for the transportation route, including:

[0053] Step S21: Obtain the continuous location data sequence reported by the driver's mobile terminal. The location data includes latitude and longitude coordinates and timestamps. Specifically, in the logistics-specific mobile terminal used by the driver (such as an APP bound to the TMS system), the location reporting rules are pre-configured: a fixed reporting frequency is set (usually once every 20-30 seconds, which can be adjusted according to the complexity of the transportation route. The frequency can be appropriately reduced on highways and increased on complex urban road sections) to ensure that the location data can continuously reflect the vehicle's driving trajectory.

[0054] After the driver initiates the mission, the mobile terminal uses the built-in GPS / BeiDou positioning module to collect the latitude and longitude coordinates of the current location in real time (the accuracy must reach 6 decimal places to ensure that the positioning error is within 10 meters). At the same time, it automatically records the timestamp of the location data collection (accurate to the second, in the format of year-month-day hour:minute:second), forming a single complete location data (including latitude and longitude and timestamp).

[0055] The mobile terminal pushes single location data to the TMS system backend in real time via 4G / 5G network or offline caching (temporarily stored locally when the network is interrupted, and uploaded in batches after the network is restored). After receiving the data, the TMS system first verifies the data integrity (checking whether it contains valid latitude and longitude and timestamp, the latitude and longitude must be within a reasonable geographical range, and the timestamp must be later than the timestamp of the previous data). If the data is missing or the format is incorrect, the data is automatically marked as pending review, and a reminder is sent to the driver's terminal that the location data upload is abnormal and to check the positioning function. The verified location data is sorted in chronological order by timestamp and stored in the in-transit location database to form a continuous location data sequence, ensuring that the sequence has no time breaks and no duplicate data.

[0056] Step S22, transforming the latitude and longitude coordinates in the location data sequence to a two-dimensional plane coordinate system through map projection to obtain a set of plane coordinate points, including:

[0057] Step S221: Based on the latitude and longitude coordinate distribution range of the continuous location data sequence, identify the projection zone range covering all latitude and longitude coordinates from the continuous location data sequence, and determine the central meridian parameter of the projection zone range. This includes: extracting the latitude and longitude coordinates of all location data from the continuous location data sequence generated in step S21, filtering out the maximum longitude, minimum longitude, maximum latitude, and minimum latitude values ​​respectively, and determining the geographical area covered by this batch of location data (i.e., the rectangular geographical area formed by longitude min-longitude max and latitude min-latitude max).

[0058] Refer to the national or industry-standard map projection zone division (such as the commonly used Gauss-Kruger projection, which is divided into 6-degree zones and 3-degree zones), and determine the projection zone to which the above-mentioned geographical region belongs based on its longitude range: For example, the Gauss-Kruger 6-degree zone starts from the 0° meridian and is divided into zones every 6 degrees, with zone numbers from 1 to 60. If the longitude range of a certain batch of data is 114°-120°, then it belongs to zone 20 (114° = 6° × 19, the longitude range of zone 20 is 114°-120°).

[0059] Based on the determined projection zone number, look up the corresponding formula for calculating the central meridian of the projection zone (e.g., the central meridian longitude of the Gauss-Kruger 6-degree zone = 6° × zone number - 3°), calculate the central meridian parameter of the projection zone (e.g., the central meridian of zone 20 is 6° × 20 - 3° = 117°), and record the longitude value of the central meridian as the core reference parameter for subsequent coordinate transformation.

[0060] Step S222: Using the central meridian parameter, convert the latitude and longitude coordinates of each location data point into a first coordinate value and a second coordinate value in a Cartesian coordinate system in meters. This includes: for each location data (including latitude and longitude coordinates) obtained in step S21, calling the map projection conversion algorithm corresponding to the projection zone (such as the Gauss-Kruger projection conversion algorithm), and substituting the central meridian parameter determined in step S221; during the conversion process, first convert the latitude and longitude coordinates (unit: degrees) to radians (to ensure calculation accuracy), and then, according to the principle of the projection algorithm, map the latitude and longitude coordinates in the spherical coordinate system to the two-dimensional Cartesian coordinate system, and calculate the two coordinate values ​​corresponding to the location data: the first coordinate value (corresponding to the X-axis of the Cartesian coordinate system, representing the distance in the east-west direction, unit: meters) and the second coordinate value (corresponding to the Y-axis of the Cartesian coordinate system, representing the distance in the north-south direction, unit: meters).

[0061] Verify the conversion result of each location data point, checking whether the converted X and Y values ​​are within the reasonable coordinate range of the projection zone (e.g., in the Gauss-Kruger projection, the X value of a 6-degree zone is usually between 200,000 and 800,000 meters, and the Y value needs to be added with the offset corresponding to the zone number to distinguish different projection zones). If it exceeds the reasonable range, recheck the latitude and longitude data and the central meridian parameters, and perform the conversion again to ensure that the plane coordinate values ​​of each location data point are accurate.

[0062] Step S223: Combine all the first and second coordinate values ​​obtained from the transformation to generate a corresponding planar coordinate point for each original location data point, so as to form a set of planar coordinate points. This includes: For all the location data transformed in step S222, according to the correspondence between the original location data points and the planar coordinate values, combine the first coordinate value (X) and the second coordinate value (Y) of each original location data to form the planar coordinate point corresponding to the original location data (in the format of (X value, Y value)).

[0063] Maintaining the same timestamp order as the continuous location data sequence in step S21, arrange all planar coordinate points sequentially to form a set of planar coordinate points. At the same time, retain the association information between each planar coordinate point and the original location data (such as the timestamp and data number of the original data) in the point set to ensure that the actual travel time and original location corresponding to the planar coordinate point can be traced during subsequent analysis, providing support for subsequent evaluation of transportation dynamic characteristics (such as travel speed and dwell time) in combination with the time dimension.

[0064] Step S23: Based on the set of planar coordinate points, a spatial reference plane for the transportation route is generated using a planar fitting algorithm. This spatial reference plane represents the overall spatial orientation of the transportation route and includes:

[0065] Step S231 involves outlier detection and removal processing on the planar coordinate point set to obtain an optimized planar coordinate point set. This includes: using a dual method of statistical analysis and trend judgment to identify outliers in the planar coordinate point set obtained in step S223.

[0066] Calculate the average X-coordinate and average Y-coordinate of all points in the plane coordinate point set, and then calculate the Euclidean distance from each point to the average point (Xavg, Yavg). Set a distance threshold (usually 3 times the average distance of all points, which can be adjusted according to the actual situation of the transportation section). If the distance of a point exceeds the threshold, it is initially judged as an outlier (such as a long-distance jump point caused by positioning signal drift).

[0067] The movement trajectory of the planar coordinate points is viewed in the order of timestamps. The distance and direction of travel between two adjacent points are calculated. If the distance between a point and the previous time point is much greater than the distance corresponding to the normal driving speed (e.g., the distance moved within 10 seconds exceeds 1000 meters, which is much greater than the driving speed of a normal truck), or the direction of travel suddenly reverses and there is no reasonable record of stopping (e.g., no loading / unloading or maintenance notes), it is judged as an abnormal value (e.g., falsely reported duplicate points or erroneous points caused by equipment failure).

[0068] For outliers initially identified, the corresponding original location data (such as timestamps, driver operation records at the time, and mobile terminal network status) is retrieved through the TMS system for manual verification. If it is confirmed that the invalid data is caused by positioning errors, equipment failures, etc., the outlier is removed from the set of plane coordinate points. If it is a special case (such as the driver temporarily deviating from the route to refuel, and there is a corresponding APP operation note), the point is retained and marked as a special reasonable point. After removing outliers, the remaining plane coordinate points are reorganized to form an optimized set of plane coordinate points to ensure that the set of points can truly reflect the actual driving trajectory of the vehicle and avoid outliers interfering with the accuracy of subsequent plane fitting, so that the spatial reference plane cannot represent the actual transportation route.

[0069] Step S232, based on the optimized planar coordinate point set, calculate the arithmetic mean coordinates of all points to obtain the geometric center point of the point set, including: extracting the X coordinate values ​​of all points in the optimized planar coordinate point set, adding these X values ​​and dividing by the total number of points to obtain the arithmetic mean of the X coordinates (denoted as Xcenter); similarly, extracting the Y coordinate values ​​of all points and calculating the arithmetic mean of the Y coordinates (denoted as Ycenter); combining the calculated Xcenter and Ycenter to form the geometric center point (Xcenter, Ycenter) of the planar coordinate point set; this center point represents the center position of the entire transportation route on the two-dimensional plane, and subsequent plane fitting will be carried out around this center point to ensure that the fitted spatial reference plane can fit the overall distribution of the route, rather than being biased towards a certain local section of the route.

[0070] Step S233: Subtract the coordinates of the geometric center point from the coordinates of each point in the optimized planar coordinate point set to obtain a centralized coordinate point set with the geometric center point as the origin, including:

[0071] For each plane coordinate point (Xi, Yi) in the optimized plane coordinate point set, perform centering calculation:

[0072] The X-coordinate value of the point is obtained by subtracting the Xcenter of the geometric center from the X-coordinate value Xi of the point (denoted as Xi' = Xi - Xcenter); the Y-coordinate value of the point is obtained by subtracting the Ycenter of the geometric center from the Y-coordinate value Yi of the point (denoted as Yi' = Yi - Ycenter); the Xi' and Yi' of each point are combined to form the corresponding centralized coordinates (Xi', Yi'); all these centralized coordinates are arranged in chronological order according to the original planar coordinate point set to form a centralized coordinate point set with the geometric center as the origin; the purpose of the centering process is to eliminate the influence of the overall positional offset of the planar coordinate point set on subsequent calculations: for example, if all points are biased towards a certain area, the covariance matrix will be affected by the overall position when uncentered, and cannot accurately reflect the distribution characteristics of the point set; after centering, the origin is the route center, and the covariance matrix can more accurately describe the dispersion and spatial correlation of the point set around the center, providing accurate data for subsequent analysis of the concentration and dispersion of transportation routes.

