CN112700190B - Improved method for distributing tray materials by scanning method and genetic simulation annealing method - Google Patents

Improved method for distributing tray materials by scanning method and genetic simulation annealing method Download PDF

Info

Publication number
CN112700190B
CN112700190B CN202011603056.8A CN202011603056A CN112700190B CN 112700190 B CN112700190 B CN 112700190B CN 202011603056 A CN202011603056 A CN 202011603056A CN 112700190 B CN112700190 B CN 112700190B
Authority
CN
China
Prior art keywords
vehicle
scanning
genetic
list
route
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.)
Active
Application number
CN202011603056.8A
Other languages
Chinese (zh)
Other versions
CN112700190A (en
Inventor
王吉栋
刘源
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.)
CETC 15 Research Institute
Original Assignee
CETC 15 Research Institute
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 CETC 15 Research Institute filed Critical CETC 15 Research Institute
Priority to CN202011603056.8A priority Critical patent/CN112700190B/en
Publication of CN112700190A publication Critical patent/CN112700190A/en
Application granted granted Critical
Publication of CN112700190B publication Critical patent/CN112700190B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

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

Abstract

The invention discloses an improved method for distributing tray materials by a scanning method and a genetic simulation annealing method. The invention firstly uses an improved scanning method, takes a hair object unit as an original point, respectively scans all receiving units with reasonable different initial degrees to obtain a scanning list for starting scanning with different initial degrees, and fully considers the whole solution space; then, the genetic algorithm and the simulated annealing algorithm are combined to make up for deficiencies, the respective defects of the genetic algorithm and the simulated annealing algorithm are overcome, the capability of grasping the whole is stronger, the capability of searching for parts is stronger, a hybrid optimization algorithm genetic simulated annealing method is formed, the final distribution scheme with the lowest cost is selected from a list, the contradiction between the quality of understanding and the long solving time is balanced, the phenomena of premature convergence, possible searching of the optimal solution and then dispersion are avoided, and the optimal solution with the global approximation of the distribution scheme can be obtained.

