Route planning method and device based on basic data, equipment and storage medium

Abstract

The invention relates to the technical field of route planning, in particular to a route planning method and device based on basic data, equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing and feature extraction processing of obtained historical basic data, and obtaining historical preprocessing data; constructing a first basic model based on an improved genetic algorithm and constructing a second basic model based on a multi-target genetic algorithm by adopting historical preprocessing data chalk training, and respectively obtaining a route planning model and a route optimization model; performing preprocessing and feature extraction processing on the obtained real-time basic data to obtain real-time key data; inputting real-time key data into the route planning model, and inputting a result output by the route planning model into a route optimization model to obtain multiple pieces of optional route information; sorting the plurality of pieces of selectable route information according to the obtained pre-selected target to obtain sorting information; according to the method disclosed by the invention, the accuracy and reliability of the generated selectable route can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of routing planning, and in particular to a routing planning method, device, equipment and storage medium based on basic data. Background Art

[0002] In the modern logistics industry, efficient distribution center operations play a vital role in the smooth operation of the entire logistics network. With the continuous growth of logistics business volume and the increasing diversification of market demand, how to carry out distribution planning scientifically and accurately has become a key challenge facing logistics companies.

[0003] For logistics distribution centers, basic data include business volume acquisition methods, special personnel operation conditions, originating outlets and destination distribution information, etc., which are crucial for the planning of distribution centers; however, traditional distribution planning often relies on experience and judgment, lacks in-depth analysis and utilization of basic data, that is, there is currently no effective method to introduce basic data into distribution planning for integration and analysis; further, existing route planning methods are usually relatively simple and do not fully consider the impact of various practical factors; for example, factors such as traffic conditions, transportation costs, and transportation time in different regions are not fully considered, resulting in insufficient optimization of route planning, reduced logistics efficiency and increased logistics costs; further, the planning of vehicle lines also has the problem of a single planning method and reliance on experience and judgment, that is, the current determination of vehicle lines often lacks scientific basis and cannot be reasonably planned based on actual business needs and basic data; unreasonable vehicle line planning may lead to waste of transportation resources, increased transportation costs, and reduced logistics efficiency.

[0004] It can be seen that the existing technology still needs to be improved and enhanced. Summary of the invention

[0005] In order to overcome the deficiencies of the prior art, an object of the present invention is to provide a route planning method based on basic data, which can effectively improve the accuracy and reliability of the generated optional routes.

[0006] The first aspect of the present invention provides a route planning method based on basic data, including: obtaining historical basic data, preprocessing and feature extraction processing on the historical basic data, obtaining first historical key data, second historical key data and historical constraint information; constructing a first basic model based on an improved genetic algorithm, training the first basic model with historical key data to obtain a route planning model, inputting the second historical key data into the route planning model to obtain route data; constructing a second basic model based on a multi-objective genetic algorithm, training the second basic model with route data and historical constraint information to obtain a route optimization model; obtaining real-time basic data, preprocessing and feature extraction processing on the real-time basic data to obtain real-time key data; inputting the real-time key data into the route planning model to obtain multiple optional routes, and inputting the multiple optional routes into the route optimization model to obtain multiple optional route information; obtaining a pre-selected target, sorting the multiple optional route information according to the pre-selected target to obtain sorting information, and using a visualization tool to convert the sorting information into a route planning table.

[0007] Optionally, in a first implementation method of the first aspect of the present invention, the historical basic data is obtained, and preprocessing and feature extraction are performed on the historical basic data to obtain first historical key data, second historical key data and historical constraint information, including: obtaining historical basic data, performing data cleaning on the historical basic data to obtain cleaned historical basic data; performing data type conversion and standardization on the cleaned historical basic data to obtain standard historical basic data; obtaining a preset analysis target, and based on the preset analysis target, extracting feature data related to the analysis target from the standard historical basic data in combination with a time series analysis method and a text mining method; dividing the feature data into first historical key data and second historical key data according to a preset first division ratio, and extracting historical constraint information from the feature data.

[0008] Optionally, in a second implementation method of the first aspect of the present invention, the first basic model is constructed based on an improved genetic algorithm, the first basic model is trained using historical key data to obtain a route planning model, and the second historical key data is input into the route planning model to obtain route data, including: constructing the first basic model based on a genetic algorithm improved by a simulated annealing algorithm, selecting a global minimum responsible selection method as an initialization method for the first basic model, and selecting a three-exchange heuristic crossover operator as a crossover operator for the first basic model; based on a preset second division ratio, dividing the historical key data into a first training set, a first validation set, and a first test set; inputting the first training set and the first validation set into the first basic model for training to obtain a first model to be tested; inputting the first test set into the first model to be tested for testing, evaluating the performance of the first model to be tested during the test process based on a cross-validation method, and optimizing the parameters of the first model to be tested based on a Bayesian optimization method according to the evaluation results to obtain a route planning model; inputting the second historical key data into the route planning model to obtain route data.

