A Timely Delivery Method and System Based on Path Planning

Abstract

The present invention discloses a timely delivery method and system based on path planning, which relates to the field of logistics and supply chain management. The method includes: receiving order information and inputting it into the distribution management center; obtaining a standard path by using the simulated annealing algorithm based on the order information; connecting to the urban intelligent transportation center based on the standard path to collect real-time traffic data, and using a traffic flow prediction model to analyze the real-time traffic data to obtain traffic condition changes; adjusting the standard path by using a multi-objective evolutionary algorithm through the traffic condition changes to obtain an optimal path; continuously tracking the delivery progress of the optimal path; and based on tracking the delivery progress of the optimal path, dynamically calling the multi-objective evolutionary algorithm to adjust the optimal path. By obtaining a standard path through the simulated annealing algorithm and dynamically adjusting the path using the multi-objective evolutionary algorithm according to real-time traffic condition changes, the present invention enhances the global search ability and robustness, and improves the delivery efficiency and resource utilization rate.

Description

Technical Field

[0001] The present invention relates to the field of logistics and supply chain management, and particularly to a timely delivery method and system based on path planning. Background Art

[0002] With the rapid development of the e-commerce and logistics industries, timely delivery has become a key factor in improving customer satisfaction and service quality. Path planning technology plays a crucial role in this process. Traditional path planning methods mainly rely on static map data and predefined routes. Although these methods can optimize the delivery path to a certain extent, they are inadequate in dealing with real-time traffic condition changes. In recent years, with the development of intelligent transportation systems (ITS), dynamic path planning methods based on real-time traffic data have gradually become a research hotspot. By combining real-time traffic information, historical data, and prediction models, these methods can more accurately predict traffic conditions and adjust the delivery path accordingly, thereby improving the delivery efficiency and on-time rate.

[0003] However, the existing dynamic path planning methods still have some deficiencies. First, many methods have low computational efficiency when dealing with large-scale orders and complex traffic conditions, resulting in long response times and unable to meet the requirements of immediate delivery. Second, the existing methods often lack support for multi-objective optimization, that is, considering multiple objectives such as minimizing the total travel time and total travel distance simultaneously. This makes the generated path may perform well in one aspect, but not ideal in other aspects, affecting the overall delivery effect. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a timely delivery method and system based on path planning, which solves the problems of low computational efficiency and insufficient multi-objective optimization.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a timely delivery method based on path planning, which includes,

[0008] Receiving order information and entering it into the distribution management center; based on the order information, obtaining a standard path using the simulated annealing algorithm; based on the standard path, connecting to the urban intelligent transportation center to collect real-time traffic data, analyzing the real-time traffic data using a traffic flow prediction model to obtain traffic condition changes; through the traffic condition changes, using a multi-objective evolutionary algorithm to adjust the standard path to obtain an optimal path; continuously tracking the delivery progress of the optimal path; based on tracking the delivery progress of the optimal path, calling the multi-objective evolutionary algorithm in real time to adjust the optimal path.

[0009] As a preferred solution of the real-time delivery method based on path planning according to the present invention, wherein: the order information is received and entered into the distribution management center, and the specific steps are as follows.

[0010] Receive customer order information, automatically verify the inventory situation and payment status, and determine that the order is valid.

[0011] For valid orders, the real-time order information is automatically entered into the distribution management center through an automated interface.

[0012] As a preferred solution of the real-time delivery method based on path planning according to the present invention, wherein: based on the order information, the standard path is obtained by using the simulated annealing algorithm, and the specific steps are as follows.

[0013] Extract vehicle information through the distribution management center.

[0014] Based on the order information and vehicle information, use random initialization to generate a random initial path combination.

[0015] Through the simulated annealing algorithm, use the inversion method to select two points in the initial path combination and exchange the path segments between them to generate a new solution, and decide whether to accept the new solution according to the Metropolis criterion. The expression is:

[0016]

[0017] Where P is the probability value, ΔF is the fitness difference between the new solution and the current solution, T is the current temperature, ΔF max is the maximum value of the fitness difference, T a is the actual driving time, T e is the estimated driving time, t is the iteration time step, λ is the fitness difference adjustment coefficient, μ is the time deviation adjustment coefficient, and k is the time decay coefficient.