Description

Improved method for distributing tray materials by scanning method and genetic simulation annealing method
Technical Field
The invention relates to the technical field of logistics distribution, in particular to a method for distributing tray materials by an improved scanning method and a genetic simulation annealing method.
Background
In recent years, the logistics industry has been developed vigorously, and distribution systems have become more and more important in the whole logistics related system. The transportation system is the most important subsystem in the distribution system, and the transportation cost accounts for about 50% of the whole logistics cost, so the logistics cost reduction is started from the logistics distribution transportation cost reduction. Among them, whether the distribution scheme reasonably and directly affects the distribution speed, cost and benefit, and especially the determination of the multi-user distribution line is a complicated system engineering. The appropriate distribution scheme can accelerate the response speed to the customer requirements, improve the service quality, enhance the satisfaction degree of the customer to the logistics link and reduce the operation cost of the service provider.
Along with the distribution links are more and more mechanized and informationized, more and more materials are stored and transported in a tray mode, and the mode reduces the times of goods stacking operation, is favorable for improving the box transporting efficiency, shortens the freight time and reduces the labor intensity.
2) At present, in the field of material distribution problem research, schemes such as a scanning method, a simulated annealing method, a genetic algorithm and the like are mainly provided. The scanning method is that the route of the vehicle is assumed to be located on a geometric plane, all points on the plane are represented by polar coordinates, a distribution center is used as an origin, an unused vehicle is selected at first, and on the premise of not violating the vehicle capacity limit, scanning is performed in a clockwise direction or a counterclockwise direction from a node with the minimum angle and which is not assigned yet. When the vehicle capacity is exceeded, the line is ended. And repeating the steps to generate a new line until all nodes are arranged. Finally, optimizing each route by using an algorithm for solving the problem of the traveling salesman; the simulated annealing method is a random search algorithm based on the thermodynamic principle. When the simulated annealing method is used for solving the material distribution problem, the energy obtained by atoms in physical annealing is equivalent to the optimal node distribution, the atom vibration is simulated as the random search of the line optimization space, the search process is completed by executing the pairwise exchange of distribution, the initial solution is the distribution total cost, the existing distribution condition is subjected to any pairwise exchange to generate new distribution, and therefore the distribution route is continuously improved until the optimal route is found. The genetic algorithm is an algorithm for carrying out local search improvement, when solving a material distribution problem, the genetic algorithm mainly utilizes the commonality between the biological evolution generation and the optimal route search iteration to search the optimal solution of a distribution route in a global scope, if the same consumption condition of the same-level distribution node branch is met, the optimal solution is considered, the optimal solution participates in the iteration of the next step respectively, the optimal distribution route is finally determined, if the different consumption conditions of the same-level distribution node branch are met, an adaptor survives, the right of continuous iteration is obtained by the distribution branch with less consumption, but the optimal qualification can not be determined, and the optimal distribution route is obtained by participating in the subsequent elimination step until the iteration is finished.
Although several solutions to the material distribution problem, such as scanning method, simulated annealing method, genetic algorithm, etc., have achieved good results, some disadvantages still exist:
(1) For the scanning method, during calculation, the search efficiency is low, the local optimal solution is easy to be trapped, and the global approximate optimal solution is difficult to obtain.
(2) For the simulated annealing method, the contradiction between the quality of the solution and the long solving time exists, repeated iterative operation is needed to obtain a good approximate optimal solution, and a feasible solution way is lacked when the scale of the problem is inevitably increased.
(3) For genetic algorithm, the problem of premature convergence exists, and in the initial stage of search, because excellent individuals are increased rapidly, the population loses diversity, so that the program falls into local parts and the global optimal solution cannot be achieved. Another drawback of genetic algorithms is the fact that during the search it is possible to search for the optimal solution and then to diverge again.
In conclusion, in the calculation process of the scanning method, the simulated annealing method and the genetic algorithm, the actual requirements cannot be well met, the generated loading scheme of the tray materials is not reasonable enough and accurate enough, and the situation that the loading cannot be carried out according to the calculated scheme in practice is caused directly. Furthermore, the priority setting of the delivery vehicles is not supported, and an additional constraint condition to explain the complicated problem is defined.
