A dynamic scheduling optimization method for a picking and placing problem and related devices

Abstract

The embodiment of the application discloses a dynamic scheduling optimization method for pickup and delivery problems and related equipment, relates to the technical field of image recognition, and comprises the following steps: mechanism modeling is performed on a dynamic pickup and delivery problem corresponding to first order data according to a preset constraint condition, and a mathematical model is obtained; data analysis is performed on road condition information, the first order data and external parking lot position information, a first analysis result is obtained, and then a first target algorithm is selected; the mathematical model is solved according to the first target algorithm, and a first planning scheme is obtained; when a new order is acquired, data analysis is performed on second order data containing new order data, second road condition information, a current vehicle position and current vehicle state data, a second analysis result is obtained, and then a second target algorithm is selected; the second order data, the second road condition information, the current vehicle position and the current vehicle state data are brought into the mathematical model, and the mathematical model is solved according to the second target algorithm, and a second planning scheme is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of operational optimization and real-time dynamic scheduling, and in particular to a dynamic scheduling optimization method for pickup and delivery problems and related equipment. BACKGROUND

[0002] Dynamic pickup and delivery problem (DPDP) is also known as real-time PDP, which considers the changes of customer demand and road conditions, and has a large number of practical applications in the fields of logistics distribution and sharing economy, such as real-time delivery of food delivery companies, dynamic pickup and drop-off services, dynamic scheduling of emergency supplies, and path optimization of online carpooling, etc. It has become a frontier topic in the fields of operations research and intelligent transportation.

[0003] In view of real-time dynamics, DPDP is usually divided into a series of periodical discrete static sub-problems for solving, which has good computational efficiency and feasibility in dynamic environment. However, the unified deployment of original vehicles (vehicles inside the yard) and vehicles outside the yard has not been fully studied, and there is still a lack of mathematical model for this problem.

[0004] Dynamic pickup and delivery problem has the characteristics of real-time dynamics and large-scale solution, which requires short solution time and high solution quality. Many scholars have made a series of researches on data analysis and optimization of dynamic pickup and delivery problem combined with optimization theory and intelligent algorithm, and have made certain progress. With the vigorous development of intelligent algorithm and mathematical modeling theory, the dynamic integration of actual vehicle data and order data by using mechanism mathematical modeling process not only speeds up the modeling speed, but also facilitates the dynamic solution of actual problems each time.

[0005] In recent years, the research on intelligent algorithm for solving pickup and delivery problem is in full swing, and continues to delve into different characteristic variant problems. However, swarm intelligence optimization, evolutionary algorithm and neighborhood optimization will show different optimization quality on different characteristic data. Dynamic pickup and delivery problem has large amount of data and complex data quality problems, and the existing technology cannot guarantee high solution efficiency. SUMMARY

[0006] Therefore, the present application provides a dynamic scheduling optimization method for pickup and delivery problem and related equipment to solve the problem that the existing technology cannot guarantee high solution efficiency.

[0007] In a first aspect, the embodiments of the present application provide a dynamic scheduling optimization method for pickup and delivery problem, which comprises:

[0008] acquire first road condition information, first order data and external parking lot position information where the delivery vehicle is located, model a dynamic pick-up and delivery problem corresponding to the first order data according to a preset constraint condition with a minimum distance as a target, and obtain a mathematical model representing a dynamic pick-up and delivery process, the first road condition information being road condition information of an area where the first order data is located;

[0009] perform data analysis on the road condition information, the first order data and the external parking lot position information to obtain a first analysis result, select a first preset algorithm corresponding to the first analysis result from a preset algorithm library according to the first analysis result, and use the first preset algorithm as a first target algorithm;

[0010] solve the mathematical model according to the first target algorithm to obtain a first planning scheme, and control the delivery vehicle to perform delivery according to the first planning scheme;

[0011] when a new order is acquired, acquire new order data of the new order, second road condition information when the new order is acquired, a current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle, perform data analysis on second order data containing the new order data, the second road condition information, the current vehicle position and the current vehicle state data to obtain a second analysis result, select a second preset algorithm corresponding to the second analysis result from a preset algorithm library according to the second analysis result, and use the second preset algorithm as a second target algorithm;

[0012] bring the second order data, the second road condition information, the current vehicle position and the current vehicle state data into the mathematical model to obtain a to-be-solved model, solve the to-be-solved model according to the second target algorithm to obtain a second planning scheme, control the delivery vehicle to perform delivery according to the second planning scheme, and complete dynamic scheduling optimization.

[0013] Optionally, the step of acquiring first road condition information, first order data and external parking lot position information where the delivery vehicle is located, modeling a dynamic pick-up and delivery problem corresponding to the first order data according to a preset constraint condition with a minimum distance as a target, and obtaining a mathematical model representing a dynamic pick-up and delivery process includes:

[0014] The pickup point and the delivery point in the first order data are taken as demand points, the external yard position information of the delivery vehicle is taken as a yard point, each delivery task is taken as a demand, and a mechanism modeling is performed on a dynamic pickup and delivery problem corresponding to the first order data according to the road condition information, with a minimum distance as a target and under preset constraint conditions, to obtain a mathematical model representing a dynamic pickup and delivery process, wherein the preset constraint conditions at least include: a first constraint condition for ensuring that each demand must be served and can only be served once, a second constraint condition for ensuring that the delivery vehicle returns to the same position, a mileage constraint for limiting the driving distance of the delivery vehicle, a load constraint for limiting the loading weight of the delivery vehicle, a pickup point flow constraint for ensuring that the in and out are equal, a delivery point flow constraint for ensuring that the in and out are equal, a time accumulation constraint, a service time constraint and a vehicle scheduling total number constraint.

[0015] Optionally, the step of performing data analysis on the road condition information, the first order data and the external yard position information to obtain a first analysis result, and selecting a first preset algorithm corresponding to the first analysis result in a preset algorithm library according to the first analysis result, and taking the first preset algorithm as a first target algorithm, comprises:

[0016] Performing statistical analysis on the road condition information, the first order data and the external yard position information to obtain data distribution characteristics, and performing data display on the data distribution characteristics.