[0073] Step S234: Based on the centralized coordinate point set, calculate the covariance matrix of all its points along the three coordinate axes. The covariance matrix represents the spatial distribution characteristics of the point set, including: First, clarifying the dimension of the centralized coordinate point set: Since the planar coordinate points are two-dimensional (X', Y'), but in order to fully describe the spatial distribution characteristics during subsequent eigenvalue decomposition, it needs to be extended to a three-dimensional coordinate system (the Z-axis is set to 0, since all points are in a two-dimensional plane, the Z-coordinate value is always 0), that is, each centralized coordinate point is represented as (Xi', Yi', 0); Based on the three-dimensional centralized coordinate point set, calculate the covariance matrix. The covariance matrix is ​​a 3×3 matrix, and its elements represent the covariance between different coordinate axes. The specific calculation method is as follows:

[0074] To calculate the covariance Cov(X', X') along the X' axis, sum the squares of all points Xi' and divide by the total number of points (since the mean of X' after centering is 0, the covariance equals the average of the sums of squares), reflecting the dispersion of points along the X' axis. Similarly, to calculate the covariance Cov(Y', Y') along the Y' axis, sum the squares of all points Yi' and divide by the total number of points, reflecting the dispersion of points along the Y' axis.

[0075] Calculate the covariance Cov(Z', Z') along the Z' axis. Since Zi' is 0 for all points, this covariance is 0. Calculate the covariance Cov(X', Y') between the X' and Y' axes. Multiply Xi' and Yi' for all points, sum them, and divide by the total number of points. This reflects the correlation between the point distributions along the X' and Y' axes (positive correlation means that Y' increases as X' increases, and negative correlation means the opposite). Calculate the covariances (Cov(X', Z') and Cov(Y', Z')) between the X' and Z' axes and between the Y' and Z' axes. Since Zi' is always 0, both of these covariances are 0.

[0076] The covariance values ​​calculated above are arranged in a 3×3 matrix format to form a complete covariance matrix. This matrix fully describes the distribution characteristics (dispersion and correlation between coordinate axes) of the centered coordinate point set in three-dimensional space, and is the core data for extracting the geometric features of the route through eigenvalue decomposition.

[0077] Step S235 involves performing eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors of the covariance matrix. This includes calling a matrix eigenvalue decomposition algorithm (such as QR decomposition or Jacobi iteration) to perform decomposition operations on the 3×3 covariance matrix obtained in step S234.

[0078] After decomposition, three sets of results are obtained: three eigenvalues ​​(denoted as λ1, λ2, λ3, usually sorted in descending order of value, i.e., λ1≥λ2≥λ3), and three eigenvectors corresponding to each eigenvalue (denoted as v1, v2, v3, all of which are unit vectors and are mutually orthogonal).

[0079] The physical meaning of the eigenvalue is: the larger the eigenvalue, the higher the degree of dispersion of the centralized coordinate point set in the direction of the corresponding eigenvector (i.e., the wider the extension range of the route in this direction); the physical meaning of the eigenvector is: it represents the main direction of the spatial distribution of the point set (e.g., v1 corresponding to λ1 is the direction of the most dispersed distribution of the point set, i.e., the main extension direction of the transportation route).

[0080] Verify the decomposition results: Calculate the product of each eigenvector and its corresponding eigenvalue, and check if it is equal to the product of the covariance matrix and the eigenvector (i.e., Av = λv, where A is the covariance matrix). If the equation holds, the decomposition result is correct; if not, re-examine the calculation process of the covariance matrix, correct it, and perform eigenvalue decomposition again to ensure the accuracy of the result.

[0081] Step S236: Select the eigenvector corresponding to the smallest eigenvalue as the initial normal vector direction of the final fitting plane. This includes: selecting the eigenvalue with the smallest value (denoted as λmin) from the three eigenvalues ​​obtained in step S235, and finding the eigenvector corresponding to λmin (denoted as vmin).

[0082] Analyze the physical meaning of the eigenvector corresponding to the minimum eigenvalue: Since λmin is the smallest, it represents the lowest degree of dispersion of the centered coordinate point set in the direction of vmin (i.e., the distribution of all points in this direction is the most concentrated and the deviation is the smallest). The spatial reference plane of the transportation route needs to fit the overall direction of all points as closely as possible. Therefore, the direction of vmin is exactly consistent with the normal vector direction that the plane should have (the normal vector is perpendicular to the plane, and the dispersion of points in the plane in the direction of the normal vector is the lowest). Determine vmin as the initial normal vector direction of the final fitted plane, and record the three components of this eigenvector (such as (a, b, c)) to provide a directional reference for subsequent calculation of plane parameters.

[0083] Step S237: Calculate the plane constant term using the coordinates of the geometric center point and the initial normal vector to construct the final fitted plane parameters. This includes: defining the basic form of the plane equation: In a three-dimensional coordinate system, the plane equation can be expressed as ax + by + cz + d = 0, where (a, b, c) is the plane normal vector (i.e., vmin determined in step S236), (x, y, z) are the coordinates of any point on the plane, and d is the plane constant term (which needs to be calculated and determined); Since the geometric center point (Xcenter, Ycenter, 0) is the center of the transportation route, it must be located on the fitted spatial reference plane. Therefore, substitute the coordinates of this center point and the normal vector (a, b, c) into the plane equation:

[0084] Substituting the values, we get: a×Xcenter+b×Ycenter+c×0+d=0.

[0085] Solve the equation to get d: d = -(a×Xcenter + b×Ycenter).

[0086] After calculating the plane constant term d, the normal vector (a, b, c) is combined with the constant term d to form the final fitting plane parameters (a, b, c, d). These parameters fully describe the position (determined by d) and orientation (determined by the normal vector) of the spatial reference plane, ensuring that the plane can accurately fit the overall spatial orientation of the transportation route.

[0087] Step S238: Construct a spatial reference plane based on the final fitted plane parameters, including: defining the spatial reference plane in a two-dimensional plane coordinate system (since the Z-axis is always 0, it can be simplified to two dimensions) based on the fitted plane parameters (a, b, c, d) obtained in step S237.

[0088] Since all plane coordinate points are in the two-dimensional plane Z=0, the plane equation is simplified to ax+by+d=0 (c×0 can be ignored); based on the simplified plane equation, the corresponding plane is drawn in the two-dimensional coordinate system: the plane covers the entire range of the plane coordinate point set obtained in step S223, and all optimized plane coordinate points are as close to the plane as possible (with the lowest degree of discretization).

[0089] The validity of the constructed spatial reference plane is verified by calculating the distance from each point in the optimized plane coordinate point set to the plane (i.e., the distance from the point to the line, since the plane is represented as a straight line in a two-dimensional plane), and calculating the average and maximum distances. If the average distance is much smaller than the overall length of the transportation route (e.g., the average distance is less than 100 meters, while the route length is greater than 10 kilometers), it indicates that the plane can effectively represent the overall direction of the route, and the verification is successful. If the average distance is too large, the steps of outlier removal, covariance matrix calculation, and eigenvalue decomposition are re-examined, and the plane is reconstructed after correction. After successful verification, the spatial reference plane is stored in the route geometry model library of the TMS system and bound to the corresponding transportation task (associated through the task number). This provides a unified spatial benchmark for subsequent mapping of location data to the analysis grid, calculation of the minimum bounding sphere volume, and generation of comprehensive adjustment parameters, thus solving the defect of the existing system that cannot comprehensively characterize the geometric features of the transportation route.

[0090] In this embodiment, step S21 collects continuous location data containing latitude, longitude and timestamp to ensure that the data has spatiotemporal continuity and avoid information gaps at discrete points; step S22 converts spherical latitude and longitude into two-dimensional planar coordinates through map projection, eliminating geometric deviations when directly analyzing spherical coordinates (such as complex latitude and longitude distance conversion and inconsistent scales in different regions), making the location data more suitable for the needs of subsequent planar modeling and mesh analysis. Step S23 generates a spatial reference plane using a plane fitting algorithm, breaking the limitation of existing systems that can only replay single trajectory points. This reference plane can intuitively reflect the overall direction of the transportation route (such as the approximate path trend from point A to point B, and whether there is an overall offset direction), rather than an isolated set of points. Managers can quickly grasp the spatial geometric characteristics of the route through the reference plane (such as whether it extends along the planned route as a whole, and whether there is a concentrated direction in a certain area). The spatial reference plane is the core carrier for dividing the analysis grid and mapping the location data in step S2. If there is no unified reference plane, the subsequent grid division will lead to analysis deviations due to the chaotic coordinate system. The standardized reference plane can ensure that all location data is mapped to a unified analysis dimension, thereby ensuring the accuracy of subsequent calculations of comprehensive adjustment parameters such as the minimum bounding sphere volume and spatial centroid coordinates, and providing a reliable spatial benchmark for accurately assessing dynamic characteristics such as driving behavior and route deviation.

[0091] In a preferred embodiment of the present invention, based on the distribution characteristics of the location data, the reference plane is divided into several analysis grids, and each location data is mapped to the corresponding analysis grid, including:

[0092] Step S24: Based on the spatial distribution range of the optimized planar coordinate point set, calculate the length and width of the circumscribed rectangle of the spatial distribution range; based on the length and width of the circumscribed rectangle, initialize a uniform meshing scheme covering the entire spatial distribution range to obtain the initial mesh cell size; on the spatial reference plane, perform the first meshing based on the initial mesh cell size to generate multiple initial mesh cells covering the entire spatial distribution range, including:

[0093] From the optimized set of planar coordinate points, extract the maximum X-coordinate (Xmax), minimum X-coordinate (Xmin), maximum Y-coordinate (Ymax), and minimum Y-coordinate (Ymin) of all points. The rectangular area formed by these values ​​is the spatial distribution range of the location data points. This rectangle is the smallest bounding rectangle that can completely cover all points.

[0094] Calculate the length and width of the circumscribed rectangle. The length is the span in the X direction, i.e., length = Xmax - Xmin; the width is the span in the Y direction, i.e., width = Ymax - Ymin.

[0095] Initialize a uniform grid partitioning scheme: Based on the total length of the transportation route and the total number of location data points, set a baseline for the number of initial grid cells (for example, if the total number of points is 1000, the total number of initial grid cells can be set to approximately 100, balancing computational efficiency and accuracy); divide the length of the circumscribed rectangle by the number of initial grid cells in the X direction to obtain the X-direction dimension (△X) of the initial grid cell; divide the width by the number of initial grid cells in the Y direction to obtain the Y-direction dimension (△Y) of the initial grid cell. △X and △Y together constitute the initial grid cell size.

[0096] Using the spatial reference plane as a baseline, and starting from the lower left corner (Xmin, Ymin) of the circumscribed rectangle, the grid is divided sequentially along the X-axis by ΔX and along the Y-axis by ΔY, forming multiple uniformly sized rectangular initial grid cells. Each initial grid cell is distinguished by a unique identifier (such as grid row number-column number), and its boundary coordinates (Xstart, Xend, Ystart, Yend) are recorded to ensure that all initial grid cells do not overlap and completely cover the circumscribed rectangle, i.e., the entire spatial distribution range from (Xmin, Ymin) to (Xmax, Ymax).

