CN111428931B

CN111428931B - Logistics distribution line planning method, device, equipment and storage medium

Info

Publication number: CN111428931B
Application number: CN202010214789.6A
Authority: CN
Inventors: 衡鹤瑞; 李培吉; 李斯
Original assignee: Dongpu Software Co Ltd
Current assignee: Dongpu Software Co Ltd
Priority date: 2020-03-24
Filing date: 2020-03-24
Publication date: 2022-07-08
Anticipated expiration: 2040-03-24
Also published as: CN111428931A

Abstract

The invention discloses a logistics distribution line planning method, a logistics distribution line planning device, logistics distribution line planning equipment and a storage medium. The method continuously adjusts the planning of the transfer center in the distribution line by calculating the mathematical relationship between the distribution line and the constraint conditions of the mathematical model, and particularly continuously screens and replaces the transfer center by adopting a genetic crossing algorithm and an iterative neighborhood search algorithm, so that the solution space can be increased, the optimal transfer center is selected to find the optimal logistics distribution route, the optimal logistics distribution route can be quickly found, the obtained route length and the total length are shortest, and the logistics distribution cost is effectively reduced.

Description

Logistics distribution line planning method, device, equipment and storage medium

Technical Field

The invention relates to the technical field of logistics distribution, in particular to a method, a device, equipment and a storage medium for planning logistics distribution lines.

Background

Along with the continuous development of market economy and logistics technology, logistics distribution is rapidly developed, particularly, the popularization of internet shopping at present, the express delivery amount is increased year by year, and the increase of the workload necessarily affects the overall distribution efficiency of the salesmen to the express.

In the prior art, aiming at the problem of overall delivery efficiency, a salesman who is skilled in road conditions is selected for delivery, and the delivery is carried out by adopting an intelligent express cabinet in a collection and delivery mode, so that the efficiency can be improved to a certain extent, but the relatively high cost is also brought to an express company, and meanwhile, the transportation cost in delivery and delivery is not reduced by the two delivery modes, and other problems are brought.

Disclosure of Invention

The invention mainly aims to solve the technical problems that the rapid distribution cannot be well realized and the distribution cost cannot be reduced in the conventional logistics distribution planning mode.

The first aspect of the present invention provides a method for planning a logistics distribution route, including:

when a distribution task is triggered, planning an initial distribution line based on a distribution demand according to the distribution task, wherein the initial distribution line comprises a line starting point, a line ending point and N transfer centers, and N is a natural number;

coding a transfer center in the initial distribution line to obtain an initial solution;

combining the transit centers in the initial solution based on a genetic crossing algorithm, and planning a new distribution line based on the combination and the line starting point and the line ending point;

calculating distribution conditions in the new distribution line;

judging whether the distribution conditions meet constraint conditions in a pre-established mathematical model for the optimization of the logistics distribution path;

and adjusting the new distribution line through neighborhood iterative search according to the judgment result until the new distribution line is adjusted to meet the constraint condition, and outputting the optimal distribution line.

Optionally, in a first implementation manner of the first aspect of the present invention, the encoding a transit center in the initial distribution route to obtain an initial solution includes:

based on a genetic algorithm, carrying out gene individual extraction processing on the initial distribution line to obtain a plurality of transfer centers;

and sequentially encoding the extracted multiple transfer centers to form an initial solution, wherein the transfer centers in the initial solution and the transfer centers are in a neighborhood relationship.

Optionally, in a second implementation manner of the first aspect of the present invention, the combining the transit centers in the initial solution based on the genetic crossing algorithm, and planning a new distribution route based on the combination and the route origin and destination includes:

based on a genetic crossing algorithm, each transfer center in the initial solution is respectively butted with the initial point and the final point of the line to form a distribution network; planning a new distribution line based on the distribution network;

alternatively, the first and second liquid crystal display panels may be,

selecting a transfer center from the initial solution as a primary transfer center; combining the first transfer center with other transfer centers in the initial solution based on a genetic crossover algorithm to obtain at least one transfer center combination; the at least one transit center combination is respectively butted with the line starting point and the line ending point to form a distribution network; and planning a new distribution line based on the distribution network.

Optionally, in a third implementation manner of the first aspect of the present invention, the adjusting, according to the result of the determination, the new distribution route through neighborhood iterative search until the new distribution route is adjusted to meet the constraint condition, and outputting an optimal distribution route includes:

if the judged result is that the distribution condition does not meet the constraint condition, selecting a transfer center from the neighborhood of the transfer centers in the combination to replace the transfer center in the combination through an iterative neighborhood search algorithm, and planning a corresponding distribution line for the replaced distribution network again;

judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

if so, outputting the optimal distribution line by taking the replaced distribution network as a reference;

if not, continuing to select the next transfer center from the neighborhood to replan the distribution line.

Optionally, in a fourth implementation manner of the first aspect of the present invention, the adjusting, according to the result of the determination, the new distribution route through neighborhood iterative search until the new distribution route is adjusted to meet the constraint condition, and outputting an optimal distribution route includes:

if the judged result is that the distribution condition meets the constraint condition, outputting a distribution line by taking the distribution network as a reference, and calculating the fitness value of each transfer center according to the position of each transfer center in the distribution network as a fitness value function;

judging whether the fitness value is a convergence solution;

if the fitness value is a convergence solution, taking a distribution line corresponding to the distribution network as an optimal distribution line;

if the fitness value is not a convergence solution, recording the fitness value, selecting a transfer center from the neighborhood of the transfer center in the combination to replace the transfer center in the combination through an iterative neighborhood search algorithm, and planning a corresponding distribution line for the replaced distribution network again;

if so, calculating the current fitness value of the transfer center;

judging whether the current fitness value is superior to the last fitness value or not, and judging whether the current fitness value is a convergence solution or not;

and if the current fitness value is superior to the last fitness value and is a convergence solution, outputting the re-planned distribution line as the optimal distribution line.

Optionally, in a fifth implementation manner of the first aspect of the present invention, the constraint condition includes at least one of:

the freight volume constraint conditions of the corresponding distribution tasks under various journey levels;

vehicle quantity constraint conditions corresponding to different freight volumes;

a difference constraint condition of a linear distance between a distribution starting point and a distribution end point and a total distance of a distribution line;

deciding the uniqueness constraint condition of the variable in the distribution process;

a delivery time window constraint;

loading limit constraints on the distribution lines;

transportation cost constraints.

Optionally, in a sixth implementation manner of the first aspect of the present invention, the determining whether the distribution condition satisfies a constraint condition in a pre-established mathematical model for logistics distribution path optimization includes:

judging whether the freight volume contained in the distribution condition meets the freight volume constraint condition in the constraint condition;

if yes, sequentially judging whether the number of corresponding vehicles, the uniqueness of decision variables, the time window, the loading limit, the transportation cost and the difference between the straight-line distance from the distribution starting point to the distribution end point in the new distribution line and the total distance of the distribution line in the distribution condition all meet the constraint condition in the mathematical model.