Disclosure of Invention
In view of the above, the present invention provides an improved method for delivering tray materials by using a scanning method and a genetic simulation annealing method, which solves the problem of delivering tray materials with the least cost in a transportation system.
The invention discloses a method for distributing tray materials by an improved scanning method and a genetic simulation annealing method, which comprises the following steps:
step one, acquiring a scanning list for starting scanning at different initial degrees by using an improved scanning method:
setting a vehicle distribution list according to the vehicle use priority; starting to scan all receiving units by using the transmitting unit as an origin and different initial degrees to obtain a scanning list starting to scan by using different initial degrees;
when scanning and loading, starting from a vehicle with high priority, starting from a receiving unit of the current starting degree, firstly loading palladable tray materials which are not loaded, meet the size requirement of the vehicle, are not mutually exclusive with the currently loaded materials and meet the requirements of the load factor of the vehicle and the load weight factor of the vehicle after loading on the vehicle, and modifying the state of the loaded tray materials into the loaded state; if the vehicle is not full, loading non-palletizable pallet materials which are not loaded, are not mutually exclusive with currently loaded materials, meet the additional size constraint condition of the vehicle, meet the requirements of the area multiplication factor of the vehicle bearing body and the weight multiplication factor of the vehicle bearing body after loading, and modifying the loaded pallet material state into loaded materials from the current receiving unit; loading all the tray materials of all the receiving units to obtain a scanning list of the initial degree; the scanning list comprises information of all vehicles and tray materials loaded by the vehicles; obtaining scanning lists corresponding to all different initial degrees;
step two, aiming at each scanning list, calculating the vehicle path and cost of each vehicle in the scanning list by adopting a genetic simulation annealing method; specifically, for a certain vehicle in a certain scan list:
firstly, genetic optimization is carried out on the current vehicle path by adopting a genetic algorithm; an initial vehicle path is randomly generated; then, taking the vehicle path after genetic optimization as an initial condition, and carrying out annealing optimization on the vehicle path by adopting a simulated annealing method; calculating the length of the vehicle route after annealing optimization, comparing the length with the length of the current shortest vehicle route, and if the length is smaller than the length of the current shortest vehicle route, taking the length of the vehicle route after annealing optimization as the length of the current shortest vehicle route, and taking the length of the vehicle route as the current vehicle route to perform genetic optimization and annealing optimization again; if the length of the vehicle route after the annealing optimization is larger than the current shortest vehicle route, the length of the current shortest vehicle route is not changed, and the vehicle route after the annealing optimization is used as the current vehicle route to perform genetic optimization and annealing optimization again; stopping circulation until the lengths of the vehicle routes after N times of continuous annealing optimization are all larger than or equal to the length of the current shortest vehicle route, and taking the vehicle route corresponding to the length of the current shortest vehicle route as the final route of the vehicle in the scanning list;
obtaining the final paths of all vehicles in the scanning list; at the cost of the length of the final path of all vehicles and as the present scan list.
The genetic-simulated annealing method specifically comprises the following steps:
step 1, giving population scale pop-size in a genetic algorithm, and parameters Pc and Pm during cross and mutation operations; setting a parameter T0 at initial temperature and a parameter a at annealing time in a simulated annealing method; giving the target optimization function as the shortest path; q in the termination rule is given.
Randomly generating Pop-size chromosomes as an initial population Pop (0) of a genetic algorithm, wherein the chromosomes represent vehicle paths of vehicles; calculating objective function values (namely vehicle path values S) of all chromosomes, and enabling the initial optimal solution Smin to be the minimum vehicle path value fmin in the initial population or a set value; let p =0.
And 3, calculating adaptive function values of the chromosomes, selecting groups by methods such as roulette and the like, and implementing an optimal retention strategy (namely, the best individual in the middle group is copied to the next generation group unconditionally, so that the best solution in the middle group is retained, the algorithm can be converged to the global optimal solution with the probability 1, and the convergence of the algorithm is ensured).
And 4, executing the cross operation of the genetic algorithm according to the cross probability Pc, calculating the objective function value of the chromosome, and implementing the optimal retention strategy.