[0009] Optionally, in a third implementation manner of the first aspect of the present invention, the second basic model is constructed based on a multi-objective genetic algorithm, and the second basic model is trained using route data and historical constraint information to obtain a route optimization model, including: constructing a second basic model based on a multi-objective genetic algorithm, setting constraints of the second basic model according to historical constraint information, and taking the lowest cost as the objective function of the second basic model; randomly combining route data and historical constraint information to obtain combined data; dividing the combined data into a second training set, a second validation set, and a second test set based on a preset second division ratio; inputting the second training set and the second validation set into the first basic model for training to obtain a second model to be tested; inputting the second test set into the second model to be tested for testing, evaluating the performance of the second model to be tested during the test process based on a cross-validation method, and optimizing the parameters of the second model to be tested based on a Bayesian optimization method according to the evaluation results to obtain a route optimization model.

[0010] Optionally, in a fourth implementation method of the first aspect of the present invention, the real-time basic data is obtained, and preprocessing and feature extraction are performed on the real-time basic data to obtain real-time key data, including: obtaining real-time basic data, performing data cleaning on the real-time basic data to obtain cleaned real-time basic data; performing data type conversion and standardization on the cleaned real-time basic data to obtain standard real-time basic data; based on preset analysis objectives, combining time series analysis methods and text mining methods, performing feature extraction on the standard real-time basic data to obtain real-time key data.

[0011] Optionally, in a fifth implementation of the first aspect of the present invention, the real-time key data is input into the route planning model to obtain multiple optional routes, and the multiple optional routes are input into the route optimization model to obtain multiple optional route information, including: inputting real-time key data into the route planning model to obtain multiple optional routes, the multiple optional routes are multiple optional routes connecting a departure point and a destination; obtaining known route information pre-stored in a route database, the known route information including multiple known one-way routes, and determining one by one whether the optional route is consistent with any known one-way route, and if inconsistent, storing the inconsistent optional routes in the route database; inputting multiple optional routes into the route optimization model to obtain multiple optional route information, the multiple optional route information including multiple optional routes and cost information corresponding to the optional routes.

[0012] Optionally, in a sixth implementation of the first aspect of the present invention, the obtaining of a preselected target, sorting multiple optional route information according to the preselected target to obtain sorting information, and using a visualization tool to convert the sorting information into a route planning table, includes: obtaining a preselected target, wherein the preselected target includes the lowest cost and the shortest path; when the preselected target is the lowest cost, using a bubble sort method to sort the multiple optional routes included in the multiple optional route information from low to high according to cost, to obtain sorting information; when the preselected target is the shortest path, using a bubble sort method to sort the multiple optional routes included in the multiple optional route information from short to long according to path, to obtain sorting information; using a BI tool to draw a route planning table based on the sorting information.

[0013] The second aspect of the present invention provides a route planning device based on basic data, including: a first processing module, used to obtain historical basic data, preprocess and feature extract the historical basic data, and obtain first historical key data, second historical key data and historical constraint information; a first construction module, used to construct a first basic model based on an improved genetic algorithm, use historical key data to train the first basic model, obtain a route planning model, input the second historical key data into the route planning model, and obtain route data; a second construction module, used to construct a second basic model based on a multi-objective genetic algorithm, use route data and historical constraint information to train the second basic model, and obtain a route optimization model; a second processing module, used to obtain real-time basic data, preprocess and feature extract the real-time basic data, and obtain real-time key data; a planning module, used to input the real-time key data into the route planning model, obtain multiple optional routes, and input the multiple optional routes into the route optimization model to obtain multiple optional route information; a sorting module, used to obtain a pre-selected target, sort the multiple optional route information according to the pre-selected target, obtain sorting information, and use a visualization tool to convert the sorting information into a route planning table.

[0014] Optionally, in a first implementation method of the second aspect of the present invention, the first processing module includes: a first cleaning unit, used to acquire historical basic data, perform data cleaning processing on the historical basic data, and obtain cleaned historical basic data; a first processing unit, used to perform data type conversion processing and standardization processing on the cleaned historical basic data, and obtain standard historical basic data; a first extraction unit, used to acquire a preset analysis target, and based on the preset analysis target, extract feature data related to the analysis target from the standard historical basic data in combination with a time series analysis method and a text mining method; a first division unit, used to divide the feature data into first historical key data and second historical key data according to a preset first division ratio, and extract historical constraint information from the feature data.

[0015] Optionally, in a second implementation of the second aspect of the present invention, the first construction module includes: a first construction unit, used to construct a first basic model based on a genetic algorithm improved by a simulated annealing algorithm, select a global minimum responsible selection method as the initialization method of the first basic model, and select a three-exchange heuristic crossover operator as the crossover operator of the first basic model; a second division unit, used to divide the historical key data into a first training set, a first verification set and a first test set based on a preset second division ratio; a first training unit, used to input the first training set and the first verification set into the first basic model for training to obtain a first model to be tested; a first testing unit, used to input the first test set into the first model to be tested for testing, evaluate the performance of the first model to be tested during the test process based on a cross-validation method, and optimize the parameters of the first model to be tested based on a Bayesian optimization method according to the evaluation results to obtain a route planning model; a first planning unit, used to input the second historical key data into the route planning model to obtain route data.