A second aspect of the present invention provides a logistics distribution route planning apparatus, including:

the system comprises an initial planning module, a distribution module and a scheduling module, wherein the initial planning module is used for planning an initial distribution line based on a distribution demand according to a distribution task when the distribution task is triggered, the initial distribution line comprises a line starting point, a line finishing point and N transfer centers, and N is a natural number;

the encoding module is used for encoding a transfer center in the initial distribution line to obtain an initial solution;

the genetic crossing module is used for combining the transfer centers in the initial solution based on a genetic crossing algorithm and planning a new distribution line based on the combination, the line starting point and the line ending point;

the calculation module is used for calculating the distribution conditions in the new distribution line;

the judging module is used for judging whether the distribution conditions meet the constraint conditions in a pre-established mathematical model for the logistics distribution path optimization;

and the adjustment output module is used for adjusting the new distribution line through neighborhood iterative search according to the judgment result until the adjustment meets the constraint condition, and outputting the optimal distribution line.

Optionally, in a first implementation manner of the second aspect of the present invention, the encoding module includes a decoding unit and an encoding unit, where:

the decoding unit is used for extracting gene individuals from the initial distribution line based on a genetic algorithm to obtain a plurality of transfer centers;

the encoding unit is used for sequentially encoding the extracted multiple transfer centers to form an initial solution, wherein the transfer centers in the initial solution and the transfer centers are in a neighborhood relationship.

Optionally, in a second implementation form of the second aspect of the invention, the genetic crossover module comprises a first genetic unit and a second genetic crossover unit, wherein:

the first genetic crossing unit is used for butting each transfer center in the initial solution with the line starting point and the line ending point respectively based on a genetic crossing algorithm to form a distribution network; planning a new distribution line based on the distribution network;

the second genetic crossing unit is used for randomly selecting one transfer center from the initial solution as a primary transfer center; combining the first transfer center with other transfer centers in the initial solution based on a genetic crossover algorithm to obtain at least one transfer center combination; the at least one transit center combination is respectively butted with the line starting point and the line ending point to form a distribution network; and planning a new distribution line based on the distribution network.

Optionally, in a third implementation manner of the second aspect of the present invention, the adjustment output module includes a first adjustment unit and a first output unit, where:

the first adjusting unit is used for selecting a transfer center from the neighborhood of the transfer centers in the combination to replace the transfer center in the combination through an iterative neighborhood search algorithm and planning a corresponding distribution line for the replaced distribution network when the judged result is that the distribution condition does not meet the constraint condition; judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

the first output unit is used for outputting the optimal distribution line by taking the replaced distribution network as a reference when the judgment result is that the optimal distribution line is met;

and the first adjusting unit is further used for continuing to select the next transfer center from the neighborhood to replan the distribution line when the first adjusting unit judges that the first adjusting unit is not satisfied.

Optionally, in a fourth implementation manner of the second aspect of the present invention, the adjustment output module further includes a second adjustment unit, a second output unit, and a convergence calculation unit, where:

the second adjusting unit is configured to output a distribution route based on the distribution network when the determined result is that the distribution condition satisfies the constraint condition, and calculate a fitness value of each relay center according to a fitness value function based on a position of each relay center in the distribution network; judging whether the fitness value is a convergence solution;

the second output unit is used for taking the distribution line corresponding to the distribution network as an optimal distribution line when the adaptability value is a convergence solution;

the second adjusting unit is further configured to record the fitness value when the fitness value is not a convergence solution, select a relay center from neighborhoods of relay centers in the combination to replace the relay center in the combination through an iterative neighborhood search algorithm, and plan a corresponding distribution line for the replaced distribution network again; judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

the convergence calculating unit is used for calculating the current fitness value of the transfer center when the judgment result is that the current fitness value is met; judging whether the current fitness value is superior to the last fitness value or not, and judging whether the current fitness value is a convergence solution or not;

and the second output unit is also used for outputting the re-planned distribution line as the optimal distribution line when the current fitness value is superior to the last fitness value and is a convergence solution.

Optionally, in a fifth implementation manner of the second aspect of the present invention, the constraint condition includes at least one of:

the difference constraint condition of the linear distance between the distribution starting point and the distribution terminal point and the total distance of the distribution line is set;

deciding a uniqueness constraint condition of a variable in a distribution process;

a delivery time window constraint;

loading limit constraints on the distribution lines;

transportation cost constraints.

Optionally, in a fifth implementation manner of the second aspect of the present invention, the determining module includes a first constraining unit and a second constraining unit, where:

the first constraint unit is used for judging whether the freight volume contained in the distribution condition meets the freight volume constraint condition in the constraint condition;

and the second constraint unit is used for sequentially judging whether the number of the corresponding vehicles, the uniqueness of the decision variable, the time window, the loading limit, the transportation cost and the difference between the straight-line distance between the distribution starting point and the distribution end point in the new distribution line and the total distance of the distribution line in the distribution condition meet the constraint condition in the mathematical model when the second constraint unit judges that the number of the corresponding vehicles, the uniqueness of the decision variable, the time window, the loading limit, the transportation cost and the difference between the straight-line distance between the distribution starting point and the distribution end point in the new distribution line meet the constraint condition in the mathematical model.

A third aspect of the present invention provides a logistics distribution route planning apparatus, including: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line; the at least one processor invokes the instructions in the memory to cause the logistics route planning apparatus to perform the logistics route planning method described above.

A fourth aspect of the present invention provides a computer-readable storage medium having stored therein a computer program which, when run on a processor on a computer, causes the computer to execute the logistics distribution route planning method described above.

According to the technical scheme provided by the invention, an initial distribution line is planned according to distribution tasks and distribution requirements, the initial distribution line is decoded to obtain an initial solution, a transfer center in the initial solution is subjected to cross combination by using a genetic cross algorithm, a new distribution line is re-planned, whether distribution conditions in the new distribution line meet constraint conditions of a mathematical model is judged, and the new distribution line is continuously adjusted according to a judgment result and neighborhood iterative search to obtain an optimal distribution line. The planning of the transfer center in the distribution line is continuously adjusted by calculating the mathematical relation between the distribution line and the constraint conditions of the mathematical model, specifically, the transfer center is continuously screened and replaced by adopting a genetic crossing algorithm and an iterative neighborhood search algorithm, so that the solution space can be increased, the optimal transfer center is selected to find the optimal logistics distribution route, the optimal logistics distribution route can be quickly found, the obtained route length and the total length are shortest, the logistics distribution cost is effectively reduced, and the method has a good application prospect.

Drawings

Fig. 1 is a schematic diagram of a first embodiment of a logistics distribution route planning method according to an embodiment of the invention;

fig. 2 is a schematic diagram of a second embodiment of the logistics distribution route planning method according to the embodiment of the invention;

FIG. 3 is a schematic diagram of a domain search iterative optimization circuit in an embodiment of the present invention;

fig. 4 is a schematic diagram of a third embodiment of the logistics distribution route planning method in the embodiment of the invention;

fig. 5 is a schematic diagram of a fourth embodiment of the logistics distribution route planning method according to the embodiment of the invention;

fig. 6 is a schematic diagram of a fifth embodiment of the logistics distribution route planning method according to the embodiment of the invention;

FIG. 7 is a flow chart of a domain search iterative optimization circuit in an embodiment of the present invention;

fig. 8 is a schematic diagram illustrating an example of a logistics distribution route planning according to an embodiment of the invention;

fig. 9 is a schematic diagram of a first embodiment of a logistics distribution route planning device according to an embodiment of the invention;

fig. 10 is a schematic diagram of a second embodiment of the logistics distribution route planning apparatus according to the embodiment of the invention;

fig. 11 is a schematic structural diagram of a logistics distribution route planning apparatus according to an embodiment of the invention.

