A method and system for intelligent dispatching and configuring shared vehicle maintenance personnel

Abstract

This invention provides an intelligent scheduling and configuration method and system for shared vehicle maintenance personnel, relating to the field of shared vehicle technology. Based on a preset time period, it retrieves historical scheduling data of shared vehicles, constructs a scheduling task matching module and a scheduling path generation module, generates a vehicle maintenance scheduling model, obtains user vehicle demand applications and integrates them as batch target application information, generates a single set of scheduling tasks and multiple scheduling paths, wherein each single set of scheduling tasks corresponds one-to-one with the multiple scheduling paths, and achieves intelligent vehicle scheduling and maintenance based on the single set of scheduling tasks and multiple scheduling paths. This invention solves the technical problem that existing shared vehicle maintenance personnel scheduling and configuration methods cannot establish reasonable indicators based on the demand for shared vehicle resources in different times and spaces, leading to unreasonable allocation of maintenance personnel. It achieves multi-model coordination and optimization, performs optimal allocation of maintenance tasks, and achieves the technical effect of improving vehicle utilization and reducing the workload of maintenance personnel.

Description

Technical Field

[0001] This invention relates to the field of shared vehicle technology, specifically to a method and system for intelligent scheduling and configuration of shared vehicle maintenance personnel. Background Technology

[0002] Shared vehicles refer to the provision of bicycles, e-bikes, and cars by companies in locations such as campuses, subway stations, bus stops, residential areas, commercial areas, and public service areas. It's a time-sharing rental model and a new type of green and environmentally friendly sharing economy. Essentially, it's a new type of transportation rental business—shared vehicle rental—primarily relying on vehicles as its carrier. It effectively utilizes the decline in bicycle travel caused by rapid urban economic development, maximizing the utilization of public roads. Shared vehicles have increasingly attracted attention, and with the advent of the sharing economy, their rapid development has greatly improved the convenience of life. However, the allocation of shared vehicle resources still has some irrationalities, and the current methods for scheduling and configuring shared vehicle maintenance personnel have certain drawbacks. There is still room for improvement in the scheduling and configuration of shared vehicle maintenance personnel.

[0003] The existing methods for scheduling and configuring shared vehicle maintenance personnel cannot establish reasonable indicators based on the demand for shared vehicle resources in different times and spaces, resulting in unreasonable allocation of maintenance personnel. Summary of the Invention

[0004] This application provides a method and system for intelligent scheduling and configuration of shared vehicle maintenance personnel, which addresses the technical problem that existing methods for scheduling and configuring shared vehicle maintenance personnel cannot establish reasonable indicators based on the demand for shared vehicle resources in different times and spaces, resulting in unreasonable allocation of maintenance personnel.

[0005] In view of the above problems, this application provides a method and system for intelligent scheduling and configuration of shared vehicle maintenance personnel.

[0006] In a first aspect, embodiments of this application provide an intelligent scheduling and configuration method for shared vehicle maintenance personnel. The method includes: retrieving historical scheduling data of shared vehicles based on a preset time period to obtain historical scheduling information; using the historical scheduling information as training information to construct a scheduling task matching module and a scheduling path generation module; using the scheduling task matching module as a pre-module and the scheduling path generation module as a post-module to generate the vehicle maintenance scheduling model; obtaining user vehicle demand applications and integrating the vehicle demand applications as batch target application information; inputting the batch target application information into the model, and generating a single set of scheduling tasks and multiple scheduling paths through module linkage analysis, wherein the single set of scheduling tasks and the multiple scheduling paths correspond one-to-one; and realizing intelligent vehicle scheduling and maintenance based on the single set of scheduling tasks and the multiple scheduling paths.

[0007] Secondly, embodiments of this application provide an intelligent dispatch and configuration system for shared vehicle maintenance personnel. The system includes: a historical dispatch data retrieval module, used to retrieve historical dispatch data of shared vehicles based on a preset time period to obtain historical dispatch information; a historical dispatch information training module, used to construct a dispatch task matching module and a dispatch path generation module using the historical dispatch information as training information; and an maintenance dispatch model construction module, used to use the dispatch task matching module as a pre-module and the dispatch path generation module as a post-module. The system comprises the following modules: a vehicle operation and maintenance scheduling module, which generates the vehicle operation and maintenance scheduling model; a vehicle demand application integration module, which obtains user vehicle demand applications and integrates them as batch target application information; a linkage analysis module, which inputs the batch target application information into the model and generates a single set of scheduling tasks and multiple scheduling paths through module linkage analysis, wherein the single set of scheduling tasks and the multiple scheduling paths correspond one-to-one; and a vehicle dispatching and operation and maintenance implementation module, which implements intelligent vehicle dispatching and operation and maintenance based on the single set of scheduling tasks and the multiple scheduling paths.