And 5, executing a copy strategy based on the Markov chain judgment criterion to generate a next generation group Pop (n + 1).
And 6, executing simulated annealing operation on the basis of the population Pop (n + 1) to obtain a new population, wherein the annealing function is as follows: t (n + 1) = α T (n), let n = n +1.
And 7, calculating a new target optimization function value S' of the population.
Step 8, determining whether S ' is smaller than Smin, if yes, letting Smin = S ', S ' = fmin, p =0, and executing step 9. Otherwise p = p +1, step 9 is performed.
And 9, judging whether p is greater than or equal to q, and if so, taking Smin as an output result of the item in the list. Otherwise, returning to the step 3.
And step three, selecting the scanning list with the minimum cost and the vehicle path thereof from all the scanning lists corresponding to different starting degrees as a final distribution scheme.
Preferably, in the first step, all the receiving units are scanned clockwise and/or counterclockwise.
Preferably, in the genetic algorithm, a roulette method is used for population selection.
Preferably, in the genetic algorithm, an optimal retention strategy is implemented.
Preferably, N/q =100.
Has the advantages that:
(1) Compared with a scanning method, a simulated annealing method and a genetic algorithm, the method provided by the invention firstly utilizes an improved scanning method, takes a transmitting object unit as an original point, respectively scans all receiving units with reasonably different initial degrees to obtain a scanning list starting scanning with different initial degrees, and fully considers the whole solution space; and then combining the genetic algorithm with the simulated annealing algorithm to make up for the deficiencies, overcoming the respective defects of the genetic algorithm and the simulated annealing algorithm, having stronger capability of grasping the whole body and stronger local search capability, forming a genetic simulated annealing method of a hybrid optimization algorithm, and selecting the final distribution scheme with the minimum cost from the list. The contradiction between the quality of understanding and the long time of solving is balanced, the phenomena of premature convergence and the possibility of searching the optimal solution and then diverging are avoided, and the optimal solution of the global approximation of the distribution scheme can be obtained.
(2) Compared with a scanning method, a simulated annealing method and a genetic algorithm, the tray material loading scheme which actually meets the requirements of workers can be generated. The loading according to the scheme can not occur, the utilization rate of the carriage is seriously insufficient or the carriage can not load the calculated materials. In the process of calculating the distribution scheme, the stackable tray material needs to meet the requirements that the material and the material which are loaded on the current vehicle are not exclusive, the length, the width and the height are met, other self-defined additional constraint conditions are met, and after the stackable tray material is loaded, the loaded tray material meets the requirements of the bearing weight multiplication coefficient of the vehicle and the bearing weight multiplication coefficient of the vehicle. For non-palletizable pallet materials, the requirements of the carriage area factor of the vehicle need to be met in addition to the conditions of palletizable pallet materials. Finally, a reasonable and accurate tray material loading scheme which can be used for loading by workers in practice according to the calculated scheme can be generated. At the same time, the priority setting of the delivery vehicles is supported, and additional constraints are defined that explain complex problems.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention is described in detail below by way of example with reference to the accompanying drawings.
The invention provides an improved method for distributing tray materials by a scanning method and a genetic simulation annealing method, which is used for solving the problem of material distribution in a logistics system for managing materials by using trays in a full flow. The invention designs a proper route for a series of tray material receiving units, so that vehicles pass through the proper route in order, and the aim of calculating and selecting a proper tray material distribution scheme is achieved under the condition of meeting certain constraint conditions.
The flow chart of the invention is shown in figure 1 and comprises the following steps:
step 1, inputting relevant attributes of tray materials, vehicles, delivery units and receiving units.
Inputting the type, number, length, width and height of a carriage, bearing weight, repeated use times and priority information of the vehicle; inputting the name and longitude and latitude information of a delivery and delivery unit; inputting names and longitude and latitude information of all receiving units for distribution; inputting the length, width, height, weight, quantity, receiving units, stacking possibility, mutually exclusive tray material list and custom material constraint conditions of all distribution tray materials.
And 2, acquiring a scanning list for starting scanning at different initial degrees by using an improved scanning method.
According to the using priority of the vehicle, sequentially traversing the vehicle-to-vehicle distribution frequency list, taking a delivery unit (namely a distribution center) as an original point, scanning clockwise (or anticlockwise or clockwise and anticlockwise) at certain degrees, for example, 15 degrees as a starting degree, scanning tray materials of each receiving unit from the current starting degree, loading the materials according to a certain rule until the vehicles are not loaded, and then starting to repeat the loading operation by using a new idle vehicle.