Detailed Description

The embodiment of the invention provides a logistics distribution route planning method, a logistics distribution route planning device, logistics distribution route planning equipment and a storage medium, wherein, the method screens the transit centers selected by the intelligent algorithm through the constraint conditions in the mathematical modeling to select the optimal transit centers, to form an optimal distribution route, i.e. after obtaining a preliminary distribution route based on the input data, the initial distribution line is optimized and adjusted through the mathematical model and the intelligent algorithm to generate the optimal distribution line, and the processing mode based on the mathematical model and the intelligent algorithm can not only improve the logistics planning efficiency, but also realize the logistics economic benefit and the logistics scientization, moreover, the urban traffic pressure can be relieved, energy sources are saved, pollution is reduced, the inherent unification of the aspects of efficiency, resources, environment and value concepts is realized, and the progress of logistics industry and the sustainable development of social economy are promoted.

Furthermore, the construction of a transfer center in the logistics line can be realized through the planning mode, so that the high-efficiency utilization rate of the transfer center is ensured.

The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

For convenience of understanding, a specific flow of an embodiment of the present invention is described below, and reference is made to fig. 1, which is a flow chart of a logistics distribution route planning method according to the present invention. The order of the steps in the flow chart may be changed and some steps may be omitted according to different needs. The method is mainly applied to a logistics distribution system, the system comprises at least one merchant platform and a logistics distribution route planning device, the logistics distribution system is a system with logistics route planning and express tracking functions, each merchant platform generates a distribution task after receiving an order, and the distribution route is planned based on the distribution task, and the specific implementation flow is as follows:

101. when a distribution task is triggered, planning an initial distribution line based on a distribution demand according to the distribution task;

in this embodiment, the initial distribution route includes at least one transit center, it is understood that the initial distribution route includes a route starting point, a route ending point, and N transit centers, where N is a natural number, and when the initial distribution route is constructed, the initial distribution route may be specifically constructed according to an existing navigation technology, and of course, the navigation herein should refer to a logistics distribution navigation system capable of knowing all distribution centers in a logistics network of a logistics company below the logistics distribution navigation system, and specifically, the initial distribution route may be constructed in the following two ways in the construction process: one is sectional planning, that is, the distribution flow is divided into multiple sections for distribution according to the distance between a distribution starting point and a distribution end point, a separate route planning is carried out based on each section of the distribution starting point and the distribution end point, and then each section is connected in series to obtain a complete distribution route; the other is non-segmented planning, that is, the transportation section between the delivery starting point and the delivery ending point is directly planned, a transit center may exist or may not exist in the middle, and if the transit center exists, the transit center should be a check point for driver to trim or register a trip.

In practical applications, the relay center includes two types, which are task string points and non-task string points, but the relay center can be used as the relay center regardless of the existence or nonexistence of the task.

In this embodiment, when an initial distribution route is constructed, the specific implementation may be:

judging the cargo quantity of the distribution tasks by utilizing a pre-constructed mathematical model, and if the cargo quantity does not meet the requirement of a distribution route, selecting a transfer center with the transfer center as a task serial point to construct an initial distribution route; in practical application, the mathematical model is specifically established by the following way:

selecting a delivery off-road of a historical delivery record in a delivery network to train and learn, for example, a total number of M customer points are selected in a selected route, knowing that the position and the demand of each customer point i are qi, at most K vehicles can reach all the customer points from a delivery center, starting from the delivery center and finally returning to the delivery center, the maximum load of each vehicle h is Ph, wherein h is 1, 2.

102. Coding a transfer center in the initial distribution line to obtain an initial solution;

in this step, the initial solution includes a plurality of relay centers, where the relay centers also include an initial line point and an end point, and the relay centers are in a neighborhood relationship with each other, for example, there are 5 relay centers in the initial solution, which are respectively encoded as 1 to 5, and 2 to 5 relay centers are both neighborhoods of 1 relay center, and 1 and 3 to 5 relay centers are both neighborhoods of 2 relay centers.

In this step, during decoding, preferably, a genetic algorithm is used to decode the initial distribution route, and a transfer center in the initial distribution route is extracted to form a transfer station set; and then, neighborhood construction is carried out on a transfer center in the transfer station set, and the neighborhood can be constructed through a mathematical model when being constructed, for example, the mathematical model is solved through a chaotic particle swarm optimization algorithm according to the established mathematical model, the particle swarm scale n and relevant parameters are determined, the relevant parameters comprise an inertia weight factor, a learning factor and the maximum iteration number, and the particle swarm is used as the neighborhood of the transfer station set.

103. Combining the transit centers in the initial solution based on a genetic crossing algorithm, and planning a new distribution line based on the combination and the line starting point and the line ending point;

the method mainly comprises the steps of optimizing initial distribution lines, determining the neighborhood of each transfer center through coding of an initial solution, and then optimizing the lines based on the transfer centers in the field so as to select the transfer centers meeting constraint conditions in a mathematical model from the neighborhoods to optimize and plan the transfer centers in the initial distribution lines.

In the present embodiment, when the combined transfer center is considered as a transfer center without task, the transfer center is understood as a transfer station for replacing the transportation strategy, for example, replacing the transportation vehicle or adding the transportation vehicle, and even after sorting the goods by region, the goods are transported by separate vehicles.

In this embodiment, before planning a new distribution line, the method further includes calculating a fitness value of the initial distribution line, and determining whether the initial distribution line belongs to the optimal solution based on the fitness value, if so, directly ending the optimization process of the distribution line, otherwise, continuing to execute step 103.

In practical applications, the calculation of the fitness value can be achieved by:

calculating the fitness value of each transfer center according to the position of each transfer center in the initial distribution line as a fitness value function to obtain the total driving distance of the vehicle, checking whether the constraint condition in the mathematical module is met, and if so, keeping the distribution line; otherwise, the distribution line is deleted and step 103 is performed.

In addition, after the distribution lines are reserved, chaotic optimization is carried out on the global optimal extreme value, the global optimal extreme value is mapped to a domain of an equation, iteration is carried out according to a formula to generate a plurality of chaotic variable series, finally the chaotic variable series are returned to a value interval of an optimization variable through inverse mapping to obtain a plurality of particles, the fitness value of each particle is calculated to obtain an optimal solution, and the optimal solution replaces the position of a transfer center in the current distribution line; and judging whether the current distribution line is premature convergence, and if so, outputting the optimal distribution line.

104. Calculating distribution conditions in the new distribution line;

in this embodiment, the distribution condition is substantially a constraint condition in the mathematical model, and the transportation cost distributed according to the current route is determined based on the actual distribution cargo amount, the distribution time, the distribution route, and the like in the new distribution route by calculating these parameters.

105. Judging whether the distribution conditions meet constraint conditions in a pre-established mathematical model for the optimization of the logistics distribution path;

that is, the actual parameters in the new distribution line are compared with the constraint conditions corresponding to the mathematical model, and if the actual parameters are not greater than the constraint conditions, the distribution line can be used while satisfying the conditions, but is a suboptimal target line, and further optimization processing is required.