[0008] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0009] This application provides an intelligent scheduling and configuration method for shared vehicle maintenance personnel, relating to the field of shared vehicle technology. Based on a preset time period, it retrieves historical scheduling data for shared vehicles, constructs a scheduling task matching module and a scheduling path generation module, generates a vehicle maintenance scheduling model, obtains and integrates user vehicle demand requests as batch target request information, and generates a single set of scheduling tasks and multiple scheduling paths through module linkage analysis. Each single set of scheduling tasks corresponds one-to-one with the multiple scheduling paths, achieving intelligent vehicle scheduling and maintenance based on these single sets of tasks and multiple scheduling paths. This solves the technical problem of existing shared vehicle maintenance personnel scheduling and configuration methods failing to establish reasonable indicators based on the demand for shared vehicle resources in different times and spaces, leading to unreasonable allocation of maintenance personnel. It achieves multi-model coordination and optimization, performing optimal allocation of maintenance tasks, thereby improving vehicle utilization and reducing the workload of maintenance personnel.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] Figure 1 This application provides a schematic flowchart of a method for intelligent scheduling and configuration of shared vehicle maintenance personnel.

[0012] Figure 2 This application provides a schematic diagram illustrating the process of obtaining historical scheduling information in an intelligent scheduling configuration method for shared vehicle maintenance personnel.

[0013] Figure 3 This application provides a schematic diagram of the process for determining batch target application information in an intelligent scheduling and configuration method for shared vehicle maintenance personnel.

[0014] Figure 4 This application provides a schematic diagram of the structure of a shared vehicle maintenance personnel intelligent dispatch and configuration system.

[0015] Figure labeling: Historical scheduling data retrieval module 10, historical scheduling information training module 20, operation and maintenance scheduling model construction module 30, vehicle demand application integration module 40, linkage analysis module 50, vehicle scheduling operation and maintenance implementation module 60. Detailed Implementation

[0016] This application provides an intelligent scheduling and configuration method for shared vehicle maintenance personnel, which addresses the technical problem that existing shared vehicle maintenance personnel scheduling and configuration methods cannot establish reasonable indicators based on the demand for shared vehicle resources in different times and spaces, resulting in unreasonable allocation of maintenance personnel.

[0017] Example 1

[0018] like Figure 1 As shown in the figure, this application provides a method for intelligent scheduling and configuration of shared vehicle maintenance personnel. This method is applied to a shared vehicle maintenance personnel intelligent scheduling and configuration system, and includes:

[0019] Step S100: Based on a preset time period, retrieve historical scheduling data of shared vehicles to obtain historical scheduling information;

[0020] Specifically, the intelligent scheduling and configuration method for shared vehicle maintenance personnel provided in this application is applied to an intelligent scheduling and configuration system for shared vehicle maintenance personnel. First, a preset time period is set based on the usage of shared vehicles, used to limit the time limit for data retrieval from shared vehicles. Based on the preset time period, multiple sets of scheduling data are retrieved, where each set of scheduling data includes multi-dimensional data types. Data identification and analysis are performed on the multiple sets of scheduling data to obtain missing data. Adjustability analysis is performed on the missing data. Based on the adjustability analysis results, the multiple sets of scheduling data are preprocessed. When the adjustability analysis result is positive, missing data is supplemented using the n-nearest neighbor mean method; when the adjustability analysis result is negative, the scheduling data group corresponding to a single missing data item is removed, and the data preprocessing result is obtained. Since the demand for shared vehicles varies at different times, multiple data division levels are determined based on holidays, weekdays, rest days, etc., to access each data item in the data preprocessing result. The data is then categorized into multiple levels according to the multiple data division levels to generate historical scheduling information. By retrieving historical scheduling data of shared vehicles, a preliminary understanding of the usage of shared vehicles was achieved, laying the foundation for the subsequent generation of a vehicle operation and maintenance scheduling model.