[0016] Optionally, in a third implementation of the second aspect of the present invention, the second construction module also includes: a second construction unit, used to construct a second basic model based on a multi-objective genetic algorithm, set the constraints of the second basic model according to historical constraint information, and take the minimum cost as the objective function of the second basic model; a combination unit, used to randomly combine route data and historical constraint information to obtain combined data; a third division unit, used to divide the combined data into a second training set, a second verification set, and a second test set based on a preset second division ratio; a second training unit, used to input the second training set and the second verification set into the first basic model for training to obtain a second model to be tested; a second testing unit, used to input the second test set into the second model to be tested for testing, evaluate the performance of the second model to be tested during the test process based on the cross-validation method, and optimize the parameters of the second model to be tested based on the Bayesian optimization method according to the evaluation results to obtain a route optimization model.

[0017] Optionally, in a fourth implementation of the second aspect of the present invention, the second processing module includes: a second cleaning unit, used to acquire real-time basic data, perform data cleaning on the real-time basic data, and obtain cleaned real-time basic data; a second processing unit, used to perform data type conversion and standardization on the cleaned real-time basic data, and obtain standard real-time basic data; a second extraction unit, used to perform feature extraction on the standard real-time basic data based on preset analysis objectives, combined with time series analysis methods and text mining methods, and obtain real-time key data.

[0018] Optionally, in a fifth implementation of the second aspect of the present invention, the planning module includes: a second planning unit, used to input real-time key data into the route planning model to obtain multiple optional routes, wherein the multiple optional routes are multiple optional routes connecting a departure point and a destination; a judgment unit, used to obtain known route information pre-stored in a route database, wherein the known route information includes multiple known one-way routes, and judge one by one whether the optional route is consistent with any known one-way route. If not, the inconsistent optional routes are stored in the route database; an optimization unit, used to input multiple optional routes into the route optimization model to obtain multiple optional route information, wherein the multiple optional route information includes multiple optional routes and cost information corresponding to the optional routes.

[0019] Optionally, in a sixth implementation of the second aspect of the present invention, the sorting module includes: an acquisition unit, used to acquire a pre-selected target, wherein the pre-selected target includes the lowest cost and the shortest path; a first sorting unit, used to, when the pre-selected target is the lowest cost, use a bubble sort method to sort the multiple optional routes included in the multiple optional route information from low to high according to cost, to obtain sorting information; a second sorting unit, used to, when the pre-selected target is the shortest path, use a bubble sort method to sort the multiple optional routes included in the multiple optional route information from short to long according to path, to obtain sorting information; a visual unit, used to use a BI tool to draw a route planning table based on the sorting information.

[0020] The third aspect of the present invention provides a routing planning device based on basic data, and the routing planning device based on basic data includes: a memory and at least one processor, wherein instructions are stored in the memory; at least one of the processors calls the instructions in the memory so that the routing planning device based on basic data executes each step of the routing planning method based on basic data described above.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the various steps of any of the above-mentioned basic data-based routing planning methods are implemented.

[0022] In the technical solution of the present invention, basic data is introduced when performing route planning, the impact of basic data on route planning is fully considered, and route planning is performed by integrating route planning model and route optimization model, which can effectively improve the accuracy and reliability of the generated optional routes and provide reliable data support for route planning of the distribution center. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A first flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0024] Figure 2 A second flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0025] Figure 3 A third flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0026] Figure 4 A fourth flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0027] Figure 5 A fifth flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0028] Figure 6 A sixth flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0029] Figure 7 A seventh flow chart of a routing planning method based on basic data provided by an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of a structure of a route planning device based on basic data provided by an embodiment of the present invention;

[0031] Fig. 9 A schematic diagram of the structure of a route planning device based on basic data provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention provides a method, apparatus, device and storage medium for routing planning based on basic data. In the present invention, the terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , an embodiment of the routing planning method based on basic data in an embodiment of the present invention includes:

[0034] 101. Obtain historical basic data, perform preprocessing and feature extraction on the historical basic data, and obtain first historical key data, second historical key data, and historical constraint information;

[0035] In this embodiment, by preprocessing and feature extraction of historical basic data, a key data set can be quickly formed, providing a high-quality data foundation for subsequent model training.

[0036] 102. Constructing a first basic model based on an improved genetic algorithm, using historical key data to train the first basic model to obtain a route planning model, and inputting second historical key data into the route planning model to obtain route data;

[0037] 103. Construct a second basic model based on a multi-objective genetic algorithm, and use route data and historical constraint information to train the second basic model to obtain a route optimization model;

[0038] In this embodiment, the route planning model constructed based on the improved genetic algorithm and the route optimization model constructed based on the multi-objective genetic algorithm are combined to effectively improve the accuracy and effectiveness of the output optional routes.

