CN109447349A - A kind of manufacturing service supply chain optimization method of Based on Networked correlation perception - Google Patents
A kind of manufacturing service supply chain optimization method of Based on Networked correlation perception Download PDFInfo
- Publication number
- CN109447349A CN109447349A CN201811268270.5A CN201811268270A CN109447349A CN 109447349 A CN109447349 A CN 109447349A CN 201811268270 A CN201811268270 A CN 201811268270A CN 109447349 A CN109447349 A CN 109447349A
- Authority
- CN
- China
- Prior art keywords
- candidate
- service
- supply chain
- scheme
- manufacturing service
- 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.)
- Granted
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 84
- 238000005457 optimization Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000008447 perception Effects 0.000 title claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims abstract description 34
- 238000012546 transfer Methods 0.000 claims abstract description 16
- 238000013178 mathematical model Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 14
- 239000002131 composite material Substances 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 9
- 230000008520 organization Effects 0.000 description 8
- 239000000203 mixture Substances 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 238000011835 investigation Methods 0.000 description 3
- 230000006855 networking Effects 0.000 description 3
- 102000002274 Matrix Metalloproteinases Human genes 0.000 description 2
- 108010000684 Matrix Metalloproteinases Proteins 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 210000003027 ear inner Anatomy 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000005242 forging Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06Q10/0633—Workflow analysis
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of manufacturing service supply chain optimization method of Based on Networked correlation perception, the manufacturing service combined mathematical model of Based on Networked correlation perception is proposed to describe manufacturing service supply chain optimization problem；Using the encoding scheme based on matrix, to formally indicate manufacturing service supply chain candidate scheme, the composite structure type (selection or parallel) of first element representation vertical cartel service group of each column in matrix, and the selected basic candidate service of other element representations in arranging, corresponding subtask is completed in collaboration, biogeography is optimized into (Biogeography-based Optimization, BBO) transfer operator of algorithm is combined with ABC algorithm, to improve manufacturing service supply chain optimization efficiency.By the comprehensive comparison to three aspects such as practicability, validity and efficiency, method of the invention is better than the other methods of the prior art on solving the problems, such as manufacturing service supply chain optimization.
Description
Technical field
The invention belongs to manufacturing service supply chain optimization technical fields more particularly to a kind of Based on Networked correlation to perceive
Manufacturing service supply chain optimization method.
Background technique
With the continuous hair of Service-Oriented Architecture Based (Service-oriented Architecture, SOA) and cloud computing
Exhibition, Distributed Manufacturing Resources can be packaged into manufacturing service.Over the past decade, manufacturing service recommended method has obtained widely
Using, an optimal manufacturing service can be selected from numerous candidate manufacturing services come complete a single manufacture appoint
Business.But when in face of complicated manufacturing operation, enterprise prefers to obtain a manufacturing service supply chain.For example, gear manufacture
Usually it is made of several subtasks (such as forging and normalizing).There is a respective candidate service collection in each subtask, in set
The functional identical but non-functional (service quality, Quality of Service-QoS) of candidate service is different.Manufacturing service
Supply chain scheme is to select corresponding candidate service from the candidate service of each subtask concentration and be composed, to realize whole
A manufacturing operation.
Manufacturing service supply chain (Manufacturing Service Supply Chain) optimization problem has received
Extensive concern, its purpose is to find the service group with optimal service quality (Quality of Service, QoS) value
Conjunction scheme.But traditional manufacturing service supply chain optimization method usually assumes that candidate service is independent from each other, traditional clothes
Business combined optimization method select single suitably candidate service only to complete single subtask, and the identical service of other functionality
By idle.Therefore, potential preferably manufacturing service supply chain candidate scheme may be due to not considering correlation pair between service
The influence of qos value and be ignored, thus obtain Services Composition be not be optimal.
Summary of the invention
The object of the present invention is to provide a kind of Based on Networked correlation perception manufacturing service supply chain optimization method,
Comprehensively considered between service in manufacturing service supply chain optimization problem vertically and horizontally (networking) correlation to entire manufacture
The QoS of services supply chain influences, with the manufacturing operation for finding optimal manufacturing service supply chain to complete complicated.