[0021] Step S200: Use the historical scheduling information as training information to construct a scheduling task matching module and a scheduling path generation module;

[0022] Specifically, historical scheduling information includes historical vehicle demand locations, demand quantities, historical vehicle backlog locations, and backlog quantities, along with corresponding historical maintenance scheduling routes and scheduling time data. The scheduling time data and corresponding historical vehicle information form the first set of data, while the historical maintenance scheduling routes and corresponding historical vehicle information form the second set. For the first set of data, the data is divided into a first training set and a first validation set in an 8:2 ratio. The training set is used to estimate the model, while the validation set is used to determine the network structure or as parameters to control the complexity of complex models. The training and validation sets are used in supervised learning. Supervised learning refers to the process of adjusting the parameters of a classifier using a set of samples with known categories to achieve the required performance; it is also called supervised training or teacher-led learning. In supervised learning, each instance consists of an input object and a desired output value. The specific algorithm analyzes the training data and generates an inference function to map new instances. For the first set of data, this means inferring scheduling time data based on historical vehicle demand locations, demand quantities, historical vehicle backlog locations, and backlog quantities, thereby constructing a scheduling task matching module. Using the same method, the second set of data was divided into a second training set and a second validation set, and a scheduling path generation module was constructed. By dividing the original data, the model with the best performance and highest accuracy was constructed.

[0023] Step S300: Using the scheduling task matching module as a front-end module and the scheduling path generation module as a back-end module, generate the vehicle operation and maintenance scheduling model.

[0024] Specifically, the vehicle operation and maintenance scheduling model is a BP neural network model. Structurally, it has an input layer, hidden layers, and an output layer, with one or more hidden layers. The computation process of the BP neural network consists of forward and backward computation. During forward propagation, the input pattern is processed layer by layer from the input layer through the hidden layers and then forward to the output layer. Each neuron in one layer only affects the state of the neuron in the next layer. If the desired output cannot be obtained in the output layer, backward propagation begins. The error signal returns along the original connection path, and the weights of each neuron are modified to minimize the error signal. The scheduling task matching module is used as a front-end module. After the shared vehicle data is input through the input layer, it is processed first in the scheduling task matching module, i.e., scheduling task matching is performed based on the shared vehicle data first. The scheduling path generation module is used as a back-end module. After the scheduling task matching is completed, the scheduling task matching result is used as input data and processed by the scheduling path generation module. The scheduling path is matched according to the scheduling task matching result, and the matching result is output as the model's output data.

[0025] Step S400: Obtain the user's vehicle demand application and integrate the vehicle demand applications as batch target application information;

[0026] Specifically, an application statistics time zone is set, a dispatch control area is determined, and the user's vehicle demand applications are statistically analyzed based on the application statistics time zone to obtain the application statistics results. Based on the application statistics results, batch target application information is determined. By integrating vehicle demand applications, duplicate data statistics are avoided, and reasonable resource allocation is achieved.

[0027] Step S500: Input the batch target application information into the model, and generate a single set of scheduling tasks and multiple scheduling paths through module linkage analysis, wherein the single set of scheduling tasks and the multiple scheduling paths correspond one-to-one;

[0028] Specifically, based on the historical scheduling information, a parking space matching module is constructed using the dbscan clustering algorithm. The batch target application information is input into the model, and information clustering is performed based on location identifiers to obtain clustering results. Based on the clustering results, a contour coefficient is calculated as the clustering evaluation result. A coefficient threshold is set, and it is determined whether the absolute value of the clustering evaluation result meets the coefficient threshold. If it does not meet the threshold, parking space clustering is performed again until the clustering result meets the contour coefficient threshold. Based on the clustering results, a single group of scheduling tasks is generated, and the single group of scheduling tasks is used as the module output result.

[0029] Based on the single group of scheduling tasks, one task is randomly selected for path planning to determine multiple feasible paths. Path optimization is performed based on the multiple feasible paths to obtain the shortest time path as the best scheduling path. Path optimization is performed on each of the N tasks in the single group of scheduling tasks to obtain N best scheduling paths. The N best scheduling paths are marked as tasks and used as the output results of the scheduling path generation module.

[0030] Step S600: Implement intelligent vehicle scheduling and maintenance based on the single group of scheduling tasks and the multiple scheduling paths.

[0031] Specifically, the system matches maintenance personnel with the single set of scheduling tasks and the multiple scheduling paths to obtain personnel matching results. Multiple task lists are then generated based on the single set of scheduling tasks and the multiple scheduling paths. Finally, vehicle scheduling and maintenance are implemented based on the personnel matching results and the task lists. This achieves intelligent scheduling configuration of shared vehicle maintenance personnel, improving the rationality and efficiency of personnel scheduling.

[0032] Furthermore, such as Figure 2 As shown, step S100 of this application further includes:

[0033] Step S110: Retrieve multiple sets of scheduling data based on the preset time period, wherein each set of scheduling data includes multi-dimensional data types;

[0034] Step S120: Perform data identification and analysis on the multiple sets of scheduling data to obtain missing data;

[0035] Step S130: Perform a adjustability analysis on the missing data, and preprocess the multiple sets of scheduling data based on the adjustability analysis results;

[0036] Step S140: When the adjustability analysis result is yes, missing data is supplemented based on the n nearest neighbor mean method; when the adjustability analysis result is no, the scheduling data group corresponding to the single missing data is removed, and the data preprocessing result is obtained.