106. And adjusting the new distribution line through neighborhood iterative search according to the judgment result until the new distribution line is adjusted to meet the constraint condition, and outputting the optimal distribution line.

In this embodiment, the result of the determination includes two types, one is that the constraint condition is satisfied, and the other is that the constraint condition is not satisfied, and for the first type, the delivery line is directly output, but after the delivery line is output, the determination of re-constraint may be performed according to the constraint condition in the mathematical model, the optimal constraint condition is calculated based on the mathematical model, and if the optimal constraint condition is satisfied, the delivery line is output as the optimal delivery line.

In practical applications, when optimizing the distribution route, a comparison consideration may be made according to the priority in the constraint conditions in the mathematical model, for example: and evaluating the conditions in the path according to the sequence of time window > cargo capacity > distance > cost, and directly outputting the corresponding distribution lines after the evaluation is passed.

Through the implementation, the transfer center in the initial distribution line is optimized and evaluated based on the constraint conditions and the algorithm in the mathematical model, and the optimal transfer center is obtained to be replaced, so that the optimal distribution line is obtained.

In this embodiment, each step in the logistics distribution route planning method provided above is not implemented in a fixed order, and another execution manner may also exist, which also solves the above technical problem, in practical application, there may be one or more initial distribution routes that are constructed, and in case of a plurality of initial distribution routes, the following process may be implemented specifically, as shown in fig. 2:

201. planning a distribution line by using a preset distribution line planning rule based on the received distribution starting point and distribution end point to obtain an initial distribution line set;

in this step, the distribution route planning rule may be understood as a rule that, according to the route, the freight volume, and the cost constraint, in addition to this rule, a rule for screening the constructed route is included, and the screening route is shortest and satisfies the cost constraint.

In this embodiment, the initial distribution route set includes a plurality of initial distribution routes, one of which is selected as an initial optimal distribution route and at least one of which is selected as a candidate distribution route, where the initial optimal distribution route is shortest in route, and the distribution routes include at least one transit center, that is, the transit center includes at least one transit center in the middle, in addition to a distribution starting point and a distribution ending point, and the transit center may be a task cluster point or a station for replacing a transportation policy.

In practical applications, the planning of the initial distribution route includes two plans, one is the planning of the road on which the vehicle runs, and the other is the planning of the transportation strategy, such as what vehicle and how many vehicles need to be used for transportation when the vehicle does not pass through a transit center, and what the driving route of the vehicle needs to be used for transportation.

202. Decoding the at least one alternative distribution line, and generating a transfer station set comprising a plurality of transfer centers through a group intelligent algorithm;

in this embodiment, in addition to decoding the alternative delivery routes, the initial optimal delivery route is also decoded, and here, the initial optimal delivery route may also be considered to be automatically decoded, and since the alternative delivery routes exist here, when generating the neighborhood in the relay center in the initial optimal delivery route, the neighborhood can be obtained by directly decoding the alternative delivery routes, and optionally, the set of relay stations is a set of neighborhoods belonging to all the relay centers in the initial optimal delivery route.

203. Constructing a neighborhood from the transit station set by using a iterative neighborhood search algorithm by taking the transit center in the initial optimal distribution line as a reference, and calculating an optimal solution to obtain a corresponding transit center;

in this embodiment, when a neighborhood is constructed from the relay station set, the neighborhood may be constructed through a mathematical model, specifically, each time a corresponding relay center is searched from the relay station set, the mathematical model is used to compare constraint conditions of the mathematical model with the relay center, and the constraint conditions of the mathematical model of the relay center corresponding to the initial optimal distribution route are compared, if the conditions are met, an optimal solution is determined, after all searches are completed, a better solution is selected from the optimal solutions, and the relay center corresponding to the better solution is used as a substitute relay center.

204. Calculating the fitness value of the transfer center in the initial optimal distribution line;

the step is to place the transfer center in the initial optimal distribution line to construct a new distribution line, and then to perform adaptive evaluation on the new distribution line by using the constraint conditions of the mathematical model to obtain an adaptive value.

205. Judging whether the fitness value is a convergence solution;

in this step, if the solution is a convergence solution, step 206 is performed, otherwise, step 202 is returned to.

206. If so, replacing the transfer center with the corresponding transfer center in the initial optimal distribution line to form a final optimal distribution line;

207. checking and accepting the final optimal distribution line through a pre-established mathematical model for logistics distribution path optimization;

208. and if the acceptance passes, outputting the final optimal distribution line as the distribution line of the current express.

In this embodiment, in the process of calculating the fitness of the substitute relay center, the method specifically includes the following steps:

1. constructing an initial solution;

in this embodiment, when an initial solution is constructed, it is optional to use the transfer centers in the initial optimal distribution route as the initial solution, specifically, perform initial solution encoding based on a genetic algorithm, extract the transfer centers in the distribution route, and perform numbering combination, for example, when 64 transfer centers are obtained by encoding, 1 to 64 transfer centers respectively represent their combinations, for example, 3, 1, and 5 (respectively represent three route combinations of hangzhou, south jing, and china, and north river).

2. Carrying out genetic crossing on the obtained combination, and selecting a better transfer center;

this step, in order to prevent the occurrence of too many infeasible solutions, does not perform crossover between individuals in the conventional genetic algorithm. Designing point cross of the same sequence number position of adjacent gene segments is beneficial to generating more feasible solutions and better solutions, and the point cross considers that the possibility of combining different transit centers is increased.

3. Calculating the fitness of the better transit center by using a hybrid large-scale neighborhood search algorithm;

in the step, after genetic algorithms are crossed, one current chromosome Z is randomly selected, a large-scale neighborhood search algorithm is added, genes of other chromosomes are used as a selection set of the current chromosome to carry out iterative search, once the y gene position of any x chromosome is selected, the y gene code and a certain position of Z dyeing are replaced, the solution space is increased, the algorithm is prevented from entering local optimization, and the disturbance degree of the algorithm is increased. And calculating the current optimal target value once searching, and replacing if the current optimal target value is better than the current value. So that it produces more feasible solutions, as shown in particular in fig. 3.

Further, after the optimal distribution route is output, planning for displaying the distribution route is included, and the optimal distribution route can be compiled into codes for display through a webpage compiling technology.

In practical application, the path planning route display technology is specifically that the rear end of a main body adopts a c #. netcore programming language and is programmed in an object-oriented mode. The algorithm package is divided into an algorithm layer and a model layer, wherein the model layer is connection data and comprises various data inputs introduced above. The front end adopts ASP.

Furthermore, a navigation function can be formed for displaying, for example, a longitude and latitude file of all express delivery distribution routes can be returned by adopting an interface calling mode through c # webapi, then a map api is called by javascript and front-end programming, map navigation data is accessed, and an intelligent navigation function is provided.

In conclusion, by the implementation of the embodiment, the problem of intelligently planning the express delivery scheme, improving the delivery efficiency, effectively solving the cost control problem of the express company is solved, a scientific basis is provided for selecting a site of a transfer center of the express company as a reference model, a delivery route is displayed through a web in time through algorithm training, and a path is optimized.

Simultaneously, still realized wisdom planning express delivery route, optimized the circuit selection for the traffic is more smooth and easy, makes freight more convenient, faster, improves the quality of service of express delivery trade, improves resident's good experience of life.

