CN112633729B - Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm - Google Patents

Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm Download PDF

Info

Publication number
CN112633729B
CN112633729B CN202011600765.0A CN202011600765A CN112633729B CN 112633729 B CN112633729 B CN 112633729B CN 202011600765 A CN202011600765 A CN 202011600765A CN 112633729 B CN112633729 B CN 112633729B
Authority
CN
China
Prior art keywords
lambda
station
cargo space
compartment
solution
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
CN202011600765.0A
Other languages
Chinese (zh)
Other versions
CN112633729A (en
Inventor
蔡敏
沈琼玮
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.)
Hangzhou Dianzi University
Original Assignee
Hangzhou Dianzi 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 Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN202011600765.0A priority Critical patent/CN112633729B/en
Publication of CN112633729A publication Critical patent/CN112633729A/en
Application granted granted Critical
Publication of CN112633729B publication Critical patent/CN112633729B/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
    • G06Q10/06—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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • 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
    • G06Q10/06313—Resource planning in a project environment
    • 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)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Quality & Reliability (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a multi-compartment material vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm, which comprises the following steps of: s1, determining parameters related to the goods space of a multi-compartment material vehicle; s2, establishing a multi-compartment material vehicle cargo space optimization model based on the determined parameters; and S3, solving the established multi-compartment material vehicle cargo space optimization model by adopting an Epsilon greedy algorithm, and outputting a cargo space assignment scheme. The loading scheme of the multi-compartment material vehicle can reduce energy consumption in loading work.