[0037] Step S150: Set data partitioning levels, traverse the data preprocessing results to perform multi-level classification, and generate the historical scheduling information.

[0038] Specifically, the preset time period is the frequency of retrieving scheduling data in advance according to the actual situation. For example, for areas with high vehicle demand, data needs to be retrieved every day, while for areas with low vehicle demand, data only needs to be retrieved once every three days. Multiple sets of scheduling data are data obtained after retrieving scheduling data according to different preset time periods. Among them, a set of scheduling data includes multi-dimensional data types, including historical vehicle demand locations, demand information, historical vehicle accumulation and parking locations, accumulation information, along with corresponding historical operation and maintenance scheduling routes and scheduling time data.

[0039] Multi-dimensional data types are used as data identification indicators. For example, historical vehicle demand location (A), demand quantity information (B), historical vehicle accumulation and parking location (C), accumulation quantity information (D), historical operation and maintenance scheduling routes (E), and scheduling time data (F) are used. Data sets containing these indicators are marked in uppercase, and those without are marked in lowercase. This method is used to identify and analyze multiple sets of scheduling data. The labeled sets are then categorized based on the lowercase letter markings. For scheduling data marked with a lowercase 'a', it indicates missing historical vehicle demand location information. The missing information in each set is analyzed to determine if the data identification indicator can be inferred or supplemented based on other information. If not, it means the data set lacks key information and cannot be used; therefore, the data set is discarded. If it can, it means that values ​​for certain moments or consecutive time periods cannot be collected, but the value can be derived or supplemented based on the preceding and following relationships. Missing data is supplemented using the n-nearest neighbor mean method, which considers the "distance" between pairs of samples and selects the average or distance-weighted average of the closest observations as the filling value for the missing samples. The data after removing and supplementing missing data is used as the data preprocessing result.

[0040] By supplementing missing data and removing missing key information, complete data was obtained and redundant data was removed, achieving the effect of simplifying data and improving data accuracy.

[0041] Furthermore, such as Figure 3 As shown, step S400 of this application further includes:

[0042] Step S410: Set the application statistics time zone;

[0043] Step S420: Determine the dispatch control area, and statistically analyze the user's vehicle demand applications based on the application statistics time zone to obtain the application statistics results;

[0044] Step S430: Based on the application statistics, determine the batch target application information.

[0045] Specifically, the application statistics time zone is the time period for implementing applications based on application frequency. If the demand for vehicles in a region is high and the application frequency is frequent, the statistics period is short, and applications within the same statistics period are considered as a single batch. The statistical results are then integrated over a period, for example, by generating an application sequence using name-location-time, thus improving information organization. The dispatch control area is a region where centralized and unified control is implemented based on related elements. The standard for dividing the control area is based on the application statistics time zone. Vehicle demand applications within the dispatch control area are statistically analyzed, and the application statistics results are integrated. In other words, all applications within the same dispatch control area within the same statistics period are classified as a single batch of applications.

[0046] Furthermore, step S500 of this application also includes:

[0047] Step S510-1: Based on the historical scheduling information, construct a parking space matching module using the dbscan clustering algorithm;

[0048] Step S510-2: Input the batch target application information into the model, perform information clustering based on location identifiers, and obtain the clustering results;

[0049] Step S510-3: Calculate the silhouette coefficient based on the clustering results, as the clustering evaluation result;

[0050] Step S510-4: Set a coefficient threshold and determine whether the absolute value of the clustering evaluation result meets the coefficient threshold;

[0051] Step S510-5: If the condition is not met, perform parking spot clustering again until the clustering result meets the contour coefficient threshold.

[0052] Step S510-6: Generate a single group of scheduling tasks based on the clustering results, and output the single group of scheduling tasks as the module output results.

[0053] Specifically, the dbscan clustering algorithm is a density-based spatial clustering algorithm. This algorithm divides regions with sufficient density into clusters and discovers clusters of arbitrary shapes in a noisy spatial database. It defines a cluster as the largest set of density-connected points. To illustrate, the system randomly selects a point from a large number of sample points and draws a circle around it, specifying the radius of the circle and the minimum number of sample points it must contain. If enough sample points are within the specified radius, the center of the circle moves to that inner sample point, and the circle continues to expand to include other nearby sample points. This process continues until the number of sample points enclosed by the rolling circle falls below a pre-specified value. The initial point is called the core point, the point where the circle stops is called the boundary point, and the point that stops rolling is called the outlier. Based on the results of this clustering process, a single set of scheduled tasks is generated, and this single set of scheduled tasks is the module output.