Furthermore, site selection and path planning of a transit center can be made for the construction of a logistics network, which is the key of logistics cost control, good logistics site selection is a long-term strategy, and if no scientific and systematic planning is adopted to hasten selection, a lot of cost which is difficult to control is inevitably generated in the future; meanwhile, after address selection, the path of the newly-built allocated call to the main cities in the country needs to be scientifically planned, and a path planning scheme is more scientifically provided.

Referring to fig. 4, a server is taken as a logistics distribution route planning device for detailed description, and another embodiment of the logistics distribution route planning method in the embodiment of the present invention includes:

401. when a distribution task is triggered, planning an initial distribution line based on a distribution demand according to the distribution task;

402. based on a genetic algorithm, carrying out gene individual extraction processing on the initial distribution line to obtain a plurality of transfer centers;

403. sequentially encoding the extracted multiple transfer centers to form an initial solution;

404. based on a genetic crossing algorithm, each transfer center in the initial solution is respectively butted with the initial point and the final point of the line to form a distribution network;

405. planning a new distribution line based on the distribution network;

in this embodiment, a transfer center in an initial solution is randomly combined with the delivery starting point and the delivery ending point by a genetic crossover algorithm to obtain M line combinations; where M is an integer greater than or equal to 2, a line combination includes one or more neighborhoods, and a neighborhood here may be understood as another combination or another relay center.

406. Calculating distribution conditions in the new distribution line;

407. judging whether the distribution conditions meet constraint conditions in a pre-established mathematical model for the optimization of the logistics distribution path;

408. and adjusting the new distribution line through neighborhood iterative search according to the judgment result until the new distribution line is adjusted to meet the constraint condition, and outputting the optimal distribution line.

In this embodiment, in the process of determining whether the constraint condition is satisfied, the comparison consideration may be performed according to the priority in the constraint condition in the mathematical model, for example: and evaluating the conditions in the path according to the sequence of time window > cargo capacity > distance > cost, and directly outputting the corresponding distribution lines after the evaluation is passed.

In this embodiment, if the constraint condition is set to include at least one of the following:

a delivery time window constraint;

loading limit constraints on the distribution lines;

transportation cost constraints.

Further, the step of performing search calculation by the genetic cross algorithm and selecting the optimal combination in the neighborhood is specifically realized by the following method:

randomly selecting one combination from the combinations formed in the neighborhood as a main combination, and using other combinations as a selection set of the main combination;

calculating the fitness sub-value of a distribution line formed by the main combination, the distribution starting point and the distribution end point;

through iterative search, selecting a combination from the selection set to be replaced with the main combination or a certain section of the distribution line, and calculating a corresponding fitness sub-value;

the two fitness sub-values are compared and an optimal combination is selected based on the result of the comparison.

The same is true for selecting the optimal transit center from the initial solution.

For example, as shown in fig. 8, it is assumed that a logistics distribution route from kunming to shanghai needs to be planned, and in the initial planning, the initial distribution route includes a transit center with kunming, noble yang, Changsha, Nanchang, Hangzhou and shanghai;

further, in the initial solution of "kunming, noble yang, Changsha, Nanchang, Hangzhou, Shanghai", the noble yang is selected as the transfer center of the first optimized distribution line plan, then whether the constraint condition of the distribution line of "kunming-noble yang-Shanghai" meets the constraint condition in the mathematical model is calculated, if the constraint condition of the mathematical model is not met, the transfer center of the noble yang needs to be adjusted, the adjustment is to enlarge the planned path through genetic crossing and neighborhood search, the mathematical model is met in the process of enlarging the plan, for example, the "Changsha" is selected from neighua, Nanchang, Hangzhou "in the neighborhood of the transfer center of the noble yang" to replace the noble yang ", the replacement graph is shown in FIG. 3, the constraint condition of" Kunming-Changsha-Shanghai "is further calculated, and after the constraint condition is met, the fitness value is calculated, if the fitness value is a convergence solution, outputting 'Kunming-Changsha-Shanghai' as an optimal distribution line, if the fitness value is not the convergence solution, recording the fitness value, selecting from the rest neighborhoods, repeatedly calculating the constraint condition and the fitness value, and outputting the corresponding line until the constraint condition and the fitness value are met.

In this example, after the convergence solution is not obtained, the number of the transfer centers of the distribution lines may be increased by selecting one from the upper and lower neighborhoods, and the increased transfer center should have a distribution task, so that the distribution lines may have more serial points and low transportation cost.

In this embodiment, a specific process of selecting a relay center from an initial solution is described by taking the selection of a single relay center as an example:

firstly, randomly selecting a transfer center from an initial solution, combining the transfer center with an initial point and an end point to form a distribution line, calculating a constraint condition of the distribution line, comparing the constraint condition with a constraint condition specified in a mathematical model, and specifically judging whether the freight volume comprising the distribution line meets the freight volume constraint condition in the constraint condition;

if so, sequentially judging whether the number of corresponding vehicles in the distribution line, the uniqueness of the decision variable, the time window, the loading limit and the transportation cost all meet the constraint conditions in the mathematical model; if the two routes are satisfied, the current distribution route can be determined as a better distribution route, then based on the same method, a neighborhood transfer center of the current transfer center is selected from the initial solution to replace the current transfer center to continue calculation based on an iterative neighborhood search algorithm, and until the iterative search of the transfer center in the initial solution is finished, a distribution route with the best fitness value is selected from the multiple distribution routes to be the most optimal distribution route for output.

In practical application, the mathematical model is obtained by performing training modeling based on a distribution history record, specifically, M customer points are obtained in an existing distribution network, the position and the demand of each customer point i are known to be qi, at most K vehicles can reach all the customer points from a distribution center, each vehicle starts from the distribution center and finally returns to the distribution center, the maximum load of each vehicle h is Ph, wherein h is 1, 2,.. K, and a vehicle driving route is required to be arranged to minimize the total driving distance of the vehicle, and a constraint condition of the established mathematical model is increased, specifically as follows:

(A1) the location of the distribution center is known and unique;

(A2) the distribution center only has one vehicle type, and the requirement of each customer point can be completed by only one vehicle;

(A3) the sum of the customer demand on each line does not exceed the vehicle load capacity;

(A4) the total length of each distribution path is not more than the maximum distance of one-time distribution driving of the automobile;

besides the constraint conditions, the method can be expanded according to actual requirements, and a mathematical model is created based on the data to obtain a final mathematical model.

In the using process of the embodiment, specifically, according to a mathematical model, the mathematical model is solved through a chaotic particle swarm optimization algorithm, and the particle swarm scale n and relevant parameters are determined, wherein the relevant parameters comprise an inertia weight factor, a learning factor and the maximum iteration number; then, carrying out fitness evaluation on the selected particles, optionally carrying out coding and decoding on the particles to generate a vehicle distribution scheme, calculating a fitness function value of each particle according to the position of each customer point as a fitness function to obtain a total vehicle driving distance, checking whether a constraint condition in a formula is met, and if so, reserving the vehicle distribution scheme; otherwise, the vehicle delivery scenario is deleted.