To achieve the goals above, technical solution of the present invention is as follows:
A kind of manufacturing service supply chain optimization method of Based on Networked correlation perception, which is characterized in that it is described towards
The manufacturing service supply chain optimization method of networking correlation perception, comprising:
Step S1, manufacturing service supply chain mathematical model is constructed based on vertical cartel and horizontal combination, establishes overall situation QoS mesh
Scalar functions；
Step S2, it is based on artificial bee colony algorithm initialization population, it includes vertical cartel that each candidate scheme, which uses, in population
The matrix of structure and basic candidate service is encoded；
Step S3, it employs the bee stage to complete the generation of new candidate scheme based on artificial bee colony algorithm, passes through greedy algorithm
It is selected, retains more preferably candidate scheme；
Step S4, it has been combined using the transfer operator of biogeography algorithm with the bee stage of looking on of artificial bee colony algorithm
It at the generation of new candidate scheme, is selected by greedy algorithm, retains more preferably candidate scheme；
Step S5, the search bee stage based on artificial bee colony algorithm completes the generation of new candidate scheme, passes through greedy algorithm
It is selected, retains more preferably candidate scheme；
Step S6, judge whether to reach termination iterated conditional, optimal candidate scheme is returned if reaching, is otherwise returned
Continue iteration to step S3.
Wherein, the vertical cartel forms one by selecting several functional identical candidate services to be combined
Vertical cartel service group completes corresponding subtask with cooperation.
The horizontal combination is to be combined candidate service or vertical cartel service group, to obtain qos value global optimum
Manufacturing service supply chain scheme.
Further, each candidate scheme uses the square including vertical cartel structure and basic candidate service in the population
Battle array is encoded, wherein the first row of the matrix indicates to complete the vertical cartel structure of each subtask, the matrix its
He manages it the basic candidate service for indicating selection.
Further, the search bee stage based on artificial bee colony algorithm completes the generation of new candidate scheme, using such as
Lower formula generates:
Wherein,Indicate the basic candidate service matrix of p-th of candidate scheme selection,Indicate new candidate scheme selection
Basic candidate service matrix, riIt is the vector being made of random digit,Indicate a candidate scheme selection of v (v ≠ p)
Basic candidate service matrix.
Further, the transfer operator using biogeography algorithm and artificial bee colony algorithm look on bee stage phase
In conjunction with the generation for completing new candidate scheme, p-th of candidate scheme XpMove into probability ηpWith the probability ξ that moves outpCalculation formula such as
Under:
Wherein Ig and Eg respectively indicates highest probability of moving into and move out；K (p) represents the sequence sequence of p-th of candidate scheme
Number；P indicates population scale.
The manufacturing service supply chain optimization method of a kind of Based on Networked correlation perception proposed by the present invention, it is contemplated that wait
Correlation of the choosing service in vertical cartel and horizontal combination, proposes the manufacturing service of Based on Networked correlation perception
Combine (Networked Correlation-aware Manufacturing Service Composition, NCMSC) mathematics
Model describes manufacturing service supply chain optimization problem；It also proposed a kind of new encoding scheme based on matrix, thus form
Changing ground indicates manufacturing service supply chain candidate scheme, the combination of first element representation vertical cartel service group of each column in matrix
Structure type (selection or parallel), and the selected basic candidate service of other element representations in arranging, collaboration are completed corresponding
Subtask.In addition, in order to improve the performance of traditional ABC algorithm, biogeography is optimized (Biogeography- by the present invention
Based Optimization, BBO) transfer operator of algorithm combines with ABC algorithm, to improve manufacturing service supply chain optimization
Efficiency.By the comprehensive comparison to three aspects such as practicability, validity and efficiency, method of the invention is solving manufacturing service
It is better than the other methods of the prior art in supply chain optimization problem.
Detailed description of the invention
The manufacturing service supply chain optimization method flow diagram of Fig. 1 Based on Networked correlation perception of the present invention；
Fig. 2 is manufacturing service of embodiment of the present invention supply chain mathematical model schematic diagram；
Fig. 3 is the vertical cartel schematic diagram that the embodiment of the present invention selects structure；
Fig. 4 is the dependency diagram between four kinds of candidate services or vertical cartel service group of the invention；
Fig. 5 is the transfer operator operation chart based on matrix；
Fig. 6 is that the investigation bee based on matrix searches for schematic diagram.
Specific embodiment
Technical solution of the present invention is described in further details with reference to the accompanying drawings and examples, following embodiment is not constituted
Limitation of the invention.
The general thought of technical solution of the present invention is to comprehensively consider service in manufacturing service supply chain optimization problem
Between vertically and horizontally (networking) correlation, propose the manufacturing service combination of a Based on Networked correlation perception thus
(Networked Correlation-aware Manufacturing Service Composition, NCMSC) mathematical model
To describe manufacturing service supply chain optimization problem.Technical solution of the present invention NCMSC model is by vertical cartel and horizontal combination two
Divide and constitutes.Vertical cartel is that several functional identical candidate services are combined into a vertical cartel service group to realize phase
The subtask answered, the composite structure of vertical cartel service group include selection or parallel organization.Vertical correlation between candidate service
Property it is stronger, then this it is relevant service be more easy to be combined into a more preferably vertical cartel service group.Horizontal combination is according to each
Subtask selects a suitable candidate service or vertical cartel service, and is combined into manufacturing service supply chain scheme.
