Optimization method and system for solving shortest logistics distribution path through cloud computing

Abstract

The invention discloses an optimization method for solving the shortest path of logistics distribution through cloud computing, and the method comprises the steps: obtaining logistics address information and vehicle information, enabling the logistics address information to comprise a distribution center and a delivery place, taking the body address of the distribution center as a starting region, determining a plurality of collection regions according to the delivery place, and carrying out the collection of the collection regions; obtaining a plurality of corresponding collection area specific addresses at the same time; performing area division according to the specific addresses of the plurality of collection areas to obtain a plurality of collection division areas; carrying out logistics path planning on the starting area and the plurality of collection division areas to generate a plurality of collection transportation paths, and optimizing the plurality of collection transportation paths to obtain a determined path; and collecting the weight of delivered goods in the determined path, and matching a proper transport vehicle according to the weight of the delivered goods to arrange a transport task. According to the invention, the distribution time and cost of a logistics enterprise can be reduced, the vehicle path planning efficiency of the logistics enterprise can be improved, the logistics transportation distance can be shortened to the maximum extent, and the logistics transportation efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a method and system for solving the shortest path optimization problem of logistics distribution through cloud computing. Background Art

[0002] Distribution is an important part of logistics operations, which is related to the operational efficiency of the entire logistics system and customer service satisfaction. Logistics distribution route optimization has always been the focus of corporate research.

[0003] In logistics distribution activities, the optimization of logistics distribution paths is a key part of the optimization of logistics distribution systems. With the increasing complexity of distribution network and the increasing distribution costs, how to plan reasonable distribution routes in logistics distribution, avoid roundabout transportation and repeated transportation, so as to further save distribution costs, reduce logistics costs, and improve the efficiency and economic benefits of logistics distribution has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a cloud computing solution to the shortest path optimization method and system for logistics distribution, so as to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: Cloud computing solves the shortest path optimization method for logistics distribution, including: Obtaining logistics address information and vehicle information, the logistics address information includes the distribution center and the delivery location, taking the distribution center address as the starting area, determining several collection areas according to the delivery location, and obtaining several corresponding specific addresses of the collection areas; Divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; Carry out logistics route planning for the starting area and multiple collection areas, generate multiple collection and transportation routes, and optimize the multiple collection and transportation routes to obtain a determined route; Collect and determine the weight of the goods to be delivered along the route, and arrange the transportation task according to the weight of the goods to be delivered and the appropriate transport vehicle.

[0006] Further, according to the delivery location, a number of collection areas are determined, and the specific addresses of a number of corresponding collection areas are obtained, which specifically includes: Obtain logistics address information and preset collection area information; Screening the collection area according to the logistics address information to determine a number of collection areas; According to the collection area information, specific addresses of the collection areas corresponding to the plurality of collection areas are obtained.

[0007] Furthermore, the regions are divided according to the specific addresses of the several collection areas, and the multiple collection areas are specifically: Determine the collection area according to the specific addresses of several collection areas; According to the preset allocation standard information, the receiving location is allocated and the collection allocation data is generated; According to the agency collection allocation data, the agency collection area is divided to obtain a plurality of agency collection divided areas.

[0008] Furthermore, logistics path planning is performed for the starting area and multiple collection areas, multiple collection transportation paths are generated, and optimization of the multiple collection transportation paths to obtain a determined path specifically includes: The starting area is recorded as the starting transportation node, and each collection area is recorded as the terminal transportation node. The distance between the starting transportation node and each terminal transportation node is defined as the direct transportation mileage; The road resistance function is used to describe the relationship between the travel time from the starting transport node to the end transport node and the section flow and maximum capacity of the through transport mileage; According to the observed through transport mileage data and the calculated travel time data, the travel matrix and travel time matrix are established respectively; According to the direct transport mileage and mutual relationship, the transport mileage matrix between adjacent nodes is calculated and constructed. The adjacency matrix in the transport mileage matrix is ​​used as the weight matrix. The weight matrix is ​​inserted several times to construct several shortest transport mileage matrices in turn. The shortest transport mileage matrix is ​​used to find the shortest transport path between any adjacent starting transport node and each end transport node.