[0054] Furthermore, step S500 of this application also includes:

[0055] Step S520-1: Based on the single group of scheduling tasks, randomly select one task for path planning and determine multiple feasible paths;

[0056] Step S520-2: Optimize the path based on the multiple feasible paths to obtain the shortest time path as the best scheduling path;

[0057] Step S520-3: Perform path optimization on each of the N tasks in the single group of scheduling tasks to obtain N optimal scheduling paths;

[0058] Step S520-4: Identify the N optimal scheduling paths as tasks, and use them as the output results of the scheduling path generation module.

[0059] Specifically, based on the single set of scheduling tasks, one task is randomly selected for path planning. By comparing historical data, multiple feasible paths are determined. One path is randomly selected from the multiple feasible paths for scheduling time limit evaluation to obtain the first path scheduling time limit. The randomly selected path is taken as the current optimal path. Another path is randomly selected from the multiple feasible paths to obtain the second path scheduling time limit. The scheduling time limit of the first path and the scheduling time limit of the second path are compared. The path with the shorter time limit is iterated as the current optimal path. The path optimization iteration is repeated until a predetermined number of iterations is reached to obtain the best scheduling path.

[0060] Furthermore, step S520-2 of this application includes:

[0061] Step S520-21: Based on the single group of scheduling tasks, randomly extract one task for path planning, and determine the multiple feasible paths by comparing historical data;

[0062] Step S520-22: Randomly extract one path from the multiple feasible paths for scheduling time limit evaluation, and obtain the scheduling time limit of the first path;

[0063] Step S520-23: Take the randomly extracted path as the current optimal path, and extract another path based on the multiple feasible paths to obtain the second path scheduling time limit;

[0064] Step S520-24: Compare the scheduling time limit of the first path with the scheduling time limit of the second path, and iterate the path with the shorter time limit as the current optimal path;

[0065] Step S520-25: Repeat the path optimization iteration until the predetermined number of iterations is reached to obtain the optimal scheduling path.

[0066] Specifically, a path planning process is performed on a randomly selected task from a single scheduling task. This involves connecting the starting and ending points, generating multiple paths in between. These paths are compared with historical data, and those similar to historical data are considered feasible paths. From these feasible paths, one path is randomly selected. Based on factors such as holidays, weekdays, and rest days, a multi-level data segmentation is used to determine date matching data as a reference. Real-time road traffic flow and traffic light conditions are analyzed to determine the travel time for this path, which serves as the first path's scheduling time limit. The same method is used to obtain the second path's scheduling time limit. The first and second path scheduling time limits are compared, and the path with the shorter time limit is iterated as the current optimal path. This iterative process is repeated until all path scheduling time limits have been iterated, resulting in the optimal scheduling path.

[0067] Furthermore, step S600 of this application also includes:

[0068] Step S610: Match maintenance personnel with the single group of scheduling tasks and the multiple scheduling paths, and obtain the personnel matching results;

[0069] Step S620: Generate multiple task lists based on the single group of scheduling tasks and the multiple scheduling paths;

[0070] Step S630: Implement vehicle scheduling and maintenance based on the personnel matching results and the task list.

[0071] Specifically, a scheduling task includes multiple scheduling paths and the demand for shared vehicles. Maintenance personnel are matched based on the number of scheduling paths, their directions, and the size of the demand for shared vehicles. If there are multiple scheduling paths and none of them are in the same direction, the number of maintenance personnel needs to be increased accordingly. For the demand for shared vehicles, the greater the demand, the more maintenance personnel are needed. This process is used to obtain the personnel matching results. A task list is constructed based on the matched maintenance personnel and their corresponding scheduling paths, scheduling times, and task loads. The task list is sent to the mobile terminals of the corresponding maintenance personnel, who can then carry out vehicle scheduling and maintenance according to the task list, such as battery swapping, scheduling, and placement.

[0072] Furthermore, step S600 of this application also includes:

[0073] Step S640: Based on the shared vehicle dispatch process, perform vehicle quality inspection synchronously and obtain the inspection results;

[0074] Step S650: Determine the vehicle demand, rank the dispatchable vehicles according to the quality inspection results, and determine the vehicles to be dispatched.

[0075] Step S660: Mark the remaining vehicles in the dispatchable vehicles with quality inspection results and location, and obtain the vehicle marking results;

[0076] Step S670: Perform factory repair based on the vehicle identification result.