[0017] Obtaining user request information for the data distribution characteristics, and selecting a first preset algorithm corresponding to the first analysis result in a preset algorithm library according to the user request information, and taking the first preset algorithm as a first target algorithm.

[0018] Optionally, the preset algorithm library includes: heuristic / meta-heuristic algorithm, adaptive large neighborhood algorithm, genetic algorithm, ant colony algorithm and large neighborhood search algorithm.

[0019] Optionally, the step of obtaining new order data of a new order when the new order is obtained, second road condition information when the new order is obtained, current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle, comprises:

[0020] Based on the new pickup point position and the new delivery point position in the new order data, collecting second road condition information corresponding to the new pickup point position and the new delivery point position when the new order is obtained;

[0021] The demand point and the parking lot point are both taken as nodes, and the last node from which the delivery vehicle departs is determined according to the first planning scheme; the current vehicle position of the delivery vehicle is determined according to the first planning scheme, the time of the last node from which the delivery vehicle departs, and the travel speed of the delivery vehicle;

[0022] The load state of the delivery vehicle is acquired as the current vehicle state data.

[0023] Optionally, the step of selecting, according to the second analysis result, a second preset algorithm corresponding to the second analysis result from a preset algorithm library comprises:

[0024] A historical analysis result and a historical target algorithm corresponding to the historical analysis result are acquired, and a training set and a verification set are constructed according to a preset proportion based on the historical analysis result and the historical target algorithm;

[0025] The training of the prediction model is completed according to the training set and the verification set, and a target prediction model is obtained;

[0026] According to the second analysis result, a second preset algorithm corresponding to the second analysis result is selected from a preset algorithm library through the target prediction model.

[0027] Optionally, the mathematical model representing the dynamic taking and delivering process comprises:

[0028] comprises:

[0029] a target function:

[0030]

[0031] a vehicle mileage constraint condition:

[0032]

[0033]

[0034] a task division constraint condition:

[0035]

[0036] a vehicle load constraint condition:

[0037]

[0038] a taking and delivering amount balance constraint condition:

[0039]

[0040] a taking and delivering (taking) arc balance constraint condition:

[0041]

[0042]

[0043] Pickup (delivery) arc balance constraints:

[0044]

[0045]

[0046] Time constraints:

[0047]

[0048]

[0049] Service time constraints:

[0050]

[0051] Departure-return constraints:

[0052]

[0053]

[0054] where K is a set of vehicles, P is a set of pickup points, x is a decision variable that judges whether vehicle k travels from i to j, D is a set of delivery points, L is a maximum driving distance limit of a vehicle, Q is a maximum load limit of a vehicle, and respectively represent the start time and end time of the time window of node i, delivery time and pickup time is the demand of node i, is positive to represent pickup, is negative to represent delivery, is a set of nodes visited by vehicle k, is the service time of node i, is the travel time from node i to node j, is the transportation cost, i.e., transportation distance, from node i to node j.

[0055] On the other hand, the application provides a dynamic scheduling optimization system for a pickup and delivery problem, the system comprising:

[0056] The data collection module is configured to acquire first road condition information, first order data, and external parking lot position information of the delivery vehicle, perform mechanism modeling on a dynamic pick-and-delivery problem corresponding to the first order data according to a preset constraint condition, and obtain a mathematical model representing a dynamic pick-and-delivery process, the first road condition information being road condition information of an area through which the first order data passes.

[0057] The first calculation module is configured to perform data analysis on the road condition information, the first order data, and the external parking lot position information to obtain a first analysis result, select a first preset algorithm corresponding to the first analysis result from a preset algorithm library according to the first analysis result, and use the first preset algorithm as a first target algorithm.

[0058] The first planning module is configured to solve the mathematical model according to the first target algorithm to obtain a first planning scheme, and control the delivery vehicle to perform delivery according to the first planning scheme.

[0059] The second calculation module is configured to, when a new order is acquired, acquire new order data of the new order, second road condition information acquired when the new order is acquired, a current vehicle position of the delivery vehicle, and current vehicle state data of the delivery vehicle, perform data analysis on second order data containing the new order data, the second road condition information, the current vehicle position, and the current vehicle state data to obtain a second analysis result, select a second preset algorithm corresponding to the second analysis result from the preset algorithm library according to the second analysis result, and use the second preset algorithm as a second target algorithm.

[0060] The second planning module is configured to bring the second order data, the second road condition information, the current vehicle position, and the current vehicle state data into the mathematical model to obtain a to-be-solved model, solve the to-be-solved model according to the second target algorithm to obtain a second planning scheme, and control the delivery vehicle to perform delivery according to the second planning scheme, thereby completing dynamic scheduling optimization.

[0061] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the dynamic scheduling optimization method for a pick-and-delivery problem as described above when executing the computer program.

[0062] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executable on a processor to implement the dynamic scheduling optimization method for a pick-and-delivery problem as described above.