[0018] Generate a random number r through random.random().

[0019] When P > r, it means accepting the new solution.

[0020] When P ≤ r, the new solution is not accepted and the current solution remains unchanged.

[0021] For the new solution that is not accepted, gradually reduce the temperature parameter through the exponential cooling method, repeat the inversion method and the Metropolis criterion, and generate an optimized path combination.

[0022] Define constraint conditions such as vehicle capacity limit, order delivery time, and single-trip driving distance limit through problem requirements and business rules.

[0023] Filter and adjust the optimized path combination through defined constraints to obtain the standard path.

[0024] As a preferred solution of the real-time delivery method based on path planning according to the present invention, wherein: based on the standard path, connect to the urban intelligent transportation center to collect real-time traffic data, analyze the real-time traffic data using a traffic flow prediction model, and obtain traffic condition changes. The specific steps are as follows.

[0025] Based on the standard path, the distribution management center connects to the urban intelligent transportation center through the API interface, sends a data request, obtains real-time traffic data, and stores the collected real-time traffic data in the database.

[0026] Construct a traffic flow prediction model with the LSTM model as the basic model.

[0027] Use time series to convert the real-time traffic data into serialized data.

[0028] Match the shape of the serialized data by setting the input_shape parameter of the input layer of the LSTM model.

[0029] Add an LSTM layer by using the Sequential API method, and add a fully connected layer after the LSTM layer.

[0030] Use a linear activation function in the output layer to predict the traffic flow index.

[0031] Train and validate the traffic flow prediction model, and test the final performance.

[0032] Input the preprocessed real-time data into the traffic flow prediction model to obtain the traffic flow index. The expression is:

[0033]

[0034] Wherein, is the traffic flow index, X is the preprocessed real-time traffic data, θ is the parameter of the LSTM model, α is the data deviation adjustment coefficient, X avg is the average value of the input data, σ X is the standard deviation of the input data, β is the time decay adjustment coefficient, t' is the time variable, t 0 is the reference time point, and τ is the time decay constant.

[0035] Divide the traffic conditions into different levels according to the traffic flow index.

[0036] Continuously update the traffic flow prediction model.

[0037] By predicting and classifying traffic conditions in real time, combining visual display and dynamic route adjustment, the traffic change situation is finally obtained.

[0038] As a preferred embodiment of the real-time delivery method based on path planning according to the present invention, wherein: by changing the traffic conditions, the multi-objective evolutionary algorithm is used to adjust the standard path to obtain the optimal path, and the specific steps are as follows.

[0039] Define the constraint conditions of path planning according to the real-time traffic condition changes.

[0040] The constraint conditions include dynamic time window constraint, dynamic driving time constraint, dynamic driving distance constraint, weather impact constraint and congestion section avoidance constraint.

[0041] Use the multi-objective evolutionary algorithm. The initial population consists of a set of randomly generated path combinations. Decompose the multi-objective problem into multiple single-objective sub-problems. Each sub-problem corresponds to a weight vector, and the weight vectors are evenly distributed on the unit hyperplane.

[0042] Assign neighbors to each sub-problem based on the Euclidean distance between weight vectors, set an initial reference point for each sub-problem, and set it to the maximum value of each objective.

[0043] For each sub-problem, randomly select individuals from the neighbors as parents, perform crossover operations and mutation operations on the parent individuals to generate new offspring individuals, update the reference point of each sub-problem, which is the minimum value of the objective function values of all individuals in the current population, and retain the better individuals by comparing the individuals in the neighbors with the newly generated offspring individuals.

[0044] When the stop condition is met, the algorithm terminates, and the final population will converge to a set of non-dominated solution sets.

[0045] Based on the constraint conditions, screen out the conforming path combinations from the non-dominated solution set, and use the lexicographic method to determine the optimal route from the conforming path combinations.

[0046] As a preferred embodiment of the real-time delivery method based on path planning according to the present invention, wherein: the specific steps for continuously tracking the delivery progress of the optimal path are as follows.

[0047] Continuously track the vehicle delivery progress through the in-vehicle GPS device and the cellular network, and continuously receive the latest traffic data from the urban intelligent transportation center.