[0097] Step S25: Count the number of location data points falling into each initial grid cell to obtain the point density of each initial grid cell, and filter them to obtain densely distributed and sparsely distributed regions. This includes: traversing all points in the optimized planar coordinate point set, determining which initial grid cell each point belongs to for its planar coordinates (X, Y) (by comparing whether X is between Xstart and Xend of the grid, and whether Y is between Ystart and Yend of the grid), and counting the number of points in the corresponding grid cell until all points have been counted, thus obtaining the number of points in each initial grid cell.

[0098] Calculate point density: The point density of each initial grid cell = the number of points in the grid ÷ the area of ​​the grid (area = △X × △Y). Point density reflects the number of location data points per unit area. The higher the value, the denser the location data distribution in that area.

[0099] Set the filtering threshold: Calculate the average point density of all initial grid cells (total number of points ÷ total area of ​​the bounding rectangle), set 1.5 times the average point density as the dense threshold, and set 0.5 times the average point density as the sparse threshold (the threshold can be adjusted according to the transportation scenario, such as increasing the dense threshold for urban routes and decreasing it for suburban routes).

[0100] Region division: The continuous region where the initial grid cell with point density is higher than the dense threshold is marked as a dense distribution region (such as a city with frequent lane changes or congested road sections, where the location points are densely distributed); the continuous region where the initial grid cell with point density is lower than the sparse threshold is marked as a sparse distribution region (such as a highway with constant speed driving sections, where the location points are sparsely distributed).

[0101] Step S26: Recursively subdivide the initial grid cells in densely distributed regions; merge the initial grid cells in sparsely distributed regions to form several analysis grids, including: recursively subdividing the initial grid cells in densely distributed regions:

[0102] Select an initial grid cell within a densely distributed region, and divide it into two equal parts along both the X and Y axes, further subdividing it into four equal-sized sub-grid cells (each sub-grid has an X dimension of ΔX / 2 and a Y dimension of ΔY / 2).

[0103] Count the number of points in each subgrid and calculate the point density. If the point density of a subgrid is still higher than the density threshold, repeat the above subdivision operation on the subgrid (divide it into 4 smaller subgrids again).

[0104] Recursively subdivide until the stopping condition is met: the point density of the subgrid is lower than the density threshold, or the size of the subgrid reaches the preset minimum threshold (e.g., the size in the X or Y direction is less than 50 meters to avoid excessive computation due to the small size of the grid); perform the above recursive subdivision on all initial grid cells in the densely distributed area to form multiple smaller grid cells to accurately capture the location data distribution details of the dense area.

[0105] Merge the initial grid cells in sparsely distributed regions:

[0106] Select initial grid cells in sparsely distributed regions and merge them into groups of 2×2 adjacent grid cells (i.e., merge two consecutive initial grid cells horizontally and two consecutive initial grid cells vertically into one large grid cell). The merged grid cell has an X size of 2×△X and a Y size of 2×△Y. The boundary is the minimum Xstart, maximum Xend, minimum Ystart, and maximum Yend of the original four grid cells. Calculate the point density of the merged grid (total number of points after merging ÷ area after merging). If it is still lower than the sparse threshold, the adjacent 2×2 merged grid cells can be merged again (to form a larger grid of 4×4 initial grid size).

[0107] Merge until the stopping condition is met: the point density of the merged mesh is higher than the sparsity threshold, or the size of the merged mesh reaches the preset maximum threshold (such as the size in the X or Y direction is greater than 1000 meters, to avoid the loss of details due to the mesh being too large); perform the above merging operation on all initial mesh cells in the sparsely distributed region to form multiple larger mesh cells, reduce the number of meshes in the sparse region and improve computational efficiency.

[0108] The subdivided dense region grid cells are integrated with the merged sparse region grid cells, and overlapping boundaries (if any) are removed. Each grid cell is assigned a unique identifier (such as region type-subdivision / merging level-row number-column number) to form several final analysis grids. These grids can accurately reflect the details of dense regions and efficiently cover sparse regions.

[0109] Step S27: For each location data point, based on its planar coordinates, traverse all analysis grids to determine the target analysis grid in which it belongs; record the unique identifier of the location data point into the data structure of its target analysis grid to complete the mapping process, including: preparing the boundary data of the analysis grid: organizing the boundary coordinate information of all analysis grids to form a mapping table of grid identifier-boundary coordinates (Xstart, Xend, Ystart, Yend) to facilitate quick query of the spatial range of each grid.

[0110] Traverse all location data points: For each location data point in the optimized set of planar coordinate points, extract its planar coordinates (X, Y) and unique identifier (such as the data collection timestamp + sequence number to ensure that each point is unique and traceable).

[0111] Determine the target analysis grid: For a single location data point (X, Y), sequentially query each grid in the analysis grid mapping table to determine if the point is within the grid's boundary (i.e., Xstart ≤ X ≤ Xend and Ystart ≤ Y ≤ Yend). Find the first grid that meets the conditions, which is the target analysis grid to which the point belongs. Add a unique identifier for the location data point to the target analysis grid's data structure (such as a list or dictionary) to establish a point-grid association record. If a point falls into multiple grids simultaneously (due to boundary calculation errors), use the grids of adjacent points before and after the point's timestamp as a reference to select consecutive grids as the target grid to avoid mapping confusion. Repeat the above steps until all location data points are mapped to their corresponding analysis grids. Ultimately, the data structure of each analysis grid contains unique identifiers for all location data points falling into that grid, laying the foundation for subsequent calculations of the spatial distribution characteristics of location data within each grid (such as the minimum enclosing sphere volume and spatial centroid coordinates).

[0112] In a preferred embodiment of the present invention, for each set of location data within an analysis grid, by calculating the volume of the smallest enclosing sphere formed by its spatial distribution range and calculating the spatial centroid coordinates of the set, comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process are generated, including:

[0113] Step S28: Based on the corresponding analysis grid, obtain all position data points mapped to the analysis grid to form the position data set of the analysis grid; based on the three-dimensional spatial coordinates of all points in the position data set, calculate the arithmetic mean of all points in the three coordinate axes, and use the obtained average coordinate point as the spatial barycenter coordinates of the position data set; using the spatial barycenter coordinates as the initial sphere center, calculate the spatial distance from each point in the position data set to the initial sphere center, and find the maximum value among all distances, using the maximum value as the initial sphere radius, including:

[0114] For a single analysis grid, all associated location data points are retrieved from the optimized planar coordinate point set using the grid's data structure (which records the unique identifiers of the location data points falling into the grid). Since subsequent three-dimensional spatial calculations are required, the two-dimensional planar coordinates (X, Y) of each location data point need to be supplemented into three-dimensional spatial coordinates (X, Y, 0) (the Z-axis coordinate is set to 0, ensuring the consistency of spatial calculation logic since all location points are in the same two-dimensional plane). All the supplemented three-dimensional coordinate points together constitute the location data set of the analysis grid. At the same time, the integrity of the set is verified to confirm that the number of points in the set is consistent with the number of unique identifiers recorded in the grid data structure, avoiding data omissions or duplications.

[0115] The spatial barycentric coordinates of the location dataset reflect the center position of the dataset in three-dimensional space. The calculation process is performed separately for each of the three coordinate axes:

[0116] To calculate the centroid coordinates of the X-axis, extract the X-coordinate values ​​of all points in the location data set, add these X-values ​​together to get the sum, and then divide the sum by the total number of points in the set to obtain the arithmetic mean in the X-axis direction, denoted as Gx.

[0117] Y-axis centroid coordinate calculation: Similarly, extract the Y-coordinate values ​​of all points and sum them. Divide the sum by the total number of points to obtain the arithmetic mean in the Y-axis direction, denoted as Gy.

[0118] Z-axis centroid coordinate calculation: Since the Z-axis coordinates of all points are 0, the summation result is 0, and dividing by the total number of points is still 0, denoted as Gz;

[0119] By combining Gx, Gy, and Gz, we obtain the spatial centroid coordinates (Gx, Gy, Gz) of the location data set, and use them as the initial center of the subsequent initial enclosing sphere.

[0120] The initial radius of the enclosing sphere is determined using the spatial centroid coordinates (Gx, Gy, Gz) as the initial sphere center. The three-dimensional spatial distance from each point in the position data set to this initial sphere center is calculated (since the Z-axis is 0, the actual distance is equivalent to the distance from a point in the two-dimensional plane to the centroid, but it is calculated according to three-dimensional logic to maintain consistency):

[0121] For each point (Xi, Yi, 0) in the set, calculate its spatial distance to the initial sphere center (Gx, Gy, Gz); iterate through all the distance values ​​of all points, select the largest distance value, and determine the largest value as the initial sphere radius. This radius can ensure that the initial enclosing sphere completely contains all points in the position data set (because the maximum distance is the radius, the distance from all points to the sphere center does not exceed the radius).

[0122] Step S29: Based on the initial sphere center and initial sphere radius, construct an initial bounding sphere that encloses all data points in the location data set; iteratively optimize the initial bounding sphere by gradually adjusting the sphere center position and reducing the sphere radius until the sphere surface contacts at least three non-coplanar points in the location data set, and the sphere contains all data points, to obtain the minimum bounding sphere of the location data set; calculate the volume of the minimum bounding sphere, combine the spatial centroid coordinates with the volume of the minimum bounding sphere, and perform weighted fusion calculation using preset weight coefficients to generate comprehensive adjustment parameters, including:

[0123] Based on the initial sphere center (Gx, Gy, Gz) and initial sphere radius R0 determined in step S28, a complete sphere is constructed in three-dimensional space, i.e., the initial enclosing sphere. The boundary of this sphere satisfies the following: all points in the location data set are located inside the sphere or on the surface of the sphere (because the radius is the maximum distance), ensuring that the core objective of enclosing all points is initially achieved; at the same time, the parameters of the initial enclosing sphere (sphere center coordinates, radius) are recorded as the basis for subsequent iterative optimization.

[0124] The core objective of iterative optimization of the initial bounding sphere is to adjust the position of the sphere's center and reduce its radius, ultimately obtaining a minimum bounding sphere with the smallest volume that still encloses all points. The specific process is as follows:

[0125] Step 1: Identify boundary points

[0126] Calculate the distance from all points in the location dataset to the current surface of the sphere (i.e., the distance from the point to the center of the sphere - the current radius of the sphere), and filter out points whose distance is less than or equal to a preset small threshold (e.g., 0.1 meters, used to determine if a point is close to the surface of the sphere). These points are called boundary points. Boundary points are the key to affecting the volume of the sphere, and optimization should be carried out around the boundary points.