Description

Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm
Technical Field
The invention relates to the technical field of human factors engineering and goods space optimization, in particular to a multi-compartment material vehicle goods space optimization method based on human factors and an Epsilon greedy algorithm.
Background
The multi-compartment traction material vehicle is a common material transport tool in a manufacturing workshop, and a plurality of compartments can also feed more stations in one-time distribution, so that the efficiency of material distribution is improved. Meanwhile, compared with a large tray, the small material box applicable to the multi-compartment material vehicle can well reduce the risk of injury in the carrying operation of workers.
The idea of the greedy algorithm is that a new local optimal solution is selected in each iteration from local, and then the global optimal solution is approached, so that a plurality of unnecessary exhaustive steps for searching the global optimal solution are omitted. Therefore, the invention provides a multi-compartment material vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm.
Disclosure of Invention
The invention aims to provide a multi-compartment material vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm, aiming at the defects of the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a multi-compartment material and vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm comprises the following steps:
s1, determining parameters related to the goods space of a multi-compartment material vehicle;
s2, establishing a multi-compartment material vehicle cargo space optimization model based on the determined parameters;
and S3, solving the established multi-compartment material vehicle cargo space optimization model by adopting an Epsilon greedy algorithm, and outputting a cargo space assignment scheme.
Further, the parameters determined in step S1 include the number and weight of the material boxes, the number of carriages of the multi-carriage material vehicle, the number of shelf layers, and the height of the shelf.
Further, in step S2, a multi-compartment material truck cargo space optimization model is established, which is represented as:
Figure BDA0002868764350000011
wherein MinEp(x) The method comprises the steps of establishing a multi-compartment material vehicle cargo space optimization model by taking the minimum total assembly required energy as an objective function; δ (k, l) represents the energy required by a worker to place a bin k on a shelf l;
s.t.
Figure BDA0002868764350000021
Figure BDA0002868764350000022
Figure BDA0002868764350000023
Figure BDA0002868764350000024
wherein, the formula (1) is an objective function; formula (2) is a decision variable; formula (3) is that each material box will be assigned to one material shelf; the formula (4) shows that two material boxes with the same target station can be arranged in the same carriage; the formula (5) is that each layer of the goods shelf is not overloaded.
Further, the step S3 specifically includes:
s31, randomly generating a plurality of feasible station division schemes, calculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, calculating the corresponding lowest energy consumption, and selecting a feasible solution with the minimum required energy as an initial solution of a greedy algorithm;
and S32, exchanging the stations contained in the sub station set randomly, and carrying out iterative optimization by adopting an Epsilon greedy strategy to obtain an optimal solution.
Further, the step S31 is specifically:
s311, initializing, and enabling
Figure BDA0002868764350000025
ru＝mu·μ；
Figure BDA0002868764350000026
Order to
Figure BDA0002868764350000027
Wherein λ isuE lambda represents the set of workstations assigned to car u,
Figure BDA0002868764350000028
rurepresents the current remaining capacity of the car U, and U ∈ U ═ 1, …, W };
Figure BDA0002868764350000029
representing the remaining set of unassigned workstations;
s312, processing all the compartments U e U ═ 1, …, W;
s313, obtaining a random feasible solution lambda ═ lambda1,……,λW}。
Further, step S312 specifically includes:
s3121, finding all station sets temp ═ i | | B placed in the carriage ui|≤ru}；
S3122, if the temp is not empty, randomly selecting a station i from the temp, and enabling all material boxes B to be needediPut into the carriage u, i.e. i is added into lambdauIn (b) to obtain λu′＝λu+ i, with the remaining capacity of the car at this time being ru′＝mu·μ-|BiL, set of remaining unassigned sites
Figure BDA00028687643500000210
If temp is empty, go to step S3123;
s3123, if there are residual unallocated site sets