[0009] Furthermore, it is determined whether the through transport mileage is blocked. If so, optimization conditions are introduced to optimize the indirect mileage between the starting transport node and each end transport node.

[0010] Furthermore, the indirect mileage is transport marked or monitored to generate transport monitoring results.

[0011] Furthermore, load constraints and quantity constraints are imposed on the transport vehicles, and the number of transport vehicles or the number of transport times used for the weight of the delivered goods is calculated.

[0012] To achieve the above object, the present invention also provides the following technical solutions: The cloud computing solution for the shortest path optimization system for logistics distribution includes an acquisition module, a division module, an optimization module, and a distribution module, among which: An acquisition module is used to acquire logistics address information and vehicle information. The logistics address information includes a distribution center and a receiving location. The distribution center address is used as the starting area. According to the receiving location, several collection areas are determined, and the specific addresses of several corresponding collection areas are acquired at the same time. A division module is used to divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; The optimization module is used to plan the logistics path for the starting area and multiple collection areas, generate multiple collection and transportation paths, and optimize the multiple collection and transportation paths to obtain a determined path; The distribution module is used to collect the weight of the goods to be delivered along the determined route, and to match the appropriate transport vehicles according to the weight of the goods to be delivered and arrange the transportation tasks.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention performs logistics path planning for a starting area and multiple collection areas, generates multiple collection and transportation paths, optimizes the multiple collection and transportation paths to obtain a determined path, and finds the optimal value of a feasible solution for the logistics and transportation path as the determined path from aspects such as path planning. This can reduce the delivery time and cost of logistics companies while improving the path planning efficiency of their vehicles, and can shorten the logistics and transportation distance to the greatest extent, thereby improving the logistics and transportation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a schematic diagram of the process of solving the shortest path optimization method for logistics distribution by cloud computing of the present invention.

[0015] Figure 2 This is a block diagram of the cloud computing solution for the shortest path optimization system for logistics distribution in the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 to Figure 2 , the present invention provides a technical solution: like Figure 1 As shown in the figure, the cloud computing solution to the shortest path optimization method for logistics distribution includes: Obtaining logistics address information and vehicle information, the logistics address information includes the distribution center and the delivery location, taking the distribution center address as the starting area, determining several collection areas according to the delivery location, and obtaining several corresponding specific addresses of the collection areas; Divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; Carry out logistics route planning for the starting area and multiple collection areas, generate multiple collection and transportation routes, and optimize the multiple collection and transportation routes to obtain a determined route; Collect and determine the weight of the goods to be delivered along the route, and arrange the transportation task according to the weight of the goods to be delivered and the appropriate transport vehicle.

[0018] Specifically, according to the delivery location, several collection areas are determined, and at the same time, several corresponding specific addresses of the collection areas are obtained, which specifically includes: Obtain logistics address information and preset collection area information; Screening the collection area according to the logistics address information to determine a number of collection areas; According to the collection area information, specific addresses of the collection areas corresponding to the plurality of collection areas are obtained.

[0019] Specifically, the regions are divided according to the specific addresses of several collection areas, and the multiple collection areas include: Determine the collection area according to the specific addresses of several collection areas; According to the preset allocation standard information, the receiving location is allocated and the collection allocation data is generated; According to the agency collection allocation data, the agency collection area is divided to obtain a plurality of agency collection divided areas.

[0020] When the goods enter the distribution station, the labels on the goods are scanned to obtain the corresponding receipt information of each goods, and then the receipt information of all goods is recorded and organized to obtain the logistics delivery address data, and the collection area information of several preset collection areas within the distribution range corresponding to the distribution center is obtained. Through comprehensive analysis of the delivery address data and the collection area information, the preset collection areas that do not need to be reached for this distribution and transportation are removed, and the remaining preset collection areas are screened and marked as collection areas for receipt, and according to the collection area information, the specific addresses of the collection areas corresponding to the several collection areas are obtained.