Horizontal correlation between vertical cartel service group is stronger, then the relevant vertical cartel service group, which is more easily formed one, has
The manufacturing service supply chain scheme of more excellent qos value.
As shown in Figure 1, a kind of embodiment of the manufacturing service supply chain optimization method of Based on Networked correlation perception, packet
It includes:
Step S1, manufacturing service supply chain mathematical model is constructed based on vertical cartel and horizontal combination, establishes overall situation QoS mesh
Scalar functions.
Fig. 2 illustrates an example of NCMSC model, a complicated manufacturing operation is decomposed into five subtasks, often
There are 10 candidate services that functional performance is identical but QoS performance is different in a subtask.Wherein stiIndicate i-th of subtask (i=
1 ..., I), I indicates the subtask quantity after complicated manufacturing operation decomposition.CSSiIndicate i-th of candidate service set, the set
In candidate service function having the same, achievable subtask sti。JiIndicate CSSiThe quantity of middle candidate service.Table
Show CSSiMiddle j-th candidates service (j=1 ..., Ji), VCSGiIt indicates to realize subtask stiVertical cartel service group.
Vertical cartel is to form a vertical cartel by selecting several functional identical candidate services to be combined
Service group completes corresponding subtask with cooperation.The composite structure of vertical cartel service group includes selection or parallel organization.It is hanging down
In straight combination, if the vertical correlation between candidate service is stronger, it is easier to be grouped together to form vertical cartel service
Group.For example, being concentrated in first candidate service, serviceWithThe vertical cartel clothes an of parallel organization can be combined into
Business group.ServiceWithThe vertical cartel clothes of selection structure (select probability is respectively 0.3 and 0.7) can be combined into
Business group.According to the qos value of vertical cartel service group, subtask st1By service group VCSG (service(0.3) it and services
(0.7)) cooperation is completed.But subtask can also such as be serviced by single candidate service complete independentlySon can be completed to appoint
Be engaged in st3。
Horizontal combination is to be combined candidate service or vertical cartel service, to obtain the manufacture of qos value global optimum
Services supply chain scheme.Wherein, candidate service or vertical cartel service group complete single subtask.Candidate service or vertical cartel
Horizontal correlation between service group is stronger, then it is higher to be easier to combination formation for the relevant service or vertical cartel service group
The manufacturing service supply chain scheme of qos value.For example, manufacturing service supply chain scheme [2 (0.3), 8 (0.7)]-[5,10]-[10]-
[2,4,5,8,9]-[2 (0.4), 7 (0.6)] are the entire manufacturing operations completed in example by the scheme that horizontal combination is formed.
Assuming that the service qos value of the latter subtask depends on the service qos value of previous subtask, therefore do not deposited in horizontal correlation
In recurrence relation, the present embodiment horizontal combination only considers that the level between neighboring candidate service or vertical cartel service group is related
Property, put aside the horizontal correlation between non-adjacent candidate service or vertical cartel service group.
Present embodiment is with two kinds of vertical cartel structures (selecting and parallel) and three kinds of QoS attributes (times, cost
And validity) be illustrated as embodiment, the present invention is not limited to the specific combining forms of manufacturing service supply chain.
The present embodiment VCSGiGlobal vertical correlation influence calculation formula it is as follows:
VCorInf(VCSGi) indicate VCSGiGlobal vertical correlation influence, with the vertical cartel of the selection structure of Fig. 3
For, three basic three basic candidate services of candidate serviceWithRespectively with probabilityWithSubtask st is realized in cooperationi。VCSGiIn three service there are three groups of relationships pair, i.e.,With
Indicate a pair of of related serviceWithThree degrees of correlation between them are respectivelyWithAccording to formula
(3) VCSG can be obtainediGlobal vertical correlation influence be VCorInf (VCSGi)=0.67.
According to VCSGiGlobal vertical correlation influence, calculate three kinds of QoS attributes (time, cost and validity):
(1) time of vertical cartel.
When calculating the qos value of vertical cartel, by the calculating of time and execution time and the calculating of haulage time point
Set the exam worry, because the haulage time that different types of correlation combines subsequent horizontal has different influences.Select structure and simultaneously
VCSG under row structureiAssembly time TCom (VCSGi) and haulage time TTr (VCSGi) calculation formula difference it is as follows:
(2) cost of vertical cartel.
The calculation method of vertical cartel cost is similar to the calculation method of above-mentioned time, the cost of each basic service according to
Time efficiency is divided.Therefore, the VCSG under structure and parallel organization is selectediOverall cost CCom (VCSGi) and transportation cost
CTr(VCSGi) calculation formula difference it is as follows:
(3) validity of vertical cartel.