Based on the implementation of the method, the transfer centers selected by the intelligent algorithm are screened through the constraint conditions in the mathematical modeling, the optimal transfer centers are selected to form the optimal distribution lines, namely, after the initial distribution lines are obtained based on the input data, the initial distribution lines are optimized and adjusted through the mathematical model and the intelligent algorithm to generate the optimal distribution lines, so that the optimal distribution lines of the logistics distribution route can be quickly found, the obtained total length of the routes is shortest, the logistics distribution cost is effectively reduced, and the method has a good application prospect.

Referring to fig. 5, another implementation flow in the method of the present invention is provided, where the flow is described in a manner of task sequence points, in this embodiment, at least two relay centers are selected from an initial solution to splice to form a distribution route, and in the following flow, a last relay center is selected to perform continuous neighborhood replacement to calculate an optimal route for explanation, but in practical application, a manner of integrally replacing multiple relay centers may also be selected to optimize calculation, and the implementation steps of the method include:

501. when a distribution task is triggered, planning an initial distribution line based on a distribution demand according to the distribution task;

502. coding the initial distribution line based on a genetic algorithm to form an initial solution;

503. selecting a transfer center from the initial solution as a primary transfer center;

504. combining the first transfer center with other transfer centers in the initial solution based on a genetic crossover algorithm to obtain at least one transfer center combination;

505. the at least one transit center combination is respectively butted with the line starting point and the line ending point to form a distribution network;

506. planning a new distribution line based on the distribution network;

507. Calculating distribution conditions in the new distribution line;

508. judging whether the distribution conditions meet constraint conditions in a pre-established mathematical model for logistics distribution path optimization or not;

in this step, if the determination result is satisfied, step 511 is executed, otherwise, step 509 is executed.

509. If not, selecting a transfer center from the neighborhood of the transfer centers in the combination to replace the transfer centers in the combination through an iterative neighborhood search algorithm, and planning corresponding distribution lines for the replaced distribution network again;

510. judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

in this step, if yes, step 511 is executed, otherwise, step 509 is returned to.

511. If yes, calculating the fitness value of each transfer center according to the position of each transfer center in the distribution network as a fitness value function;

512. judging whether the fitness value is a convergence solution;

in this step, if yes, step 517 is executed, otherwise, step 513 is executed.

513. If not, recording the fitness value, selecting a transfer center from the neighborhood of the transfer centers in the combination to replace the transfer centers in the combination through an iterative neighborhood search algorithm, and planning corresponding distribution lines for the replaced distribution network again;

514. judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

in this step, if yes, step 515 is performed, otherwise, step 517 is performed.

515. If so, calculating the current fitness value of the transfer center;

516. judging whether the current fitness value is superior to the last fitness value or not, and judging whether the current fitness value is a convergence solution or not;

in this step, if the current fitness value is better than the previous fitness value and is a convergence solution, step 517 is executed to output an optimal distribution line; if not, the process returns to step 513 to continue to select the next transit center from the neighborhood to re-plan the distribution line.

517. And taking the distribution line corresponding to the current distribution network as the optimal distribution line.

In the present embodiment, in the process of selecting the transit center from the initial solution, step 509, it may also plan a new delivery route by adding the transit center, so that the delivery route satisfies the constraint condition, and then continue to calculate the fitness value.

In practical application, the above process can be understood as three major processes, namely creating a mathematical model, planning and optimizing a distribution route, and displaying a navigation of the distribution route, and the following description will be made by combining a scenario and a service, where the process is shown in fig. 6:

601. carrying out mathematical model modeling according to scene and business requirement abstraction;

acquiring M customer points in an existing distribution network, knowing that the position and the demand of each customer point i are qi, at most K available vehicles arrive at all the customer points from a distribution center, each vehicle starts from the distribution center and finally returns to the distribution center, the maximum load of each vehicle h is Ph, wherein h is 1, 2.

a. The cargo quantity control constraint condition Q is that the cargo quantity is greater than 4930 kg to meet the requirement of dispatching a delivery task, and of course, the requirement of the round trip and the return trip are included, and the expressions are as follows (1) and (2):

wherein, o represents a distribution starting point, j is a task string point, and j' is a distribution end point;

in this embodiment, if the cargo volumes from the delivery starting point to the delivery ending point in the delivery task do not satisfy the requirements (1) and (2), it is necessary to select a relay center of a task having a task between the delivery starting point and the delivery ending point, that is, a task string point i, where the cargo volume constraint expressions are as follows (3) and (4):

wherein, the cargo capacity between any two task points is more than 1000 kg, namely the expression (5):

b. the vehicle control constraint condition V, which is a vehicle load amount constraint, is as follows expressions (6) and (7):

wherein, V₅₆，V₁₃₀Is a 0, 1 variable;

c. the distance constraint D, which is understood to mean that the total length of each delivery route is not greater than the maximum distance traveled by a single delivery of the automobile, is expressed by the following expressions (8) to (10):

D_oi+D_ij-D_oj＜＝200 (8)

D_ij'+D_jo-D_io＜＝200 (9)，

the round-trip constraint is:

D_oij＜＝300 (10)，

wherein D is_oj＞D_ij,D_oj＞D_oi，D_io＞D_jo,D_io＞D_ij'；

d. Decision variable uniqueness constraints, which are divided into a go-way and a return-way, are expressed as (11) and (12):

going to the journey:

and (3) returning:

wherein, X_oj，X_oij，X_oij，X_jo'，X_ijo，X_oijAll variables are 0 and 1, o represents a distribution starting point, j is a task string point, and j' is a distribution terminal point;

e. a time window constraint T comprising the transit time plus the clear time control and delivery time control of the operational time cap phishing allotment, as in expressions (13) and (14), respectively:

f. the stowage constraint condition that the cargo volume is smaller than the loading limit of the stowage car insurance, and the stowage car line also meets the time window constraint condition that the stowage car line reaches the stowage terminal point before the clearing time, are respectively expressed as (15) and (16):

X_oij*D_oij＜line^k _limit (15)

wherein, k, X_oijAre all 0, 1 variables;

g. the index constraint is finally completed, as expressed by (17- (20):

cost:

when in use:

the number of vehicles used is as follows:

task string points:

in this embodiment, the constraint condition may be expanded according to actual requirements, and a mathematical model may be created based on the data to obtain a final mathematical model.

602. Planning a first distribution route according to the distribution tasks;

603. coding a transfer center on the first distribution route, and calculating an optimal substitute transfer center to obtain an optimal distribution route;

in this embodiment, in the process of optimization calculation, a group intelligence algorithm and an iterative domain search hybrid algorithm are used for execution, and a specific optimization flow is as follows:

1. constructing an initial solution;

in this embodiment, when an initial solution is constructed, it is optional to use the transfer centers in the first optimal distribution line as the initial solution, specifically, perform initial solution encoding based on a genetic algorithm, extract the transfer centers in the distribution line, and perform numbering combination, for example, when 64 transfer centers are obtained by encoding, 1 to 64 transfer centers respectively represent their combinations, for example, 3, 1, and 5 (respectively represent three route combinations of hangzhou, south jing, and china, and north river).