[0127] Step 2: Adjust the position of the ball's center

[0128] Select three non-coplanar points from the boundary points (non-coplanarity avoids optimization deviations caused by the linear distribution of points on the sphere). Calculate the circumcenter of the triangle formed by these three points (which is the center of the circumsphere in 3D space). Adjust the current center of the sphere surrounding the sphere to the position of this circumcenter. If there are fewer than three boundary points (e.g., only one or two), select the one or two boundary points farthest from the current center and move the center of the sphere towards these points (the movement distance is the current radius minus half the distance from the point to the center), ensuring that the sphere still contains all points after the movement.

[0129] Step 3: Reduce the radius of the sphere

[0130] Using the adjusted new sphere center as a reference, recalculate the maximum distance from all position data points to the new sphere center, and set this maximum distance as the new sphere radius. At this time, the new radius will be smaller than the radius before adjustment (because the sphere center is closer to the boundary point, the maximum distance is shortened).

[0131] Step 4: Determine the stopping condition for iteration

[0132] Compare the difference between the adjusted sphere radius and the previous radius: if the difference is less than the preset accuracy threshold (e.g., 0.05 meters, indicating that the radius reduction space is extremely small), and the current sphere surface is in contact with at least 3 non-coplanar boundary points (ensuring the sphere shape is stable and there is no possibility of further reduction), and all position data points are inside the sphere, then stop the iteration; if the conditions are not met, repeat the process of identifying boundary points → adjusting the sphere center → reducing the radius until the stopping condition is met; the sphere obtained after the iteration stops is the minimum bounding sphere of the position data set, and its final sphere center coordinates and final radius are recorded.

[0133] The volume of the minimum bounding sphere is calculated based on its final radius (denoted as Rmin). Following the logic of sphere volume calculation (the volume of a 3D sphere is proportional to the cube of its radius), the volume of the minimum bounding sphere (denoted as Vmin) is calculated. The size of the volume value directly reflects the spatial distribution dispersion of the location data set. A larger volume indicates a more dispersed distribution of location points within the grid (e.g., frequent lane changes by drivers, route deviations); a smaller volume indicates a more concentrated distribution of location points (e.g., uniform straight-line driving, stable route).

[0134] Generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process. These parameters require the fusion of two types of information: spatial centroid coordinates and the volume of the smallest enclosing sphere. Weighted fusion is achieved through preset weighting coefficients. Specific steps include:

[0135] Determine the weighting coefficients: Set weights according to the assessment requirements of the dynamic characteristics of the transportation process. For example, if the focus is on assessing the degree of route deviation, the weight of the minimum enclosing sphere volume (denoted as Wv) is set to 0.6, and the weight of the spatial center of gravity coordinates (denoted as Wg) is set to 0.4 (the center of gravity coordinates reflect the concentrated position of the points, and the center of gravity will deviate from the reference direction when there is a deviation). If the focus is on assessing driving stability, the weight of the center of gravity coordinates can be appropriately increased. The total weighting coefficients are 1, and they need to be verified and adjusted through historical transportation data to ensure the accuracy of the assessment.

[0136] Weighted fusion calculation: Multiply the three components (Gx, Gy, Gz) of the spatial barycenter coordinates by the weight Wg respectively to obtain the weighted barycenter coordinates (Gx×Wg, Gy×Wg, Gz×Wg); multiply the minimum enclosing sphere volume Vmin by the weight Wv to obtain the weighted volume (Vmin×Wv).

[0137] Forming a comprehensive adjustment parameter: The weighted centroid coordinates and the weighted volume are combined to form a parameter containing four-dimensional information (Gx×Wg, Gy×Wg, Gz×Wg, Vmin×Wv). This parameter is used to evaluate the dynamic characteristics of the transportation process (such as route deviation, driving stability, and regional traffic difficulty). Subsequent steps (such as reimbursement detail calculation and audit process optimization) can be based on this parameter to quantitatively judge the actual situation of the transportation process, ensuring that the decision-making is supported by accurate data.

[0138] In a preferred embodiment of the present invention, step S3 includes:

[0139] Step S31: When the TMS system detects that the driver has completed the loading confirmation operation and updated the task status through the mobile terminal, it automatically triggers the prepayment process, including: After the driver completes the vehicle loading, he needs to find the loading confirmation function button for the corresponding task on the task execution page of the logistics-dedicated mobile terminal (an APP bound to the TMS system). After clicking, he needs to upload the loading vouchers (such as photos of the vehicle after loading, and photos of paper documents signed by the loading and unloading contact person). After confirming that everything is correct, he submits the loading confirmation operation. The TMS system monitors the task status change data of the driver's mobile terminal in real time. When it receives the operation signal of loading confirmation submission, it automatically verifies the completeness of the vouchers (such as whether the photos are clear and whether the document information is consistent with the loading and unloading requirements in the dispatch order). If the vouchers are compliant, the system updates the status of the task from pending loading to loaded and ready to depart. The TMS system has built-in rules for the association between status and process. When the task status is updated to loaded and ready to depart for the first time, no manual intervention is required. The system automatically triggers the preset prepayment process. After the process starts, the system transmits key information such as the task number and driver ID to the finance module to prepare for the subsequent calculation of the prepayment amount.

[0140] Step S32: Calculate the prepayment amount for this task according to the preset prepayment rules, and generate a prepayment instruction containing the prepayment amount, task number, and driver account information. This includes: The TMS system's fee rule library pre-stores prepayment rules for different types of transportation tasks. Rule dimensions include: task type (order transportation / short-distance transfer / express transportation), transportation mileage (based on map distance measurements of the origin and destination locations in the dispatch order), vehicle type (flatbed truck / closed truck, different vehicle types have different prepayment ratios), and driver level (drivers with excellent historical performance can enjoy higher prepayment ratios). The system automatically matches the corresponding prepayment rule based on the dispatch information of the current task (e.g., for long-distance order transportation tasks, the prepayment ratio is 30% of the basic freight).

[0141] To calculate the prepayment amount for this task, first obtain the basic freight cost for this task from the fee standard associated with the dispatch order (basic freight cost = transportation mileage × unit mileage freight cost, the unit mileage freight cost is preset according to the route type (highway / national highway)); then calculate the prepayment amount according to the matched prepayment ratio (e.g., basic freight cost of 10,000 yuan × 30% = 3,000 yuan); if the task includes special requirements (e.g., additional equipment deposit is required for the transportation of fragile vehicles), then add the corresponding fees to the basic prepayment amount to obtain the final prepayment amount.

[0142] Generate a prepayment instruction. The instruction must contain complete transaction information, including: a unique task identifier (task number, associated with the dispatch order and driver), prepayment amount (accurate to the cent, indicating the currency unit), driver account information (retrieved from the driver database, including the bank name, bank account number, and account name, ensuring consistency with the driver's real-name authentication information), purpose of payment (indicating prepayment for XX task (task number)), and payment deadline (e.g., payment to be completed within 24 hours). After the system automatically generates the instruction, it verifies the account information (e.g., whether the number of digits in the account number conforms to the corresponding bank rules, and whether the account name matches the driver's name). Once the verification is successful, the instruction content is locked to prevent modification.

[0143] Step S33: The prepayment instruction is sent to the associated third-party payment platform, which then completes the prepayment to the driver's designated account. This includes: the TMS system establishing an encrypted communication channel with the cooperating third-party payment platform (such as a bank's corporate online banking interface or a third-party payment institution's interface) through a preset API interface (Application Programming Interface) to ensure data transmission security (using SSL encryption protocol to prevent account information leakage); the TMS system encapsulates the prepayment instruction generated in step S32 in the format required by the third-party payment platform (such as JSON format) and pushes it to the payment platform through the API interface. The system's authentication information (such as API key and enterprise ID) is included with the push notification to verify the legitimacy of the instruction's source and prevent unauthorized access. Upon receiving the instruction, the third-party payment platform first verifies its completeness (whether key information is missing) and legitimacy (whether identity authentication is successful and account information is valid). If verification is successful, the corresponding prepayment amount is automatically deducted from the enterprise's designated payment account and transferred to the driver's designated account. During the payment process, the payment platform records transaction logs in real time (such as payment time and transaction serial number). After completing the payment, the payment platform sends a callback to the TMS system via API to confirm the payment result (success / failure): if successful, it provides the transaction serial number, actual payment time, and estimated arrival time; if unsuccessful (e.g., driver's account frozen, account information incorrect), it provides the reason for failure (e.g., abnormal account status). Upon receiving the result, the TMS system binds it to the corresponding task and stores it in the prepayment record ledger, then pushes a payment result notification to the driver's mobile terminal (displaying the amount and transaction number upon success, and providing the reason and an entry point for resubmitting account information upon failure).

[0144] Step S34: When the TMS system detects that the driver has completed the arrival confirmation operation via mobile terminal and updated the task status to "completed," it automatically pushes a task completion notification and reimbursement application portal to the driver's mobile terminal. This includes: After the driver delivers the vehicle to the destination and completes unloading, they need to click the arrival confirmation button on the task execution page of the mobile terminal and upload unloading vouchers (such as a signed receipt from the recipient or a photo of the empty vehicle). If the recipient has an electronic signature system, the TMS system can automatically obtain electronic signature information through its connection with the recipient's system. After the TMS system verifies the compliance of the unloading vouchers, it updates the task status from "in transit" to "task completed."

[0145] The system automatically generates task completion notifications. Once the task status is updated to "completed," the TMS system automatically generates notification content including: task number, task name (e.g., vehicle transport from OEM to dealer), actual completion time (the time the system records the status update), transport mileage (actual mileage, calculated from on-the-go location data), and cargo condition (default is "intact"; if damaged, the recipient's reported damage information is displayed). Push notifications and reimbursement application entry are provided to the driver's mobile terminal in two ways: first, an in-app pop-up notification (highest priority, displayed directly when the driver opens the app, cannot be ignored); second, an SMS notification (as a backup, containing a brief message about task completion and a link to view the reimbursement entry, which redirects to the corresponding page in the app). Simultaneously, in the driver's app's Task Center - Completed Tasks list, a new reimbursement application button (i.e., reimbursement application entry) is added under this task entry, and the button is set to clickable, prompting the driver to initiate a reimbursement process.