[0021] According to the specific addresses of the collection areas corresponding to the several collection areas, the collection areas are marked in the map to determine the collection areas, and according to the preset allocation standard information, multiple collection locations are allocated so that the multiple collection locations allocated together are adjacent collection points, and the collection allocation data is generated, and then the collection area is divided according to the collection allocation data to obtain multiple collection divided areas. For example: the preset allocation standard information can be "8 collection locations are grouped as a group for allocation processing", then multiple collection locations are allocated as a group of 8 adjacent collection locations, and the number of collection locations in the last allocated group can be less than 8.

[0022] Specifically, logistics path planning is performed for the starting area and multiple collection areas, multiple collection and transportation paths are generated, and the optimization of the multiple collection and transportation paths to obtain a determined path specifically includes: The starting area is recorded as the starting transport node i, and the corresponding collection area is recorded as the terminal transport node j (including j1, j2, j3...). The distance between the starting transport node i and each terminal transport node j is defined as the direct transport mileage d ij ; Through transport mileage ij The calculation is: (1) in, It means that (with other conditions unchanged in the transportation task) the delivery distance between the starting transportation node i and each end transportation node j increases by 1%, and the distance between i and j ij Increase accordingly . is the time coefficient. (When other conditions in the transportation task remain unchanged), the delivery time increases by 1%, d ij Increase( )%. is an adjustable parameter, and is greater than zero and less than one; θ is the transport vehicle performance index, and θ is greater than zero and less than one. The larger θ is, the better the transport vehicle performance is, and vice versa.

[0023] When logistics companies pay more attention to saving mileage costs than saving time costs, Greater than 0.5, accordingly, when logistics companies pay more attention to saving time costs than saving mileage costs, Less than 0.5; t ij It represents the time actually required for the transport vehicle to pass through the section between i and j.

[0024] If the transport vehicle encounters traffic congestion during the transportation task, the time used to pass the road section will be longer than the free driving time (smooth road). Therefore, the road resistance function is used to describe the relationship between the travel time from the starting transportation node to the end transportation node and the section flow and maximum traffic capacity of the direct transportation mileage, which is expressed as: (2) Among them, α and β are parameters to be calibrated, and the general values ​​are α is 0.15 and β is 4.0; t ij 0 represents the free travel time of the transport vehicle through the road section between i and j; Q ij (t) is the traffic volume of the road section between i and j; C ijis the traffic capacity of the road section between i and j, and Q ij (t) has the following relationship: (3) According to the observed through transport mileage data and the calculated travel time data, the travel matrix S and travel time matrix T are established respectively. S and T are expressed as follows:

[0025] (4) Among them, S ij =S ji , t ij =t ji , given appropriate parameters, according to equations (1) and (2), the transportation mileage matrix D between adjacent i and each j is calculated and constructed.

[0026] (5) Among them, d ij =d ji , the reduction of delivery time and distance will produce benefits in an exponential form, measured by direct transportation mileage data, and the goal of route optimization is to minimize the direct transportation mileage between the distribution center and any distribution site.

[0027] Taking the effective distance as the weight, the optimal path between a distribution center and multiple collection areas should be solved, and the direct transportation mileage from the distribution center to each collection area should be calculated to determine the path. The detailed steps are as follows: Using the adjacency matrix in formula (5) as the weight matrix, insert the vertex v times and construct v shortest transport mileage matrices in turn. The matrix is ​​expressed as follows: , in , It is the minimum transport mileage in the path from i to j that only allows 1 as the intermediate point; , in , It is the minimum transport mileage in the path from i to j that only allows 2 as the intermediate point; Insert vertices one by one according to the above method until all vertices are inserted and calculated: , in , , is the minimum transport mileage in the path from i to j that only allows 1, 2, ..., v as intermediate points, that is, the minimum transport mileage in the path from i to j where any vertex can be inserted, so E (v) is the shortest effective distance matrix between i and j, which can be obtained from E (v) To find the shortest transportation distance between any two points (i and j1; i and j2; i and j3...).