[0063] The technical scheme of the embodiment of the present application obtains first road condition information, first order data and external parking lot position information of a delivery vehicle, targets at minimum distance, performs mechanism modeling on a dynamic pick-and-delivery problem corresponding to the first order data according to a preset constraint condition, and obtains a mathematical model representing a dynamic pick-and-delivery process; data analysis is performed on the road condition information, the first order data and the external parking lot position information to obtain a first analysis result, a first preset algorithm corresponding to the first analysis result is selected from a preset algorithm library according to the first analysis result, and the first preset algorithm is used as a first target algorithm; the mathematical model is solved according to the first target algorithm to obtain a first planning scheme, and the delivery vehicle is controlled to perform delivery according to the first planning scheme; when a new order is obtained, new order data of the new order, second road condition information obtained when the new order is obtained, a current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle are obtained, and data analysis is performed on second order data containing the new order data, the second road condition information, the current vehicle position and the current vehicle state data to obtain a second analysis result, a second preset algorithm corresponding to the second analysis result is selected from the preset algorithm library according to the second analysis result, and the second preset algorithm is used as a second target algorithm; the second order data, the second road condition information, the current vehicle position and the current vehicle state data are brought into the mathematical model to obtain a to-be-solved model, and the to-be-solved model is solved according to the second target algorithm to obtain a second planning scheme, and the delivery vehicle is controlled to perform delivery according to the second planning scheme, thereby completing dynamic scheduling optimization. The current dynamic pick-and-delivery problem is solved iteratively and periodically, real-time data is updated, the problem is solved again to re-optimize vehicle path planning, so as to realize the comprehensive path planning goal of minimum pick-and-delivery distance, minimum pick-and-delivery time, maximum logistics efficiency and minimum transportation cost. BRIEF DESCRIPTION OF DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0065] Among them:

[0066] Figure 1 A flowchart of a dynamic scheduling optimization method for a pick-and-delivery problem in an embodiment of the present application;

[0067] Figure 2This is a framework diagram of the dynamic retrieval and delivery end-to-end problem analysis and model solution method in a dynamic scheduling optimization method for retrieval and delivery problems according to an embodiment of the present invention;

[0068] Figure 3 This is a schematic diagram of the structure of a dynamic scheduling optimization system for the retrieval and delivery problem in an embodiment of the present invention;

[0069] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;

[0070] Figure 5 This is a schematic diagram of the structure of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] like Figures 1-2 As shown in the figure, a dynamic scheduling optimization method for the pick-up and delivery problem according to an embodiment of the present invention specifically includes the following steps:

[0073] S110. Obtain first road condition information, first order data and the location information of the external parking lot where the delivery vehicle is located. With the minimum distance as the objective, and according to preset constraints, perform mechanism modeling for the dynamic pickup and delivery problem corresponding to the first order data to obtain a mathematical model characterizing the dynamic pickup and delivery process. The first road condition information is the road condition information of the area traversed by the first order data.

[0074] S120. Perform data analysis on the road condition information, the first order data and the external parking lot location information to obtain a first analysis result, and select a first preset algorithm corresponding to the first analysis result from the preset algorithm library according to the first analysis result, and use the first preset algorithm as the first target algorithm.

[0075] S130. Solve the mathematical model according to the first target algorithm to obtain a first planning scheme, and control the delivery vehicle to perform delivery according to the first planning scheme;

[0076] S140, when a new order is obtained, obtaining new order data of the new order, second road condition information when the new order is obtained, current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle, performing data analysis on second order data containing the new order data, the second road condition information, the current vehicle position and the current vehicle state data, obtaining a second analysis result, selecting a second preset algorithm corresponding to the second analysis result in a preset algorithm library according to the second analysis result, and taking the second preset algorithm as a second target algorithm;

[0077] S150, the second order data, the second road condition information, the current vehicle position and the current vehicle state data are brought into the mathematical model to obtain a to-be-solved model, and the second target algorithm is used to solve the to-be-solved model to obtain a second planning scheme, and the delivery vehicle is controlled to perform delivery according to the second planning scheme to complete dynamic scheduling optimization.

[0078] By obtaining the first road condition information, the first order data and the external parking lot position information of the delivery vehicle, a mathematical model representing the dynamic taking and delivering process is obtained by mechanism modeling for the dynamic taking and delivering problem corresponding to the first order data with the minimum distance as the target and under the preset constraint condition; the first analysis result is obtained by performing data analysis on the road condition information, the first order data and the external parking lot position information, and a first preset algorithm corresponding to the first analysis result is selected in a preset algorithm library according to the first analysis result, and the first preset algorithm is taken as a first target algorithm; the first planning scheme is obtained by solving the mathematical model according to the first target algorithm, and the delivery vehicle is controlled to perform delivery according to the first planning scheme; when a new order is obtained, new order data of the new order, second road condition information when the new order is obtained, current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle are obtained, and second order data containing the new order data, the second road condition information, the current vehicle position and the current vehicle state data are analyzed to obtain a second analysis result, and a second preset algorithm corresponding to the second analysis result is selected in a preset algorithm library according to the second analysis result, and the second preset algorithm is taken as a second target algorithm; the second order data, the second road condition information, the current vehicle position and the current vehicle state data are brought into the mathematical model to obtain a to-be-solved model, and the second target algorithm is used to solve the to-be-solved model to obtain a second planning scheme, and the delivery vehicle is controlled to perform delivery according to the second planning scheme to complete dynamic scheduling optimization. Iterative and periodic solving of the current dynamic taking and delivering problem updates the real-time data, and the problem is solved again to re-optimize the vehicle path planning, so as to realize the comprehensive path planning target of the lowest taking and delivering distance, the shortest taking and delivering time, the highest logistics efficiency and the lowest transportation cost.

[0079] In a possible implementation, the first order data, the first road condition information, and the external parking lot position information of the delivery vehicle are acquired, a mechanism model is established for a dynamic pickup and delivery problem corresponding to the first order data under a preset constraint condition with a minimum distance as a target, and a mathematical model representing a dynamic pickup and delivery process is obtained.

[0080] The pickup point and the delivery point in the first order data are taken as demand points, the external parking lot position information of the delivery vehicle is taken as a parking lot point, each delivery task is taken as a demand, a mechanism model is established for a dynamic pickup and delivery problem corresponding to the first order data under a preset constraint condition with a minimum distance as a target according to the road condition information, and a mathematical model representing a dynamic pickup and delivery process is obtained, where the preset constraint condition at least includes a first constraint condition for ensuring that each demand must be served and can only be served once, a second constraint condition for ensuring that the delivery vehicle returns to the same position, a mileage constraint for limiting the driving distance of the delivery vehicle, a load constraint for limiting the loading weight of the delivery vehicle, a pickup point flow constraint for ensuring that the in-flow and out-flow are equal, a delivery point flow constraint for ensuring that the in-flow and out-flow are equal, a time accumulation constraint, a service time constraint, and a vehicle scheduling total number constraint.