The validity of vertical cartel is also classified into three steps, including prepares validity, executes validity and transport validity.
VCSG under selection and parallel organizationiCombination efficiency ACom (VCSGi) and transport validity ATr (VCSGi) calculation formula difference
It is as follows:
It is based on above-mentioned vertical cartel as a result, horizontal combination can find optimal manufacturing service supply chain scheme.It is horizontal
Combination is to be combined candidate service or vertical cartel service, to obtain the manufacturing service supply chain of qos value global optimum
Scheme.
In horizontal combination, the present embodiment only considers the sequential organization of subtask, because other labyrinths (such as select
Select, parallel and loop structure) it can be converted to sequential organization.Different types of correlation is to horizontal combination between candidate service
In transport have opposite impacts on, therefore the present embodiment discusses previous vertical cartel service group to the latter vertical cartel
The influence of service group correlation, and horizontal combination mathematical model is established to assess the global QoS of manufacturing service supply chain scheme
Value.
According to vertical cartel as a result, there are four types of correlations, i.e. (a) a pair between vertical cartel service group or candidate service
One, (b) many-one, (c) be one-to-many and (d) multi-to-multi, as shown in Figure 4.
In horizontal combination, it still can choose single candidate service and come some subtask of complete independently.For ease of calculation, exist
During level composition, single candidate service as a special vertical cartel service group.Therefore, horizontal correlation pair
The influence formula of the latter vertical cartel service group is as follows:
HCorInf(VCSGi) indicate horizontal correlation to VCSGiInfluence value, it is assumed that the adjacent candidate service of any two
Between horizontal correlation type beVCSGi-1And VCSGiBetween horizontal correlation type be
CorTypei=1.Otherwise, VCSGi-1And VCSGiBetween horizontal correlation type be CorTypei=2.Wherein, VCSGi-1With
VCSGiBetween correlation type: 1 indicate correlation typeIndicate another correlation type
(1) time of horizontal combination.
Correlation between different types of basic candidate service has not the qos value and transport that combine in horizontal combination
Same influence.If CorTypei=1, the haulage time of horizontal combination must become apparent from than time and execution saving of time.
If but CorTypei=2, the haulage time of horizontal combination and the saving degree of other two steps are similar.Therefore, horizontal group
The calculation formula for closing time T (HC) is as follows:
(2) cost of horizontal combination.
The cost calculation of horizontal combination is similar with time calculating.If CorTypei=1, transportation cost ratio in horizontal combination
Setting up cost and executory cost save to become apparent from.If but CorTypei=2, the transportation cost of horizontal combination is walked with other two
Rapid saving degree is similar.Therefore, the calculation formula of horizontal combination cost C (HC) is as follows:
(3) validity of horizontal combination.
Influence of the horizontal correlation to validity between vertical cartel service group, is not with the influence to time and cost
With.If CorTypei=1, the raising degree for transporting validity is more significant than preparation process and execution step.But if
CorTypei=2, the raising degree for transporting validity is suitable with other two steps.Therefore, horizontal combination validity A is calculated
(HC) formula is as follows:
On the basis of the above embodiments, the purpose of the present invention is find the manufacturing service supply chain of an optimal qos value.
Time and cost are negative QoS attribute, but validity is positive QoS attribute.Therefore, it is impossible to which three of the above QoS attribute is directly integrated
Into QoS utility function.Positive and negative QoS attribute value needs advanced row standardization processing, and formula difference is as follows:
In view of the total time of manufacturing service supply chain scheme and totle drilling cost are necessarily less than estimated time T0And estimated cost
C0, and total validity has to be larger than minimum value A0, therefore the QoS objective function of manufacturing service supply chain scheme is as follows:
It it should be noted that the QoS attribute that the present invention considers can also be more, such as further include customer evaluation etc., this hair
The bright form for being not limited to specific objective function.The global QoS objective function that this step is established, will be used in subsequent step
It calculates the global QoS of each candidate scheme, repeats no more below.
In above-mentioned formula, relevant parameter is described as follows:
Table 1
Step S2, it is based on artificial bee colony algorithm initialization population, it includes vertical cartel that each candidate scheme, which uses, in population
The matrix of structure and basic candidate service is encoded.
Artificial bee colony algorithm (ABC algorithm) is a kind of heuritic approach of imitation bee colony search food, it includes four ranks
Section: bee is employed in initialization, is looked on bee and investigation bee, is not repeated one by one here.
In order to completely express and store complicated manufacturing service supply chain candidate scheme information, the present embodiment is proposed
A kind of candidate scheme encoding scheme based on matrix.Candidate scheme based on matrix includes two parts information: vertical cartel structure
With basic candidate service.