For any one of the vehicles in the series,
firstly, traversing stackable tray materials which are not loaded by a current receiving unit and are not accessed by a current vehicle, if the current tray materials and the current vehicle loaded materials are not mutually exclusive materials, meet additional constraint conditions (under a specific scene), meet the length, width and height requirements of the vehicle, and meet the requirements of the bearing weight multiplication coefficient of the vehicle and the bearing weight multiplication coefficient of the vehicle after the tray materials are loaded, loading the stackable tray materials, and modifying the state of the stackable tray materials into the loaded state to generate a tray stackable material scanning result.
And then, if the vehicle is not full, traversing non-palletizable tray materials which are not loaded by the current receiving unit and are not accessed by the current vehicle, if the current tray materials and the current vehicle loaded materials are non-mutually exclusive tray materials, meet additional constraint conditions (under a specific scene), meet the requirements of length, width and height of the vehicle, and meet the requirements of the area multiplication coefficient of the compartment of the vehicle and the bearing weight multiplication coefficient of the vehicle after the tray materials are loaded, loading the non-palletizable tray materials, modifying the state of the non-palletizable tray materials into the loaded state, and generating a scanning result of the tray non-palletizable materials.
And when the loading of the palletizable tray materials and the non-palletizable tray materials of all receiving units is finished, adding the scanning results of the palletizable and non-palletizable tray materials of each vehicle into the scanning list to obtain the scanning list of the initial angle.
And scanning all different initial degrees once to obtain a scanning list of each initial angle.
For example, for the delivery task: two vehicles of 'car 1 by 1' and 'car 2 by 1' are provided to complete 'receiving unit 1' to distribute 'material A' and 'material B'; the receiving unit 4 distributes the material A and the material B; "receiving unit 3": distributing 'material A' and 'material B'; the receiving unit 2 distributes the tasks of the materials C and D. "multiply by 1" indicates that the vehicle can be used 1 time.
There are two scan lists of starting angles:
(a)
[{
"vehicle 1": [ { "receiving unit 1": { "material a", "material B" } }, and { "receiving unit 2": { "material C", "material D" } } ],
the 'vehicle 2': [ { "receiving unit 3": { "material a", "material B" } }, and { "receiving unit 4": { "Material A", "Material B" } } ]
}],
(b)
[{
"vehicle 2": [ { "receiving unit 1": { "material a", "material B" } }, and { "receiving unit 4": { "Material A", "Material B" } ],
the 'vehicle 1': [ { "receiving unit 3": { "material a", "material B" } }, and { "receiving unit 2": { "Material C", "Material D" } } ]
}]
The scan list shows: starting from this starting point, all the vehicles needed to complete the task, and what materials these vehicles carry to which unit; each entry in the scan list refers to what material a vehicle is responsible for delivering to which unit in the initial angle scan.
The two scanning lists (a) and (b) can independently complete the distribution task.
And 3, aiming at the scanning list of each starting angle in the step 1, calculating the vehicle path and cost of each vehicle in the scanning list by adopting a hybrid optimization algorithm combining a genetic algorithm and a simulated annealing algorithm. The route is a traveling route of the vehicle route from "receiving unit a" - > "receiving unit B" - > "receiving unit C" - > "receiving unit a"; the cost is the sum of the vehicle paths of all vehicles in the scan list.
For a certain vehicle, the specific method for optimizing the vehicle path based on the hybrid optimization algorithm of 'genetic-annealing' comprises the following sub-steps:
s1, the group size pop-size and other parameter values in the algorithm are given as follows: the parameters T0 at the initial temperature, a at the annealing, pc, pm at the crossover and mutation operations and q in the termination rule are selected.
Generation of initial population: randomly generating Pop-size chromosomes with the length of n as an initial population Pop (0), wherein n is the number of cities, namely the cities are receiving units; within each chromosome is a random sequence of city numbers (the sequence being the vehicle delivery path for that vehicle). In order to make the initial population spread over the entire solution space, and reflect the behavior of the search space as much as possible, the pop-size cannot be made too small, and becomes larger as the number of nodes increases, but too large increases the computation time.
Determining initial temperature and performing annealing operation: determination of initial temperature selection T 0 A form of = k δ, where k is a sufficiently large number, and k =10,20,100.. Et al test values may be selected; δ = f max -f min ，f max The largest objective function value (sum of delivery routes of vehicles) in the initial population, and fmin is the smallest objective function value in the initial population. The temperature-reducing function adopts a common T (n + 1) = alpha T (n), wherein 0 < alpha < 1.