Figure BDA0002868764350000031
The total number of the corresponding material boxes is less than or equal to the total capacity of the rest compartments, i.e.
Figure BDA0002868764350000032
Continuing from step S3121 with the probability α, and performing the next loop with the probability 1- α; if not, the process continues to step S3121.
Further, the step S32 is specifically:
s321, generating T feasible solutions Lambda1,……,ΛTCalculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, calculating the corresponding lowest energy consumption, selecting the feasible solution with the minimum required energy from the T feasible solutions, taking the feasible solution as the initial solution of the greedy algorithm, and recording the total energy consumption of the optimal solution as EpThe initial solution is Λ ═ λ1,……,λW}；
S322, setting the maximum iteration number as M, and stopping when the maximum iteration number is reached;
s323, outputting the obtained optimal station division scheme Lambda*＝{λ1,……,λWH, and corresponding material box allocation scheme, optimal objective function value (minimum energy consumption) Ep *。
Further, step S322 specifically includes:
s3221. randomly selecting the sub-station set lambda in the work station set1，λ2；
S3222. from lambda1From a random selection of a station i1From λ2In the process, a station i is randomly selected2To i, pair1And i2Exchange is carried out, i.e. i2Adding lambda1I is to1Adding lambda2；
S3223. if after switching, lambda1，λ2The total number of material boxes required by the stations contained in the material box container does not exceed the maximum capacity, then:
a. distributing the material boxes contained in all the carriages to the shelves of each layer by using the Hungarian algorithm, and obtaining the minimum energy consumption E which can be reached by each carriagep(λ1),Ep(λ2),……,Ep(λW) The objective function value of the current solution is Ep′＝∑u∈{1,…,w}Ep(λu)；
b. If the random probability α < ∈ or Ep′＜EpLet Ep＝Ep', and updating the current optimal solution;
s3224. if Q times are accumulated, EpIs no longer decreasing, then exit.
Compared with the prior art, the loading scheme of the multi-compartment material vehicle can reduce energy consumption in loading work.
Drawings
Fig. 1 is a flowchart of a multi-compartment material-vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm according to an embodiment;
fig. 2 is an iteration diagram of total energy consumption in loading and unloading under the Epsilon-greedy strategy provided by the second embodiment.
Detailed Description
The following embodiments of the present invention are provided by way of specific examples, and other advantages and effects of the present invention will be readily apparent to those skilled in the art from the disclosure herein. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
The invention aims to provide a multi-compartment material vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm, aiming at the defects of the prior art.
Example one
The embodiment provides a multi-compartment material and vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm, as shown in fig. 1, the method comprises the following steps:
s1, determining parameters related to the goods space of a multi-compartment material vehicle;
s2, establishing a multi-compartment material vehicle cargo space optimization model based on the determined parameters;
and S3, solving the established multi-compartment material vehicle cargo space optimization model by adopting an Epsilon greedy algorithm, and outputting a cargo space assignment scheme.
In step S1, parameters relating to the multi-car material yard are determined.
Determining the number and the weight of the material boxes; determining parameters such as the number of carriages, the number of layers of goods shelves, the height of the goods shelves and the like of the multi-carriage material vehicle; determining the energy required for a worker to place a material box k on a shelf l as delta (k, l), the value of which is related to the weight of the material box and the height of the shelf; the target station of each material box is determined, one material box is only sent to one station, and one station can receive a plurality of material boxes.
In step S2, a multi-car material yard optimization model is established based on the determined parameters.
And (3) establishing a multi-compartment material vehicle cargo space optimization model by taking the minimum total assembly required energy as an objective function:
Figure BDA0002868764350000041
wherein MinEp(x) Representing the establishment of multi-compartment materials with the objective function of minimizing the total energy requiredA vehicle cargo space optimization model; δ (k, l) represents the energy required by the worker to place a material box k on a shelf l;
