CN113344267A - Logistics network resource allocation optimization method based on cooperation - Google Patents

Logistics network resource allocation optimization method based on cooperation Download PDF

Info

Publication number
CN113344267A
CN113344267A CN202110602907.5A CN202110602907A CN113344267A CN 113344267 A CN113344267 A CN 113344267A CN 202110602907 A CN202110602907 A CN 202110602907A CN 113344267 A CN113344267 A CN 113344267A
Authority
CN
China
Prior art keywords
algorithm
nsga
data
logistics
cooperation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110602907.5A
Other languages
Chinese (zh)
Inventor
王勇
樊建新
罗思妤
许茂增
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing Jiaotong University
Original Assignee
Chongqing Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing Jiaotong University filed Critical Chongqing Jiaotong University
Priority to CN202110602907.5A priority Critical patent/CN113344267A/en
Publication of CN113344267A publication Critical patent/CN113344267A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G06Q10/08355—Routing methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • General Business, Economics & Management (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Development Economics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Physiology (AREA)
  • Genetics & Genomics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a cooperation-based logistics network resource allocation optimization method, which comprises the following steps of firstly, constructing a dual-objective optimization model for minimizing the total network operation cost and the vehicle use number; and secondly, providing a mixed heuristic algorithm combining a k-means clustering algorithm and a CW-NSGA-II algorithm, designing the k-means clustering algorithm to reduce the calculation difficulty of the mixed algorithm, designing a greedy algorithm in the CW-NSGA-II algorithm to generate an initial feasible solution, and improving the convergence performance and the optimization performance of the algorithm by adopting an elite retention strategy. And finally, distributing the profits of the multi-center distribution and collection cooperative alliance by using an MCRS method, determining an optimal alliance joining sequence according to a strict monotone path principle in order to further improve the stability of the cooperative alliance, and comparing the calculated profit distribution scheme with the calculated profit distribution scheme, wherein the verified profit distribution scheme is more optimal.