The objective optimization function is determined as the vehicle path value of the vehicle Samin.
Let the initial optimal solution Smin = fmin and let p =0.
Termination rule: by monitoring the minimum objective function value f in each generation of the evolutionary population min To determine whether the algorithm is terminated. When the successive q generations do not change, the algorithm is considered to be converged, and the calculation is stopped.
S2, genetic optimization is carried out on the current vehicle path by adopting a genetic algorithm; the genetic algorithm comprises the following steps: selection, coding, crossing, mutation;
the adaptation function: the fitness function is the basis of genetic evolution operation, and the construction of the fitness function is the key of genetic algorithm. The reasonable fitness function can guide the search to the optimization direction, construct the fitness function based on the order, its characteristic is that the probability that the individual is chosen is independent of the concrete value of the objective function, therefore dispel many unreasonable results caused by objective function. All individuals in the population are arranged in a descending order according to the magnitude of the objective function value, the parameter alpha is set to be (0,1), and an adaptive function based on the order is defined as
eval(v i )＝α(1-α) i＝1 ，i＝1，2，...，popsize.
Wherein Vi is the ith individual after the population is sorted, and alpha is 0.01-0.3, which is beneficial to keeping the diversity of the population.
The selection process comprises the following steps: the selection process is based on a spin betting round pop-size number, each spin selecting a chromosome for a new population. The betting round selects chromosomes according to the fitness of each chromosome. First, the cumulative q is calculated for each chromosome i ,q 0 ＝0；q i ＝eval(v j ) J = 1.. ·, i; i =1,. Pop = size. Probability; then, a random number r is generated from the (0,pop-size) interval; then, if q is i-1 <r<q i Selecting the ith chromosome; finally, the two previous steps are repeated for pop-size times, so that pop-size copies of chromosomes can be obtained.
Of course, the selection may also be performed by a "tournament selection method", a "random traversal selection method", or the like.
And (3) an encoding process: the Grefenstette code is adopted, and for a city list w, an access sequence of T for each city is assumed to be T: stipulation T = (T) 1 ,t 2 ,...t n ) Each time a city is visited, it is removed from the city list W and the i-th (i =1,2,3.., n) city t is visited i At all W- { t 1 ,t 2 ,...t i-1 Corresponding position number g of city list i (1≤g i N-i + 1) may indicate which city is specifically visited, and so on until all cities in w have been processed. A list is obtained by putting together all the sequences: g = (G) 1 ,g 2 ,...g n ) A circuit route is represented, which is the gene of an individual in the genetic algorithm.
And (3) a crossing process: the crossover process is the primary method of generating new individuals. A conventional single point crossover is used. To determine the parent of the crossover operation, the following process is repeated from 0 to pop-size: a random number r is generated from [0,1], and Vi is selected as a parent if r < Pc. And (4) grouping the selected parents into two teams, and randomly generating a position to cross.
And (3) mutation process: the mutation operator aims to improve the genetic searching capability, maintain the diversity of the population and prevent the premature phenomenon. Uniform multipoint variation is employed.
New individual replication strategies: the population which is selected to be copied, crossed and mutated by a genetic algorithm is used as an initial population, and a copy strategy based on Markov chain and discrimination criteria is applied to generate a next generation population. Firstly, an optimal retention strategy is implemented (namely, the individuals with the best performance in the middle population are copied into the next generation population unconditionally, so that the best solution in the middle population is retained, the algorithm can be converged to the global optimal solution with the probability 1, and the convergence of the algorithm is ensured). Then, new individuals j are randomly generated in the field of chromosome i, i and j compete into the next generation, generating a next generation population Pop (n + 1).
S3, taking the genetic optimization result of the S2 as an initial condition, and performing secondary annealing optimization by adopting a simulated annealing method; calculating S' of the new generation population after annealing optimization.
S4, judging whether S ' is smaller than Smin, if so, making Smin = S ', S ' = fmin, and p =0, and executing the step S5; otherwise p = p +1, step S5 is performed.
S5, judging whether p is larger than or equal to q, and if so, taking Smin as an output result of the item in the list; otherwise, return to S2. Generally, q is 100.
Each item in the initial angle scanning list is calculated according to the method to obtain the output result of the item, and the optimal path of the vehicle is obtained; the length of the optimal path for all vehicles in the scan list and the cost as the scan list.
And 4, selecting the scanning list with the minimum replacement price and the optimal vehicle path from the scanning lists of all the initial angles as a final distribution scheme.
Thus, an improved method for distributing tray materials by a scanning method and a genetic simulation annealing method is completed.
In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (6)