[0081] For example, the integer programming model of the dynamic pickup and delivery problem can be applied to most kinds of dynamic pickup and delivery problems based on the mechanism modeling. In a static pickup and delivery problem, the related information (pickup point, delivery point, vehicle, pickup and delivery request, scheduling, and other related information) before the optimization of the logistics pickup and delivery path is known and fixed; in a dynamic pickup and delivery optimization problem, many pieces of information are uncertain and unpredictable, and part of the information can be fuzzy and random. The vehicle path is planned and optimized in real time according to real-time information when the path is planned. In a dynamic pickup and delivery problem, the number, position, and cargo volume of part of the pickup points and corresponding delivery points are known, part of the information is unknown and unpredictable, and part of the information can be random and fuzzy. How to design a reasonable pickup and delivery route so that the vehicle departs from the distribution center, the same vehicle sequentially visits the pickup point and the corresponding delivery point, meets the related constraint conditions (weight, volume, time window, etc.), and deals with the new order received during the vehicle service (new pickup point and delivery point), and finally returns to the distribution center is the main research content of the pickup and delivery problem. The research purpose of the pickup and delivery problem is to reduce the number of vehicles used, reduce the pickup and delivery distance, reduce the pickup and delivery time, improve the logistics efficiency, save the transportation cost, and the like.

[0082] According to the problem abstract characteristics, mechanism modeling is performed based on point-arc relationship in graph theory: all points are divided into demand points and depot points, wherein the demand points are divided into pickup points and delivery points. Vehicles need to serve all customer orders. Each customer order can only be served by one vehicle, that is, a pickup point and a corresponding delivery point can only be served by the same vehicle, and the pickup point needs to be visited before the delivery point. The vehicle service start time must be within the demand point time window. If the vehicle arrives at the demand point location earlier than the time window start time, the vehicle must stop and wait. If the vehicle arrives at the demand point location later than the time window end time, the customer order cannot be served. Meanwhile, all vehicles are of the same type, that is, the vehicle capacity is the same, and the vehicle load during service cannot exceed the maximum vehicle capacity. All vehicles are dispatched from the same depot, and the depot also has a time window. All vehicles must complete the pickup and delivery of all goods within the time window.

[0083] In a possible implementation, the step of performing data analysis on the road condition information, the first order data and the external depot location information to obtain a first analysis result, and selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library according to the first analysis result, and taking the first preset algorithm as a first target algorithm, includes:

[0084] Performing statistical analysis on the road condition information, the first order data and the external depot location information to obtain data distribution characteristics, and performing data display on the data distribution characteristics;

[0085] Obtaining user request information for the data distribution characteristics, and selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library according to the user request information, and taking the first preset algorithm as a first target algorithm.

[0086] For example, a visualization operation is performed, and imported data is analyzed by a suitable statistical analysis method (such as a traditional KNN fast clustering method) to summarize, understand and digest them so as to maximize the development of the function of the data and play the role of the data. A data visualization page is provided to facilitate the observation of the data by the user and understand the characteristics of the data.

[0087] In a possible implementation, the preset algorithm library includes heuristic / meta-heuristic algorithms, self-adaptive large neighborhood algorithms, genetic algorithms, ant colony algorithms and large neighborhood search algorithms.

[0088] For example, the preset algorithm library includes a plurality of heuristic / meta-heuristic algorithms suitable for solving pickup and delivery problems, and different solutions can be obtained in each iteration according to the user's preference.

[0089] Adaptive Large Neighborhood Search Algorithm Introduction: Adaptive Large Neighborhood Search Algorithm (ALNS), proposed by Ropke and Pisinger in 2006, is a heuristic method that is an extension of large neighborhood search. ALNS is not limited to a certain destruction and repair method, but selects from the operator pool according to the effect of the last search in each iteration, that is, it measures the effect of the operator on the basis of neighborhood search, so that the algorithm can automatically select good operators to destroy and repair the solution, thereby having a certain probability of getting a better solution.

[0090] Variable Neighborhood Search Algorithm Introduction: Variable Neighborhood Search Algorithm (VNS), first proposed by Hansen and Mladenovic in 1997, is an improved local search algorithm. It realizes a good balance between concentration and dispersion by dynamically switching between different neighborhood structures for alternating search. VNS is based on the following perception: the local minimum of one neighborhood structure is not necessarily the local minimum of another neighborhood structure; for all possible neighborhood structures, the global minimum is a local minimum; for many problems, the local minimum of one or more neighborhoods is relatively close to each other.

[0091] Genetic Algorithm Introduction: Genetic Algorithm (GA), first proposed by John Holland in 1975, is a method of searching for optimal solution by simulating the natural selection and genetic mechanism of biological evolution process of Darwin's biological evolution theory. By drawing on biological evolution theory, genetic algorithm simulates the biological evolution process of the problem to be solved, generates the next generation of solutions through replication, crossover, mutation and other operations, and gradually eliminates solutions with low fitness function values and increases solutions with high fitness function values. After N generations of evolution, it is very likely that individuals with high fitness function values will evolve.

[0092] Ant Colony Algorithm Introduction: Ant Colony Algorithm (ACO), first proposed by Italian scholar Dorigo M in 1991, is a brand-new bionic simulation evolution algorithm. The idea of ACO is to simulate the foraging behavior of ants, that is, to use a large number of "ants" to search randomly in the search space, and to use pheromone to strengthen the search route and guide the search of other "ants", while introducing a volatile mechanism to avoid falling into local optimum. This positive feedback of introducing volatile mechanism makes the algorithm able to find multiple high-quality global solutions without falling into local optimal solution as quickly as other search algorithms, and because of its own parallelism, it can easily realize parallel computing.