[0077] Specifically, while implementing vehicle dispatch and maintenance according to the task list, maintenance personnel also conduct quality inspections on shared vehicles. The current quality of shared vehicles is graded based on their usage, wear and tear, and missing parts. For example, vehicles with only simple wear and no damaged parts are classified as Grade S; vehicles with minor parts damage but still usable are classified as Grade A; vehicles with many damaged parts requiring repair are classified as Grade B; and vehicles with severely damaged parts affecting usability are classified as Grade C. Depending on demand, when demand is high, Grades S, A, and B vehicles are prioritized for supply. When demand is low, Grades S and A vehicles are retained, while Grades B and C vehicles are repaired. Vehicles requiring repair are marked with their grade and location for return to the factory for repair.

[0078] Example 2

[0079] Based on the same inventive concept as the intelligent scheduling and configuration method for shared vehicle maintenance personnel in the foregoing embodiments, such as Figure 4 As shown, this application provides an intelligent dispatch and configuration system for shared vehicle maintenance personnel, the system comprising:

[0080] The historical scheduling data retrieval module 10 is used to retrieve historical scheduling data of shared vehicles based on a preset time period to obtain historical scheduling information.

[0081] Historical scheduling information training module 20, which is used to use the historical scheduling information as training information to construct a scheduling task matching module and a scheduling path generation module;

[0082] The operation and maintenance scheduling model construction module 30 is used to generate the vehicle operation and maintenance scheduling model by taking the scheduling task matching module as the front module and the scheduling path generation module as the back module.

[0083] Vehicle demand application integration module 40 is used to obtain users' vehicle demand applications and integrate the vehicle demand applications as batch target application information.

[0084] Linkage analysis module 50 is used to input the batch target application information into the model and generate a single set of scheduling tasks and multiple scheduling paths through module linkage analysis, wherein the single set of scheduling tasks corresponds one-to-one with the multiple scheduling tasks.

[0085] The vehicle dispatching and maintenance implementation module 60 is used to realize intelligent vehicle dispatching and maintenance based on the single group of dispatching tasks and the multiple dispatching paths.

[0086] Furthermore, the system also includes:

[0087] The multi-group vehicle dispatching and maintenance implementation module is a data retrieval module used to retrieve multiple groups of dispatching data based on the preset time period, wherein each group of dispatching data includes multi-dimensional data types.

[0088] The data identification and analysis module is used to perform data identification and analysis on the multiple sets of scheduling data to obtain missing data;

[0089] The adjustability analysis module is used to perform adjustability analysis on the missing data and preprocess the multiple sets of scheduling data based on the adjustability analysis results.

[0090] The data preprocessing result acquisition module is used to supplement missing data based on the n nearest neighbor mean method when the adjustability analysis result is yes; and to remove the scheduling data group corresponding to the single missing data when the adjustability analysis result is no, thereby obtaining the data preprocessing result.

[0091] The historical scheduling information generation module is used to set data division levels, traverse the data preprocessing results to perform multi-level classification, and generate the historical scheduling information.

[0092] Furthermore, the system also includes:

[0093] The application statistics time zone acquisition module is used to set the application statistics time zone;

[0094] The demand application statistics module is used to determine the dispatch control area, and to perform statistics on the user's vehicle demand applications based on the application statistics time zone to obtain the application statistics results;

[0095] The batch target application information determination module is used to determine the batch target application information based on the application statistics results.

[0096] Furthermore, the system also includes:

[0097] The parking space matching module construction module is used to construct the parking space matching module based on the historical scheduling information and the dbscan clustering algorithm.

[0098] The information clustering module is used to input the batch target application information into the model, perform information clustering based on location identifiers, and obtain clustering results;

[0099] The silhouette coefficient calculation module is used to calculate the silhouette coefficient based on the clustering results, as a clustering evaluation result.

[0100] The coefficient threshold acquisition module is used to set a coefficient threshold and determine whether the absolute value of the clustering evaluation result meets the coefficient threshold.

[0101] The parking spot clustering module is used to re-cluster the parking spots when the conditions are not met, until the clustering result meets the contour coefficient threshold.

[0102] A single-group scheduling task generation module is used to generate a single-group scheduling task based on the clustering results, and to output the single-group scheduling task as the module output result.

[0103] Furthermore, the system also includes:

[0104] A multiple feasible path determination module is used to randomly extract one of the single group of scheduling tasks for path planning and determine multiple feasible paths.

[0105] The shortest time-limited path acquisition module is used to perform path optimization based on the multiple feasible paths and obtain the shortest time-limited path as the best scheduling path.