[0146] Step S35: In response to the reimbursement application submitted by the driver via mobile terminal, retrieve the basic reimbursement items and corresponding standard fees for this task, including: After the driver clicks the reimbursement application button, he / she needs to confirm the task information (such as task number, start and end locations) on the pop-up reimbursement application page, and fill in the necessary supplementary reimbursement information (such as the number of toll receipts generated during this transportation, and whether there are temporary repair costs), and submit the reimbursement application after confirmation; After receiving the application signal, the TMS system records the application time and marks the task as entering the reimbursement processing status to prevent duplicate applications.

[0147] The TMS system retrieves the basic reimbursement items for this task. Based on the task number, it automatically matches the basic reimbursement items for this task by associating the transportation type, route information, and pre-set reimbursement item database in the dispatch order. Basic reimbursement items are fixed categories and commonly include: basic freight (the core item, consistent with the basic freight at the time of prepayment), fuel subsidy (calculated based on transportation mileage), toll subsidy (calculated based on the pre-set highway / national road ratio for the route, which can be adjusted later based on receipts), driver labor subsidy (calculated based on transportation time, such as 20 yuan per hour), and basic vehicle wear and tear fee (a fixed standard based on vehicle type, such as 500 yuan per flatbed truck).

[0148] Obtain the standard costs for basic reimbursement items. Retrieve the calculation basis and standard amount for each basic reimbursement item from the cost standard library. For example, the standard for fuel subsidy is 0.8 yuan per kilometer. Combined with the actual transportation mileage of this task (the driving distance calculated from the on-the-road location data), the standard amount for fuel subsidy is obtained (e.g., 1000 kilometers × 0.8 yuan / kilometer = 800 yuan). The standard for toll subsidy is 1000 yuan based on the default toll fee for the route (which can be increased or decreased later based on actual receipts). The standard amount for basic freight is 10,000 yuan (consistent with the prepayment). All standard costs are bound to the task number and stored in the temporary reimbursement details library.

[0149] Step S36: Obtain the comprehensive adjustment parameters of all analysis grids generated in this task, and calculate their arithmetic mean to obtain the global comprehensive adjustment coefficient of this task. This includes: The TMS system extracts the comprehensive adjustment parameters of all analysis grids generated in step S2 from the transportation dynamic parameter library according to the task number; the comprehensive adjustment parameters of each analysis grid are four-dimensional information (Gx×Wg, Gy×Wg, Gz×Wg, Vmin×Wv), where Vmin×Wv (weighted minimum enclosing sphere volume) is the core indicator reflecting transportation dynamics, and Gx×Wg and Gy×Wg (weighted centroid coordinates) help reflect the route offset.

[0150] Calculate the arithmetic mean of the overall adjustment parameters. Since the overall adjustment parameters are four-dimensional data, the arithmetic mean needs to be calculated for the values ​​of each dimension separately:

[0151] Extract the Gx×Wg values ​​from all analysis grid parameters, sum them, and divide by the total number of analysis grids to obtain the average value of the Gx dimension (denoted as AvgGx); similarly, calculate the average value of the Gy dimension (AvgGy), the average value of the Gz dimension (AvgGz, since the Z-axis is 0, the average value is still 0), and the average value of the Vmin×Wv dimension (AvgV).

[0152] The average values ​​of the four dimensions are integrated to form the global comprehensive adjustment coefficient for this task, expressed as (AvgGx, AvgGy, AvgGz, AvgV). This coefficient reflects the overall dynamic characteristics of the entire transportation process: for example, the larger the value of AvgV, the higher the degree of dispersion of the vehicle's driving route during this transportation process (such as frequent detours and lane changes); the more AvgGx or AvgGy deviates from the preset route center, the more serious the overall route deviation.

[0153] Step S37: The standard cost of the basic reimbursement items is weighted and calculated with the global comprehensive adjustment coefficient, and the dynamic cost items are corrected to generate a preliminary reimbursement detail that includes detailed items, calculation process and final amount. This includes: for each basic reimbursement item, the standard cost is weighted and adjusted in combination with the core indicator (AvgV, as this indicator best reflects the impact of transportation dynamics on costs) in the global comprehensive adjustment coefficient; the adjustment rules are preset in the system, for example: the basic freight is weighted by the AvgV coefficient. If AvgV≤1.0 (stable transportation, no obvious detour), the basic freight remains at the standard amount; if 1.0<AvgV≤1.2 (minor detour or lane change), the basic freight = standard amount × 1.05; if AvgV>1.2 (serious detour or frequent lane change), the basic freight = standard amount × 1.1 (additional compensation for fuel costs and time costs is required). According to this rule, the standard cost for each basic reimbursement item is calculated to obtain the adjusted basic reimbursement amount (e.g., if the basic freight standard is 10,000 yuan and AvgV = 1.05, the adjusted amount is 10,000 × 1.05 = 10,500 yuan).

[0154] Dynamic expense items are non-fixed categories and need to be adjusted based on actual transportation conditions and overall adjustment coefficients. Common dynamic items include: overtime allowance (if the transportation time exceeds the preset time of the dispatch order, it is calculated as the number of overtime hours × the hourly allowance standard; a high AvgV indicates that the overtime may be due to detours, which needs to be confirmed and adjusted), detour fuel allowance (if AvgV > 1.0, it is calculated as the excess coefficient × the basic fuel allowance; for example, if AvgV = 1.05, the detour fuel allowance = 800 yuan × 0.05 = 40 yuan), and temporary repair allowance (if the driver submits a repair invoice, based on the reason for the repair (such as a breakdown due to a bumpy route; a high AvgV may reflect poor road conditions), 80% of the invoice amount will be reimbursed). When making adjustments, the system automatically links the corresponding supporting documents (such as overtime records and repair invoices) to ensure that the expense adjustments are based on evidence.

[0155] The adjusted basic reimbursement items and the revised dynamic reimbursement items are compiled into a detailed list in the format of item name-calculation basis-standard amount-adjustment coefficient-adjusted amount-remarks; for example: Basic freight - 1000 km × 10 yuan / km (standard) - 10000 yuan - global coefficient 1.05 - 10500 yuan - adjusted due to slight detour; Fuel subsidy - 1000 km × 0.8 yuan / km (standard) - 800 yuan - no adjustment - 800 yuan - no remarks; Detour fuel subsidy - fuel standard × 0.05 (coefficient exceeding part) - 40 yuan - no adjustment - 40 yuan - compensation due to AvgV = 1.05. At the same time, the total reimbursement amount (the sum of the adjusted amounts of all items) is calculated at the end of the details, and the prepaid amount (such as 3000 yuan) is deducted to obtain the amount to be reimbursed this time; finally, a preliminary reimbursement detail containing all detailed items, calculation process and final amount is formed and stored in the reimbursement detail database of the TMS system.

[0156] Step S38 involves pushing the preliminary reimbursement details to the driver's mobile terminal for verification and confirmation. This includes: the TMS system pushing the preliminary reimbursement details generated in step S37 to the driver's mobile terminal in the form of a PDF preview or an in-app list; simultaneously sending an in-app message reminder (e.g., "Your XX task (task number) reimbursement details have been generated, please check") and notifying the driver via SMS (to prevent the driver from missing information if they are not logged into the app); when displaying the details, it is necessary to ensure that the calculation process for each item is clearly visible (e.g., clicking on the item name can expand to see how the standard amount is calculated and why the adjustment factor is 1.05), so that the driver can understand it easily.

[0157] Drivers should verify the details: When viewing the details on their mobile devices, drivers should focus on verifying the following: whether the reimbursement items are complete (e.g., whether toll subsidies are missing), whether the calculation basis is accurate (e.g., whether the transport mileage is consistent with the actual driving), whether the adjustment coefficient is reasonable (e.g., whether AvgV matches their driving conditions), and whether the reimbursement amount is correct (whether the total amount minus the prepaid amount is correct). If a driver has any questions about a certain item, they can click the question mark button on the details page, fill in the question (e.g., the toll subsidy standard should be 1200 yuan, not 1000 yuan), and upload relevant supporting documents (e.g., photos of toll receipts).

[0158] Driver Confirmation or Rejection of Detailed Claims: After verifying the details, the driver clicks the "Confirm" button at the bottom of the details page. The system records the confirmation time and updates the claim details status to "Driver Confirmed," then pushes it to the next stage (Regional Manager Review). If the driver believes the details are incorrect, they click the "Reject and Modify" button, submitting the questions and supporting documents to the TMS system's finance module. Upon receiving the feedback, the finance staff re-verifies the expense standards and adjustment coefficients, corrects the details, and pushes them back to the driver for confirmation. This process continues until the driver confirms the details are correct, completing the initial confirmation process for the claim details.

[0159] In a preferred embodiment of the present invention, step S4, after the reimbursement details are confirmed by the driver, involves a multi-level review process by the regional manager and finance department, optimized in conjunction with the comprehensive adjustment parameters, including:

[0160] Step S41: After the driver's mobile terminal returns a confirmation instruction for the preliminary expense reimbursement details, the TMS system locks the expense reimbursement details and automatically assigns it to the corresponding regional manager's review node based on the task's region. This includes: When the TMS system receives the preliminary expense reimbursement details confirmation instruction from the driver's mobile terminal, it immediately triggers the details locking mechanism. The system marks the status of the expense reimbursement details as pending regional manager review - locked. After locking, all items in the details (such as basic freight, adjustment coefficient, and total amount) cannot be edited. At the same time, a lock log is automatically generated, recording the lock time, the operation subject (automatically executed by the system), and the task number, to prevent any subsequent role (including the driver and finance) from modifying the details and to ensure the uniqueness of the review basis.

[0161] Determine the task's region: The TMS system extracts the main driving area of ​​the task from the dispatch information (determined according to the administrative region of the start and end points in the dispatch order, such as a task from area A to area B, the main driving area is East China). The system has a preset task region-regional manager mapping table (such as North China region corresponding to North China regional manager, East China region corresponding to East China regional manager). If there are multiple managers in a region, the system further matches the regional managers of the corresponding regions according to the core areas the task passes through (such as Jiangsu region and Zhejiang region under East China region), ensuring that the reviewer is familiar with the transportation conditions and cost standards of the region.

[0162] Assignment to review node and notification: The system will package the locked reimbursement details (including the details list and driver confirmation record) with the basic information of the task (dispatch order number and transport vehicle information) and automatically push it to the matching regional manager's TMS management backend pending review task list; at the same time, the regional manager will be notified in two ways: one is a pop-up reminder in the regional manager's dedicated management APP (showing that there is 1 reimbursement details of XX task (task number) pending review), and the other is a work SMS notification (including a link to the review entry, which can directly jump to the TMS review interface), and indicate the review time limit (e.g., please complete the review within 2 working days).