[0039] 104. Obtain real-time basic data, perform pre-processing and feature extraction on the real-time basic data, and obtain real-time key data;

[0040] 105. Input the real-time key data into the route planning model to obtain multiple optional routes, and input the multiple optional routes into the route optimization model to obtain multiple optional route information;

[0041] 106. Obtain a pre-selected target, sort the multiple optional route information according to the pre-selected target, obtain sorting information, and use a visualization tool to convert the sorting information into a route planning table;

[0042] In this embodiment, a visualization tool is used to convert the sorting information into a route planning table, so as to realize a visual and intuitive display of the data, facilitate decision makers to understand the route planning table, and make the decision-making process more efficient.

[0043] The present application discloses a route planning method based on basic data. Basic data is introduced when performing route planning, the influence of basic data on route planning is fully considered, and route planning is performed by integrating route planning model and route optimization model. This can effectively improve the accuracy and reliability of the generated optional routes, and provide reliable data support for route planning of distribution centers.

[0044] See also Figure 2 , a second embodiment of the routing planning method based on basic data in an embodiment of the present invention includes:

[0045] 201. Obtain historical basic data, perform data cleaning on the historical basic data, and obtain cleaned historical basic data;

[0046] In this embodiment, the historical basic data is cleaned by removing missing values, abnormal values ​​and duplicate records from the historical basic data.

[0047] 202. Perform data type conversion and standardization processing on the cleaned historical basic data to obtain standard historical basic data;

[0048] In this embodiment, the data type conversion processing of the cleaned historical basic data is achieved by converting non-numeric data into numeric data, and the standardization processing of the cleaned historical basic data is completed by converting the historical basic data after data type conversion into a unified format, thereby laying a solid foundation for subsequent data analysis and modeling.

[0049] 203. Obtaining a preset analysis target, and based on the preset analysis target, extracting feature data related to the analysis target from standard historical basic data in combination with a time series analysis method and a text mining method;

[0050] In this embodiment, a preset analysis target is first obtained, and the analysis target may be the length of the analysis path, the time consumption of the analysis path, or the cost of the analysis path; then, a time series analysis method is applied to identify patterns and trends in standard historical basic data, such as using moving average, exponential smoothing, or ARIMA model; at the same time, text mining techniques are used, such as keyword extraction, topic modeling, or sentiment analysis, to extract valuable information from the annotated historical basic data; finally, the feature data obtained by time series analysis and text mining are combined to obtain feature data.

[0051] 204. Divide the feature data into first historical key data and second historical key data according to a preset first division ratio, and extract historical constraint information from the feature data;

[0052] In this embodiment, the preset first division ratio can be 8:2; when extracting historical constraint information, the historical constraint conditions can be confirmed according to the preset analysis objectives, and the constraint data related to the historical constraint conditions can be extracted from the standard historical basic data, and the historical constraint conditions and constraint data can be integrated to obtain the historical constraint information.

[0053] See also Figure 3 , a third embodiment of the routing planning method based on basic data in an embodiment of the present invention includes:

[0054] 301. Construct a first basic model based on a genetic algorithm improved by a simulated annealing algorithm, select a global minimum responsible selection method as an initialization method of the first basic model, and select a three-exchange heuristic crossover operator as a crossover operator of the first basic model;

[0055] In this embodiment, the global minimum responsible selection method is used to initialize the population, that is, a set of initial solutions is selected from the possible solution space, and these solutions are as close to the global optimal solution as possible, which helps the algorithm converge to a high-quality solution faster; a three-exchange inspired crossover operator is selected, and the three-exchange inspired crossover operator generates offspring by performing three exchange operations in the parent individuals, which helps to maintain the diversity of the population and inherit the excellent characteristics of the parent; when the population is iterated, the fitness of each individual in the population is evaluated to determine their degree of adaptation to the problem, and individuals are selected for reproduction according to the fitness by roulette selection to ensure that individuals with high fitness have a higher chance of being selected; then, the individuals are mutated with a certain probability to introduce new genetic information to prevent the algorithm from converging to the local optimal solution too early; further, a simulated annealing strategy is introduced in the iterative process of the genetic algorithm, and the randomness and meticulousness of the search process are controlled by setting a "temperature" parameter and gradually reducing this temperature in each generation; in the early stage of the algorithm, a higher temperature allows more random searches, which helps to jump out of the local optimum; as the temperature decreases, the algorithm gradually focuses on the area of ​​high-quality solutions currently found.

[0056] 302. Based on a preset second division ratio, divide the historical key data into a first training set, a first verification set, and a first test set;

[0057] In this embodiment, the second division ratio is: 60% of the data is used for training, 20% for verification, and 20% for testing; in order to ensure the representativeness of the data set, the historical key data can be randomly shuffled to avoid the impact of data order on model training.