[0048] As a preferred embodiment of the real-time delivery method based on path planning according to the present invention, wherein: based on tracking the delivery progress of the optimal path, the multi-objective evolutionary algorithm is called in real time to adjust the optimal path, and the specific steps are as follows.

[0049] Based on the vehicle delivery progress and the latest traffic data, if it is found that the actual travel time exceeds the expected time, serious traffic jams and unexpected events occur, the route replanning process will be automatically started;

[0050] Use the multi-objective evolutionary algorithm to adjust the existing route combinations to obtain the optimal route.

[0051] In a second aspect, the present invention provides a timely delivery system based on route planning, including an order management module, a route planning module, a real-time traffic analysis module, a dynamic route adjustment module, a delivery tracking and adjustment module, and a customer feedback analysis module; the order management module is used to receive order information and input it into the delivery management center; the route planning module is used to obtain a standard route based on the order information by using the simulated annealing algorithm; the real-time traffic analysis module is used to connect to the urban intelligent traffic center based on the standard route to collect real-time traffic data, and analyze the real-time traffic data by using the traffic flow prediction model to obtain changes in traffic conditions; the dynamic route adjustment module is used to adjust the standard route by using the multi-objective evolutionary algorithm based on the changes in traffic conditions to obtain the optimal route; the delivery tracking module is used to continuously track the delivery progress of the optimal route; the real-time adjustment module is used to call the multi-objective evolutionary algorithm in real time to adjust the optimal route based on tracking the delivery progress of the optimal route.

[0052] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the timely delivery method based on route planning as described in the first aspect of the present invention is implemented.

[0053] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the timely delivery method based on route planning as described in the first aspect of the present invention is implemented.

[0054] The beneficial effects of the present invention are as follows: By using the simulated annealing algorithm to obtain the standard route, the present invention realizes high-quality route planning, enhances the global search ability and robustness of the system, and ensures the generation of the optimal route under complex constraint conditions; at the same time, by using the multi-objective evolutionary algorithm to dynamically adjust the route according to the changes in real-time traffic conditions, the multi-objective optimization of the total travel time and the total travel distance is realized, and the delivery efficiency and resource utilization rate are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0056] Figure 1 This is a flow chart of the timely delivery method based on path planning in Example 1.

[0057] Figure 2 This is a system diagram of the timely delivery system based on path planning in Example 1. DETAILED DESCRIPTION

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0060] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0061] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a timely delivery method based on path planning, comprising the following steps:

[0062] S1. Receive order information and enter it into the distribution management center.

[0063] S1.1. Receive customer order information, automatically verify inventory status and payment status, and confirm that the order is valid.

[0064] It should be noted that when customers place orders through the platform, the platform receives customer order information including the customer’s detailed address, type of goods and weight.

[0065] The above information is obtained with the user's consent and is used for legal purposes.

[0066] S1.2. For valid orders, real-time order information is automatically entered into the distribution management center through an automated interface.

[0067] It should be noted that the automation interface (API) is developed and configured by the technical teams of the e-commerce platform and the distribution management center to achieve data exchange between the two.

[0068] S2. Obtain the standard path using the simulated annealing algorithm based on the order information.

[0069] S2.1. Extract vehicle information through the distribution management center.

[0070] It should be noted that the vehicle information is extracted through the internal management system of the distribution management center.

[0071] S2.2. Based on the order information and vehicle information, generate a random initial path combination using random initialization.

[0072] Specifically, the vehicle information includes vehicle location information, vehicle scheduling information, and vehicle capacity information.

[0073] S2.3. Through the simulated annealing algorithm, select two points in the initial path combination using the inversion method and exchange the path segments between them to generate a new solution. Decide whether to accept the new solution according to the Metropolis criterion. The expression is:

[0074]

[0075] where P is the probability value, ΔF is the fitness difference between the new solution and the current solution, T is the current temperature, ΔF max is the maximum value of the fitness difference, T a is the actual driving time, T e is the estimated driving time, t is the iteration time step, λ is the fitness difference adjustment coefficient, μ is the time deviation adjustment coefficient, and K is the time decay coefficient;

[0076] Generate a random number r through random.random();

[0077] When P > r, it means accepting the new solution;

[0078] When P ≤ r, do not accept the new solution and keep the current solution unchanged.