Description

Logistics network resource allocation optimization method based on cooperation
Technical Field
The invention relates to the field of experimental equipment, in particular to a cooperation-based logistics network resource allocation optimization method.
Background
The configuration and optimization of logistics network resources is an important decision problem in the field of logistics systems. At present, the acceleration of the urbanization process and the rapid development of electronic commerce promote the increase of the demand of the logistics service and the expansion of the range, and put forward higher requirements on the timeliness and the specialty of the logistics service. However, in the current independent operation mode of the logistics enterprise, it is difficult for a single logistics enterprise to satisfy all the demands (time demand and goods demand) of customers by means of its own logistics channel and transportation resources, and the problems of low utilization rate of the logistics network resources, high total operation cost of the logistics network, and the like are frequent. The rise of the sharing economy provides a new idea and a new opportunity for cooperation among logistics enterprises. A cooperative logistics network system is constructed on the basis of the cooperative relationship, and the defects of the existing logistics service mode are overcome by optimizing and integrating logistics network resources, so that the logistics network service level and the resource utilization rate can be improved.
The cooperation-based logistics network resource allocation optimization is a typical NP-hard problem, and a single intelligent optimization solving algorithm has the defects of low convergence speed and poor optimizing capability. Aiming at a cooperation-based logistics network resource allocation optimization model, a hybrid intelligent optimization algorithm of a solution model is designed, and the hybrid algorithm has advantages in algorithm convergence and global space searching capability obtained by optimization solution. In the aspect of improving the convergence speed of the algorithm, a clustering algorithm is adopted to split the multi-center logistics network into a plurality of single-center logistics networks, so that the complexity of the algorithm is reduced; in the aspect of initial population generation, a saving algorithm is designed to construct a local optimal solution, and the convergence speed is accelerated; in the aspect of global space searching capability, the mixed algorithm designs an elite reservation strategy, effectively avoids trapping in local optimum, and ensures that the algorithm converges to a global optimum solution. The application of the hybrid intelligent optimization algorithm not only can solve a plurality of targets, but also can help logistics enterprises to save a large amount of calculation cost, and support is provided for improving the operation efficiency of a logistics network.
Unlike the independent logistics operation mode, in the cooperative mode, the logistics network shares network resources by using digitization technologies such as big data and the internet. On one hand, the customers are distributed to logistics enterprises with closer distances, and the logistics service areas are reasonably divided, so that unreasonable phenomena of cross transportation, roundabout transportation, no-load transportation and the like in a network can be effectively reduced; on the other hand, large trucks are adopted to finish goods dispatching tasks among logistics enterprises, customer demands and transportation resources of logistics service areas can be effectively coordinated, intensification and precision of the logistics resources are achieved, meanwhile, vehicle transportation routes are optimized in the logistics service areas, and network total operation cost and vehicle quantity can be minimized under the conditions that customer time window requirements are responded quickly and customer service levels are guaranteed. Therefore, the cooperative logistics network resource allocation optimization can effectively solve the contradiction between the limited transportation resources of the logistics enterprises and the excessive number of the clients, exert the complementary advantages of the logistics networks, ensure the logistics service competition capability of the logistics enterprises, reduce the operation cost of the logistics networks and improve the operation benefits of the logistics enterprises.
Disclosure of Invention
The present invention is directed to overcome the above problems in the prior art, and to provide a method for optimizing configuration of logistics network resources based on cooperation, so as to solve the above problems in the background art.
Therefore, the invention provides a cooperation-based logistics network resource allocation optimization method, which comprises the following steps:
acquiring logistics cost data, customer point data and distribution center data;
respectively establishing two objective functions F1 and F2 according to the relationship among the logistics cost data, the customer point data and the distribution center data;
inputting the two objective functions F1 and F2 into a CW-NSGA-II algorithm;
using a customer clustering algorithm to the customer point data and the distribution center data to obtain a clustering result;
and optimizing the clustering result by using the CW-NSGA-II algorithm.
In this embodiment, the customer clustering algorithm is a k-means clustering algorithm.
In this embodiment, when using the k-means clustering algorithm, the method includes the following steps:
setting an initial clustering result according to the number of distribution centers;
distributing the customer data with the distribution center;
and outputting the final clustering result through multiple times of iterative optimization.
In this embodiment, when the CW-NSGA-ii algorithm is used for optimization, the following steps are included:
setting parameters of the CW-NSGA-II algorithm, wherein the parameters comprise the total population number and the total number of elite individuals;
the CW-NSGA-II algorithm is used for operation in an iteration mode, and each iteration of the CW-NSGA-II algorithm evaluates the two objective functions F1 and F2 to obtain an elite individual;
and (4) iteratively calculating the elite individual by using a pareto frontier algorithm, wherein the maximum number of iterative calculations is the total number of the elite individual.
Meanwhile, in this embodiment, after obtaining the elite individual, the method includes the following steps:
carrying out genetic operation on non-elite individuals in the population to generate new filial individuals, carrying out cross operation on two parent individuals by adopting a single-point cross operator according to cross probability, and carrying out mutation operation on individual genes by adopting a partial mapping mutation operator according to mutation probability;
generating a new offspring population according to the evaluation of the two target functions F1 and F2;
and taking the new filial generation population as the elite individual.
The cooperation-based logistics network resource allocation optimization method provided by the invention has the following beneficial effects: the invention starts from the practical background of the urban logistics network, carries out overall design on the cooperative logistics network, adds a client information sharing and vehicle resource sharing mode into the construction of the existing logistics network, aims to reduce the overall operation cost of the logistics network and the number of distributed vehicles, simultaneously considers practical factors such as client time window constraint, vehicle loading capacity constraint and the like, establishes a logistics network optimization model, carries out decision-making on vehicle paths and distribution schemes, realizes the intensive configuration of the logistics resources of the cooperative logistics network, and obtains a reasonable logistics network system based on cooperation. The method effectively avoids a plurality of defects of the traditional logistics network, improves the overall operation efficiency and the customer service level of the logistics network, reduces the operation cost of the logistics network, and provides powerful support for scientific decision of the resource allocation optimization of the logistics network.
Drawings
FIG. 1 is a schematic block diagram of the overall process of the present invention;
fig. 2 is a schematic block diagram of a detailed process of the present invention.
Detailed Description
An embodiment of the present invention will be described in detail below with reference to the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the embodiment.
Specifically, as shown in fig. 1-2, an embodiment of the present invention provides a cooperation-based method for optimizing configuration of resources of a logistics network, including the following steps:
acquiring logistics cost data, customer point data and distribution center data;
(II) respectively establishing two objective functions F1 and F2 according to the relationship among the logistics cost data, the customer point data and the distribution center data;
(III) inputting the two objective functions F1 and F2 into a CW-NSGA-II algorithm;
fourthly, clustering results are obtained by using a customer clustering algorithm on the customer point data and the distribution center data;
and (V) optimizing the clustering result by using the CW-NSGA-II algorithm.
In this embodiment, the customer clustering algorithm is a k-means clustering algorithm.
In this embodiment, when using the k-means clustering algorithm, the method includes the following steps:
(1) setting an initial clustering result according to the number of distribution centers;
(2) distributing the customer data with the distribution center;
(3) and outputting the final clustering result through multiple times of iterative optimization.
In this embodiment, when the CW-NSGA-ii algorithm is used for optimization, the following steps are included:
(-1-) setting parameters of the CW-NSGA-II algorithm, wherein the parameters comprise population total number and elite individual total number;
the (-2-) iteration uses the CW-NSGA-II algorithm to carry out operation, and each iteration of the CW-NSGA-II algorithm evaluates the two objective functions F1 and F2 to obtain elite individuals;
and (-3-) iteratively calculating the elite individual by using a pareto frontier algorithm, wherein the maximum iterative calculation times are the total number of the elite individual.
Meanwhile, in this embodiment, after obtaining the elite individual, the method includes the following steps:
(a) carrying out genetic operation on non-elite individuals in the population to generate new filial individuals, carrying out cross operation on two parent individuals by adopting a single-point cross operator according to cross probability, and carrying out mutation operation on individual genes by adopting a partial mapping mutation operator according to mutation probability;
(b) generating a new offspring population according to the evaluation of the two target functions F1 and F2;
(c) and taking the new filial generation population as the elite individual.
In this embodiment, when the logistics cost data, the customer point data, and the distribution center data are acquired, the constraint condition of the logistics cost data, the constraint condition of the customer point data, and the constraint condition of the distribution center data are acquired at the same time.
Specifically, we verified the above method through a specific experiment.
Firstly, data acquisition is carried out, namely data in the step (I) is acquired, and the acquired data specifically comprises cost data, customer point data and distribution center data; the cost data comprises transportation cost and penalty cost; the client point data comprises client requirements and a client time window; the distribution center data comprises vehicle load and the like; these data include, but are not limited to, these data.
And then, establishing a double objective function, and establishing constraint conditions according to the requirements of customers. The method comprises the following specific steps:
a dual objective function:
Figure BDA0003093200990000061
Figure BDA0003093200990000062
constraints (we now count general customer needs):
Figure BDA0003093200990000071
Figure BDA0003093200990000072
Figure BDA0003093200990000073
Figure BDA0003093200990000074
Figure BDA0003093200990000075
Figure BDA0003093200990000076
Figure BDA0003093200990000077
Figure BDA0003093200990000078
Figure BDA0003093200990000079
Figure BDA00030932009900000710
Figure BDA00030932009900000711
Figure BDA00030932009900000712
Figure BDA00030932009900000713
Figure BDA00030932009900000714
Figure BDA00030932009900000715
Figure BDA00030932009900000716
Figure BDA00030932009900000717
Figure BDA00030932009900000718
Figure BDA00030932009900000719