[0058] 303. Input the first training set and the first validation set into the first basic model for training to obtain a first model to be tested;

[0059] 304. Input the first test set into the first model to be tested for testing, evaluate the performance of the first model to be tested in the testing process based on a cross-validation method, and optimize the parameters of the first model to be tested based on a Bayesian optimization method according to the evaluation result to obtain a route planning model;

[0060] In this embodiment, the use of the cross-validation method can more accurately evaluate the performance of the model on unknown data, reduce the risk of overfitting, and improve the generalization ability of the model; the Bayesian optimization method takes into account the uncertainty of parameter adjustment in the parameter optimization process, and can more efficiently find the optimal parameter combination and improve model performance; by inputting the first test set into the first model to be tested for actual testing, the performance of the first model to be tested in actual applications can be intuitively observed, thereby ensuring the practicability of the route planning model.

[0061] 305. Input the second historical key data into the route planning model to obtain route data.

[0062] See also Figure 4 , a fourth embodiment of the routing planning method based on basic data in an embodiment of the present invention includes:

[0063] 401. Construct a second basic model based on a multi-objective genetic algorithm, set constraint conditions of the second basic model according to historical constraint information, and use the minimum cost as the objective function of the second basic model;

[0064] In this embodiment, the minimum cost is taken as the main objective function, and other objectives are considered at the same time, such as time efficiency, cargo capacity, vehicle type and the number corresponding to each type of vehicle; reasonable constraints are set according to historical constraint information, and the constraints may include resource limitations, time windows, cargo requirements, etc.; when iterating, a group of possible solutions are randomly generated as the initial population, and each individual (solution) in the population is evaluated, and its fitness is calculated according to the objective function and the constraints. The better individuals are selected for reproduction according to the fitness, and the selected individuals are cross-over (pairing and gene exchange) to produce new offspring, and the offspring are mutated to increase the diversity of the population, and some or all individuals in the original population are replaced with offspring generated after selection, crossover and mutation operations, and it is checked whether the termination conditions are met, such as reaching a preset number of iterations or fitness convergence, etc. When the termination conditions are met, the current optimal solution is output.

[0065] In this embodiment, the MOGA multi-objective optimization algorithm is used to construct a path optimization model, which can consider multiple objectives at the same time to find the optimal solution that meets multiple conditions; the genetic algorithm can efficiently search the solution space by simulating natural selection and genetic mechanisms, and can easily adapt to changes in the problem. It can be applied to various optimization problems by adjusting parameters and operations.

[0066] 402. Randomly combine route data and historical constraint information to obtain combined data;

[0067] In this embodiment, a combination algorithm is used to randomly combine route data and historical constraint information to obtain combined data.

[0068] 403. Based on a preset second division ratio, divide the combined data into a second training set, a second verification set, and a second test set;

[0069] 404. Input the second training set and the second validation set into the first basic model for training to obtain a second model to be tested;

[0070] 405. Input the second test set into the second model to be tested for testing, evaluate the performance of the second model to be tested during the testing process based on a cross-validation method, and optimize the parameters of the second model to be tested based on a Bayesian optimization method according to the evaluation result to obtain a route optimization model.

[0071] See also Figure 5 , a fifth embodiment of the routing planning method based on basic data in the embodiment of the present invention includes:

[0072] 501. Acquire real-time basic data, perform data cleaning on the real-time basic data, and obtain cleaned real-time basic data;

[0073] 502. Perform data type conversion processing and standardization processing on the cleaned real-time basic data to obtain standard real-time basic data;

[0074] 503. Based on the preset analysis objectives, combined with time series analysis methods and text mining methods, feature extraction and processing are performed on standard real-time basic data to obtain real-time key data.

[0075] See also Figure 6 , a sixth embodiment of the routing planning method based on basic data in the embodiment of the present invention includes:

[0076] 601. Inputting the real-time key data into the route planning model to obtain a plurality of optional routes, wherein the plurality of optional routes are a plurality of optional routes connecting a departure point and a destination;

[0077] 602. Acquire known route information pre-stored in a route database, wherein the known route information includes a plurality of known one-way routes, and determine whether the optional routes are consistent with any of the known one-way routes one by one. If not, store the inconsistent optional routes in the route database;

[0078] In this embodiment, the KMP algorithm is used to determine whether the optional route is consistent with any known one-way route. If inconsistent, the inconsistent optional route is stored in the route database; the route database can be continuously expanded to improve the comprehensiveness and accuracy of route information, and avoid repeated planning of known routes, saving computing resources and time. Further, the interference of erroneous route information can be reduced and the accuracy of route planning can be improved.

[0079] 603. Input the plurality of optional routes into the route optimization model to obtain a plurality of optional route information, wherein the plurality of optional route information includes the plurality of optional routes and cost information corresponding to the optional routes.