[0079] It should be noted that the fitness difference between the new solution and the current solution is calculated. The expression is:

[0080] ΔF = F new - F current ;

[0081] where ΔF is the fitness difference between the new solution and the current solution, F new is the fitness value of the newly generated solution, and F current is the fitness value of the current solution;

[0082] If ΔF < 0, that is, the new solution is better, then directly accept the new solution;

[0083] If ΔF > 0, that is, the new solution is worse, it is necessary to decide whether to accept the new solution according to P.

[0084] S2.4. Gradually reduce the temperature parameter of the new solution that is not accepted through the exponential cooling method, repeat the inversion method and the Metropolis criterion, and generate an optimized path combination.

[0085] It should be noted that the exponential cooling method is a commonly used temperature reduction strategy in the simulated annealing algorithm. By simulating the gradual reduction of temperature in the metal annealing process, the optimization algorithm can widely explore the solution space in the initial stage and perform fine optimization in the later stage, so as to balance global search and local search and increase the possibility of finding the global optimal solution.

[0086] S2.5. Define constraint conditions such as vehicle capacity limit, order delivery time, and single-trip driving distance limit through problem requirements and business rules.

[0087] It should be noted that the business rules include the maximum load capacity of the vehicle, the delivery time window of the order, and the maximum allowed driving distance of each vehicle.

[0088] S2.6. Filter and adjust the optimized path combination through the defined constraint conditions to obtain the standard path.

[0089] S3. Based on the standard path, connect to the urban intelligent transportation center to collect real-time traffic data, use the traffic flow prediction model to analyze the real-time traffic data, and obtain traffic condition changes.

[0090] S3.1. Based on the standard path, the distribution management center connects to the urban intelligent transportation center through the API interface, sends a data request, obtains real-time traffic data, and stores the collected real-time traffic data in the database.

[0091] Specifically, the real-time traffic data includes traffic flow, average vehicle speed, traffic incidents, weather conditions, and traffic lights.

[0092] S3.2. Construct a traffic flow prediction model with the LSTM model as the basic model.

[0093] Use time series to convert the real-time traffic data into serialized data;

[0094] Match the shape of the serialized data by setting the input_shape parameter of the input layer of the LSTM model;

[0095] Add an LSTM layer by using the Sequential API method, and add a fully connected layer after the LSTM layer;

[0096] Use a linear activation function in the output layer to predict the traffic flow index.

[0097] It should be noted that the traffic flow prediction model is constructed based on the LSTM model because LSTM can effectively capture the long-term dependencies and complex temporal patterns in time series data, thus predicting future traffic flow more accurately. Through its unique memory cells and gating mechanisms, LSTM can handle trends, seasonality, and periodic changes in traffic data and maintain good prediction performance even when there are long intervals in the data.

[0098] S3.3. Train and validate the traffic flow prediction model and test its final performance.

[0099] Specifically, use the train_test_split function to divide the historical traffic data into a training set, a test set, and a validation set.

[0100] Use the training set data to set appropriate batch sizes and the number of training epochs to train the traffic flow prediction model. During the training process, use the validation set to monitor the performance of the traffic flow prediction model, and use the trained traffic flow prediction model to evaluate the final performance of the traffic flow prediction model based on the test set data.

[0101] S3.4. Input the preprocessed real-time data into the traffic flow prediction model to obtain the traffic flow index, and the expression is:

[0102]

[0103] where is the traffic flow index, X is the preprocessed real-time traffic data, θ is the parameter of the LSTM model, α is the data deviation adjustment coefficient, X avg is the average value of the input data, σ X is the standard deviation of the input data, β is the time decay adjustment coefficient, t' is the time variable, t 0 is the reference time point, and τ is the time decay constant.

[0104] It should be noted that the calculation process of the expression is as follows:

[0105] Input the preprocessed real-time traffic data X into the LSTM model to obtain the basic prediction result f LSTM (X; θ), calculate the Euclidean distance between the input data X and the average value X avg , and divide it by the standard deviation σ X of the input data, take the sine function, and multiply it by the data deviation adjustment coefficient α. Then, calculate the difference between the time variable t' and the reference time point t 0Take the difference, divide it by the time decay constant τ, apply the hyperbolic tangent function, and multiply by the time decay adjustment coefficient β. Add the above two results, add 1, and then multiply by the basic prediction result f LSTM (X; θ) of the LSTM model to finally obtain the adjusted traffic flow prediction result

[0106] S3.5. Classify traffic conditions into different levels according to the traffic flow index.