For example, p-th candidate scheme can be by 6 × I'sMatrix composition.The first row indicate to complete it is each
The vertical cartel structure of subtask, including single candidate service, selection and parallel organization.The present embodiment is using integer 1-3 come table
Show three kinds of vertical cartel structures, as follows:
MatrixThe rear five-element (from the 2nd row to the 6th row) indicate the basic candidate service of selection, completed with collaboration corresponding
Subtask.For simplicity, the maximum quantity by vertical cartel service is limited to 5.If the quantity of vertical cartel service is few
In 5, then remaining vacancy is filled up with 0.For example, matrixIndicate that general assignment is decomposed into the candidate scheme of five subtasks, such as
Shown in lower:
Wherein first row indicates single candidate serviceIt is selected to complete first subtask；Secondary series indicates basic and waits
Choosing serviceWithVertical cartel service group will be combined into cooperate with 0.3,0.5 and 0.2 select probability respectively
Complete second subtask, other column and so on.
In initial phase, population scale is set as P.Indicate p-th of time
Scheme is selected, wherein arrangingFor a subsolution (the i.e. realization subtask st in candidate schemeiVertical cartel scheme).The first row
ElementIt is to be generated at random from the vector for the integer that range is [1,3].IfSo in i-th of services set
In only select a basic candidate service, and be put into corresponding columnIfEqual to 2 or 3,In basic candidate service
The range of quantity is [2,5].In addition, if(i.e. selection structure), select probability will be randomly generated and its summation is 1.WhenIn basic candidate service quantity determine after, other elements in column can be produced at random by formula (22) at:
Wherein(i.e. matrixIn an element) indicate from services set CSSiThe candidate service selectedSequence
Number, and range necessarily is between Lb and Ub；Lb and Ub respectively indicate CSSiLower bound and the upper bound；Rand (0,1) is in range
Uniformly distributed function in (0,1)；Round (x) indicates round function.
It should be noted that initial population generation method of the formula 22 referring to ABC algorithm, belongs to mature technology, here
It repeats no more.The difference is that the present embodiment is using the matrix coder including vertical cartel structure and basic candidate service come table
Show candidate scheme.
IfThe element that middle any two generate at random is identical, then recurring formula (22) is predetermined until meeting
Candidate service quantity is to complete subtask sti.Finally, the candidate scheme that population scale is P will be obtained.
Step S3, it employs the bee stage to complete the generation of new candidate scheme based on artificial bee colony algorithm, passes through greedy algorithm
It is selected, retains more preferably candidate scheme.
The bee stage is being employed, is each employing bee that will distribute a candidate scheme to search for more optimal solution.The present embodiment
By according to the needs of candidate scheme matrix coder, original operator is improved.
Still by taking Fig. 2 as an example, in subsolutionIn, first partRemain unchanged (because being randomly generated).
Second part by several (but be no more than five) candidate service set of serial numbers at completing subtask st to cooperate withi。
According toIt is as follows after candidate scheme adjustment based on matrix:
Wherein riIt is the vector being made of five random digits.Indicate the basic time of p-th of candidate scheme selection
Choosing service matrix,Indicate the basic candidate service matrix of new candidate scheme selection, riBe one from random digit form to
Amount,Indicate the basic candidate service matrix of a candidate scheme selection of v (v ≠ p).Wherein v (v ≠ p) a candidate scheme from
It is randomly choosed in population.
If matrixIn some element be nonzero digit, then its corresponding position riIt will be by [- 1, a 1] range
Interior equally distributed random number filling；Otherwise, riCorresponding position be 0.IfGenerate certain elements (in addition to zero with
Outside) identical, then formula (23) can repeat, untilIn nonzero element it is different.Then, original candidate schemeAnd new departureIt is selected by greedy algorithm, retains more optimal solution.All employs bee
After the search for completing new solution, honeycomb can be returned to and employ bee sharing information to seek more optimal solution with non-.
Step S4, it has been combined using the transfer operator of biogeography algorithm with the bee stage of looking on of artificial bee colony algorithm
It at the generation of new candidate scheme, is selected by greedy algorithm, retains more preferably candidate scheme.
The bee stage is being looked on, looking on bee will be according to certain one candidate scheme of probability selection as further exploring
Object.New candidate scheme can be scanned for according to the strategy of formula (23).But since the solution of manufacturing service supply chain is one
A discrete optimization problems of device, the strategy of formula (23) are unable to fully using from employing bee stage candidate scheme information obtained.Cause
This, the solution performance of ABC algorithm still has improved space.The principle of the transfer operator of biogeography algorithm (BBO algorithm) is:
Preferably candidate scheme gives effective information sharing to lower candidate scheme, to improve solution performance.Therefore, BBO algorithm
Transfer operator can effectively alleviate the limitation of traditional ABC algorithm.