s.t.
Figure BDA0002868764350000042
Figure BDA0002868764350000051
Figure BDA0002868764350000052
Figure BDA0002868764350000053
wherein, the formula (1) is an objective function; equation (2) is the decision variable, x when an item cassette k is placed on shelf l lk1 is ═ 1; otherwise, xlk0; formula (3) ensures that each bin will be and will only be assigned to one shelf; the formula (4) ensures that two material boxes with the same target station can be arranged in the same carriage; the formula (5) ensures that each layer of goods shelf is not overloaded. This means that each car is not overloaded. It is assumed here that a multi-compartment material cart only limits the number of loadable cassettes, but does not take into account whether the total weight of the cassettes is overloaded or not, since a material cart for parts distribution in a workshop is usually not loaded with large, heavy parts.
In step S3, an Epsilon greedy algorithm is used to solve the established multi-compartment material-vehicle cargo space optimization model, and a cargo space assignment scheme is output.
Using Λ ═ λ1,……,λWDenotes a station division scheme in which λu∈Λ，
Figure BDA0002868764350000058
Figure BDA0002868764350000059
Is a set of workstations assigned to the car u, i.e. each sub-set of workstations λuCorresponds to one carriage. Definition of ruAnd U ∈ U ═ {1, …, W }, which is the current remaining capacity of the car U. Definition of
Figure BDA0002868764350000057
The remaining unassigned workstation sets.
S31, randomly generating a plurality of feasible station division schemes, calculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, calculating the corresponding lowest energy consumption, and selecting a feasible solution with the minimum required energy as an initial solution of a greedy algorithm;
algorithm a can be used to randomly generate a feasible station partitioning scheme Λ. Repeating the algorithm A for T times to obtain a plurality of feasible solutions Lambda1,……,ΛT. And for each feasible solution, calculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, and calculating the corresponding lowest energy consumption. And finally, selecting the feasible solution with the minimum required energy from the t feasible solutions to serve as the initial solution of the greedy algorithm.
Algorithm A: the randomly generated feasible station division scheme specifically comprises the following steps:
s311. initialization
Firstly, initialization is carried out: order to
Figure BDA0002868764350000054
ru＝mu·μ；
Figure BDA0002868764350000055
Order to
Figure BDA0002868764350000056
S312, for all cars U e U ∈ U {1, …, W }, the following operations are performed in sequence starting from U ═ 1:
s3121, finding all station sets temp ═ i | | B in which the cars u can be placedi|≤ru}。
S3122, if the temp is not empty, randomly selecting a station i from the temp, and putting all material boxes B required by the station i into the material box BiPut into the car u, i.e. add i to lambdauIn (b) to obtain λu′＝λu+ i, with the remaining capacity of the car at this time being ru′＝mu·μ-|BiI, set of remaining unallocated sites
Figure BDA0002868764350000061
If temp is empty, then c is executed.
S3123. if there are remaining unallocated site sets
Figure BDA0002868764350000062
The total number of the corresponding material boxes is less than or equal to the total capacity of the rest compartments, i.e.
Figure BDA0002868764350000063
Continuing to execute the step a with the probability alpha, and executing the next loop with the probability 1-alpha; otherwise, execution continues from step a.
S313, obtaining a random feasible solution lambda ═ lambda1,……,λW}。
And S32, exchanging the stations contained in the sub station set randomly, and carrying out iterative optimization by adopting an Epsilon greedy strategy to obtain an optimal solution.
From the feasible solution generated in the first stage, it may happen that the car allocated earlier is full, and the car allocated later has fewer material boxes, and the rack which is less laborious to load and unload is not utilized. Thus, the second stage will randomly pair the set of child workstations λ1，λ2E, exchanging the stations contained in the lambda, and adopting an EpsilonGreedy strategy to prevent the iteration from falling into local optimization. Specifically by algorithm B.
And algorithm B: the Epsilon greedy optimized material box position allocation specifically comprises the following steps:
s321, generating an initial solution