[0107] The different levels are specifically unobstructed, slightly congested, moderately congested, and severely congested.

[0108] S3.6. Continuously update the traffic flow prediction model to maintain an accurate understanding of the current and upcoming traffic conditions.

[0109] S3.7. Through real-time prediction and classification of traffic conditions, combined with visual display and dynamic route adjustment, finally obtain the traffic change situation.

[0110] S4. Adjust the standard route using a multi-objective evolutionary algorithm through traffic condition changes to obtain the optimal route.

[0111] It should be noted that the multi-objective evolutionary algorithm originated from the field of evolutionary computing. By simulating natural selection and genetic mechanisms and combining the requirements of multi-objective optimization, it is used to solve optimization problems with multiple conflicting objectives. For example, in route planning, it may be necessary to optimize multiple objectives simultaneously, such as minimizing the total travel time, minimizing the total travel distance, minimizing fuel consumption, maximizing customer satisfaction, etc. These objectives often conflict with each other, that is, optimizing one objective may sacrifice the performance of another objective. The multi-objective evolutionary algorithm can effectively handle such multi-objective optimization problems and find a set of non-dominated solutions (Pareto optimal solutions), thereby finding the best solution among multiple objectives.

[0112] S4.1. Define the constraint conditions for route planning through real-time traffic condition changes;

[0113] The constraint conditions include dynamic time window constraint, dynamic travel time constraint, dynamic travel distance constraint, weather impact constraint, and congested section avoidance constraint.

[0114] It should be noted that the dynamic time window constraint: constrains the delivery time of each order; the dynamic travel time constraint: constrains the actual travel time of each section of the route; the dynamic travel distance constraint: constrains the single travel distance of each vehicle; the weather impact constraint: constrains the weather factors in route planning; the congested section avoidance constraint: constrains the traffic congestion situation in the route.

[0115] S4.2. Use a multi-objective evolutionary algorithm. The initial population consists of a set of randomly generated path combinations. Decompose the multi-objective problem into multiple single-objective sub-problems, with each sub-problem corresponding to a weight vector that is uniformly distributed on the unit hyperplane.

[0116] It should be noted that the multi-objective problem refers to minimizing the total travel time and minimizing the total travel distance.

[0117] S4.3. Assign neighbors to each sub-problem based on the Euclidean distance between weight vectors. Set an initial reference point for each sub-problem, which is set to the maximum value of each objective.

[0118] S4.4. For each sub-problem, randomly select individuals from the neighbors as parents. Perform crossover operations and mutation operations on the parent individuals to generate new offspring individuals. Update the reference point for each sub-problem, which is the minimum value of the objective function values of all individuals in the current population. Retain the better individuals by comparing the individuals in the neighbors with the newly generated offspring individuals.

[0119] It should be noted that for the crossover operation: use the partially mapped crossover (PMX); for the mutation operation: randomly swap the positions of two nodes in the path.

[0120] S4.5. When the stopping condition is met, the algorithm terminates, and the final population will converge to a set of non-dominated solutions.

[0121] It should be noted that the stopping conditions are the maximum number of iterations, fitness convergence, time limit, and solution set stability.

[0122] S4.6. Based on the constraint conditions, screen out the eligible path combinations from the set of non-dominated solutions, and use the lexicographic method to determine the optimal route from the eligible path combinations.

[0123] It should be noted that the lexicographic method determines the optimal route from the eligible path combinations by comparing the objective function values of the solutions in the multi-objective optimization problem one by one, usually sorted in the preset priority order.

[0124] S5. Continuously track the delivery progress of the optimal path.

[0125] S5.1. Continuously track the vehicle delivery progress through in-vehicle GPS devices and cellular networks, and continuously receive the latest traffic data from the urban intelligent transportation center.

[0126] It should be noted that continuously receive the latest traffic data from the urban intelligent transportation center through an automated interface (API).

[0127] S6. Based on tracking the delivery progress of the optimal path, real-time call the multi-objective evolutionary algorithm to adjust the optimal path.