According to transfer operator, the present embodiment arranges in descending order again after calculating the global qos values of all candidate schemes.With
Afterwards, according to probability PmodTo determine whether to select the candidate scheme to be migrated.The candidate scheme of selection is known as moving into object, will
It is adjusted.The present embodiment explores more preferably candidate scheme using moving into probability and moving out probability for BBO algorithm.P-th of time
Select scheme XpMove into probability ηpWith the probability ξ that moves outpCalculation formula it is as follows:
Wherein Ig and Eg respectively indicates highest probability of moving into and move out；K (p) represents the sequence sequence of p-th of candidate scheme
Number；P indicates population scale.
The candidate scheme X to move outeIt will be according to the probability ξ that moves outeIt is obtained by roulette selection strategy, which can be with
The candidate scheme of other suboptimums is helped to be evolved into more optimal solution.According to moving into probability ηp, transfer operator is according to formula Xp←XeIt is real
It is existing, generate a new candidate schemeThen, X is compared by greedy algorithmpWithQos value and retain one it is more excellent
Scheme.
It is required according to the matrix coder of candidate scheme as can be seen that original transfer operator not can be used directly in new candidate
The search of scheme.Therefore, the present embodiment improves transfer operator using binary mask character string.Fig. 5 indicates that is based on a square
The transfer operator example of battle array.
As shown in figure 5, a complicated task is broken down into five subtasks.P-th of candidate scheme XpAs moving into pair
As being adjusted.E-th of candidate scheme XeIt will be according to the probability ξ that moves outeIt is chosen as object of moving out.Then, one group generate at random two
System mask character string will be according to moving into probability ηpDetermine XpIn vertical vector (vertical cartel service group) whether should move
It moves.Assuming that the binary mask character string generated is (0-1-1-0-1), X is indicatedpIn second, third and the 5th vertically to
Amount will be by XeIn corresponding vertical vector replace (including vertical cartel structure type and selected candidate service).Therefore, it adjusts
Result after wholeShow to serviceWithIt is respectively completed subtask st1With subtask st2,WithRespectively
It is composed with 0.1,0.2 and 0.7 select probability to cooperate with and complete subtask st3, andWithIt is made of parallel organization,
Subtask st is completed with collaboration4。
Step S5, the search bee stage based on artificial bee colony algorithm completes the generation of new candidate scheme, passes through greedy algorithm
It is selected, retains more preferably candidate scheme.
The investigation bee stage is mainly used for alleviating the defect that ABC algorithm falls into local optimum.In view of candidate scheme matrix is compiled
The needs of code, original search bee stage are also required to do corresponding adjustment.Fig. 6 indicates the reality of a search bee search based on matrix
Example.
Assuming that XpIt can not still be improved after being improved at limit times, carry XpTo employ bee that will become search bee new to find
Candidate scheme.The binary mask character string (1-0-0-1-1) generated at random indicates first, the 4th and the 5th vertical cartel
Structure needs to change.Therefore, initiallyIt will randomly be converted intoCorresponding base
This candidate service is randomly regenerated by formula (22) for realizing subtask st1、st4And st5.With subtask st1For
Example, original vertical composite structure (only one candidate service) will be converted into selection structure.Corresponding candidate service and it
Select probability regenerated by formula (22), and the sum of select probability is 1.Therefore, candidate serviceWithSubtask st is completed into probability collaboration respectively with 0.2,0.1,0.3 and 0.41.As new candidate
After schemes generation, a preferably candidate scheme will be retained by greedy algorithm.
Step S6, judge whether to reach termination iterated conditional, optimal candidate scheme is returned if reaching, is otherwise returned
Continue iteration to step S3.
Finally, returning to the candidate scheme of near-optimization after reaching termination condition.If not up to termination condition is returned
It returns to and the bee stage is employed to continue iteration.
The present embodiment select three kinds of algorithms such as tradition ABC algorithm, GA algorithm and PSO algorithm and optimization method of the invention into
Row compares, the experimental results showed that no matter how population scale changes, the manufacturing service supply chain that optimization method of the invention obtains
The average qos value of scheme is better than other three kinds of algorithms.The experimental results showed that network correlation is to manufacturing service supply chain scheme
Play a significant role.Method of the invention can reduce time and the cost consumption of Services Composition, and find one and more preferably dive
In scheme, main reason is that the vertical cartel service group of cooperation completes a sub- required by task than single independent service
Time and cost are less.In addition, excellent using the resulting average value of optimization method of the invention, optimal value, the qos value of worst-case value
In other three kinds of algorithms.Standard deviation value the experimental results showed that, method of the invention solve every group of manufacturing service supply chain
There is better stability, but runing time is slightly above other three kinds of algorithms, is primarily due to of the invention excellent when optimization problem
The transfer operator of change method needs bigger calculation amount to find more excellent solution.In addition, other three kinds of algorithms need less calculating
Cost but it is easy to fall into local optimum.Therefore, method of the invention has good on solving the problems, such as manufacturing service supply chain optimization
Good practicability.