[0028] As an optional but not necessary solution: To improve The minimum transportation mileage in the path is obtained during the insertion calculation process, and the improved ant algorithm is used for optimization. The specific method is as follows: The optimal solution among multiple global collection and transportation routes is regarded as U ant, the second-best solution is regarded as G ant, the third-best solution is regarded as H ant, and the other candidate solutions are regarded as B ants. The optimization of the ant algorithm is a process of continuously iterating feasible solutions and gradually leaning towards the optimal solution, which is expressed as follows:

[0029]

[0030]

[0031]

[0032]

[0033] Where: DD is the distance between the ant colony individual and the prey (shortest transportation distance); CC is the corresponding coefficient vector; XX ▽,tt is the position of the prey at the ttth generation; XX tt is the position of the ant in the ttth generation; tt is the current iteration number; XX tt+1 is the updated position of the ant as the number of iterations increases; AA is the coefficient vector; is the convergence factor, which decreases linearly from 2 to 0 as the number of iterations increases; r1 and r2 are both random vectors between [0,1].

[0034] The ant colony has the ability to identify the location of marked prey and capture them. Once the ant colony identifies the location of the prey, the ant colony will surround the prey under the leadership of U ants, G ants and H ants. In order to simulate the encirclement process of ants, the present invention assumes that the higher the leadership level of U ants, G ants and H ants, the stronger their ability to identify the location of marked prey. In each iteration process, U ants, G ants and H ants are retained, and the location information of other ants in the ant colony is updated based on the location information of the above-mentioned dominant ants. The mathematical model of ant encirclement is constructed as follows:

[0035]

[0036]

[0037] Where: DD U ,DD G and DD H The distances between U ant, G ant and H ant and other ants respectively; CC1, CC2 and CC3 are random coefficient vectors; XX1, XX2 and XX3 are the current positions of U ant, G ant and H ant respectively; AA1, AA2 and AA3 are coefficient vectors; XX is the final position of B ant. Determine whether the iteration termination condition is met. If so, output the last updated position of U ant as the optimal solution. If not, return to continue iteration.

[0038] While obtaining the transport mileage matrix, establish the transport mileage matrix , w ij Indicates that the shortest path from i to j goes through w ij Point (w ij is also the distance weight of i and j): initial value , , in each case a transport mileage matrix E is obtained (k) When , the corresponding new path matrix is ​​generated as follows : , when When , point k is a point passed through the path from i to j, it will be recorded in the matrix W (k) middle; when When , point k is not a point passed by the path from i to j. Insert other points to calculate and verify whether they are points passed by the path from i to j.

[0039] In order to obtain E (V) The transport mileage matrix W is obtained in the process (V) , can be obtained by W (V) To find the shortest transportation distance between any two points, in the matrix W (V) The path can be traced by the following methods: , then point q1 is a point in the transport mileage of i and j. Use the same method to search at both ends. If: Tracing back to point i, we get: , , …, ; Tracing back to point j, we get: , …, ; Then the transportation mileage from i to j [i-j] is: .

[0040] Specifically, it is determined whether the direct transport mileage is blocked. If so, the optimization condition is introduced to optimize the indirect mileage between the starting transport node and each terminal transport node. The optimization condition is introduced to avoid transport congestion when searching for the shortest path between each two points (i and j1; i and j2; i and j3...).