wherein, F1 is the total operation cost of the logistics network; f2 number of delivery vehicles; i is a set of distribution centers, I belongs to I; j is a set of clients, and J belongs to J; v is a set of distribution vehicles, and V belongs to V; s is a set of trucks, and S belongs to S; o issA distribution center set serving a vehicle to be blocked S belongs to S; o isvA set of customers servicing a delivered vehicle V, V ∈ V; qvV belongs to V for the load of the distribution vehicle; qsThe load of the truck is S belongs to S; qiI belongs to I as the capacity of the distribution center; q. q.sjJ belongs to J as the requirement of a client; q. q.schThe goods transportation volume from a distribution center c to a distribution center h, c, h belongs to I, and c is not equal to h; divThe time when the delivery vehicle V starts from the delivery center I belongs to I, and the time when the delivery vehicle V starts from the delivery center I belongs to V; drivThe time for returning the delivery vehicle V to the delivery center I belongs to I, and V belongs to V; t is tjvJ belongs to J and V belongs to V as the time of the delivery vehicle V reaching the client J; [ m ] ofj,nj]A service time window for the client, J belongs to J; [ e ] ai,li]I belongs to I as a service time window of a distribution center I; dijThe distance from a node I to a node J is represented by I, J belongs to I, U, J, I is not equal to J; dchThe distance from a distribution center c to a distribution center h is c, h belongs to I, and c is not equal to h; f. ofvA hundred kilometer fuel consumption rate for the delivery vehicle; f. ofsIs the fuel consumption rate of the truck in hundred kilometers; p is a radical ofvThe price of gasoline for the delivery vehicle; p is a radical ofsIs the diesel price of the truck; c. CiAdding cooperative cost for the distribution center i; u. ofePenalty cost for delivery vehicles arriving at the customer early; u. ofdPenalty cost for delayed arrival of delivery vehicles to customers; t is tijvTravel time spent for delivery vehicles from node i to node j; t is the maximum allowable path time for the delivery vehicle; x is the number ofijvTravel from node i to node j for delivery vehicle v; w is aijvcDriving a distribution vehicle v from a distribution center c to a node j through a node i; y ischsDriving a truck s from a distribution center c to a distribution center h; ecjhChange from being serviced by delivery vehicle c to being serviced by delivery center h for customer j; ziThe distribution center i agrees to cooperate.
In order to solve a multi-center co-distribution and collection dual-target optimization model, a CW-NSGA-II hybrid algorithm based on a k-means clustering algorithm is provided. Firstly, designing a k-means clustering algorithm to reasonably divide the service area of a multi-center distribution and collection network, and distributing customers to logistics facilities with closer distances; secondly, a greedy algorithm is designed to process the results of the k-means clustering algorithm, and initial vehicle running routes are generated between each service area and each service area; and finally, optimizing the vehicle driving route by using NSGA-II so as to generate a multi-center distribution and collection optimization route. The flow of the CW-NSGA-II hybrid algorithm based on the k-means clustering algorithm is shown in FIG. 2.
The method has the advantages that clustering analysis is carried out on customers, service areas of all centers in the multi-center common distribution and collection network are reasonably divided, on one hand, the phenomena of staggered transportation, long-distance transportation and the like in the multi-center common distribution and collection network can be reduced, the total operation cost of the multi-center common distribution and collection network is reduced, on the other hand, the multi-center vehicle path optimization problem is converted into the single-center vehicle path optimization problem, the calculation complexity of a hybrid algorithm is reduced, and the total driving distance of an initial feasible solution is reduced and the searching speed of the hybrid algorithm is accelerated by distributing the customers to facilities with closer distances. The k-means clustering algorithm process is as follows:
step 1: the number and position coordinates of the distribution centers and the collection centers, and the position coordinates of the distribution customers and the collection customers are input.
Step 2: and judging whether the logistics facilities participating in the cooperation only comprise a distribution center or a collection center, if the judgment conditions are met, turning to Step 3, and if the judgment conditions are not met, turning to Step 4.
Step 3: setting k distribution centers or collection centers as initial clustering centers.
Step 4: setting k to { k1, k2} distribution centers and collection centers as initial clustering centers, wherein k1 distribution centers and k2 collection centers are included.
Step 5: the distance from each customer to the initial cluster center is calculated and each customer is assigned to the closest cluster center.
Step 6: and updating k cluster centers, and reevaluating the distance from each client to the new cluster center.
Step 7: and repeating the steps 5-7 until the distance from each client to the cluster center is minimum and the cluster center is stable and unchanged.
Step 8: and calculating the distance from the latest clustering center to each distribution center and collection center, and allocating the customer clusters of each clustering center to the distribution center or the collection center with the closest distance.
Step 9: and outputting a clustering result.
The Step (3) is a further refinement of the expansion of the k-means clustering algorithm, wherein the Step (1) is a Step (1), the Step (2-3) is a Step (2), and the Step (4-9) is a Step (3).
Then, the following process is specific in combination with the CW-NSGA-II algorithm:
step 1: setting algorithm related parameters, a maximum optimization time tmax, a maximum iteration time runmax, a population size pop _ size, a chromosome selection probability ps, a chromosome crossing probability pcr and a chromosome variation probability pmu.
Step 2: designing a greedy algorithm to process the result of the k-means clustering algorithm and generate an initial feasible solution; an initial solution is generated using a scanning algorithm until the population size of the initial parent population Pt reaches pop _ size, at which time t is set to 1 and run to 1. Step 2.1-Step 2.3 are processes for greedy algorithm to generate initial feasible solutions.
Step 2.1: and calculating the distance of a path formed between any two delivery clients or collection client nodes, and respectively sorting the paths from small to large according to the distance value.