[0106] The module for obtaining N optimal scheduling paths is used to perform path optimization for each of the N tasks in the single group of scheduling tasks and obtain N optimal scheduling paths.

[0107] The task identification module is used to identify the N optimal scheduling paths as tasks, and the results are output by the scheduling path generation module.

[0108] Furthermore, the system also includes:

[0109] A multiple feasible path acquisition module is used to randomly extract one path based on the single group of scheduling tasks for path planning, and determine the multiple feasible paths by comparing historical data.

[0110] The first path scheduling time limit acquisition module is used to randomly extract one path from the multiple feasible paths, evaluate the scheduling time limit, and obtain the first path scheduling time limit.

[0111] The second path scheduling time limit acquisition module is used to take the randomly extracted path as the current optimal path, and then extract a path again based on the multiple feasible paths to obtain the second path scheduling time limit.

[0112] The calibration module is used to calibrate the scheduling time limit of the first path and the scheduling time limit of the second path, and iterate the path with the shorter time limit as the current optimal path;

[0113] The path optimization iteration module is used to repeatedly perform path optimization iterations until a predetermined number of iterations is reached to obtain the optimal scheduling path.

[0114] Furthermore, the system also includes:

[0115] The maintenance personnel matching module is used to match maintenance personnel with the single group of scheduling tasks and the multiple scheduling paths, and obtain the personnel matching results.

[0116] A multiple task list generation module is used to generate multiple task lists based on the single set of scheduled tasks and the multiple scheduled paths;

[0117] The scheduling and maintenance implementation module is used to implement vehicle scheduling and maintenance based on the personnel matching results and the task list.

[0118] Furthermore, the system also includes:

[0119] The vehicle quality inspection module is used to perform vehicle quality inspection synchronously based on the shared vehicle scheduling process and obtain the inspection results.

[0120] The priority ranking module is used to determine the vehicle demand, and based on the quality inspection results, to rank the dispatchable vehicles and determine the vehicles to be dispatched.

[0121] The vehicle identification result acquisition module is used to identify the quality inspection results and location of the remaining vehicles among the schedulable vehicles, and to acquire the vehicle identification results.

[0122] The return-to-factory repair module is used to perform return-to-factory repairs based on the vehicle identification results.

[0123] Through the foregoing detailed description of a method for intelligent scheduling and configuration of shared vehicle maintenance personnel, those skilled in the art can clearly understand the method and system for intelligent scheduling and configuration of shared vehicle maintenance personnel in this embodiment. As for the apparatus disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the description in the method section.

[0124] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent scheduling and configuration of shared vehicle maintenance personnel, characterized in that, The method includes: Based on a preset time period, historical scheduling data of shared vehicles is retrieved to obtain historical scheduling information. The historical scheduling information is used as training information to construct a scheduling task matching module and a scheduling path generation module; The vehicle operation and maintenance scheduling model is generated by using the scheduling task matching module as a front-end module and the scheduling path generation module as a back-end module. Obtain users' vehicle demand requests and integrate these requests as batch target request information; The batch target application information is input into the model, and a single set of scheduling tasks and multiple scheduling paths are generated through module linkage analysis, wherein the single set of scheduling tasks and the multiple scheduling paths correspond one-to-one. Intelligent vehicle scheduling and maintenance is achieved based on the single group of scheduling tasks and the multiple scheduling paths. include: Based on the historical scheduling information, a parking space matching module is constructed using the DBSCAN clustering algorithm; The batch target application information is input into the model, and information clustering is performed based on location identifiers to obtain clustering results; Based on the clustering results, the silhouette coefficient is calculated as the clustering evaluation result; Set a coefficient threshold and determine whether the absolute value of the clustering evaluation result meets the coefficient threshold. If the condition is not met, the parking spot clustering is repeated until the clustering result meets the outline coefficient threshold. Based on the clustering results, a single set of scheduling tasks is generated, and the single set of scheduling tasks is used as the module output result. include: Based on the single group of scheduling tasks, one task is randomly selected for path planning to determine multiple feasible paths. Based on the multiple feasible paths, path optimization is performed to obtain the shortest time path as the best scheduling path. For each of the N tasks in the single group of scheduling tasks, path optimization is performed to obtain N optimal scheduling paths; The N optimal scheduling paths are assigned task identifiers, which are then output as the results of the scheduling path generation module.