3. Calculating the fitness of the better transfer center by using a hybrid large-scale neighborhood search algorithm;

in the step, after genetic algorithms are crossed, one chromosome Z is randomly selected, a large-scale neighborhood search algorithm is added, genes of other chromosomes are used as a selection set of the current chromosome to carry out iterative search, once a y gene position of any x chromosome is selected, a y gene code and a certain position of Z dyeing are replaced, a solution space is increased, the algorithm is prevented from entering local optimum, and the disturbance degree of the algorithm is increased. And calculating the current optimal target value once searching, and replacing if the current optimal target value is better than the current value. So that it produces more feasible solutions, as shown in particular in fig. 3.

604. And compiling and displaying the output optimal distribution line.

In this step, the search presentation process includes planning presentation of the route and navigation presentation of the route, which are respectively implemented as follows:

1) the intelligent path planning route display technology comprises the following steps: the back end of the main body adopts c #. netcore programming language and is programmed in an object-oriented mode. The algorithm package is divided into an algorithm layer and a model layer, wherein the model layer is connection data and comprises various data inputs introduced above. The front end adopts ASP.

2) The intelligent navigation function: and returning a longitude and latitude file of all express delivery distribution routes by adopting an interface calling mode through the c # webapi, and then calling the map api by front-end programming through javascript to access map navigation data and provide an intelligent navigation function.

In this embodiment, when performing the neighborhood search in step 603, the specific implementation is as shown in fig. 7:

701. generating an initial population based on a given rule;

in this step, the given rule may be understood as a construction rule and a decomposition algorithm of the initial distribution route, initial planning of the route is performed based on the construction rule, and a neighborhood of a relay center in the initial distribution route, that is, an initial population, is determined according to a mathematical model.

702. Judging whether the individuals in the population meet the requirements or not;

in this embodiment, the individual refers to whether the relay center in the neighborhood meets the requirement of the constraint condition of the mathematical model, if not, step 5033 is executed, otherwise, step 5304 is executed.

703. Generating a new individual;

the new individual specifically searches the iterative transfer center through an iterative neighborhood search algorithm to obtain a new transfer center nearby.

704. Calculating the initial fitness value of the individual, and recording the optimal initial fitness value;

705. judging whether the initial fitness value is a convergence solution or not;

706. if not, selecting other individuals from the population to form a new distribution line;

707. selecting the difference and the same route section between the new distribution route and the previous distribution route through a genetic crossing algorithm;

708. performing neighborhood variation processing on the differential part, selecting a proper transfer center, and calculating a corresponding fitness value;

709. judging whether the fitness value is an optimal solution or not;

in this step, if yes, step 710 is executed, otherwise, step 706 is returned to.

710. And taking the individual corresponding to the optimal solution as an initial solution, continuously adopting an iterative neighborhood search algorithm to select a better transfer center from the neighborhood corresponding to the current individual to process the optimal solution, and outputting the optimal distribution line until all the neighborhoods are calculated.

Through the steps, the initial distribution route is optimized and adjusted through the mathematical model and the intelligent algorithm to generate the optimal distribution route, the optimal distribution route of the logistics distribution route can be quickly found, the obtained route length is shortest, the logistics distribution cost is effectively reduced, and the method has a good application prospect.

With reference to fig. 9, the method for planning logistics distribution routes in an embodiment of the present invention is described above, and a logistics distribution route planning apparatus in an embodiment of the present invention is described below, where the apparatus in an embodiment of the present invention includes:

the initial planning module 91 is configured to plan an initial distribution line based on a distribution demand according to a distribution task when the distribution task is triggered, where the initial distribution line includes a line starting point, a line ending point, and N relay centers, where N is a natural number;

the encoding module 92 is configured to encode a transfer center in the initial distribution route to obtain an initial solution;

a genetic crossover module 93, configured to combine the transit centers in the initial solution based on a genetic crossover algorithm, and plan a new distribution route based on the combination and the route starting point and end point;

a calculation module 94 for calculating the distribution conditions in the new distribution line;

a judging module 95, configured to judge whether the distribution condition meets a constraint condition in a pre-established mathematical model for logistics distribution path optimization;

and the adjustment output module 96 is used for adjusting the new distribution line through neighborhood iterative search according to the judgment result until the adjustment meets the constraint condition, and outputting the optimal distribution line.

By the aid of the device, the transfer centers in the adjusting process are screened by adopting a neighborhood search algorithm and constraint conditions in a mathematical model, the most suitable transfer centers are selected to find the optimal logistics distribution route, the optimal logistics distribution route can be quickly found, the obtained route length is shortest, the logistics distribution cost is effectively reduced, and the device has a good application prospect.

Based on the same description of the embodiments as the logistics distribution route planning method of the present invention, the contents of the embodiment of the logistics distribution route planning apparatus are not described in detail in this embodiment.

Referring to fig. 10, another embodiment of a logistics distribution route planning apparatus according to an embodiment of the present invention includes:

a calculation module 94, configured to calculate a distribution condition in the new distribution route;

a judging module 95, configured to judge whether the distribution condition satisfies a constraint condition in a pre-established mathematical model for logistics distribution path optimization;

Optionally, the encoding module 92 includes a decoding unit 921 and an encoding unit 922, where:

the decoding unit 921 is configured to perform extraction processing of gene individuals on the initial distribution route based on a genetic algorithm, so as to obtain a plurality of transfer centers;

the encoding unit 922 is configured to sequentially encode the extracted multiple relay centers to form an initial solution, where the relay centers in the initial solution and the relay centers are in a neighborhood relationship.

Optionally, the genetic crossover module 93 comprises a first genetic unit 931 and a second genetic crossover unit 932, wherein:

the first genetic crossing unit 931 is configured to dock each transit center in the initial solution with the line origin and destination, respectively, based on a genetic crossing algorithm, to form a distribution network; planning a new distribution line based on the distribution network;

the second genetic crossover unit 932 is configured to arbitrarily select one of the hubs from the initial solution as a primary hub; combining the first transfer center with other transfer centers in the initial solution based on a genetic crossover algorithm to obtain at least one transfer center combination; the at least one transit center combination is respectively butted with the line starting point and the line ending point to form a distribution network; and planning a new distribution line based on the distribution network.

Optionally, the adjustment output module 96 includes a first adjustment unit 961 and a first output unit 962, where:

the first adjusting unit 961 is configured to, when the determined result is that the distribution condition does not satisfy the constraint condition, select a relay center from a neighborhood of the relay centers in the combination to replace the relay center in the combination through an iterative neighborhood search algorithm, and plan a corresponding distribution line for the replaced distribution network again; judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

the first output unit 962 is configured to output the optimal distribution line based on the replaced distribution network when the determination result is that the optimal distribution line is satisfied;

the first adjusting unit 961 is further configured to, when the determination result is that the distribution route is not satisfied, continue to select a next transfer center from the neighborhood to re-plan the distribution route.