[0080] See also Figure 7 , a seventh embodiment of the routing planning method based on basic data in the embodiment of the present invention includes:

[0081] 701. Obtain a pre-selected target, where the pre-selected target includes a lowest cost and a shortest path;

[0082] 702. When the pre-selected target has the lowest cost, a bubble sort method is used to sort the multiple optional routes included in the multiple optional route information from low to high according to the cost to obtain sorting information;

[0083] In this embodiment, the bubble sort method is a simple and intuitive sorting algorithm with the advantages of high sorting efficiency and low time complexity. When the pre-selected target is the lowest cost, the bubble sort method can effectively sort multiple optional routes from low to high according to cost, thereby quickly obtaining sorting information.

[0084] 703. When the pre-selected target is the shortest path, a bubble sort method is used to sort the multiple optional routes included in the multiple optional route information from short to long paths to obtain sorting information;

[0085] 704. Use BI tools to draw a route planning table based on sorting information

[0086] In this embodiment, the sorting information is imported into the BI tool for data cleaning and integration to ensure the accuracy and completeness of the data; then the analysis function of the BI tool is used in combination with the sorting results to generate a route planning table, which includes multiple candidate routes and evaluates key indicators such as cost, time, and distance of each route; the generated route planning table is displayed in a visual way such as charts and maps to facilitate user understanding and decision-making, improve decision-making efficiency, and reduce unnecessary cost expenditures in actual transportation, such as reducing transportation costs and reducing time waste.

[0087] The above describes the routing planning method based on basic data in the embodiment of the present invention. The following describes the routing planning device based on basic data in the embodiment of the present invention. Figure 8 , an embodiment of a route planning device based on basic data in an embodiment of the present invention includes:

[0088] The first processing module 801 is used to obtain historical basic data, perform preprocessing and feature extraction on the historical basic data, and obtain first historical key data, second historical key data and historical constraint information; the first construction module 802 is used to construct a first basic model based on an improved genetic algorithm, use historical key data to train the first basic model, obtain a route planning model, input the second historical key data into the route planning model, and obtain route data; the second construction module 803 is used to construct a second basic model based on a multi-objective genetic algorithm, use route data and historical constraint information to train the second basic model, and obtain a route optimization model; the second processing module 804 is used to obtain real-time basic data, perform preprocessing and feature extraction on the real-time basic data, and obtain real-time key data; the planning module 805 is used to input the real-time key data into the route planning model to obtain multiple optional routes, and input the multiple optional routes into the route optimization model to obtain multiple optional route information; the sorting module 806 is used to obtain a pre-selected target, sort the multiple optional route information according to the pre-selected target, obtain sorting information, and use a visualization tool to convert the sorting information into a route planning table.

[0089] In this embodiment, the first processing module 801 includes: a first cleaning unit 8011, which is used to obtain historical basic data, perform data cleaning processing on the historical basic data, and obtain cleaned historical basic data; a first processing unit 8012, which is used to perform data type conversion processing and standardization processing on the cleaned historical basic data, and obtain standard historical basic data; a first extraction unit 8013, which is used to obtain a preset analysis target, and based on the preset analysis target, extract feature data related to the analysis target from the standard historical basic data in combination with a time series analysis method and a text mining method; a first division unit 8014, which is used to divide the feature data into first historical key data and second historical key data according to a preset first division ratio, and extract historical constraint information from the feature data.

[0090] In this embodiment, the first construction module 802 includes: a first construction unit 8021, which is used to construct a first basic model based on a genetic algorithm improved by a simulated annealing algorithm, select a global minimum responsible selection method as an initialization method of the first basic model, and select a three-exchange heuristic crossover operator as a crossover operator of the first basic model; a second division unit 8022, which is used to divide the historical key data into a first training set, a first validation set and a first test set based on a preset second division ratio; a first training unit 8023, which is used to input the first training set and the first validation set into the first basic model for training to obtain a first model to be tested; a first testing unit 8024, which is used to input the first test set into the first model to be tested for testing, evaluate the performance of the first model to be tested in the test process based on a cross-validation method, and optimize the parameters of the first model to be tested based on a Bayesian optimization method according to the evaluation result to obtain a route planning model; a first planning unit 8025, which is used to input the second historical key data into the route planning model to obtain route data.

[0091] In this embodiment, the second construction module 803 also includes: a second construction unit 8031, which is used to construct a second basic model based on a multi-objective genetic algorithm, set the constraint conditions of the second basic model according to the historical constraint information, and take the minimum cost as the objective function of the second basic model; a combination unit 8032, which is used to randomly combine route data and historical constraint information to obtain combined data; a third division unit 8033, which is used to divide the combined data into a second training set, a second validation set and a second test set based on a preset second division ratio; a second training unit 8034, which is used to input the second training set and the second validation set into the first basic model for training to obtain a second model to be tested; a second testing unit 8035, which is used to input the second test set into the second model to be tested for testing, evaluate the performance of the second model to be tested during the test based on the cross-validation method, and optimize the parameters of the second model to be tested based on the Bayesian optimization method according to the evaluation results to obtain a route optimization model.