[0128] S6.1. Based on the vehicle delivery progress and the latest traffic data, if it is found that the actual travel time exceeds the expectation, serious traffic jams or unexpected events occur, the path replanning process will be automatically started.

[0129] It should be noted that when the path replanning process is automatically started, the system will set a threshold. Once it detects that a certain indicator exceeds the threshold, the system will automatically call the path replanning module to recalculate the optimal path.

[0130] S6.2. Use the multi-objective evolutionary algorithm to adjust the existing path combination to obtain the optimal path.

[0131] When the vehicle arrives at the designated address, confirm the arrival through the mobile device. The customer signs on the mobile device to confirm the receipt of goods, and the mobile device will automatically pop up a customer satisfaction questionnaire to obtain customer feedback.

[0132] Based on the in-vehicle GPS device and the cellular network, continuously track the vehicle delivery progress to obtain delivery data.

[0133] Save the collected customer feedback and delivery data to the data center, regularly analyze the data, and continuously optimize the delivery algorithm and customer service process.

[0134] This embodiment also provides a just-in-time delivery system based on path planning, including: an order management module, a path planning module, a real-time traffic analysis module, a dynamic path adjustment module, a delivery tracking and adjustment module, and a customer feedback analysis module;

[0135] The order management module is used to receive order information and enter it into the delivery management center;

[0136] The path planning module is used to obtain the standard path based on the order information by using the simulated annealing algorithm;

[0137] The real-time traffic analysis module is used to connect to the urban intelligent traffic center based on the standard path to collect real-time traffic data, and analyze the real-time traffic data by using the traffic flow prediction model to obtain the changes in traffic conditions;

[0138] The dynamic path adjustment module is used to adjust the standard path by using the multi-objective evolutionary algorithm according to the changes in traffic conditions to obtain the optimal path;

[0139] The delivery tracking module is used to continuously track the delivery progress of the optimal path;

[0140] The real-time adjustment module is used to call the multi-objective evolutionary algorithm in real time to adjust the optimal path based on tracking the delivery progress of the optimal path.

[0141] This embodiment also provides a computer device applicable to the case of the on-time delivery method based on path planning, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the on-time delivery method based on path planning as proposed in the above embodiment.

[0142] The computer device may be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0143] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the on-time delivery method based on path planning as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM for short), Electrically Erasable Programmable Read-Only Memory (EEPROM for short), Erasable Programmable Read-Only Memory (EPROM for short), Programmable Read-Only Memory (PROM for short), Read-Only Memory (ROM for short), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0144] In summary, the present invention realizes high-quality path planning by using the simulated annealing algorithm, enhances the global search ability and robustness of the system, and ensures the generation of the optimal path under complex constraint conditions. At the same time, by using the multi-objective evolutionary algorithm to dynamically adjust the path according to the changes in real-time traffic conditions, the multi-objective optimization of the total travel time and the total travel distance is achieved, and the distribution efficiency and resource utilization rate are improved.

[0145] Example 2. Referring to Table 1, this is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the timely delivery method based on path planning are given.

[0146] To verify the effectiveness and advantages of the timely delivery method based on path planning, a specific example was designed. This example was carried out in a logistics distribution center in a certain city, and the peak period of a day (from 9:00 am to 11:00 am) was selected as the test time period. The test objects were 50 randomly generated orders, and each order included the detailed address, type and weight of the goods of the customer. During the test process, the simulated annealing algorithm, the multi-objective evolutionary algorithm and the traffic flow prediction model were used, and a comparison was made with the existing traditional path planning method.

[0147] First, the order information was input into the distribution management center through an automated interface, and the standard path was generated using the simulated annealing algorithm. Secondly, the urban intelligent transportation center was connected to collect real-time traffic data, and the LSTM model was used to predict the changes in traffic conditions. Then, based on the changes in traffic conditions, the multi-objective evolutionary algorithm was used to adjust the standard path to obtain the optimal path, and the distribution progress was continuously tracked and adjusted. Finally, the delivery was completed and the customer feedback and distribution data were recorded to optimize the future distribution algorithm and service process.