[0163] Step S42, at the regional manager's review node, the system displays the transportation route trajectory of this task, the distribution map of each analysis grid, and the comprehensive adjustment parameters corresponding to each analysis grid to the regional manager. This includes: in the trajectory analysis module of the regional manager's review interface, the system converts the continuous location data of this task into a visual path and overlays it on a high-precision electronic map.

[0164] The route uses different colors to distinguish driving status (blue for normal driving, red for deviating from the recommended route, and yellow for abnormal stops exceeding 30 minutes). Hovering the mouse over a segment of the trajectory displays key data for that segment (such as driving time, real-time speed, and location timestamps). Clicking on a deviated segment brings up a deviation analysis (such as distance from the recommended route, duration of deviation, and the position of the corresponding analysis grid), helping regional managers determine whether the deviation is reasonable (e.g., whether it's due to road construction and detours). In the grid distribution module, the system displays the spatial reference plane corresponding to the transportation route and all analysis grids in the form of a heatmap.

[0165] The grid is color-coded according to the density of location data points within it (dark colors indicate densely distributed areas, such as urban road sections; light colors indicate sparsely distributed areas, such as highway sections). Each grid is labeled with a unique grid ID. Clicking on a grid allows you to view its boundary coordinates and the number of location data points it contains. If the grid has been subdivided or merged, the subdivision / merging level is displayed synchronously (e.g., initial grid, 1st subdivision), helping regional managers understand the grid division logic and the complexity of regional traffic.

[0166] Displaying comprehensive tuning parameters for each analysis grid: In the grid parameter module, the system displays the parameters in a combined format of grid ID, parameter list, and visual chart.

[0167] The list includes the four-dimensional integrated adjustment parameters (Gx×Wg, Gy×Wg, Gz×Wg, Vmin×Wv) for each grid, along with descriptions of the dynamic characteristics of each parameter (e.g., Vmin×Wv = 1.2, indicating that the location points within the grid are relatively scattered, and there may be detours or route changes). A bar chart is used to compare the Vmin×Wv values ​​(volume-weighted values) of different grids, and a scatter plot is used to show the distribution of Gx×Wg and Gy×Wg (gravimeter coordinate weighted values) of each grid on the spatial reference plane, helping regional managers quickly identify abnormal grids in the transportation process (e.g., areas with Vmin×Wv values ​​far exceeding those of other grids may have serious detours).

[0168] Step S43: The regional manager enters their review comments on the review interface. If the review fails, the reimbursement application is returned to the driver with an explanation of the reasons. If the review succeeds, the reimbursement details and the regional manager's electronic signature information are transferred to the next level of financial review. This includes: After logging into the TMS review interface, the regional manager can view the reimbursement details list, transportation route trajectory, grid distribution map, and grid comprehensive adjustment parameters in sequence. They can then review the reimbursement based on their understanding of regional transportation (such as whether the route frequently experiences construction detours and the parameter range for normal transportation).

[0169] The review interface has a review comments module at the bottom, which includes a preset comments drop-down box (such as normal transportation trajectory, no abnormality in comprehensive adjustment parameters, reasonable cost calculation, abnormal deviation with reasonable reasons, cost adjustment in line with standards, abnormal parameters without explanation, and supplementary materials required) and a custom comments text box. Regional managers can select preset comments and add explanations, or directly enter comments manually.

[0170] If you need to verify details (such as the reason for a certain deviation from the route), you can use the "Contact Driver" function within the interface to send a real-time message to the driver's mobile terminal (such as "Please explain the reason for the deviation from the route during XX time period"). After receiving a reply, you can continue the review.

[0171] Handling of rejected applications: If the regional manager determines that there are problems with the reimbursement details (e.g., the comprehensive adjustment parameters show a serious detour, but the details do not explain the reason for the detour and there is no supporting documentation, or the cost adjustment coefficient exceeds the regional standard), select "Reject Review," and specify the reasons for rejection in the comments box (e.g., XX grid Vmin×Wv=1.5, exceeding the normal range (≤1.2), no supplementary proof of the detour, and the reasonableness of the cost adjustment cannot be confirmed). Supporting materials (e.g., a document on the regional normal transportation parameter standards) can be uploaded. After clicking "Submit Rejection," the system will automatically return the reimbursement application to the driver's mobile terminal reimbursement center page, indicating that the regional manager has rejected the application, and will remind the driver to check the reasons for rejection via APP pop-up and SMS. The driver needs to supplement the materials according to the reasons (e.g., photos of traffic police construction notices for the detour) and resubmit the reimbursement application.

[0172] Processing upon approval: If the regional manager confirms that the transportation route is reasonable, the overall adjustment parameters are in line with the actual situation, and the cost calculation and adjustment are correct, the approval will be selected, and then electronic signature must be completed.

[0173] After completing identity verification according to the company's security settings (such as entering a preset approval password, verifying fingerprints or facial recognition), the system automatically generates an electronic approval document (including the approver's name, approval time, approval comments, and electronic seal). The system packages the approved expense details, the regional manager's electronic approval document, and supporting materials such as transportation trajectory / grid parameters, and automatically pushes them to the next level of financial review node through the TMS review workflow module. At the same time, the expense report status is updated to "Regional Manager Approved - Pending Financial Review" and a review notification (APP + SMS) is sent to the finance staff.

[0174] Step S45: At the financial review node, the system displays the expense details approved by the regional manager, the overall adjustment coefficient, and the calculation basis for this coefficient to the finance personnel. The finance personnel enter their final review comments on the review interface. If the review fails, the reason is noted and the application is returned to the previous node or the applicant. If the review is approved, electronic approval is completed, and the expense report status is updated to "approved." This includes: After the finance personnel log into the TMS financial review interface, the system displays core information in the review materials module.

[0175] Reimbursement Details: Displays a complete list of reimbursement details, including the project name, standard amount, global comprehensive adjustment coefficient, adjusted amount, and calculation basis for each item. Clicking on an item will expand to view the associated supporting data (such as the AvgV value associated with the basic freight adjustment and the invoice upload record associated with the dynamic cost).

[0176] Global Comprehensive Adjustment Coefficient Column: Displays global coefficients (AvgGx, AvgGy, AvgGz, AvgV) with numerical values ​​and comparison charts, indicating the normal transportation coefficient range (e.g., the normal range for AvgV is 0.8-1.2), and coefficients outside the range are highlighted in yellow to help finance personnel quickly identify anomalies;

[0177] The calculation basis section provides an entry point for parameter traceability. Clicking it will allow you to view: ① the original list of comprehensive adjustment parameters for all analysis grids in this task; ② screenshots of the global coefficient calculation process (including the summation steps for each dimension's value and the calculation steps for dividing by the total number of grids); ③ associated location data collection records (such as location data reporting time, latitude and longitude, and converted planar coordinates, ensuring that the parameter source is traceable); ④ the regional manager's review comments and approval documents, ensuring that the financial audit has complete data support.

[0178] Finance personnel review process: Finance personnel focus on reviewing the compliance of expenses, the accuracy of parameters, and the completeness of materials.

[0179] Verify that each reimbursement item complies with the company's "Logistics Expense Reimbursement Management Measures" (e.g., whether the basic freight adjustment ratio is within the prescribed upper limit, and whether there are corresponding legal invoices for dynamic expense items (e.g., toll receipts, maintenance invoices)); randomly check the comprehensive adjustment parameters of 3-5 analysis grids to verify their calculation accuracy (e.g., whether the Vmin×Wv value of a certain grid is consistent with the location data distribution of that grid), and verify whether the arithmetic mean calculation of the global coefficient is correct; check whether the driver has uploaded all necessary vouchers (e.g., loading and unloading receipts, toll receipts), and whether the regional manager's approval information is complete and valid (e.g., whether the signatory is the regional manager authorized by the company, and whether the electronic signature has passed the system's anti-counterfeiting verification).

[0180] Review result processing:

[0181] Review Failure: If there are issues, the finance staff should select "Review Failure," specify the reason in the feedback box, and select the return node: ① If the problem lies in the incorrect calculation of the global comprehensive adjustment coefficient, return it to the regional manager, stating that the XX grid parameters need to be re-verified, the global coefficient corrected, and attaching a screenshot of the error; ② If the problem lies in the driver not submitting key invoices, return it to the driver, explaining the supplementary materials required (e.g., uploading all electronic toll receipts for this transport within 3 working days); After submission, the system will synchronize the status to the corresponding node and send a return notification.

[0182] Approval complete: If all approval items meet the requirements, the finance staff clicks "Final Approval" to complete the electronic signature (the process is the same as for the regional manager, but the finance staff's authorization information needs to be verified). The system automatically updates the expense report status to "Approved - Pending Payment," synchronizes it to the TMS expense management ledger, and sends a notification of approval to the driver and regional manager (APP pop-up and SMS). The notification specifies that the expense payment process will be triggered within 1-3 business days. At the same time, the approved expense details and signature documents are archived in the company's financial system to prepare for subsequent payment and reconciliation.

[0183] In a preferred embodiment of the present invention, step S5, after approval, automatically triggers the reimbursement and payment process to complete the expense settlement; simultaneously updates contract management, reconciliation management, and handling of quality damage and violations to achieve closed-loop management of the entire process, including:

[0184] Once the TMS system detects that the expense report status has been updated to "Approved - Pending Payment," it automatically triggers the preset expense payment process without manual intervention. First, it extracts core payment information from the expense report details: 1) the final reimbursement amount (total reimbursement amount minus prepaid amount, e.g., total reimbursement amount 11340 yuan - prepaid 3000 yuan = 8340 yuan, accurate to the cent); 2) the driver's receiving account information (retrieving the verified account from the driver database, including the bank, account number, and account name, consistent with the prepaid account to avoid errors); 3) payment association information (task number, expense report number, payment purpose marked as "XX task (task number) reimbursement final payment"). Before generating the payment instruction, the system performs dual checks: ① verifying whether the corresponding reimbursement amount matches the total amount of the approved expense report details (to prevent data transmission errors); ② verifying whether the driver's account status is normal (checking whether the account is frozen or canceled through a third-party payment platform interface to avoid payment failure). Once the verification is successful, a formal reimbursement payment instruction will be generated, locking the amount and account information, which cannot be modified.