[0093] Large Neighborhood Search Algorithm Introduction: Adaptive Large Neighborhood Search (LNS) algorithm is proposed by Shaw. The algorithm is mainly divided into two parts: destruction operation and repair operation. In the destruction operation, a certain amount of customers are removed from the current solution to obtain a partial solution. Then, enter the repair operation, apply the repair operator to the partial solution obtained in the destruction operation, and all possible cases of reinsertion constitute a neighborhood of the original solution.

[0094] In a possible implementation, the step of obtaining the new order data of the new order, the second road condition information when the new order is obtained, the current vehicle position of the delivery vehicle and the current vehicle state data of the delivery vehicle when the new order is obtained, comprises:

[0095] Based on the new pickup point position and the new delivery point position in the new order data, the second road condition information corresponding to the new pickup point position and the new delivery point position when the new order is obtained is collected;

[0096] The demand point and the parking lot point are both regarded as nodes, the last node from which the delivery vehicle departs is determined according to the first planning scheme, and the current vehicle position of the delivery vehicle is determined according to the first planning scheme, the time of the last node from which the delivery vehicle departs and the travel speed of the delivery vehicle;

[0097] The load state of the delivery vehicle is obtained as the current vehicle state data.

[0098] Example, because the dynamic pickup and delivery problem model is based on mechanism modeling, the use of real-time data here can ensure that the obtained solution is more robust, that is, the solution maintains its stability under the influence of certain disturbance factors. There are a lot of uncertainties in the pickup and delivery of vehicles: such as congestion, bad weather, vehicle failure, customer information change, etc., which will directly affect the change of pickup and delivery vehicle driving time or speed, leading to waiting or even changing the route, thereby causing the increase of vehicle scheduling cost. These uncertainties make the optimization of scheduling complex, but when the information is known, the dynamic pickup and delivery problem can be converted into a static pickup and delivery problem to solve. In this software, the real-time data update of the customer not only considers the decrease or increase of the original customer demand, but more considers the new customer demand. Under the premise of meeting the vehicle load, how to add new pickup points and corresponding delivery points to the existing path, if there is no such path currently, a new vehicle service needs to be arranged. For the case of reducing the demand of the original customer, the local route does not need to be adjusted. If the demand is 0, skip this customer. For the case of increasing the demand of the original customer, if the vehicle load is exceeded, the demand of the last customer serviced on the sub-path is selected as the new customer demand for processing, until the vehicle load limit is met. The algorithm library introduces the concept of external distribution center, and sets the demand point of the vehicle as the external distribution center.

[0099] According to the current time and the solution of the previous period, the current vehicle position is updated, and the position of the external distribution center is calculated. Combined with the current dynamic order data, the data updating process is completed, and after the data updating, the information is clear. At this time, the dynamic pickup and delivery problem can be temporarily converted into a static pickup and delivery problem to solve. At the end of the previous period, the algorithm is given a maximum allowed calculation time, and within this time period, the algorithm performs new vehicle path scheduling according to the data obtained at the moment of the end of the previous period, and obtains a new vehicle path planning scheme, that is, a new solution. After the maximum allowed calculation time, the new vehicle path planning is executed until the end of the next period.

[0100] According to the solution given each time, the vehicle position, remaining load, time window, customer demand change, external distribution center and other state information are updated at the current time, and the dynamic order data at this time forms a new instance of the dynamic pickup and delivery problem. On this basis, the algorithm re-optimizes the vehicle path planning to achieve the comprehensive path planning goal of the shortest pickup and delivery distance, the shortest pickup and delivery time, the highest logistics efficiency and the lowest transportation cost.

[0101] In a possible implementation, the step of selecting a second preset algorithm corresponding to the second analysis result from a preset algorithm library according to the second analysis result comprises:

[0102] Obtain historical analysis results and historical target algorithms corresponding to the historical analysis results, and construct training sets and validation sets based on the historical analysis results and the historical target algorithms according to a preset ratio;

[0103] The prediction model is trained using the training set and the validation set to obtain the target prediction model.

[0104] Based on the target prediction model, a second preset algorithm corresponding to the second analysis result is selected from the preset algorithm library.

[0105] For example, since human selection of algorithms may be subject to human error, algorithm selection can be performed using a deep learning model with predictive capabilities, which is faster and more accurate.

[0106] For example, the dynamic pickup and delivery problem can be decomposed into a virtual periodicity, and the road condition and order data for each stage can be combined with the current vehicle location to complete the data update. The algorithm planning and solution flowchart is attached. Figure 4 As shown, users first upload data on the front end, and then the back end determines the status of the raw data. Next, the front end analyzes the raw data, providing a visual data analysis page, allowing users to select a suitable intelligent optimization algorithm based on the analysis results. Some parameters in the algorithm can be adjusted by the user. Finally, the back end calculates the specific vehicle pickup and delivery route planning solution. At the end of the previous virtual cycle, a new problem instance is constructed based on the real-time updated vehicle status information and changes in customer needs. Then, within a very short timeframe, the algorithm re-optimizes the vehicle pickup and delivery route based on the real-time data, obtaining a new vehicle pickup and delivery route plan. Subsequently, the vehicle begins execution according to the new vehicle pickup and delivery route plan until the end of the next virtual cycle.

[0107] This iterative and periodic approach—solving the current dynamic retrieval and delivery problem, updating real-time data, and then resolving the problem to re-optimize vehicle routing—constitutes the solution mechanism for dynamic retrieval and delivery problems in this software. The vehicle travel paths, i.e., the dynamic retrieval and delivery plans, for each virtual cycle will be presented to the user through a visual interface.

[0108] In one possible implementation, the mathematical model representing the dynamic fetching process includes:

[0109] Objective function: used to evaluate the total mileage of the entire path solution, where It is a node i To the node j Transportation costs are the same as transportation distances;

[0110]

[0111] Vehicle mileage constraint: for any vehicle in the fleet, the total distance traveled by the vehicle is less than the maximum mileage of the vehicle L .