Optionally, the adjustment output module 96 further includes a second adjustment unit 963, a second output unit 964, and a convergence calculation unit 965, where:

the second adjusting unit 963 is configured to, when the determined result is that the distribution condition satisfies the constraint condition, output a distribution route with the distribution network as a reference, and calculate a fitness value of each relay center according to a function of the fitness value of a position of each relay center in the distribution network; judging whether the fitness value is a convergence solution;

the second output unit 964 is configured to use the distribution line corresponding to the distribution network as an optimal distribution line when the fitness value is a convergence solution;

the second adjusting unit 963 is further configured to record the fitness value when the fitness value is not a convergence solution, select a relay center from a neighborhood of the relay centers in the combination to replace the relay center in the combination through an iterative neighborhood search algorithm, and plan a corresponding distribution line for the replaced distribution network again; judging whether the re-planned distribution line meets the constraint condition in the mathematical model;

the convergence calculating unit 965 is configured to calculate a current fitness value of the transfer center when the determination is that the current fitness value is satisfied; judging whether the current fitness value is superior to the last fitness value or not, and judging whether the current fitness value is a convergence solution or not;

the second output unit 964 is further configured to output the re-planned distribution line as the optimal distribution line when the current fitness value is better than the previous fitness value and is a convergence solution.

Optionally, the constraint condition includes at least one of:

a delivery time window constraint;

loading limit constraints on the distribution lines;

transportation cost constraints.

Optionally, the determining module 95 includes a first constraint unit 951 and a second constraint unit 952, where:

the first constraint unit 951 is configured to determine whether a shipment amount included in the delivery condition satisfies a shipment amount constraint condition in the constraint condition;

the second constraint unit 952 is configured to, when the determination result is that the constraint condition is satisfied, sequentially determine whether the number of vehicles, uniqueness of decision variables, time window, loading limit, transportation cost, and a difference between a straight-line distance from a delivery starting point to a delivery ending point in the new delivery route and a total distance of the delivery route in the delivery condition all satisfy the constraint condition in the mathematical model.

In practical applications, fig. 9 to 10 describe the logistics distribution route planning apparatus in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the logistics distribution route planning apparatus in the embodiment of the present invention is described in detail from the perspective of hardware processing.

Fig. 11 is a schematic structural diagram of an entity apparatus of a logistics distribution route planning device provided by the logistics distribution route planning method of the present invention, where the logistics distribution route planning device 2000 may generate a relatively large difference according to different actual needs, actual configurations, or performances, and may include, for example, one or more processors (CPUs) 2010 (e.g., one or more processors) and a memory 2020, and one or more storage media 2030 (e.g., one or more mass storage devices) storing an application program 2033 or data 2032. The storage mode used by the memory 1020 and the storage medium 2030 may be a transient storage or a persistent storage. The program stored in the storage medium 2030 may include modules (not shown) of the functionality provided by one or more embodiments, and each module may include a series of instructions operating on the logistics distribution route planning apparatus 2000. Further, the processor 2010 may be configured to communicate with the storage medium 2030, and execute a series of instruction operations in the storage medium 2030 on the logistics distribution route planning apparatus 2000, where the series of instructions correspondingly implement the functions of the logistics distribution route planning method provided in the foregoing embodiment.

The logistics distribution route planning apparatus 2000 may also include one or more power supplies 2040, one or more wired or wireless network interfaces 2050, one or more input-output interfaces 2060, and/or one or more operating systems 2031, such as Windows server, Mac OS X, Unix, Linux, FreeBSD, and the like. Those skilled in the art will appreciate that the structure of the logistics route planning apparatus shown in fig. 11 does not constitute the only limitation of the logistics route planning apparatus, and in practical applications it may also comprise more or less components than those shown, or some components may be combined, or a different arrangement of components.

The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and may also be a volatile computer-readable storage medium, where a computer program (i.e., instructions) is stored, and when the instructions are executed on a computer, the instructions cause the computer to execute the steps of the logistics distribution route planning method, and optionally, the computer program is executed by a processor on the computer.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A logistics distribution route planning method is characterized by comprising the following steps:

when a distribution task is triggered, judging the cargo quantity of the distribution task by using a pre-constructed mathematical model, and if the cargo quantity does not meet the requirement of a distribution route, selecting a transfer center as a transfer center of a task string point to construct an initial distribution route, wherein the initial distribution route comprises a route starting point, a route ending point and N transfer centers, N is a natural number, and the mathematical model is obtained by training and learning the distribution route in the historical distribution record of a distribution network;

extracting transfer centers in the initial distribution lines to form a transfer station set, constructing neighborhood relations among the transfer centers in the transfer station set, and coding based on the neighborhood relations to obtain an initial solution;

calculating distribution conditions in the new distribution line;

judging whether the distribution conditions meet constraint conditions in a pre-established mathematical model for logistics distribution path optimization or not;

2. The logistics distribution route planning method of claim 1, wherein the combining the transit centers in the initial solution based on the genetic crossover algorithm and planning a new distribution route based on the combination and the starting point and the ending point of the route comprises:

alternatively, the first and second electrodes may be,

3. The logistics distribution route planning method of claim 2, wherein the adjusting the new distribution route through neighborhood iterative search according to the result of the judgment until the adjustment meets the constraint condition, and outputting an optimal distribution route comprises:

if yes, outputting the optimal distribution line by taking the replaced distribution network as a reference;

4. The logistics distribution route planning method of claim 2, wherein the adjusting the new distribution route through neighborhood iterative search according to the result of the judgment until the adjustment meets the constraint condition, and outputting an optimal distribution route comprises:

judging whether the fitness value is a convergence solution;

if the fitness value is not a convergence solution, recording the fitness value, selecting a transfer center from the neighborhood of the transfer centers in the combination to replace the transfer center in the combination through an iterative neighborhood search algorithm, and planning a corresponding distribution line for the replaced distribution network again;

if so, calculating the current fitness value of the transfer center;

5. The logistics distribution route planning method of any one of claims 1-4, wherein the constraints comprise at least one of:

the freight volume constraint conditions of the corresponding distribution tasks under various route levels;

a delivery time window constraint;

loading limit constraints on the distribution lines;

transportation cost constraints.

6. The logistics distribution route planning method of claim 5, wherein the determining whether the distribution conditions satisfy the constraint conditions in the pre-established mathematical model of logistics distribution path optimization comprises:

and if so, sequentially judging whether the number of the corresponding vehicles in the distribution conditions, the uniqueness of the decision variable, the time window, the loading limit, the transportation cost and the difference between the straight-line distance from the distribution starting point to the distribution end point in the new distribution line and the total distance of the distribution line meet the constraint conditions in the mathematical model.

7. A logistics distribution route planning apparatus, characterized in that the logistics distribution route planning apparatus comprises:

the system comprises an initial planning module, a distribution network management module and a data processing module, wherein the initial planning module is used for judging the cargo quantity of a distribution task by using a pre-constructed mathematical model when the distribution task is triggered, and selecting a transfer center as a transfer center of a task string point to construct an initial distribution line if the cargo quantity does not meet the requirement of a distribution route, the initial distribution line comprises a line starting point, a line finishing point and N transfer centers, N is a natural number, and the mathematical model is obtained by training and learning the distribution line in the historical distribution record of the distribution network;

the encoding module is used for extracting the transfer centers in the initial distribution line to form a transfer station set, constructing neighborhood relations among the transfer centers in the transfer station set, and encoding based on the neighborhood relations to obtain an initial solution;

8. A logistics distribution route planning apparatus, characterized in that the logistics distribution route planning apparatus comprises: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line;

the at least one processor invokes the instructions in the memory to cause the logistics route planning apparatus to perform the logistics route planning method of any of claims 1-6.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the logistics distribution route planning method of any one of claims 1-6.