[0185] The TMS system, which connects to third-party payment platforms, uses an encrypted API interface (using the same payment channel as prepayments to ensure compatibility and security) to encapsulate reimbursement instructions in the format required by the payment platform before pushing them. The push includes the company's identity authentication information (API key, company bank account number) for the payment platform to verify the instruction's legitimacy. Upon receiving the instruction, the third-party payment platform first verifies its completeness (whether it includes the reimbursement amount, account information, and purpose) and the company's account balance (ensuring sufficient funds in the company bank account): ① If the balance is sufficient, the corresponding reimbursement amount is automatically deducted from the company bank account and transferred to the driver's receiving account; ② If the balance is insufficient, the system immediately reports the insufficient balance to the TMS system, which simultaneously sends an emergency notification (APP + SMS) to the finance personnel requesting timely top-up of the reimbursement account balance. The payment is then retried after the finance department tops up the account. After the payment is executed, the payment platform generates a unique transaction serial number, records the payment time and actual arrival time (estimated 1-2 business days), and sends a callback of the payment result (success / failure) to the TMS system via the API interface.

[0186] After receiving the payment result, the TMS system binds it to the expense report and task number, and stores it in the expense report payment ledger. The ledger content includes: expense report number, task number, amount due for reimbursement, actual payment amount, transaction serial number, payment time, estimated arrival time, and payment status (success / failure). Simultaneously, the system notifies the driver in two ways: ① a pop-up notification on the driver's mobile app (showing that the final payment for your XX task (task number) has been made, amount XX yuan, transaction number XXX, expected arrival in X business days); ② an SMS notification (content is the same as the app pop-up, ensuring the driver is aware even when not logged into the app). If the payment fails (e.g., due to changes in account information), the system indicates the reason for the failure in the notification (e.g., account information does not match real-name authentication) and provides an entry point to modify account information. After updating, the driver can re-trigger the payment process.

[0187] The TMS system automatically locates the corresponding transportation service contract based on the task number associated with the expense report (retrieved from the contract management module; the contract includes core clauses such as Party A (logistics company), Party B (driver / fleet), subject matter of transportation (vehicle model / quantity), performance requirements (transportation time limit, cargo integrity rate), and fee standards (unit price, adjustment rules). The system extracts key performance information from the task execution data: ① Actual transportation completion time (the time the task status is updated to completion); ② Actual transportation volume (the number of vehicles successfully delivered, verified against the contractually agreed transportation volume); ③ Cargo integrity status (records of quality damage reported by the recipient; if no damage is found, it is marked as intact); ④ Actual expenses incurred (total reimbursement amount, verified against the fee calculation method stipulated in the contract).

[0188] The contract performance status and clause verification and update system checks the performance status item by item according to the contract clauses: ① If the actual completion time is within the delivery time limit agreed in the contract, the transportation volume is consistent with the agreement, and there is no quality damage, the performance is deemed qualified, and the contract status is updated from "Performance" to "Performance Completed"; ② If there is a slight deviation (such as a delay of 1 hour but reported in advance and without loss), the performance is still deemed qualified, but the slight delay is noted in the contract remarks and has been reported; ③ If there is a serious performance anomaly (such as a shortage of transportation volume or serious quality damage), the performance is deemed partially qualified, the performance anomaly item is marked in the contract (such as a shortage of 1 vehicle, pending verification), and the performance anomaly follow-up process is triggered (notifying the contract administrator to connect with the driver for verification). At the same time, the system will verify the total reimbursement amount against the upper limit / unit price of the expenses agreed in the contract (e.g., if the contract stipulates a unit price of 10 yuan per kilometer and the total expenses do not exceed 10,000 yuan, if the total reimbursement amount is 10,500 yuan, it is necessary to confirm whether the expenses corresponding to the adjustment coefficient meet the special circumstances expense adjustment clause in the contract). After verification, the actual settlement expense is recorded in the contract as XX yuan, forming a closed loop of contract-task-expense association.

[0189] The contract performance report generation and archiving system automatically generates transportation contract performance reports. The report content includes: contract number, information of both parties, key performance data (completion time, transportation volume, integrity rate), cost settlement details (reimbursement amount, prepayment amount, final payment amount), performance status (qualified / partially performed / abnormal), and abnormal handling status (if any). After the report is generated, it automatically links to supporting materials such as reimbursement forms, payment records, and task tracking, and archives them in the company's contract archive for subsequent auditing and querying. At the same time, it sends a notification to the contract administrator that the contract performance has been completed and the report has been archived.

[0190] The TMS system automatically collects multi-dimensional reconciliation data from various core modules, forming a reconciliation data source set: ① Expense data (expense form number, task number, total expense amount, expense details, audit record); ② Payment data (prepayment transaction number, prepayment amount, expense reimbursement transaction number, expense amount, total payment amount); ③ Contract data (contract number, contract agreed fee, actual settlement fee, performance status); ④ Task data (task completion time, transportation mileage, comprehensive adjustment coefficient, used to verify the basis of expense calculation). After data collection, the system performs uniqueness verification (ensuring that the same task / expense form corresponds to only one data entry, with no duplication or omission) and format consistency (e.g., the unit of amount is uniformly RMB, and the time format is uniformly year-month-day hour:minute:second) to avoid reconciliation errors caused by differences in data format.

[0191] The automatic reconciliation and discrepancy marking system performs multi-dimensional automatic reconciliation based on the association between task number, expense report number, and contract number:

[0192] First dimension: Verify the total reimbursement amount against the contractually agreed expenses to confirm whether the actual expenses are within the scope of the contract (e.g., the contractually agreed expenses should not fluctuate by more than 10%; if the reimbursement amount exceeds this, mark the expenses as exceeding the agreed-upon difference).

[0193] The second dimension: The total payment amount (prepayment + reimbursement) is reconciled with the total reimbursement amount to confirm whether the payment amount is completely consistent with the reimbursement amount (e.g., prepayment of 3,000 yuan + reimbursement of 8,340 yuan = 11,340 yuan, which must be completely matched with the total reimbursement amount of 11,340 yuan; otherwise, the payment amount difference is marked).

[0194] The third dimension: verify the cost details with the task data. For example, fuel subsidies need to match the actual transportation mileage × unit fuel cost standard, and adjustment costs need to match the overall comprehensive adjustment coefficient × basic cost. If they do not match, mark the difference in detailed calculation.

[0195] All discrepancies will generate a discrepancy explanation (e.g., for discrepancies exceeding the agreed expense limit: the contract stipulates a maximum of 10,000 yuan, but the actual reimbursement is 10,500 yuan, which is 5% over budget, due to a global comprehensive adjustment coefficient of 1.05), and will be associated with corresponding supporting materials (e.g., screenshots of contract terms and calculation records of the comprehensive adjustment coefficient).

[0196] If there are no discrepancies: The system automatically generates a formal reconciliation statement, which includes the reconciliation dimensions, verification results, associated data numbers (contract / reimbursement / payment numbers), and reconciliation time. After being automatically approved by the system, the statement is archived in the financial reconciliation archive and a notification is sent to the finance staff that the reconciliation is complete and there are no discrepancies.

[0197] If discrepancies are found: The system will push a list of discrepancies and explanations to the financial reconciliation specialist's management backend, and simultaneously send a notification that a reconciliation discrepancy exists and needs to be addressed. After verifying the cause of the discrepancy, if it is determined to be a reasonable discrepancy (such as cost overruns due to comprehensive adjustment coefficients, which comply with contract terms), the specialist will manually confirm that the discrepancy is negligible and generate a reconciliation statement; if it is determined to be an erroneous discrepancy (such as data entry errors), it will be returned to the corresponding module for correction (such as returning to the reimbursement module to correct expense details), and the reconciliation will be triggered again until there are no discrepancies.

[0198] The TMS system retrieves and verifies damage and violation data from two channels based on the task number: ① Damage data: Retrieves vehicle damage feedback forms submitted by the recipient from the cargo handover module (including photos of damaged parts, extent of damage, feedback time, and recipient's signature). If no feedback form is received, it is assumed there is no damage. ② Violation data: Through API interface with traffic management departments, it retrieves traffic violation records of the transport vehicle during the task execution period (from departure to task completion) (including violation time, location, violation type, fine amount, and processing status). If the interface does not return any data, it is assumed there are no violations. The system then verifies the retrieved abnormal data against the abnormal reporting records submitted by the driver's mobile terminal (such as temporary violations for minor scratches reported during transport) to confirm whether the abnormal data was reported in advance (unreported abnormalities require close follow-up).

[0199] Damage handling: ① If there is no damage, the system will update the damage handling status to "No damage," indicating the processing is complete; ② If there is damage, the status will be updated according to the damage handling agreement (the damage liability and claims rules stipulated in the contract): If the damage is caused by the driver's responsibility and the driver has already compensated (the compensation amount is deducted from the reimbursement amount, and the deduction record needs to be verified with the finance module), the status will be updated to "Damage has been compensated," indicating the processing is complete; if no compensation has been paid, the status will be updated to "Damage pending compensation," and a notification will be sent to the driver that the damage compensation has not been completed and that the driver should process it within 3 working days. At the same time, the driver's subsequent task allocation permissions will be frozen until the compensation is completed.

[0200] Traffic violation processing: ① If there is no violation, the status will be updated to "No violation," indicating the processing is complete; ② If there is a violation, the violation processing status will be checked: If the driver has submitted proof of violation processing (screenshot of fine payment, demerit point record), and the system has verified the processing through the traffic department interface, the status will be updated to "Violation processed," indicating completion; if not processed, the status will be updated to "Violation pending processing," and the driver will be notified to process it, and a violation processing deadline will be associated (e.g., processing must be completed within 15 days after the violation notification). If the violation is not processed within the deadline, it will be synchronized to the driver's credit file, affecting subsequent task allocation.

[0201] The anomaly handling report generation and archiving system automatically generates damage and violation handling reports. The report includes: task number, vehicle information, damage / violation details (time, location, type), handling status, handling result (compensation amount / fine amount), and responsible person (driver / third party). After the report is generated, it automatically links the damage photos, violation vouchers, and compensation / fine payment records, archives them in the anomaly handling archive, and sends a notification to the logistics safety management specialist that the task anomaly handling has been completed and the report has been archived, ensuring that the entire process of anomalies is traceable.

[0202] After completing the reimbursement, payment, contract update, reconciliation update, and quality / violation update, the TMS system performs a full-process closed-loop verification: checking whether the status of all four modules is complete (e.g., payment status successful, contract status fulfilled, reconciliation status no discrepancies, quality / violation status processed). If all modules are complete, the task is marked as completed, and a full-process report of the vehicle logistics task is generated (including all key data and statuses from task order issuance, in-transit monitoring, cost settlement to exception handling), and archived in the task archive for subsequent management review and data analysis. If a module is not completed (e.g., quality / violation pending claims), the process is marked as not closed, and a process breakpoint reminder is sent to the corresponding responsible person until all modules are completed, ultimately achieving seamless management of the entire vehicle logistics process from task initiation to completion.