Step 2.2: and sequentially judging whether the path is a sub-path or not, if so, adding the path into the current path, and otherwise, directly judging the next path. And (3) judging rules of the sub-paths: adding the path without making the number of the connecting edges of any node more than 2; adding this path does not close the path; the total demand of the clients in the single path generated by adding the path does not exceed the maximum loading capacity of the vehicle; and fourthly, the time window requirement of the client is met when the path is added.
Step 2.3: step 2.2 is performed until no sub-paths exist, at which time the two end points of each path are connected to a distribution or collection center, respectively, to form a closed loop.
Step 3: the dual objective function values T (total network operating cost) and V (number of delivery and collection vehicle uses) of each individual in the primary parent population Pt are evaluated, and a Pareto non-dominated solution set is constructed, non-dominated sorting operations are performed, and the crowding distance of the individual is calculated.
Step 4: executing an elite reservation strategy, and selecting a certain number of individuals with higher fitness values in the current primary parent population Pt according to the fitness function values to record as elite individuals, wherein the elite individuals are reserved without participating in subsequent genetic selection, crossing and mutation operations.
Step 5: genetic selection, crossover and mutation operations are performed on non-elite individuals to generate progeny individuals. The process mainly comprises the following steps: selecting a parent individual by using a championship selection method; and carrying out cross operation on the parent individuals by adopting a partial mapping cross operator, and carrying out mutation operation on the individual genes by adopting an inversion mutation operator.
Step 6: and (3) in the progeny individuals generated at Step 5, replacing the individuals with lower fitness values after the genetic crossing and mutation operations with the elite individuals reserved at Step4, so as to generate a new progeny population Qt.
Step 7: and combining the primary generation parent population Pt and the new offspring population Qt to generate a new generation population Rt, wherein the Rt is Pt U Qt, the size of the Rt population is 2pop _ size, and according to the Rank value and the congestion distance of the non-dominant ordering of each individual in the new population Rt, the individuals are selected to form a new generation parent population Pt +1, and the size of the Pt +1 population is pop _ size.
Step 8: if run is not more than runmax, returning to Step 6 of the k-means clustering algorithm; if run is greater than or equal to runmax, then go to the hybrid algorithm Step 9.
Step 9: if t is less than or equal to tmax, returning to the Step 3 of the hybrid algorithm; and if t is larger than or equal to tmax, ending the loop iteration operation of the hybrid algorithm.
Step 10: and outputting a Pareto optimal solution set, selecting an optimal solution from the Pareto optimal solution set, and ending the algorithm.
In the above process, Step 1 is Step (-1-), Step2 is Step (-2-), Step 2-10 is Step (-3-), and Step 2.1-2.3 are steps (a) - (c).
To verify the effectiveness of the CW-NSGA-II hybrid algorithm presented herein, the CW-NSGA-II hybrid algorithm was compared to NSGA-II and a multiple target genetic algorithm (MOGA). The experimental data were modified according to the coreau standard calculation and the multicenter co-distribution and collection network characteristics, as shown in table 1. According to the prior relevant literature, the algorithm parameters are set as shown in table 2. Considering that the heuristic algorithm results have randomness, each group of experiments is operated for 10 times, and the optimal solution in the 10 operation results is selected to be compared with the corresponding operation time, as shown in table 3.
TABLE 1 data set characterization
Figure BDA0003093200990000121
TABLE 2 hybrid Algorithm parameter set
Figure BDA0003093200990000122
TABLE 3 comparison of the results of the calculations for the different algorithms
Figure BDA0003093200990000131
As can be seen from the data in the table above, the cost results obtained by the CW-NSGA-II hybrid algorithm and the NSGA-II and MOGA solution have significant differences. In the aspect of cost, the average value of the logistics cost solved by the CW-NSGA-II hybrid algorithm is 2735.06 yuan, which is 4.12% lower than the 2852.52 yuan of the logistics cost solved by NSGA-II and 5.96% lower than the 2908.38 yuan of the logistics cost solved by MOGA; in the aspect of the vehicle use number, the average value of the vehicle use number solved by the CW-NSGA-II hybrid algorithm is 16, and is lower than the vehicle use number obtained by the NSGA-II and MOGA solution; in terms of operation time, the average operation time of the CW-NSGA-II hybrid algorithm is 155.15 seconds, which is 7.2 seconds less than the average operation time 162.35 seconds of NSGA-II and 9.5 seconds less than the average operation time 164.65 seconds of MOGA. The results show that the CW-NSGA-II hybrid algorithm proposed herein has better search and optimization capabilities than NSGA-II and MOGA.
In summary, aiming at the optimization problem of the multi-center distribution and collection network, firstly, a dual-target optimization model for minimizing the total network operation cost and the vehicle use number is constructed; and secondly, providing a mixed heuristic algorithm combining a k-means clustering algorithm and a CW-NSGA-II algorithm, designing the k-means clustering algorithm to reduce the calculation difficulty of the mixed algorithm, designing a greedy algorithm in the CW-NSGA-II algorithm to generate an initial feasible solution, and improving the convergence performance and the optimization performance of the algorithm by adopting an elite retention strategy. The effectiveness of the CW-NSGA-II hybrid algorithm is verified by carrying out comparative analysis on logistics cost, vehicle using quantity and solving running time by the CW-NSGA-II hybrid algorithm, the NSGA-II algorithm and the MOGA algorithm. And finally, distributing the profits of the multi-center distribution and collection cooperative alliance by using an MCRS method, determining an optimal alliance adding sequence according to a strict monotone path principle in order to further improve the stability of the cooperative alliance, and comparing the MCRS method with a profit distribution scheme calculated by CGA, EPM and Shapley methods, verifying that the profit distribution scheme calculated by the MCRS method is more optimal.
The above disclosure is only for a few specific embodiments of the present invention, however, the present invention is not limited to the above embodiments, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present invention.