[0148] Specifically, as shown in Table 1 below:

[0149] Table 1 Comparative analysis table of distribution efficiency and customer satisfaction

[0150]

[0151] Through the data analysis of the above table, it can be clearly seen that the method of the present invention is superior to the traditional path planning method in multiple key indicators. Compared with the traditional method, the present invention combines the simulated annealing algorithm, the multi-objective evolutionary algorithm and real-time traffic data to achieve more efficient and accurate path planning. For example, in terms of the actual driving time, the ratio of the actual driving time to the estimated driving time of the method of the present invention is 1.08 hours, compared with 1.25 hours of the traditional method, reducing the additional driving time by 13.6%. This shows that the method of the present invention can more accurately predict and respond to traffic conditions, improving the on-time delivery rate. In addition, in terms of the actual driving distance, the ratio of the actual driving distance to the estimated driving distance of the method of the present invention is 1.05 kilometers, compared with 1.18 kilometers of the traditional method, reducing the additional driving distance by 11%, effectively reducing the fuel cost and maintenance cost. In terms of the number of path adjustments, the method of the present invention is only 1.2 times, compared with 3.5 times of the traditional method, reducing the number of path adjustments by 65.7%, improving the overall delivery efficiency. In terms of the total delivery cost, the ratio of the total delivery cost of the method of the present invention to the benchmark cost is 11,000 yuan, compared with 14,000 yuan of the traditional method, reducing the cost by 21.4%, significantly improving the economic benefits.

[0152] By combining the simulated annealing algorithm, the multi-objective evolutionary algorithm and real-time traffic data, the present invention significantly improves the on-time delivery rate of distribution, reduces the operating cost, reduces the number of path adjustments, and greatly improves the customer satisfaction, thus being superior to the traditional method in multiple key indicators such as actual driving time, driving distance, path planning efficiency and total delivery cost, providing a more efficient, reliable and customer-friendly solution for the logistics and distribution industry.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A timely delivery method based on path planning, characterized by: include, Receive order information and enter it into the distribution management center; Based on the order information, the simulated annealing algorithm is used to obtain the standard path; Based on standard routes, connect to the city's intelligent transportation center to collect real-time traffic data, use traffic flow prediction models to analyze real-time traffic data, and obtain changes in traffic conditions; The specific steps of obtaining the traffic condition change are as follows: Based on the standard route, the distribution management center connects to the city intelligent transportation center through the API interface, sends data requests, obtains real-time traffic data, and stores the collected real-time traffic data in the database; The traffic flow prediction model is constructed using the LSTM model as the basic model; Use time series to convert real-time traffic data into serialized data; By setting the input_shape parameter of the LSTM model input layer to match the shape of the serialized data; Add an LSTM layer by using the Sequential API method and add a fully connected layer after the LSTM layer. A linear activation function is used in the output layer to predict the traffic flow index; Train and validate the traffic flow prediction model and test the final performance; The preprocessed real-time data is input into the traffic flow prediction model to obtain the traffic flow index, which is expressed as: ； in, is the traffic flow index, is the preprocessed real-time traffic data. are the parameters of the LSTM model, is the data bias adjustment coefficient, is the mean value of the input data, is the standard deviation of the input data, is the time decay adjustment factor, is the time variable, is the reference time point, is the time decay constant; Divide traffic conditions into different levels according to the traffic flow index; Continuously update traffic flow prediction models; By predicting and classifying traffic conditions in real time, combined with visualization and dynamic path adjustment, we can finally obtain traffic change conditions. Based on traffic condition changes, we can use a multi-objective evolutionary algorithm to adjust the standard path and obtain the optimal path. Continuously track the optimal route delivery progress; Based on tracking the delivery progress of the optimal path, the multi-objective evolutionary algorithm is called in real time to adjust the optimal path.

2. The timely delivery method based on path planning as claimed in claim 1, characterized in that: The specific steps of receiving order information and entering it into the distribution management center are as follows: Receive customer order information, automatically verify inventory and payment status, and confirm that the order is valid; For valid orders, real-time order information is automatically entered into the distribution management center through an automated interface.