T feasible solutions Lambda are obtained by repeatedly executing the algorithm A for T times1,……,ΛT. For each feasible solution, the Hungarian algorithm is usedAnd calculating the optimal material box distribution position of each compartment, and calculating the corresponding lowest energy consumption. And finally, selecting the feasible solution with the minimum required energy from the T feasible solutions to serve as the initial solution of the greedy algorithm. The total energy consumption (objective function value) of the optimal solution is recorded as EpThe initial solution (current optimal solution) is Λ ═ λ1,……,λW}。
S322. iterative optimization
The maximum number of iterations is set to M. In each iteration, the following steps a, b, c, d are executed, and the process stops after the maximum iteration number M is reached. And setting the early stop times as Q, namely after accumulating the iteration Q times, finding that the value of the current objective function value (minimum energy consumption) E is not reduced any more, and exiting to avoid invalid iteration. And setting epsilon probability as epsilon.
S3221. randomly selecting the sub-station set lambda in the work station set1，λ2
S3222. from lambda1From a random selection of a station i1From λ2In the process, a station i is randomly selected2To i, pair1And i2Exchange is carried out, i.e. i2Adding lambda1I is to1Adding lambda2。
S3223. if after switching, lambda1，λ2The total number of material boxes required by the stations contained in the material box does not exceed the maximum capacity, then
a. Distributing the material boxes contained in all the carriages to the shelves of each layer by using the Hungarian algorithm, and obtaining the minimum energy consumption E which can be reached by each carriagep(λ1),Ep(λ2),……,Ep(λW) The objective function value of the current solution is Ep′＝∑u∈{1,…,w}Ep(λu)。
b. If the random probability α < ∈ or Ep′＜EpLet Ep＝Ep', and updates the current optimal solution.
S3224. if Q times are accumulated, EpIs no longer decreasing, then exit.
S323, outputting the obtained optimal station division scheme Lambda*＝{λ1,……,λWH, and corresponding material box allocation scheme, optimal objective function value (minimum energy consumption) Ep *。
Compared with the prior art, the loading scheme of the multi-compartment material vehicle can reduce energy consumption in loading work.
Example two
The difference between the multi-compartment material vehicle cargo space optimization method based on human factors and the Epsilon greedy algorithm provided by the embodiment and the embodiment I is that:
the present embodiment is described with specific examples.
An independent material warehouse is arranged in the workshop and supplies materials for each station or assembly line. The large and medium-sized workshops can be abstracted into a 300m × 300m square, independent material warehouses are arranged in the workshops, the position coordinates are (150 ), and materials are supplied for all stations or assembly lines. There are 50 stations, and the material kind and the quantity of demand differ between the station. According to the survey analysis, the following settings were made for the parameters: the number of the stations is n equal to 50; the materials required by each station are placed in the material boxes in advance and stored in the warehouse, and N is 162 material boxes; the total number of the material cars for distribution is 6, each car is provided with 4 layers of shelves with different heights, and at most 10 material boxes can be placed on each shelf, so that the maximum load of each car is 40, and the maximum load of each material car is 240. The height of the first layer of the shelf is 1.55m, and the energy required for loading and unloading a single material box k is delta (k,1) ═ w (k) · 0.1823+4.4521 kcal; the height of the second layer of shelves is 1.25m, and the energy required for loading and unloading a single material box k is delta (k,2) ═ w (k) · 0.1643+3.6206 kcal; the height of the third layer of the shelf is 0.95m, and the energy required for loading and unloading a single material box k is delta (k,3) ═ w (k) · 0.1504+ 2.0162 kcal; the third shelf height is 0.65m, and the energy required to load and unload a single cartridge k is δ (k,3) ═ w (k) 0.1911+4.3123 kcal. The energy consumption prediction formula is obtained by regression of experimental data. And at the moment t being 0, the water spider starts to load the material trolley in the material supermarket, and after the loading task is completed, the water spider can slightly wait or immediately start to carry out material distribution, wherein the time is specifically determined by the time window of each station. . The weight w (k) of each material box and the destination station i (k) are shown in table 1.