[0041] The optimization condition is expressed as: (6) In formula (6), Y ij is the optimization condition expression; in the process from i to j, if i and j are not connected (through), that is, it is not a straight line distance, and a turning point R is required to complete the transportation, then Y ij =0; if i and j are connected, it is a straight line distance and the transportation can be completed without turning point R, then Y ij =1. Based on the mathematical description of the feasible solution of the transportation path, a mathematical model of the feasible solution of the transportation path is constructed, which is expressed as: (7) In formula (7), V is the number of selected i to j; f0 is the minimum objective function; ij =1, Y ji =1, i and j are connected, and j and i are both connected. At this time, the transportation path taken by the logistics transportation has no retreat, and f0 is the optimal value of the feasible solution of the logistics transportation path after optimization, that is, the shortest transportation path. ij = 0, i and j are not connected. The mathematical model of the feasible solution of the logistics transportation path is expressed as: (8) In formula (8), f0 ’ is the worst value of the feasible solution of the logistics transportation path under the condition that i and j are not connected, that is, the longest transportation path. At this time, the transportation path taken by the logistics transportation has more than one retreat. Assuming that there are two turning points between i and j, the shortest path at this time is: (9) In formula (9), f0 ” is the shortest transportation path under the conditions of two turning points; η, ε are weighted coefficients. Combining f0, f0', f0'' and other conditions, the optimal logistics and transportation plan established can meet the actual needs of logistics and transportation and maximize the economic benefits of the logistics and transportation industry.

[0042] Specifically, the indirect mileage is marked or monitored for transportation, and the transportation monitoring results are generated. When i and j are not connected, that is, there is a congestion between i and j. At this time, i turns to j for the indirect mileage, and the direct path between i and j needs to be marked or monitored, such as how long the congestion lasts or whether it is permanent, that is, the road is diverted or closed.

[0043] Load constraints and quantity constraints are imposed on transport vehicles, and the number of transport vehicles or transportation times used for the weight of the delivered goods is calculated. For example, if the transport vehicle has a load of 100kg and a fully loaded driving distance of 40 kilometers, the arranged goods cannot exceed 100kg. If the weight of the goods is 160kg, two transportations or two vehicles are required for transportation.

[0044] The cloud computing solution for the shortest path optimization system for logistics distribution includes an acquisition module, a division module, an optimization module, and a distribution module, among which: An acquisition module is used to acquire logistics address information and vehicle information. The logistics address information includes a distribution center and a receiving location. The distribution center address is used as the starting area. According to the receiving location, several collection areas are determined, and the specific addresses of several corresponding collection areas are acquired at the same time. A division module is used to divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; The optimization module is used to plan the logistics path for the starting area and multiple collection areas, generate multiple collection and transportation paths, and optimize the multiple collection and transportation paths to obtain a determined path; The distribution module is used to collect the weight of the goods to be delivered along the determined route, and to match the appropriate transport vehicles according to the weight of the goods to be delivered and arrange the transportation tasks.

[0045] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0046] The present invention, the undescribed parts are prior art.

[0047] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Cloud computing solution for the shortest path optimization method for logistics distribution, characterized by: include: Obtaining logistics address information and vehicle information, the logistics address information includes the distribution center and the delivery location, taking the distribution center address as the starting area, determining several collection areas according to the delivery location, and obtaining several corresponding specific addresses of the collection areas; Divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; Carry out logistics route planning for the starting area and multiple collection areas, generate multiple collection and transportation routes, and optimize the multiple collection and transportation routes to obtain a determined route; Collect the weight of the goods delivered along the determined route, and arrange the transportation task according to the weight of the goods delivered and the appropriate transport vehicle; Carry out logistics route planning for the starting area and multiple collection areas, generate multiple collection and transportation routes, and optimize the multiple collection and transportation routes to obtain the determined route, which specifically includes: The starting area is recorded as the starting transportation node, and each collection area is recorded as the terminal transportation node. The distance between the starting transportation node and each terminal transportation node is defined as the direct transportation mileage; The road resistance function is used to describe the relationship between the travel time from the starting transport node to the end transport node and the section flow and maximum capacity of the through transport mileage; According to the observed through transport mileage data and the calculated travel time data, the travel matrix and travel time matrix are established respectively; According to the direct transport mileage and mutual relationship, the transport mileage matrix between adjacent nodes is calculated and constructed. The adjacency matrix in the transport mileage matrix is ​​used as the weight matrix. The weight matrix is ​​inserted several times to construct several shortest transport mileage matrices in turn. The shortest transport mileage matrix is ​​used to find the shortest transport path between any adjacent starting transport node and each end transport node.