In the manufacturing service supply chain optimization problem of the subtask for the candidate service or different number for solving different number,
Method of the invention is better than other three kinds of algorithms always.The experimental results showed that policymaker can be true according to the difference preference of oneself
Determine the weight of multiple target, and more preferably manufacturing service supply chain scheme is obtained using method of the invention.The experimental results showed that adopting
With the overall trend of the obtained optimal qos value of four kinds of algorithms with the increase of subtask linear reduction, the main reason is that
When the quantity of subtask increases, manufacturing service supply chain scheme needs take more time and cost.It can be seen that asking
When solving manufacturing service supply chain optimization problem, effectiveness of the invention is better than other three kinds of algorithms.As candidate service or son are appointed
The increase for quantity of being engaged in, the various combination of QoS attribute weight is to the influence very little for calculating the time.The experimental results showed that with candidate
The increase of quantity of service, the runing time kept stable of each algorithm.Four kinds of algorithms are for handling manufacturing service supply chain
Optimization problem all has preferably efficiency.It is comprehensively compared in terms of practicability, validity and efficiency etc. three, the present invention
Method solve manufacturing service supply chain optimization problem on be better than other three kinds of algorithms.
The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, without departing substantially from essence of the invention
In the case where mind and its essence, those skilled in the art make various corresponding changes and change in accordance with the present invention
Shape, but these corresponding changes and modifications all should fall within the scope of protection of the appended claims of the present invention.
Claims (6)
1. a kind of manufacturing service supply chain optimization method of Based on Networked correlation perception, which is characterized in that described towards net
The manufacturing service supply chain optimization method of network correlation perception, comprising:
Step S1, manufacturing service supply chain mathematical model is constructed based on vertical cartel and horizontal combination, establishes overall situation QoS target letter
Number；
Step S2, it is based on artificial bee colony algorithm initialization population, it includes vertical cartel structure that each candidate scheme, which uses, in population
It is encoded with the matrix of basic candidate service；
Step S3, it employs the bee stage to complete the generation of new candidate scheme based on artificial bee colony algorithm, is carried out by greedy algorithm
Selection retains more preferably candidate scheme；
Step S4, completion is combined newly with the bee stage of looking on of artificial bee colony algorithm using the transfer operator of biogeography algorithm
The generation of candidate scheme, is selected by greedy algorithm, retains more preferably candidate scheme；
Step S5, the search bee stage based on artificial bee colony algorithm completes the generation of new candidate scheme, is carried out by greedy algorithm
Selection retains more preferably candidate scheme；
Step S6, judge whether to reach termination iterated conditional, optimal candidate scheme is returned if reaching, otherwise back to step
Rapid S3 continues iteration.
2. the manufacturing service supply chain optimization method of Based on Networked correlation perception as described in claim 1, feature exist
In the vertical cartel forms a vertical cartel clothes by selecting several functional identical candidate services to be combined
Business group completes corresponding subtask with cooperation.
3. the manufacturing service supply chain optimization method of Based on Networked correlation perception as described in claim 1, feature exist
In the horizontal combination is to be combined candidate service or vertical cartel service group, to obtain the system of qos value global optimum
Make services supply chain scheme.
4. the manufacturing service supply chain optimization method of Based on Networked correlation perception as described in claim 1, feature exist
In, each candidate scheme uses the matrix including vertical cartel structure and basic candidate service to be encoded in the population,
Described in the first row of matrix indicate to complete the vertical cartel structure of each subtask, other rows of the matrix indicate selection
Basic candidate service.
5. the manufacturing service supply chain optimization method of Based on Networked correlation perception as described in claim 1, feature exist
In, the search bee stage based on artificial bee colony algorithm completes the generation of new candidate scheme, it is generated using following formula:
Wherein,Indicate the basic candidate service matrix of p-th of candidate scheme selection,Indicate the base of new candidate scheme selection
This candidate service matrix, riIt is the vector being made of random digit,Indicate the base of a candidate scheme selection of v (v ≠ p)
This candidate service matrix.