Figure 1
Figure BDA0002868764350000091
Table 1 weight of each material box waiting for distribution and destination station information
Firstly, executing the T times by using the algorithm A to be 100 times to obtain 100 feasible station division schemes, solving each feasible solution by using the Hungarian algorithm to obtain the feasible solution with the minimum energy required by loading, wherein the feasible solution is as follows:
λ1＝{29,18,24}；
λ2＝{9,35,22,34,8}；
λ3＝{26,11,17,43,45,31}；
λ4＝{1,49,27,20,41,39,40,32,15,30,12,44}；
λ5＝{21,3,28,14,37,13,2,50,25,47,48,16,42}；
λ6＝{10,4,46,5,23,6,19,33,38,36,7}。
therefore, the feasible solution is used as an initial solution, and the corresponding objective function value (total energy required for loading) is 607.3704 kcal.
Then, from the initial solution, iterative optimization is performed, as described in section 4.2.2. And setting the Epsilon probability Epsilon to be 0.05 after a plurality of experiments. The algorithm is implemented by using Python and runs in visual studio code software, and the change situation of the target function value in the iteration process is shown in FIG. 2. After 200 iterations, the objective function value appears to converge.
After iterative optimization, as shown in table 2, the station division condition corresponding to the obtained optimal solution is as follows:
λ1＝{33,34,38,25,42,27,12}；
λ2＝{1,19,5,24,8,9,29}；
λ3＝{34,3,6,45,14,48,16,26,31}；
λ4＝{17,50,39,41,44,47,32}；
λ5＝{2,4,40,10,43,15,49,18,20,30}；
λ6＝{35,37,7,11,13,46,21,22,23,28}。
each sub-station set corresponds to one carriage, namely, the material boxes required by the stations in the set are placed into the same carriage. By applying the Hungarian algorithm, the position of the material box is optimized for each carriage, and the minimum energy consumption which can be reached by each carriage in the optimal solution is obtained as Ep(λ1)＝99.3979kcal， Ep(λ2)＝99.5209kcal，Ep(λ3)＝104.4860kcal，Ep(λ4)＝93.1606kcal， Ep(λ5)＝94.0009kcal，Ep(λ6) 103.7779kcal, the energy required to load and unload all the cartridges is therefore Ep＝Ep(λ1)+Ep(λ1)+Ep(λ1)+Ep(λ1)+Ep(λ1) 594.3342kcal, which is the optimal objective function value.
Figure BDA0002868764350000101
Figure BDA0002868764350000111
TABLE 2 greedy algorithm solution of cargo space assignment scheme results to minimize energy consumption for loading
When the loading scheme of the multi-compartment material vehicle described in the embodiment is not used, the loading scheme is obtained by observing the daily loading work of the spider, a common loading scheme is that the material boxes are loaded from the first floor of the first compartment according to the serial numbers of the material boxes, after the first compartment is loaded, if the material boxes which are not loaded remain, the loading is continued from the first floor of the second compartment,this is done until all the material boxes have been loaded. For the same 162 material boxes as shown in table 1, this example simulates and calculates the energy required in this way. When the method is used, the total number of the material boxes placed on the first floor of all the compartments is 42, the total weight is 140.31kg, and the consumed energy is E1140.31 · 0.1823+4.4521 ═ 212.5667 kcal; the total number of the material boxes on the second layer of all the carriages is 40, the total weight is 97.83kg, and the energy consumption is E297.83 · 0.1643+3.6206 × 40 ═ 160.8975 kcal; the total number of the material boxes on the third layer of all the compartments is 40, the total weight is 123.39kg, and the energy consumption is E3123.39 · 0.1504+2.0152 × 40 ═ 99.1659 kcal; the total number of the material boxes on the fourth layer of all the carriages is 40, the total weight is 114.93kg, and the energy consumption is E4114.93 · 0.1911+4.3123 ═ 194.4551 kcal. Therefore, when all 162 material boxes are loaded by the loading method used by the current water spider, the total energy required is E-E1+E2+E3+E4＝667.0852kcal。
Through comparison, the loading scheme of the multi-compartment material truck can reduce the energy consumption of the water spider by 10.91% in the loading process. In addition, if the water spiders transfer the materials according to the serial numbers of the material boxes, the material boxes required by the same station are dispersed in different carriages without using the loading strategy, so that the water spiders do not need to go back and forth between carriages when distributing the materials, and mental load of the water spiders when searching the material boxes is increased.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (3)