[0112]

[0113]

[0114] Task division constraint: all pickup P and delivery D tasks need to be divided to fixed vehicles.

[0115]

[0116] Vehicle load constraint: for any vehicle in the fleet, the sum of the pickup amounts is no more than the total capacity of the vehicle Q。

[0117]

[0118] Pickup and delivery amount balance constraint: for any order , the pickup amount and the delivery amount are equal.

[0119]

[0120] Pickup and delivery (pickup) arc balance constraint: for any vehicle in the fleet, the number of times of entering the pickup point and leaving the delivery point is equal.

[0121]

[0122]

[0123] Pickup and delivery (delivery) arc balance constraint: for any vehicle in the fleet, the number of times of entering the delivery point and leaving the pickup point is equal.

[0124]

[0125]

[0126] Time constraint: the difference between the delivery time and the pickup time is no less than the time consumption of directly from the pickup point to the delivery point .

[0127]

[0128]

[0129] Service time constraint: any pickup point and delivery point need to be served within the start time after, and the end of the service time before providing the service.

[0130]

[0131] Departure-return constraint: for any vehicle in the vehicle fleet, the number of times of leaving the depot and the number of times of returning to the depot are equal, and are both 1.

[0132]

[0133]

[0134] wherein K is a vehicle set, P is a pickup point set, x is a decision variable for judging whether a vehicle k travels from i to j, D is a delivery point set, L is a maximum travel distance limit of a vehicle, Q is a maximum load limit of a vehicle, and respectively represent the start time and the end time of the time window of node i, , is the demand of node i, is positive for pickup, is negative for delivery, is a node set visited by a vehicle k, is the service time of node i, is the travel time from node i to node j, is the transportation cost from node i to node j, i.e., the transportation distance.

[0135] Exemplarily, the mathematical model representing the dynamic pickup and delivery process is as follows, wherein, is a decision variable, representing whether a vehicle k travels from i to j , and is 1 if yes, and is 0 if no. K is a vehicle set, P is a pickup point set, D is a delivery point set.

[0136] On the other hand, as shown in Figure 3 , the application provides a dynamic scheduling optimization system for a pickup and delivery problem, the system comprising:

[0137] a data acquisition module 201, configured to acquire first road condition information, first order data, and external depot location information of a delivery vehicle, to perform mechanism modeling on a dynamic pickup and delivery problem corresponding to the first order data according to a preset constraint condition, so as to obtain a mathematical model representing a dynamic pickup and delivery process, the first road condition information being road condition information of an area where the first order data is located;

[0138] The first calculation module 202 is used to perform data analysis on the road condition information, the first order data and the external parking lot location information to obtain a first analysis result, and select a first preset algorithm corresponding to the first analysis result from the preset algorithm library according to the first analysis result, and use the first preset algorithm as the first target algorithm.

[0139] The first planning module 203 is used to solve the mathematical model according to the first target algorithm to obtain a first planning scheme, and control the delivery vehicle to perform delivery according to the first planning scheme.

[0140] The second calculation module 204 is used to, when a new order is obtained, acquire the new order data of the new order, the second road condition information at the time the new order is obtained, the current vehicle location of the delivery vehicle, and the current vehicle status data of the delivery vehicle, and perform data analysis on the second order data containing the new order data, the second road condition information, the current vehicle location, and the current vehicle status data to obtain a second analysis result, and select a second preset algorithm corresponding to the second analysis result from a preset algorithm library based on the second analysis result, and use the second preset algorithm as the second target algorithm;

[0141] The second planning module 205 is used to input the second order data, the second road condition information, the current vehicle location and the current vehicle status data into the mathematical model to obtain the model to be solved, and to solve the model to be solved according to the second objective algorithm to obtain the second planning scheme, and to control the delivery vehicle to perform delivery according to the second planning scheme to complete the dynamic scheduling optimization.

[0142] In one possible implementation, such as Figure 4 As shown, this application embodiment provides a terminal device 300, including: a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following: acquiring first road condition information, first order data, and the location information of the external parking lot where the delivery vehicle is located; taking the minimum distance as the objective, and according to preset constraints, performing mechanism modeling on the dynamic pickup and delivery problem corresponding to the first order data, to obtain a mathematical model characterizing the dynamic pickup and delivery process; the first road condition information is the road condition information of the area traversed by the first order data.

[0143] The road condition information, the first order data, and the external parking lot location information are analyzed to obtain a first analysis result. Based on the first analysis result, a first preset algorithm corresponding to the first analysis result is selected from a preset algorithm library, and the first preset algorithm is used as the first target algorithm.

[0144] solving the mathematical model according to the first target algorithm to obtain a first planning scheme, and controlling the delivery vehicle to perform delivery according to the first planning scheme;

[0145] When a new order is obtained, new order data of the new order, second road condition information when the new order is obtained, a current vehicle position of the delivery vehicle, and current vehicle state data of the delivery vehicle are obtained, second order data containing the new order data, the second road condition information, the current vehicle position, and the current vehicle state data are analyzed to obtain a second analysis result, a second preset algorithm corresponding to the second analysis result is selected from a preset algorithm library according to the second analysis result, and the second preset algorithm is taken as a second target algorithm;

[0146] The second order data, the second road condition information, the current vehicle position, and the current vehicle state data are brought into the mathematical model to obtain a to-be-solved model, the to-be-solved model is solved according to the second target algorithm to obtain a second planning scheme, the delivery vehicle is controlled to perform delivery according to the second planning scheme, and dynamic scheduling optimization is completed.