[0092] In this embodiment, the second processing module 804 includes: a second cleaning unit 8041, which is used to obtain real-time basic data, perform data cleaning processing on the real-time basic data, and obtain cleaned real-time basic data; a second processing unit 8042, which is used to perform data type conversion processing and standardization processing on the cleaned real-time basic data to obtain standard real-time basic data; a second extraction unit 8043, which is used to perform feature extraction processing on the standard real-time basic data based on preset analysis objectives, combined with time series analysis methods and text mining methods, to obtain real-time key data.

[0093] In this embodiment, the planning module 805 includes: a second planning unit 8051, which is used to input real-time key data into the route planning model to obtain multiple optional routes, and the multiple optional routes are multiple optional routes connecting the departure point and the destination; a judgment unit 8052, which is used to obtain known route information pre-stored in the route database, and the known route information includes multiple known one-way routes, and judge one by one whether the optional route is consistent with any known one-way route. If inconsistent, the inconsistent optional routes are stored in the route database; an optimization unit 8053, which is used to input multiple optional routes into the route optimization model to obtain multiple optional route information, and the multiple optional route information includes multiple optional routes and cost information corresponding to the optional routes.

[0094] In this embodiment, the sorting module 806 includes: an acquisition unit 8061, which is used to acquire a pre-selected target, and the pre-selected target includes the lowest cost and the shortest path; a first sorting unit 8062, which is used to use a bubble sort method to sort the multiple optional routes included in the multiple optional route information from low to high according to cost when the pre-selected target is the lowest cost, so as to obtain sorting information; a second sorting unit 8063, which is used to use a bubble sort method to sort the multiple optional routes included in the multiple optional route information from short to long according to path when the pre-selected target is the shortest path, so as to obtain sorting information; a visual unit 8064, which is used to use a BI tool to draw a route planning table based on the sorting information.

[0095] Based on the same idea as the method in the above embodiment, the device provided by the present application can implement the method in the above embodiment.

[0096] above Figure 8 The routing planning device based on basic data in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The routing planning device based on basic data in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0097] Fig. 91 is a schematic diagram of the structure of a routing planning device based on basic data provided by an embodiment of the present invention. The routing planning device 900 based on basic data may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 910 (for example, one or more processors) and a memory 920, and one or more storage media 930 (for example, one or more mass storage devices) storing application programs 933 or data 932. Among them, the memory 920 and the storage medium 930 can be short-term storage or permanent storage. The program stored in the storage medium 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the routing planning device 900 based on basic data. Furthermore, the processor 910 can be configured to communicate with the storage medium 930, and execute a series of instruction operations in the storage medium 930 on the routing planning device 900 based on basic data to implement the steps of the routing planning method based on basic data provided in the above-mentioned method embodiments.

[0098] The routing planning device 900 based on basic data may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input and output interfaces 960, and / or one or more operating systems 931, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Fig. 9 The structure of the routing planning device based on basic data shown does not constitute a limitation on the routing planning device based on basic data, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0099] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of a routing planning method based on basic data.

[0100] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0102] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A routing planning method based on basic data, characterized in that: include: Acquire historical basic data, perform preprocessing and feature extraction on the historical basic data, and obtain first historical key data, second historical key data, and historical constraint information; Building a first basic model based on an improved genetic algorithm, using historical key data to train the first basic model to obtain a route planning model, and inputting second historical key data into the route planning model to obtain route data; Building a second basic model based on a multi-objective genetic algorithm, and training the second basic model using route data and historical constraint information to obtain a route optimization model; Obtain real-time basic data, perform pre-processing and feature extraction on the real-time basic data, and obtain real-time key data; Inputting the real-time key data into the route planning model to obtain a plurality of optional routes, and inputting the plurality of optional routes into the route optimization model to obtain a plurality of optional route information; A pre-selected target is obtained, and information of multiple optional routes is sorted according to the pre-selected target to obtain sorting information, and a visualization tool is used to convert the sorting information into a route planning table.

2. The method for routing planning based on basic data according to claim 1, characterized in that: The obtaining of historical basic data, preprocessing and feature extraction of the historical basic data, and obtaining first historical key data, second historical key data, and historical constraint information include: Obtain historical basic data, perform data cleaning on the historical basic data, and obtain cleaned historical basic data; The cleaned historical basic data are respectively subjected to data type conversion and standardization processing to obtain standard historical basic data; Obtain the preset analysis target, and based on the preset analysis target, extract the feature data related to the analysis target from the standard historical basic data by combining the time series analysis method and the text mining method; The feature data is divided into first historical key data and second historical key data according to a preset first division ratio, and historical constraint information is extracted from the feature data.