1. A multi-compartment material vehicle cargo space optimization method based on human factors and an Epsilon greedy algorithm is characterized by comprising the following steps:
s1, determining parameters related to the goods space of a multi-compartment material vehicle;
s2, establishing a multi-compartment material vehicle cargo space optimization model based on the determined parameters;
s3, solving the established multi-compartment material vehicle cargo space optimization model by adopting an Epsilon greedy algorithm, and outputting a cargo space assignment scheme;
the parameters determined in the step S1 comprise the number and weight of the material boxes, the number of carriages of the multi-carriage material vehicle, the number of shelf layers and the height of the shelf;
in step S2, a multi-compartment material vehicle cargo space optimization model is established, which is expressed as:
Figure FDA0003570866090000011
wherein MinEp(x) Representing that a multi-compartment material vehicle cargo space optimization model is established by taking the energy required by the minimized total loading as an objective function; δ (k, l) represents the energy required by the worker to place a material box k on a shelf l;
s.t.
Figure FDA0003570866090000012
Figure FDA0003570866090000013
Figure FDA0003570866090000014
Figure FDA0003570866090000015
wherein, the formula (1) is an objective function; formula (2) is a decision variable; formula (3) is that each material box will be assigned to one material shelf; the formula (4) shows that two material boxes with the same target station can be arranged in the same carriage; the formula (5) is that each layer of goods shelf can not be overloaded;
step S3 specifically includes:
s31, randomly generating a plurality of feasible station division schemes, calculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, calculating the corresponding lowest energy consumption, and selecting a feasible solution with the minimum required energy as an initial solution of a greedy algorithm;
s32, exchanging the stations contained in the sub-station set randomly, and carrying out iterative optimization by adopting an Epsilon greedy strategy to obtain an optimal solution;
step S31 specifically includes:
s311, initializing, and enabling
Figure FDA0003570866090000016
ru＝mu·μ；
Figure FDA0003570866090000017
Order to
Figure FDA0003570866090000021
Wherein λ isuE lambda represents the set of workstations assigned to car u,
Figure FDA0003570866090000022
rurepresents the current remaining capacity of the car U, U ∈ U ═ 1, …, W };
Figure FDA0003570866090000023
representing the remaining unassigned set of workstations;
s312, processing all the compartments U e U ═ 1, …, W;
s313. getTo random feasible solution lambda ═ lambda { (λ)1,……,λW}；
Step S312 specifically includes:
s3121, finding all station sets temp ═ i | | B placed in the carriage ui|≤ru}；
S3122, if the temp is not empty, randomly selecting a station i from the temp, and enabling all material boxes B to be needediPut into the compartment u, i.e. i is added into lambdauIn (b) to obtain λu′＝λu+ i, with the remaining capacity of the car at this time being ru′＝mu·μ-|BiL, set of remaining unassigned sites
Figure FDA0003570866090000024
If temp is empty, go to step S3123;
s3123, if there are residual unallocated site sets
Figure FDA0003570866090000025
The total number of the corresponding material boxes is less than or equal to the total capacity of the rest compartments, i.e.
Figure FDA0003570866090000026
Continuing from step S3121 with the probability α, and performing the next loop with the probability 1- α; if not, the process continues to step S3121.
2. The multi-compartment material and vehicle cargo space optimization method based on the human factor and Epsilon greedy algorithm as claimed in claim 1, wherein the step S32 specifically comprises:
s321, generating T feasible solutions Lambda1,……,ΛTCalculating the optimal material box distribution position of each carriage by using a Hungarian algorithm, calculating the corresponding lowest energy consumption, selecting the feasible solution with the minimum required energy from the T feasible solutions, taking the feasible solution as the initial solution of the greedy algorithm, and recording the total energy consumption of the optimal solution as EpThe initial solution is Λ ═ λ1,……,λW}；
S322, setting the maximum iteration number as M, and stopping when the maximum iteration number is reached;
s323, outputting the obtained optimal station division scheme Lambda*＝{λ1,……,λW}, corresponding material box distribution scheme and optimal objective function value Ep *。
3. The multi-compartment material-vehicle goods-space optimization method based on the human factor and Epsilon greedy algorithm as claimed in claim 2, wherein the step S322 specifically comprises:
s3221. randomly selecting the sub-station set lambda in the work station set1，λ2；
S3222. from λ1From a random selection of a station i1From λ2In the process, a station i is randomly selected2To i, pair1And i2Exchange is carried out, i.e. i is2Adding lambda1I is to1Adding lambda2；
S3223. if after switching, lambda1，λ2The total number of the material boxes required by the stations in the system does not exceed the maximum capacity, then:
a. distributing the material boxes contained in all the carriages to the shelves of each layer by using the Hungarian algorithm, and obtaining the minimum energy consumption E which can be reached by each carriagep(λ1),Ep(λ2),……,Ep(λW) The objective function value of the current solution is Ep′＝∑u∈{1,…,w}Ep(λu)；
b. If the probability of randomness is alpha<E or Ep′<EpLet Ep＝Ep', and updating the current optimal solution;
s3224. if Q times are accumulated, EpIs no longer decreasing, then exit.
CN202011600765.0A 2020-12-29 2020-12-29 Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm Active CN112633729B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011600765.0A CN112633729B (en) 2020-12-29 2020-12-29 Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011600765.0A CN112633729B (en) 2020-12-29 2020-12-29 Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm

Publications (2)

Publication Number Publication Date
CN112633729A CN112633729A (en) 2021-04-09
CN112633729B true CN112633729B (en) 2022-06-10

Family

ID=75287150

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011600765.0A Active CN112633729B (en) 2020-12-29 2020-12-29 Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm

Country Status (1)

Country Link
CN (1) CN112633729B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106651089A (en) * 2016-09-19 2017-05-10 清华大学 Modeling and optimization method for production scheduling problem distribution set robust model
CN109597304A (en) * 2018-11-30 2019-04-09 北京工业大学 Die storehouse Intelligent partition storage method based on artificial bee colony algorithm
CN110909930A (en) * 2019-11-20 2020-03-24 浙江工业大学 Goods position distribution method of mobile goods shelf storage system for refrigeration house
CN111626659A (en) * 2020-04-17 2020-09-04 宏图智能物流股份有限公司 Warehouse delivery goods space recommendation method based on greedy algorithm
CN111768240A (en) * 2020-06-30 2020-10-13 杭州电子科技大学 Aero service recommendation method and system based on perceptual engineering and Kano model
CN111815233A (en) * 2020-06-24 2020-10-23 武汉理工大学 Goods position optimization method based on total logistics amount and energy consumption

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
MX2016014234A (en) * 2014-04-29 2017-02-14 Wix Com Ltd System and method for the creation and use of visually-diverse high-quality dynamic layouts.
CN103942670B (en) * 2014-04-30 2017-02-15 东北大学 Steel strand finished product storage and distribution method
CN110065515B (en) * 2019-05-08 2020-03-27 郑州铁路职业技术学院 Railway freight carriage convenient for loading and unloading goods

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106651089A (en) * 2016-09-19 2017-05-10 清华大学 Modeling and optimization method for production scheduling problem distribution set robust model
CN109597304A (en) * 2018-11-30 2019-04-09 北京工业大学 Die storehouse Intelligent partition storage method based on artificial bee colony algorithm
CN110909930A (en) * 2019-11-20 2020-03-24 浙江工业大学 Goods position distribution method of mobile goods shelf storage system for refrigeration house
CN111626659A (en) * 2020-04-17 2020-09-04 宏图智能物流股份有限公司 Warehouse delivery goods space recommendation method based on greedy algorithm
CN111815233A (en) * 2020-06-24 2020-10-23 武汉理工大学 Goods position optimization method based on total logistics amount and energy consumption
CN111768240A (en) * 2020-06-30 2020-10-13 杭州电子科技大学 Aero service recommendation method and system based on perceptual engineering and Kano model

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Ergonomic workplace design in the fast pick area;Alena Otto 等;《OR Spectrum》;20170503;第39卷;全文 *
基于贪婪算法的货位优化模型;刘志帅 等;《物流科技》;20131231;第36卷(第9期);第99-101页 *
考虑供料策略和人因的装配线平衡问题建模研究;蔡敏 等;《系统工程》;20200229(第2期);全文 *

Also Published As

Publication number Publication date
CN112633729A (en) 2021-04-09

Similar Documents

Publication Publication Date Title
CN109784566B (en) Order sorting optimization method and device
CN110723448B (en) Intelligent warehousing system and control method thereof
CN109064081B (en) Continuous batch order fulfillment
CN109102205A (en) Radio frequency, processing method, device, server and the medium of order taking responsibility
Zäpfel et al. Warehouse sequencing in the steel supply chain as a generalized job shop model
EP3995416A1 (en) Container storage system, warehousing system, and robot
CN216735894U (en) Storage scheduling system
CN109767151A (en) Storage management method, device, medium and electronic equipment
CN111754014B (en) Control method for picking workstation ex-warehouse task, storage medium and electronic equipment
CN111915257B (en) Container processing method, device, control equipment, system and storage medium
CN216188201U (en) Storage scheduling system
CN112132522A (en) Material warehousing method and device and management equipment
CN111222819B (en) Method for determining a cargo vehicle, warehouse management device, system, device and medium
CN112132520A (en) Material warehousing method and device and management equipment
CN111582781B (en) Method for distributing goods shelves according to replenishment orders and computer readable storage medium
CN110134985A (en) The modeling of Multilayer shuttle car system conveyer buffer storage length and optimization method
CN116002257A (en) Warehouse system, warehouse system scheduling method, warehouse system scheduling device and electronic equipment
CN112633729B (en) Multi-compartment material vehicle cargo space optimization method based on human factors and Epsilon greedy algorithm
JP2023531969A (en) Multi-position search
CN111382969A (en) Order processing method, device, equipment and storage medium
CN112581032A (en) Dynamic programming-based multi-compartment material vehicle cargo space optimization method
JP2023524443A (en) Optimizing the operating capacity of container handling vehicles assigned to interact with the same ports for transferring storage containers to and from automated storage and retrieval systems. good use
JP5229553B2 (en) Goods storage equipment
CN113674051B (en) Goods delivery method, device, electronic equipment and storage medium
JP2004359452A (en) Location arrangement method for flatly placing warehouse and flatly placing warehouse

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