[0147] In a possible implementation, as shown in Figure 5 The computer readable storage medium 400 stores a computer program 411, which is executed by a processor to implement the following steps: obtaining first road condition information, first order data, and external parking lot position information of a delivery vehicle, performing mechanism modeling on a dynamic pick-up and delivery problem corresponding to the first order data according to a preset constraint condition, taking a minimum distance as a target, to obtain a mathematical model representing a dynamic pick-up and delivery process, the first road condition information being road condition information of an area through which the first order data passes;

[0148] performing data analysis on the road condition information, the first order data, and the external parking lot position information to obtain a first analysis result, selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library according to the first analysis result, and taking the first preset algorithm as a first target algorithm;

[0149] solving the mathematical model according to the first target algorithm to obtain a first planning scheme, and controlling the delivery vehicle to perform delivery according to the first planning scheme;

[0150] When a new order is acquired, new order data of the new order, second road condition information when the new order is acquired, current vehicle position of the delivery vehicle and current vehicle state data of the delivery vehicle are acquired, data analysis is performed on the second order data containing the new order data, the second road condition information, the current vehicle position and the current vehicle state data, a second analysis result is obtained, a second preset algorithm corresponding to the second analysis result is selected from a preset algorithm library according to the second analysis result, and the second preset algorithm is taken as a second target algorithm;

[0151] The second order data, the second road condition information, the current vehicle position and the current vehicle state data are brought into the mathematical model to obtain a to-be-solved model, and the second target algorithm is used to solve the to-be-solved model to obtain a second planning scheme, and the delivery vehicle is controlled to perform delivery according to the second planning scheme, so that dynamic scheduling optimization is completed.

[0152] It should be noted that the information interaction, execution process and the like between the above devices / units are based on the same concept as the method embodiments of the present application, and the specific functions and the technical effects brought by the same can be referred to the method embodiments part, which will not be repeated here.

[0153] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific names of each functional unit and module are only for convenient distinction, and do not limit the protection scope of the present application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0154] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the present application can implement all or part of the processes in the above-mentioned embodiment methods through a computer program to instruct relevant hardware to complete, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium can at least include any entity or device capable of carrying the computer program code to the photographing device / terminal equipment, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, magnetic disk or optical disk, etc. In some jurisdictions, according to legislation and patent practice, the computer readable medium can not be an electrical carrier signal and a telecommunication signal.

[0155] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0156] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0157] In the embodiments provided by the present application, it should be understood that the disclosed apparatus / network device and method can be implemented in other ways. For example, the above-described apparatus / network device embodiments are merely schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed each other can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0158] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0159] The above embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

[0160] The above disclosure is only the preferred embodiment of the present application, of course, cannot limit the scope of the right of the present application, therefore, the equivalent changes made by the claims of the present application, still belong to the scope covered by the present application.

Claims

1. A dynamic scheduling optimization method for the fetching and delivery problem, characterized in that, include: The system acquires first road condition information, first order data, and the location information of the external parking lot where the delivery vehicle is located. With the goal of minimizing the distance, it performs mechanism modeling on the dynamic pickup and delivery problem corresponding to the first order data according to preset constraints, and obtains a mathematical model characterizing the dynamic pickup and delivery process. The first road condition information is the road condition information of the area traversed by the first order data. The steps of acquiring first road condition information, first order data, and the location information of the external parking lot where the delivery vehicle is located, and performing mechanism modeling on the dynamic pickup and delivery problem corresponding to the first order data under preset constraints with the goal of minimizing distance, to obtain a mathematical model representing the dynamic pickup and delivery process, include: Taking the pickup and delivery points in the first order data as demand points, the location information of the external parking lot where the delivery vehicle is located as parking lot points, and each delivery task as a demand, based on the first road condition information, with the minimum distance as the objective, and according to preset constraints, a mechanism model is performed for the dynamic pickup and delivery problem corresponding to the first order data to obtain a mathematical model characterizing the dynamic pickup and delivery process. The preset constraints include at least: a first constraint to ensure that each demand must be served and can only be served once; a second constraint to ensure that the delivery vehicle returns to the same location; a mileage constraint to limit the driving distance of the delivery vehicle; a load constraint to limit the load weight of the delivery vehicle; a pickup point flow constraint to ensure equal entry and exit; a delivery point flow constraint to ensure equal entry and exit; a time accumulation constraint; a service time constraint; and a total number of vehicle scheduling constraints. The road condition information, the first order data, and the external parking lot location information are analyzed to obtain a first analysis result. Based on the first analysis result, a first preset algorithm corresponding to the first analysis result is selected from the preset algorithm library, and the first preset algorithm is used as the first target algorithm. The step of performing data analysis on the road condition information, the first order data, and the external parking lot location information to obtain a first analysis result, and selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library based on the first analysis result, and using the first preset algorithm as the first target algorithm, includes: Statistical analysis is performed on the road condition information, the first order data, and the external parking lot location information to obtain data distribution characteristics, and the data distribution characteristics are then displayed. Obtain user request information for the data distribution characteristics, and select a first preset algorithm corresponding to the first analysis result from a preset algorithm library based on the user request information, and use the first preset algorithm as the first target algorithm; The mathematical model is solved according to the first target algorithm to obtain a first planning scheme, and the delivery vehicles are controlled to perform delivery according to the first planning scheme. When a new order is received, the system acquires the new order data, the second road condition information at the time the new order was acquired, the current vehicle location of the delivery vehicle, and the current vehicle status data of the delivery vehicle. The system then performs data analysis on the second order data containing the new order data, the second road condition information, the current vehicle location, and the current vehicle status data to obtain a second analysis result. Based on the second analysis result, the system selects a second preset algorithm corresponding to the second analysis result from a preset algorithm library and uses the second preset algorithm as the second target algorithm. The second order data, the second road condition information, the current vehicle location, and the current vehicle status data are input into the mathematical model to obtain the model to be solved. The model to be solved is then solved according to the second objective algorithm to obtain the second planning scheme. The delivery vehicles are then controlled to perform delivery according to the second planning scheme to complete the dynamic scheduling optimization.

2. The dynamic scheduling optimization method for the pick-up and delivery problem as described in claim 1, characterized in that, The preset algorithm library includes: heuristic / meta-heuristic algorithms, adaptive large neighborhood algorithms, genetic algorithms, ant colony algorithms, and large neighborhood search algorithms.