3. The routing planning method based on basic data according to claim 1, characterized in that: The method of constructing a first basic model based on an improved genetic algorithm, using historical key data to train the first basic model to obtain a route planning model, and inputting second historical key data into the route planning model to obtain route data includes: The first basic model is constructed based on the genetic algorithm improved by the simulated annealing algorithm, the global minimum responsible selection method is selected as the initialization method of the first basic model, and the three-exchange heuristic crossover operator is selected as the crossover operator of the first basic model; Based on a preset second division ratio, the historical key data is divided into a first training set, a first validation set, and a first test set; Inputting the first training set and the first validation set into the first basic model for training to obtain a first model to be tested; Inputting the first test set into the first model to be tested for testing, evaluating the performance of the first model to be tested during the testing process based on a cross-validation method, and optimizing the parameters of the first model to be tested based on a Bayesian optimization method according to the evaluation result to obtain a route planning model; The second historical key data is input into the route planning model to obtain route data.

4. The routing planning method based on basic data according to claim 1, characterized in that: The method of constructing a second basic model based on a multi-objective genetic algorithm and training the second basic model using route data and historical constraint information to obtain a route optimization model includes: A second basic model is constructed based on a multi-objective genetic algorithm, constraint conditions of the second basic model are set according to historical constraint information, and the minimum cost is used as the objective function of the second basic model; Randomly combining route data and historical constraint information to obtain combined data; Based on a preset second division ratio, dividing the combined data into a second training set, a second validation set, and a second test set; Inputting the second training set and the second validation set into the first basic model for training to obtain a second model to be tested; The second test set is input into the second model to be tested for testing. The performance of the second model to be tested during the test is evaluated based on the cross-validation method. According to the evaluation result, the parameters of the second model to be tested are optimized based on the Bayesian optimization method to obtain a route optimization model.

5. The method for routing planning based on basic data according to claim 2, characterized in that: The real-time basic data is obtained, and pre-processed and feature extracted to obtain real-time key data, including: Acquire real-time basic data, perform data cleaning on the real-time basic data, and obtain cleaned real-time basic data; The cleaned real-time basic data is respectively subjected to data type conversion and standardization processing to obtain standard real-time basic data; Based on the preset analysis objectives, combined with time series analysis methods and text mining methods, feature extraction and processing are performed on standard real-time basic data to obtain real-time key data.

6. The method for routing planning based on basic data according to claim 1, characterized in that: The real-time key data is input into the route planning model to obtain multiple optional routes, and the multiple optional routes are input into the route optimization model to obtain multiple optional route information, including: Inputting the real-time key data into the route planning model to obtain a plurality of optional routes, wherein the plurality of optional routes are a plurality of optional routes connecting a departure point and a destination; Acquire known route information pre-stored in a route database, the known route information including a plurality of known one-way routes, determine one by one whether the optional route is consistent with any of the known one-way routes, and if not, store the inconsistent optional route in the route database; A plurality of optional routes are input into the route optimization model to obtain a plurality of optional route information, wherein the plurality of optional route information includes a plurality of optional routes and cost information corresponding to the optional routes.

7. The method for routing planning based on basic data according to claim 1, characterized in that: The obtaining of the pre-selected target, sorting the plurality of optional route information according to the pre-selected target to obtain sorting information, and converting the sorting information into a route planning table using a visualization tool, includes: Acquire a pre-selected target, wherein the pre-selected target includes a minimum cost and a shortest path; When the pre-selected target is the lowest cost, a bubble sort method is used to sort the multiple optional routes included in the multiple optional route information from low to high according to the cost to obtain sorting information; When the pre-selected target is the shortest path, a bubble sort method is used to sort the multiple optional routes included in the multiple optional route information from short to long paths to obtain sorting information; Use BI tools to draw a route planning table based on the sorting information.

8. A route planning device based on basic data, characterized in that: include: A first processing module is used to obtain historical basic data, perform preprocessing and feature extraction on the historical basic data, and obtain first historical key data, second historical key data and historical constraint information; A first construction module is used to construct a first basic model based on an improved genetic algorithm, train the first basic model with historical key data to obtain a route planning model, and input second historical key data into the route planning model to obtain route data; A second construction module is used to construct a second basic model based on a multi-objective genetic algorithm, and train the second basic model using route data and historical constraint information to obtain a route optimization model; The second processing module is used to obtain real-time basic data, perform pre-processing and feature extraction on the real-time basic data, and obtain real-time key data; A planning module, used for inputting real-time key data into the route planning model to obtain multiple optional routes, and inputting the multiple optional routes into the route optimization model to obtain multiple optional route information; The sorting module is used to obtain a pre-selected target, sort multiple optional route information according to the pre-selected target, obtain sorting information, and use a visualization tool to convert the sorting information into a route planning table.

9. A route planning device based on basic data, characterized in that: The basic data-based route planning device comprises: a memory and at least one processor, wherein instructions are stored in the memory; At least one of the processors calls the instructions in the memory to enable the basic data-based routing planning device to perform each step of the basic data-based routing planning method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the routing planning method based on basic data as described in any one of claims 1 to 7 are implemented.

Images