2. The method for optimizing the shortest path for logistics distribution using cloud computing as claimed in claim 1, characterized in that: According to the delivery location, several collection areas are determined, and the specific addresses of several corresponding collection areas are obtained, including: Obtain logistics address information and preset collection area information; Screening the collection area according to the logistics address information to determine a number of collection areas; According to the collection area information, specific addresses of the collection areas corresponding to the plurality of collection areas are obtained.

3. The method for optimizing the shortest path for logistics distribution using cloud computing as claimed in claim 1, characterized in that: According to the specific addresses of several collection areas, the areas are divided into multiple collection areas, including: Determine the collection area according to the specific addresses of several collection areas; According to the preset allocation standard information, the receiving location is allocated and the collection allocation data is generated; According to the agency collection allocation data, the agency collection area is divided to obtain a plurality of agency collection divided areas.

4. The method for optimizing the shortest path for logistics distribution using cloud computing as claimed in claim 1, characterized in that: Determine whether the direct transport mileage is blocked. If so, introduce optimization conditions to optimize the indirect mileage between the starting transport node and each end transport node.

5. The method for optimizing the shortest path for logistics distribution using cloud computing as claimed in claim 4, characterized in that: Carry out transportation marking or monitoring on indirect mileage and generate transportation monitoring results.

6. The method for optimizing the shortest path for logistics distribution using cloud computing as claimed in claim 1, characterized in that: Load constraints and quantity constraints are imposed on transport vehicles, and the number of transport vehicles or the number of transport times used for the weight of the delivered goods are calculated.

7. Cloud computing solves the shortest path optimization system for logistics distribution, characterized by: It includes acquisition module, division module, optimization module and distribution module, among which: An acquisition module is used to acquire logistics address information and vehicle information. The logistics address information includes a distribution center and a receiving location. The distribution center address is used as the starting area. According to the receiving location, several collection areas are determined, and the specific addresses of several corresponding collection areas are acquired at the same time. A division module is used to divide the area according to the specific addresses of several collection areas to obtain multiple collection division areas; The optimization module is used to plan the logistics path for the starting area and multiple collection areas, generate multiple collection and transportation paths, and optimize the multiple collection and transportation paths to obtain a determined path; The distribution module is used to collect the weight of the goods to be delivered along the determined route, match the appropriate transport vehicle according to the weight of the goods to be delivered and arrange the transportation task; Carry out logistics route planning for the starting area and multiple collection areas, generate multiple collection and transportation routes, and optimize the multiple collection and transportation routes to obtain the determined route, which specifically includes: The starting area is recorded as the starting transportation node, and each collection area is recorded as the terminal transportation node. The distance between the starting transportation node and each terminal transportation node is defined as the direct transportation mileage; The road resistance function is used to describe the relationship between the travel time from the starting transport node to the end transport node and the section flow and maximum capacity of the through transport mileage; According to the observed through transport mileage data and the calculated travel time data, the travel matrix and travel time matrix are established respectively; According to the direct transport mileage and mutual relationship, the transport mileage matrix between adjacent nodes is calculated and constructed. The adjacency matrix in the transport mileage matrix is ​​used as the weight matrix. The weight matrix is ​​inserted several times to construct several shortest transport mileage matrices in turn. The shortest transport mileage matrix is ​​used to find the shortest transport path between any adjacent starting transport node and each end transport node.

Images