2. The method as described in claim 1, characterized in that, The step of retrieving historical scheduling data of shared vehicles based on a preset time period to obtain historical scheduling information includes: Multiple sets of scheduling data are retrieved based on the preset time period, wherein each set of scheduling data includes multi-dimensional data types; Data identification and analysis are performed on the multiple sets of scheduling data to obtain missing data; The missing data is subjected to adjustability analysis, and the multiple sets of scheduling data are preprocessed based on the adjustability analysis results; When the adjustability analysis result is yes, missing data is supplemented based on the n nearest neighbor mean method; when the adjustability analysis result is no, the scheduling data group corresponding to the single missing data is removed, and the data preprocessing result is obtained. The data is divided into hierarchical levels, and the data preprocessing results are traversed to perform multi-level classification to generate the historical scheduling information.

3. The method as described in claim 1, characterized in that, include: Set the application statistics time zone; Determine the dispatch and control area, and statistically analyze the user's vehicle demand applications based on the application statistics time zone to obtain the application statistics results; Based on the application statistics, the batch target application information is determined.

4. The method as described in claim 1, characterized in that, The step of optimizing the path based on the multiple feasible paths to obtain the shortest time path as the optimal scheduling path includes: Based on the single group of scheduling tasks, one task is randomly selected for path planning, and multiple feasible paths are determined by comparing historical data. Based on the multiple feasible paths, one path is randomly selected for scheduling time limit evaluation to obtain the scheduling time limit of the first path. The randomly extracted path is taken as the current optimal path, and another path is extracted again based on the multiple feasible paths to obtain the second path scheduling time limit; The scheduling time limits of the first path and the second path are compared, and the path with the shorter time limit is iterated as the current optimal path. Repeat the path optimization iteration until the predetermined number of iterations is reached to obtain the optimal scheduling path.

5. The method as described in claim 1, characterized in that, The above includes: Based on the single group of scheduling tasks and the multiple scheduling paths, maintenance personnel are matched to obtain the personnel matching results; Multiple task lists are generated based on the single group of scheduling tasks and the multiple scheduling paths; Vehicle scheduling and maintenance are implemented based on the personnel matching results and the task list.

6. The method as described in claim 5, characterized in that, include: Based on the shared vehicle dispatch process, vehicle quality inspection is carried out simultaneously, and the inspection results are obtained. Determine the vehicle demand, rank the available vehicles based on the quality inspection results, and determine the vehicles to be dispatched. The remaining vehicles among the dispatchable vehicles are identified by quality inspection results and location, and the vehicle identification results are obtained. The vehicle will be returned to the factory for repair based on the identified vehicle information.

7. A shared vehicle maintenance personnel intelligent dispatch and configuration system, characterized in that, The system includes: The historical scheduling data retrieval module is used to retrieve historical scheduling data of shared vehicles based on a preset time period to obtain historical scheduling information. A historical scheduling information training module is used to construct a scheduling task matching module and a scheduling path generation module by using the historical scheduling information as training information. The operation and maintenance scheduling model construction module is used to generate the vehicle operation and maintenance scheduling model by taking the scheduling task matching module as the front module and the scheduling path generation module as the back module. The vehicle demand application integration module is used to obtain users' vehicle demand applications and integrate the vehicle demand applications as batch target application information. The linkage analysis module is used to input the batch target application information into the model and generate a single set of scheduling tasks and multiple scheduling paths through module linkage analysis, wherein the single set of scheduling tasks corresponds one-to-one with the multiple scheduling tasks. A vehicle dispatching and maintenance implementation module, which is used to realize intelligent vehicle dispatching and maintenance based on the single group of dispatching tasks and the multiple dispatching paths; The parking space matching module construction module is used to construct the parking space matching module based on the historical scheduling information and the dbscan clustering algorithm. The information clustering module is used to input the batch target application information into the model, perform information clustering based on location identifiers, and obtain clustering results; The silhouette coefficient calculation module is used to calculate the silhouette coefficient based on the clustering results, as a clustering evaluation result. The coefficient threshold acquisition module is used to set a coefficient threshold and determine whether the absolute value of the clustering evaluation result meets the coefficient threshold. The parking spot clustering module is used to re-cluster the parking spots when the conditions are not met, until the clustering result meets the contour coefficient threshold. A single-group scheduling task generation module is used to generate a single-group scheduling task based on the clustering results, and to output the single-group scheduling task as the module output result. A multiple feasible path determination module is used to randomly extract one of the single group of scheduling tasks for path planning and determine multiple feasible paths. The shortest time-limited path acquisition module is used to perform path optimization based on the multiple feasible paths and obtain the shortest time-limited path as the best scheduling path. The module for obtaining N optimal scheduling paths is used to perform path optimization for each of the N tasks in the single group of scheduling tasks and obtain N optimal scheduling paths. The task identification module is used to identify the N optimal scheduling paths as tasks, and the results are output by the scheduling path generation module.

Images