3. The dynamic scheduling optimization method for the pick-up and delivery problem as described in claim 1, characterized in that, The step of acquiring the new order data, the second road condition information at the time of acquiring the new order, the current vehicle location, and the current vehicle status data of the delivery vehicle when a new order is acquired includes: Based on the new pickup point location and new delivery point location in the new order data, collect the second traffic information corresponding to the new pickup point location and new delivery point location when a new order is obtained; Both the demand point and the parking lot point are used as nodes. The last node where the delivery vehicle leaves is determined according to the first planning scheme. The current vehicle position of the delivery vehicle is determined according to the first planning scheme, the time of the last node where the delivery vehicle leaves, and the speed of the delivery vehicle. The load status of the delivery vehicle is obtained as the current vehicle status data.

4. The dynamic scheduling optimization method for the pick-up and delivery problem as described in claim 1, characterized in that, The step of selecting a second preset algorithm corresponding to the second analysis result from a preset algorithm library based on the second analysis result includes: Obtain historical analysis results and historical target algorithms corresponding to the historical analysis results, and construct training sets and validation sets based on the historical analysis results and the historical target algorithms according to a preset ratio; The prediction model is trained using the training set and the validation set to obtain the target prediction model. Based on the target prediction model, a second preset algorithm corresponding to the second analysis result is selected from the preset algorithm library.

5. The dynamic scheduling optimization method for the pick-up and delivery problem as described in claim 1, characterized in that, The mathematical model representing the dynamic fetching process includes: Objective function: Vehicle mileage constraints: Task partitioning constraints: Vehicle load constraints: Pickup and delivery quantity balance constraint: Pickup and delivery arc balance constraints: Pick-up and delivery arc balance constraints: Time constraints: Service time constraints: Departure and return constraints: Where K is the set of vehicles, P is the set of pickup points, x is the decision variable for determining whether vehicle k travels from i to j, D is the set of delivery points, L is the maximum driving distance limit for vehicles, and Q is the maximum load limit for vehicles. and These represent the start and end times of the time window for node i, respectively. Delivery time and pickup time It is the demand of node i. A positive number indicates that the goods have been picked up. A negative number indicates delivery. It is the set of nodes visited by vehicle k. It is the service time of node i. It is the travel time from node i to node j. It is the transportation cost, i.e., the transportation distance, from node i to node j.

6. A dynamic scheduling optimization system for the retrieval and delivery problem, characterized in that, The system includes: The data acquisition module is used to acquire first road condition information, first order data, and the location information of the external parking lot where the delivery vehicle is located. With the goal of minimizing the distance, and according to preset constraints, it performs mechanism modeling on the dynamic pickup and delivery problem corresponding to the first order data to obtain a mathematical model representing the dynamic pickup and delivery process. The first road condition information refers to the road condition information of the area traversed by the first order data. The step of acquiring the first road condition information, first order data, and the location information of the external parking lot where the delivery vehicle is located, with the goal of minimizing the distance, and according to preset constraints, performing mechanism modeling on the dynamic pickup and delivery problem corresponding to the first order data to obtain a mathematical model representing the dynamic pickup and delivery process includes: taking the pickup point and delivery point in the first order data as demand points, and taking the location information of the external parking lot where the delivery vehicle is located as the demand points. Using the location information of the vehicle depot as the depot point, and treating each delivery task as a demand, based on the first road condition information, with the minimum distance as the objective, and according to preset constraints, a mechanism model is performed on the dynamic pickup and delivery problem corresponding to the first order data to obtain a mathematical model characterizing the dynamic pickup and delivery process. The preset constraints include at least: a first constraint to ensure that each demand must be served and can only be served once; a second constraint to ensure that delivery vehicles return to the same location; a mileage constraint to limit the driving distance of delivery vehicles; a load constraint to limit the load weight of delivery vehicles; a pickup point flow constraint to ensure equal entry and exit; a delivery point flow constraint to ensure equal entry and exit; a time accumulation constraint; a service time constraint; and a total number of vehicle scheduling constraints. The first calculation module is used to perform data analysis on the road condition information, the first order data, and the external parking lot location information to obtain a first analysis result, and select a first preset algorithm corresponding to the first analysis result from a preset algorithm library based on the first analysis result, and use the first preset algorithm as the first target algorithm; the step of performing data analysis on the road condition information, the first order data, and the external parking lot location information to obtain a first analysis result, and selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library based on the first analysis result, and using the first preset algorithm as the first target algorithm includes: performing statistical analysis on the road condition information, the first order data, and the external parking lot location information to obtain data distribution characteristics, and displaying the data distribution characteristics; obtaining user request information for the data distribution characteristics, and selecting a first preset algorithm corresponding to the first analysis result from a preset algorithm library based on the user request information, and using the first preset algorithm as the first target algorithm; The first planning module is used to solve the mathematical model according to the first target algorithm to obtain the first planning scheme, and control the delivery vehicle to perform delivery according to the first planning scheme. The second calculation module is used to acquire, when a new order is acquired, the new order data of the new order, the second road condition information at the time the new order is acquired, the current vehicle location of the delivery vehicle, and the current vehicle status data of the delivery vehicle, and to perform data analysis on the second order data containing the new order data, the second road condition information, the current vehicle location, and the current vehicle status data to obtain a second analysis result, and to select a second preset algorithm corresponding to the second analysis result from a preset algorithm library based on the second analysis result, and to use the second preset algorithm as the second target algorithm; The second planning module is used to input the second order data, the second road condition information, the current vehicle location and the current vehicle status data into the mathematical model to obtain the model to be solved, and to solve the model to be solved according to the second objective algorithm to obtain the second planning scheme, and to control the delivery vehicle to perform delivery according to the second planning scheme to complete the dynamic scheduling optimization.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the dynamic scheduling optimization method for the fetching and delivery problem as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the dynamic scheduling optimization method for the fetching and delivery problem as described in any one of claims 1 to 5.

Images