[0203] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent management of vehicle logistics based on a TMS system, characterized in that, The method includes: Step S1: Generate a dispatch order according to the plan and assign the dispatch order to the designated driver's mobile terminal so that the driver can execute the task after receiving the dispatch order and update the task status in real time; Step S2: Based on the continuous location data collected from the reported locations along the route, perform spatial geometric modeling on the location data to construct a spatial reference plane for the transportation route; based on the distribution characteristics of the location data, divide the reference plane into several analysis grids and map each location data to the corresponding analysis grid; for each set of location data within the analysis grid, calculate the volume of the smallest enclosing sphere formed by its spatial distribution range and calculate the spatial centroid coordinates of the set to generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process; Step S3: The prepayment process is automatically triggered based on the task status, and the payment is completed by linking a third-party payment platform. After the task is completed, the driver submits a reimbursement application through a mobile terminal, and a reimbursement detail calculated in conjunction with the comprehensive adjustment parameters is automatically generated and pushed to the driver for confirmation. Step S4: After the reimbursement details are confirmed by the driver, the regional manager and finance department conduct a multi-level review process based on the comprehensive adjustment parameters. Step S5: After approval, the reimbursement and payment process is automatically triggered to complete the settlement of expenses; contract management, reconciliation management, and handling of quality damage and violations are updated simultaneously to achieve closed-loop management of the entire process.

2. The intelligent vehicle logistics management method based on a TMS system according to claim 1, characterized in that, Step S2, based on the continuous location data collected from the reported locations along the route, performs spatial geometric modeling on the location data to construct a spatial reference plane for the transportation route, including: Acquire a continuous sequence of location data reported by the driver's mobile terminal, wherein the location data includes latitude and longitude coordinates and a timestamp; The latitude and longitude coordinates in the location data sequence are transformed into a two-dimensional plane coordinate system through map projection to obtain a set of plane coordinate points; Based on the set of planar coordinate points, a spatial reference plane for the transportation route is generated using a planar fitting algorithm. This spatial reference plane is used to represent the overall spatial orientation of the transportation route.

3. The intelligent vehicle logistics management method based on a TMS system according to claim 2, characterized in that, Step S22, transforming the latitude and longitude coordinates in the location data sequence to a two-dimensional plane coordinate system through map projection to obtain a set of plane coordinate points, including: Based on the distribution range of latitude and longitude coordinates of the continuous location data sequence, identify the projection zone range covering all latitude and longitude coordinates from the continuous location data sequence, and determine the central meridian parameter of the projection zone range. Using the central meridian parameter, the latitude and longitude coordinates of each location data point are converted into the first and second coordinate values ​​in a Cartesian coordinate system in meters. All the transformed first and second coordinate values ​​are combined to generate corresponding planar coordinate points for each original location data point, thus forming a set of planar coordinate points.

4. The intelligent vehicle logistics management method based on a TMS system according to claim 3, characterized in that, Based on the aforementioned set of planar coordinate points, a spatial reference plane for the transportation route is generated using a planar fitting algorithm. This spatial reference plane represents the overall spatial orientation of the transportation route, including: The set of planar coordinate points is subjected to outlier detection and removal to obtain an optimized set of planar coordinate points; Based on the optimized planar coordinate point set, the arithmetic mean coordinates of all points are calculated to obtain the geometric center point of the point set. Subtract the coordinates of the geometric center point from the coordinates of each point in the optimized planar coordinate point set to obtain a centered coordinate point set with the geometric center point as the origin. Based on the centralized coordinate point set, calculate the covariance matrix of all its points in the three coordinate axes, where the covariance matrix represents the spatial distribution characteristics of the point set. The covariance matrix is ​​decomposed into eigenvalues ​​to obtain the eigenvalues ​​and corresponding eigenvectors of the covariance matrix. The eigenvector corresponding to the smallest eigenvalue is selected as the initial normal vector direction of the final fitting plane; The plane constant term is calculated using the coordinates of the geometric center point and the initial normal vector to form the final fitted plane parameters; Based on the parameters of the final fitted plane, a spatial reference plane is constructed.

5. The intelligent vehicle logistics management method based on a TMS system according to claim 4, characterized in that, Based on the distribution characteristics of the location data, the reference plane is divided into several analysis grids, and each location data point is mapped to its corresponding analysis grid, including: Based on the spatial distribution range of the optimized planar coordinate point set, calculate the length and width of the bounding rectangle of the spatial distribution range; based on the length and width of the bounding rectangle, initialize a uniform grid division scheme covering the entire spatial distribution range to obtain the initial grid cell size; on the spatial reference plane, perform the first grid division based on the initial grid cell size to generate multiple initial grid cells covering the entire spatial distribution range. The number of location data points falling into each initial grid cell is counted to obtain the point density of each initial grid cell, and then filtered to obtain densely distributed and sparsely distributed regions. For the initial grid cells in densely distributed regions, recursively subdivide them; for the initial grid cells in sparsely distributed regions, merge them to form several analysis grids. For each location data point, based on its planar coordinates, traverse all analysis grids to determine its target analysis grid; record the unique identifier of the location data point into the data structure of its target analysis grid to complete the mapping process.

6. The intelligent vehicle logistics management method based on a TMS system according to claim 5, characterized in that, For each set of location data within an analysis grid, by calculating the volume of the smallest enclosing sphere formed by its spatial distribution range and the spatial centroid coordinates of this set, comprehensive adjustment parameters are generated for evaluating the dynamic characteristics of the transportation process, including: Based on the corresponding analysis grid, obtain all position data points mapped to the analysis grid to form the position data set of the analysis grid; based on the three-dimensional spatial coordinates of all points in the position data set, calculate the arithmetic mean of all points in the three coordinate axes, and use the obtained average coordinate point as the spatial centroid coordinates of the position data set; using the spatial centroid coordinates as the initial sphere center, calculate the spatial distance from each point in the position data set to the initial sphere center, and find the maximum value among all distances, and use the maximum value as the initial sphere radius; Based on the initial center and radius of the sphere, an initial bounding sphere is constructed that encloses all data points in the location dataset. The initial bounding sphere is iteratively optimized by gradually adjusting the center position and reducing the radius until the surface of the sphere contacts at least three non-coplanar points in the location dataset, and the sphere contains all data points, thus obtaining the minimum bounding sphere for the location dataset. The volume of the minimum bounding sphere is calculated, and the spatial centroid coordinates are combined with the volume of the minimum bounding sphere. The combined volume is then weighted and fused using preset weighting coefficients to generate comprehensive adjustment parameters.

7. The intelligent vehicle logistics management method based on a TMS system according to claim 6, characterized in that, Step S3 includes: When the TMS system detects that the driver has completed the vehicle loading confirmation operation and updated the task status through the mobile terminal, it automatically triggers the prepayment process. Based on the preset prepayment rules, calculate the prepayment amount for this task and generate a prepayment instruction containing the prepayment amount, task number, and driver account information; The prepayment instruction is sent to the associated third-party payment platform, which then completes the prepayment to the driver's designated account. When the TMS system detects that the driver has completed the arrival confirmation operation through the mobile terminal and updated the task status to "completed", it will automatically push a task completion notification and reimbursement application portal to the driver's mobile terminal. In response to the reimbursement application submitted by the driver via mobile terminal, retrieve the basic reimbursement items and corresponding standard fees for this task; Obtain the comprehensive adjustment parameters of all analysis grids generated in this task, calculate their arithmetic mean, and obtain the global comprehensive adjustment coefficient for this task; The standard cost of basic reimbursement items is weighted and calculated with the overall comprehensive adjustment coefficient, and dynamic cost items are corrected to generate a preliminary reimbursement detail that includes detailed items, calculation process and final amount. The preliminary reimbursement details will be pushed to the driver's mobile device for the driver to check and confirm.

8. The intelligent vehicle logistics management method based on a TMS system according to claim 7, characterized in that, Step S4: After the driver confirms the expense details, the regional manager and finance department conduct a multi-level review process optimized based on the aforementioned comprehensive adjustment parameters, including: Once the driver's mobile terminal returns a confirmation instruction for the preliminary expense report details, the TMS system locks the expense report details and automatically assigns it to the corresponding regional manager's review node based on the region to which the task belongs. At the regional manager's review stage, show the regional manager the transportation route trajectory for this mission, the distribution map of each analysis grid, and the comprehensive adjustment parameters corresponding to each analysis grid; The regional manager enters their review comments on the review interface. If the review fails, the reimbursement application is returned to the driver with an explanation of the reasons. If the review is approved, the reimbursement details and the regional manager's electronic signature information are transferred to the next level of financial review. At the financial review stage, the financial staff are shown the expense details approved by the regional manager, the overall adjustment coefficient, and the calculation basis of the coefficient. The financial staff enters the final review comments on the review interface. If the review fails, the reason is noted and the expense is returned to the previous stage or the applicant. If the review is approved, the electronic signature is completed and the expense report status is updated to approved.

9. The intelligent vehicle logistics management method based on a TMS system according to claim 8, characterized in that, The mission status includes loading confirmation, on-the-way location reporting, and arrival confirmation.

10. A vehicle logistics intelligent management system based on a TMS system, wherein the system implements the method as described in any one of claims 1 to 9, characterized in that, include: The update module is used to generate dispatch orders according to the plan and distribute the dispatch orders to the designated driver's mobile terminal so that the driver can execute the task after receiving the dispatch order and update the task status in real time. The evaluation module is used to perform spatial geometric modeling on the location data collected from the reported locations along the route, and construct a spatial reference plane for the transportation route; based on the distribution characteristics of the location data, the reference plane is divided into several analysis grids, and each location data is mapped to the corresponding analysis grid; for the location data set within each analysis grid, the volume of the smallest enclosing sphere formed by its spatial distribution range is calculated, and the spatial centroid coordinates of the set are calculated, to generate comprehensive adjustment parameters for evaluating the dynamic characteristics of the transportation process; The push module is used to automatically trigger the prepayment process based on the task status and link with a third-party payment platform to complete the payment. After the task is completed, the driver can submit a reimbursement application via mobile terminal, and the reimbursement details calculated in combination with the comprehensive adjustment parameters will be automatically generated and pushed to the driver for confirmation. The processing module is used to optimize the multi-level review process of the regional manager and finance department after the reimbursement details are confirmed by the driver, in conjunction with the comprehensive adjustment parameters. Once approved, the reimbursement and payment process is automatically triggered to complete the settlement of expenses; contract management, reconciliation management, and handling of quality damage and violations are updated simultaneously to achieve closed-loop management of the entire process.

Images