1. An improved method for distributing tray materials by a scanning method and a genetic simulation annealing method is characterized by comprising the following steps:
step one, acquiring a scanning list for starting scanning at different initial degrees by using an improved scanning method:
setting a vehicle distribution list according to the vehicle use priority; starting to scan all receiving units by using the transmitting unit as an origin and different initial degrees to obtain a scanning list starting to scan by using different initial degrees;
when scanning and loading, starting from a vehicle with high priority, starting from a receiving unit of the current starting degree, firstly loading palladable tray materials which are not loaded, meet the size requirement of the vehicle, are not mutually exclusive with the currently loaded materials and meet the requirements of the area multiplication coefficient of the vehicle bearing body and the weight multiplication coefficient of the vehicle bearing body after loading on the vehicle, and modifying the loaded tray material state into the loaded state; if the vehicle is not full, loading non-palletizable pallet materials which are not loaded, are not mutually exclusive with currently loaded materials and meet the requirements of the area multiplication coefficient of the vehicle bearing body and the weight multiplication coefficient of the vehicle bearing body after loading on the vehicle from the current receiving unit, and modifying the loaded pallet material state into the loaded vehicle; loading all the tray materials of all the receiving units to obtain a scanning list of the initial degree; the scanning list comprises information of all vehicles and tray materials loaded by the vehicles; obtaining scanning lists corresponding to all different starting degrees;
step two, aiming at each scanning list, calculating the vehicle path and cost of each vehicle in the scanning list by adopting a genetic simulation annealing method; specifically, for a certain vehicle in a certain scan list:
step 2.1, firstly, genetic optimization is carried out on the current vehicle path by adopting a genetic algorithm; randomly generating an initial vehicle path;
2.2, using the vehicle path after genetic optimization in the step 2.1 as an initial condition, and adopting a simulated annealing method to perform annealing optimization on the vehicle path; calculating the length of the vehicle route after annealing optimization, comparing the length with the length of the current shortest vehicle route, if the length is smaller than the length of the current shortest vehicle route, taking the length of the vehicle route after annealing optimization as the length of the current shortest vehicle route, and returning to the step 2.1 as the current vehicle route for genetic optimization; if the current shortest vehicle route is larger than the current shortest vehicle route, taking the current shortest vehicle route as the current vehicle route, and returning to the step 2.1 for genetic optimization; stopping circulation until the lengths of the vehicle routes after N times of continuous annealing optimization are all larger than or equal to the length of the current shortest vehicle route, and taking the vehicle route corresponding to the length of the current shortest vehicle route as the final route of the vehicle in the scanning list;
obtaining the final paths of all vehicles in the scanning list; the length of the final path of all vehicles and the cost as the present scan list;
and step three, selecting the scanning list with the minimum cost and the vehicle path thereof from all the scanning lists corresponding to different starting degrees as a final distribution scheme.
2. The method of claim 1, wherein in step one, all receiving units are scanned clockwise and/or counterclockwise.
3. The method of claim 1, wherein in the first step, when the truck loading is scanned, the truck loading constraint is reduced after the increase according to the actual situation.
4. The method of claim 1, wherein the genetic algorithm of step 2.1 employs roulette for population selection.
5. The method of claim 1, wherein in the genetic algorithm of step 2.1, an optimal retention strategy is implemented.
6. The method of claim 1, wherein in step 2.2, the N =100.
CN202011603056.8A 2020-12-29 2020-12-29 Improved method for distributing tray materials by scanning method and genetic simulation annealing method Active CN112700190B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011603056.8A CN112700190B (en) 2020-12-29 2020-12-29 Improved method for distributing tray materials by scanning method and genetic simulation annealing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011603056.8A CN112700190B (en) 2020-12-29 2020-12-29 Improved method for distributing tray materials by scanning method and genetic simulation annealing method