Claims (6)

1. A cooperation-based logistics network resource allocation optimization method is characterized by comprising the following steps:
acquiring logistics cost data, customer point data and distribution center data;
respectively establishing two objective functions F1 and F2 according to the relationship among the logistics cost data, the customer point data and the distribution center data;
inputting the two objective functions F1 and F2 into a CW-NSGA-II algorithm;
using a customer clustering algorithm to the customer point data and the distribution center data to obtain a clustering result;
and optimizing the clustering result by using the CW-NSGA-II algorithm.
2. The cooperation-based logistics network resource allocation optimization method of claim 1, wherein the customer clustering algorithm is a k-means clustering algorithm.
3. The cooperation-based logistics network resource allocation optimization method of claim 1, wherein when using k-means clustering algorithm, the method comprises the following steps:
setting an initial clustering result according to the number of distribution centers;
distributing the customer data with the distribution center;
and outputting the final clustering result through multiple times of iterative optimization.
4. The method as claimed in claim 1, wherein the optimization using CW-NSGA-ii algorithm comprises the following steps:
setting parameters of the CW-NSGA-II algorithm, wherein the parameters comprise the total population number and the total number of elite individuals;
the CW-NSGA-II algorithm is used for operation in an iteration mode, and each iteration of the CW-NSGA-II algorithm evaluates the two objective functions F1 and F2 to obtain an elite individual;
and (4) iteratively calculating the elite individual by using a pareto frontier algorithm, wherein the maximum number of iterative calculations is the total number of the elite individual.
5. The cooperation-based logistics network resource allocation optimization method of claim 4, wherein after obtaining the elite individual, the method comprises the following steps:
carrying out genetic operation on non-elite individuals in the population to generate new filial individuals, carrying out cross operation on two parent individuals by adopting a single-point cross operator according to cross probability, and carrying out mutation operation on individual genes by adopting a partial mapping mutation operator according to mutation probability;
generating a new offspring population according to the evaluation of the two target functions F1 and F2;
and taking the new filial generation population as the elite individual.
6. The cooperation-based optimization method for logistics network resource allocation as claimed in claim 1, wherein when the logistics cost data, the customer point data and the distribution center data are obtained, the constraints of the logistics cost data, the constraints of the customer point data and the constraints of the distribution center data are obtained at the same time.
CN202110602907.5A 2021-05-31 2021-05-31 Logistics network resource allocation optimization method based on cooperation Pending CN113344267A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110602907.5A CN113344267A (en) 2021-05-31 2021-05-31 Logistics network resource allocation optimization method based on cooperation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110602907.5A CN113344267A (en) 2021-05-31 2021-05-31 Logistics network resource allocation optimization method based on cooperation