3. The timely delivery method based on path planning as claimed in claim 2, characterized in that: Based on the order information, the simulated annealing algorithm is used to obtain the standard path. The specific steps are as follows: Extract vehicle information through the distribution management center; Based on the order information and vehicle information, random initialization is used to generate a random initial route combination; Through the simulated annealing algorithm, the inversion method is used to select two points in the initial path combination and exchange the path segments between them to generate a new solution. The Metropolis criterion is used to decide whether to accept the new solution. The expression is: ； in, is the probability value, is the fitness difference between the new solution and the current solution, is the current temperature, is the maximum value of the fitness difference, is the actual driving time, is the estimated travel time, is the time step of the iteration, is the fitness difference adjustment coefficient, is the time deviation adjustment factor, is the time decay coefficient; Generate random numbers through random.random() ; when , indicating acceptance of the new solution; when , do not accept new solutions and keep the current solution unchanged; The unacceptable new solutions are treated through the exponential cooling method, gradually reducing the temperature parameters, and repeating the inversion method and Metropolis criterion to generate the optimized path combination; Define constraints such as vehicle capacity, order delivery time, and single travel distance through problem requirements and business rules; The optimized path combination is filtered and adjusted through defined constraints to obtain a standard path.

4. The timely delivery method based on path planning as claimed in claim 1, characterized in that: The multi-objective evolutionary algorithm is used to adjust the standard path according to the change of traffic conditions to obtain the optimal path. The specific steps are as follows: Define the constraints of path planning based on real-time traffic condition changes; The constraints include dynamic time window constraints, dynamic driving time constraints, dynamic driving distance constraints, weather impact constraints and congested road section avoidance constraints; Using a multi-objective evolutionary algorithm, the initial population consists of a set of randomly generated path combinations, and the multi-objective problem is decomposed into multiple single-objective sub-problems. Each sub-problem corresponds to a weight vector, and the weight vectors are evenly distributed on the unit hyperplane. Assign neighbors to each subproblem based on the Euclidean distance between weight vectors, and set an initial reference point for each subproblem, set to the maximum value of each objective; For each subproblem, randomly select individuals from the neighborhood as parents, perform crossover and mutation operations on the parent individuals to generate new offspring individuals, update the reference point of each subproblem to the minimum value of the objective function of all individuals in the current population, and retain better individuals by comparing the individuals in the neighborhood with the newly generated offspring individuals; When the stopping condition is met, the algorithm terminates, and the population eventually converges to a set of non-dominated solutions; Based on the constraints, the matching path combinations are screened out from the solution set of non-dominated solutions, and the optimal route is determined from the matching path combinations using the lexicographic order method.

5. The timely delivery method based on path planning as claimed in claim 4, characterized in that: The specific steps of continuously tracking the optimal path delivery progress are as follows: Continuously track vehicle delivery progress through on-board GPS devices and cellular networks, and continuously receive the latest traffic data from the city's intelligent transportation center.

6. The timely delivery method based on path planning as claimed in claim 5, characterized in that: Based on tracking the optimal path delivery progress, the multi-objective evolutionary algorithm is called in real time to adjust the optimal path. The specific steps are as follows: According to the vehicle delivery progress and the latest traffic data, if the actual driving time exceeds the expected time, serious traffic jams and unexpected events occur, the route re-planning process will be automatically initiated; A multi-objective evolutionary algorithm is used to adjust the optimal path for the existing path combination.

7. A timely delivery system based on path planning, based on the timely delivery method based on path planning according to any one of claims 1 to 6, characterized in that: Including order management module, route planning module, real-time traffic analysis module, dynamic route adjustment module, delivery tracking and adjustment module and customer feedback analysis module; The order management module is used to receive order information and enter it into the distribution management center; The path planning module is used to obtain the standard path based on the order information using the simulated annealing algorithm; The real-time traffic analysis module is used to connect to the city's intelligent traffic center to collect real-time traffic data based on standard routes, analyze the real-time traffic data using the traffic flow prediction model, and obtain changes in traffic conditions; Dynamic path adjustment module, which is used to adjust the standard path according to traffic conditions and obtain the optimal path using a multi-objective evolutionary algorithm; Delivery tracking module, used to continuously track the progress of delivery along the optimal route; The real-time adjustment module is used to adjust the optimal path based on tracking the delivery progress of the optimal path and calling the multi-objective evolutionary algorithm in real time.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the timely delivery method based on path planning described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the timely delivery method based on path planning described in any one of claims 1 to 6 are implemented.

Images