Publications (2)

Publication Number Publication Date
CN112700190A CN112700190A (en) 2021-04-23
CN112700190B true CN112700190B (en) 2023-04-07

Family

ID=75512221

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011603056.8A Active CN112700190B (en) 2020-12-29 2020-12-29 Improved method for distributing tray materials by scanning method and genetic simulation annealing method

Country Status (1)

Country Link
CN (1) CN112700190B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114971011B (en) * 2022-05-24 2024-04-23 燕山大学 Multi-mode intermodal route optimization method based on improved genetic simulated annealing algorithm

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102930340A (en) * 2012-10-26 2013-02-13 上海电力学院 Adaptive genetic annealing calculation method for solving zero-one knapsack problem
CN104636813B (en) * 2013-11-12 2018-02-06 中国科学院沈阳计算技术研究所有限公司 A kind of blending heredity simulated annealing for solving Job-Shop Scheduling Problem
CN109934364B (en) * 2019-03-10 2023-04-25 西北工业大学 Tool on-time distribution method based on genetic simulated annealing algorithm

Also Published As

Publication number Publication date
CN112700190A (en) 2021-04-23

Similar Documents

Publication Publication Date Title
CN111860754B (en) AGV scheduling method based on ant colony and genetic algorithm
CN108764777B (en) Electric logistics vehicle scheduling method and system with time window
Surekha et al. Solution to multi-depot vehicle routing problem using genetic algorithms
CN111178582B (en) Logistics distribution optimization method based on improved genetic algorithm
CN109919376A (en) Multi-field model and multi-vehicle-type vehicle route dispatch control method
Mohammed et al. Using genetic algorithm in implementing capacitated vehicle routing problem
CN109559062B (en) Task allocation and path planning method for cooperative logistics problem
CN111798067B (en) Automatic driving automobile distribution path planning method based on self-adaptive large neighborhood search algorithm
Ting et al. Multi-vehicle selective pickup and delivery using metaheuristic algorithms
CN107909228B (en) Dynamic vehicle goods receiving and dispatching path planning method and device based on modular factor calculation
CN110163544A (en) A method of the genetic algorithm based on optimization carries out goods delivery
CN113343575A (en) Multi-target vehicle path optimization method based on improved ant colony algorithm
CN113570112B (en) Optimization algorithm for solving vehicle path problem with time window considering cooperation
CN110276488A (en) A kind of vehicle routing optimization method based on matrix in block form and fuzzy haulage time
CN112700190B (en) Improved method for distributing tray materials by scanning method and genetic simulation annealing method
CN114444843A (en) Agricultural product green logistics distribution vehicle scheduling method and system based on large-scale variable neighborhood search strategy
CN115879657A (en) Electric vehicle power station changing location path optimization method considering multi-station capacity design
Alvarez et al. Scatter search for network design problem
Wu et al. Green vehicle routing problem: the tradeoff between travel distance and carbon emissions
Fallah et al. A green competitive vehicle routing problem under uncertainty solved by an improved differential evolution algorithm
CN116187599B (en) Multi-type intermodal railway logistics center point distribution method and system based on genetic algorithm
CN111985730A (en) Logistics path planning method and device, computer equipment and storage medium
Oonsrikaw et al. Enhanced ant colony optimization with local search
Chen et al. A hybrid algorithm for multi-depot vehicle routing problem
CN114118600A (en) Route planning method for urban garbage collection

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
GR01 Patent grant
GR01 Patent grant