Publications (1)

Publication Number Publication Date
CN113344267A true CN113344267A (en) 2021-09-03

Family

ID=77473310

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110602907.5A Pending CN113344267A (en) 2021-05-31 2021-05-31 Logistics network resource allocation optimization method based on cooperation

Country Status (1)

Country Link
CN (1) CN113344267A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114897307A (en) * 2022-04-13 2022-08-12 重庆交通大学 Logistics centralized distribution network alliance optimization method based on cooperation
CN115130787A (en) * 2022-08-29 2022-09-30 深圳市城市公共安全技术研究院有限公司 Configuration method, system, terminal equipment and medium of emergency resource scheduling scheme

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102831495A (en) * 2012-07-19 2012-12-19 浙江工商大学 Logistics supply chain coordination optimization method based on improved ant colony labor division model
CN104766188A (en) * 2014-01-02 2015-07-08 中国移动通信集团江苏有限公司 Logistics distribution method and logistics distribution system
US9690629B1 (en) * 2012-11-15 2017-06-27 Google Inc. Distributed batch matching of videos based on recency of occurrence of events associated with the videos
CN107578128A (en) * 2017-08-31 2018-01-12 南京理工大学 Across level distribution network planing method based on immunity particle cluster algorithm

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102831495A (en) * 2012-07-19 2012-12-19 浙江工商大学 Logistics supply chain coordination optimization method based on improved ant colony labor division model
US9690629B1 (en) * 2012-11-15 2017-06-27 Google Inc. Distributed batch matching of videos based on recency of occurrence of events associated with the videos
CN104766188A (en) * 2014-01-02 2015-07-08 中国移动通信集团江苏有限公司 Logistics distribution method and logistics distribution system
CN107578128A (en) * 2017-08-31 2018-01-12 南京理工大学 Across level distribution network planing method based on immunity particle cluster algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王勇 等: "基于车辆共享的多中心共同配送联盟优化", 《计算机集成制造系统》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114897307A (en) * 2022-04-13 2022-08-12 重庆交通大学 Logistics centralized distribution network alliance optimization method based on cooperation
CN115130787A (en) * 2022-08-29 2022-09-30 深圳市城市公共安全技术研究院有限公司 Configuration method, system, terminal equipment and medium of emergency resource scheduling scheme

Similar Documents

Publication Publication Date Title
Wang et al. Collaborative two-echelon multicenter vehicle routing optimization based on state–space–time network representation
CN109034465B (en) Charging station two-layer planning method considering coupling of charging station site selection and travel path
Miao et al. Autonomous connected electric vehicle (ACEV)-based car-sharing system modeling and optimal planning: A unified two-stage multi-objective optimization methodology
CN111144568A (en) Multi-target urban logistics distribution path planning method
CN110059934A (en) The method of fuel vehicle and the scheduling of new energy vehicle coperating distribution
CN107180274B (en) Typical scene selection and optimization method for electric vehicle charging facility planning
CN113344267A (en) Logistics network resource allocation optimization method based on cooperation
CN111260128B (en) Vehicle path planning method and system
CN109559062A (en) A kind of task distribution of cooperative logistical problem and paths planning method
Tucker et al. Online charge scheduling for electric vehicles in autonomous mobility on demand fleets
CN112733272A (en) Method for solving vehicle path problem with soft time window
CN113848970B (en) Multi-target cooperative path planning method for vehicle-unmanned aerial vehicle
Li et al. Deploying autonomous mobile lockers in a two-echelon parcel operation
Wang et al. Carbon reduction in the location routing problem with heterogeneous fleet, simultaneous pickup-delivery and time windows
CN115576343B (en) Multi-target vehicle path optimization method combined with unmanned aerial vehicle distribution
CN114897307A (en) Logistics centralized distribution network alliance optimization method based on cooperation
Wang et al. Cooperation and profit allocation for two-echelon logistics pickup and delivery problems with state–space–time networks
CN117035598A (en) Vehicle-machine collaborative path planning method and system under regional limitation
CN115879657A (en) Electric vehicle power station changing location path optimization method considering multi-station capacity design
CN113887782A (en) Genetic-firework mixing method and system for maintenance resource distribution scheduling
Luo et al. Two‐Echelon Multidepot Logistics Network Design with Resource Sharing
Abdallah The plug-in hybrid electric vehicle routing problem with time windows
CN115146866A (en) Multi-equivalent optimal path planning method considering actual multi-constraints
CN112036623B (en) Benefit coordination method of transverse logistics alliance
CN113222241B (en) Taxi quick-charging station planning method considering charging service guide and customer requirements

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210903