6. the manufacturing service supply chain optimization method of Based on Networked correlation perception as described in claim 1, feature exist
In the transfer operator using biogeography algorithm combines the new candidate of completion with the bee stage of looking on of artificial bee colony algorithm
The generation of scheme, p-th of candidate scheme XpMove into probability ηpWith the probability ξ that moves outpCalculation formula it is as follows:
Wherein Ig and Eg respectively indicates highest probability of moving into and move out；K (p) represents the sequence serial number of p-th of candidate scheme；P
Indicate population scale.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811268270.5A CN109447349B (en) | 2018-10-29 | 2018-10-29 | Manufacturing service supply chain optimization method facing networked relevance perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811268270.5A CN109447349B (en) | 2018-10-29 | 2018-10-29 | Manufacturing service supply chain optimization method facing networked relevance perception |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109447349A true CN109447349A (en) | 2019-03-08 |
| CN109447349B CN109447349B (en) | 2022-04-19 |
Family
ID=65549769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811268270.5A Active CN109447349B (en) | 2018-10-29 | 2018-10-29 | Manufacturing service supply chain optimization method facing networked relevance perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447349B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766279A (en) * | 2019-09-20 | 2020-02-07 | 西安文理学院 | Design and manufacture integration method based on QoS |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070043602A1 (en) * | 2005-08-22 | 2007-02-22 | Markus Ettl | Method and system for balancing asset liability and supply flexibility in extended value networks |
| CN104778523A (en) * | 2014-01-15 | 2015-07-15 | 上海亨钧科技有限公司 | Method for generating smart operation managing and scheduling plan of enterprise railway and system thereof |
| CN107180286A (en) * | 2017-07-18 | 2017-09-19 | 浙江财经大学 | Manufacturing service supply chain optimization method and system based on modified pollen algorithm |
| CN107590572A (en) * | 2017-10-24 | 2018-01-16 | 广东电网有限责任公司电力调度控制中心 | A kind of complicated Directional protection in loops MBPS acquiring methods based on IBBO |
| CN107590623A (en) * | 2016-07-07 | 2018-01-16 | 袁宏斌 | A kind of cloud manufacturing service preferred method of combination for considering region logistics service capability |
| CN107862404A (en) * | 2017-10-24 | 2018-03-30 | 浙江财经大学 | Manufacturing service supply chain optimization method based on service relevance |
| CN107944698A (en) * | 2017-11-22 | 2018-04-20 | 重庆大学 | The manufacture demand of facing cloud manufacture and capacity of equipment normalization modeling method |
| CN108256621A (en) * | 2016-12-28 | 2018-07-06 | 北京天诚同创电气有限公司 | Bee colony optimization method and device |
| CN108536536A (en) * | 2018-03-09 | 2018-09-14 | 三峡大学 | Resource consolidation and optimization system and method under a kind of cloud manufacturing environment based on quantum genetic algorithm |
| CN108711155A (en) * | 2018-04-20 | 2018-10-26 | 西京学院 | A kind of image Fast Segmentation Algorithm of artificial bee colony Optimization of Fuzzy cluster |
-
2018
- 2018-10-29 CN CN201811268270.5A patent/CN109447349B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070043602A1 (en) * | 2005-08-22 | 2007-02-22 | Markus Ettl | Method and system for balancing asset liability and supply flexibility in extended value networks |
| CN104778523A (en) * | 2014-01-15 | 2015-07-15 | 上海亨钧科技有限公司 | Method for generating smart operation managing and scheduling plan of enterprise railway and system thereof |
| CN107590623A (en) * | 2016-07-07 | 2018-01-16 | 袁宏斌 | A kind of cloud manufacturing service preferred method of combination for considering region logistics service capability |
| CN108256621A (en) * | 2016-12-28 | 2018-07-06 | 北京天诚同创电气有限公司 | Bee colony optimization method and device |
| CN107180286A (en) * | 2017-07-18 | 2017-09-19 | 浙江财经大学 | Manufacturing service supply chain optimization method and system based on modified pollen algorithm |
| CN107590572A (en) * | 2017-10-24 | 2018-01-16 | 广东电网有限责任公司电力调度控制中心 | A kind of complicated Directional protection in loops MBPS acquiring methods based on IBBO |
| CN107862404A (en) * | 2017-10-24 | 2018-03-30 | 浙江财经大学 | Manufacturing service supply chain optimization method based on service relevance |
| CN107944698A (en) * | 2017-11-22 | 2018-04-20 | 重庆大学 | The manufacture demand of facing cloud manufacture and capacity of equipment normalization modeling method |
| CN108536536A (en) * | 2018-03-09 | 2018-09-14 | 三峡大学 | Resource consolidation and optimization system and method under a kind of cloud manufacturing environment based on quantum genetic algorithm |
| CN108711155A (en) * | 2018-04-20 | 2018-10-26 | 西京学院 | A kind of image Fast Segmentation Algorithm of artificial bee colony Optimization of Fuzzy cluster |
Non-Patent Citations (2)
| Title |
|---|
| SHUAI ZHANG等: "A hybrid approach combining an extended BBO algorithm with an intuitionistic fuzzy entropy weight method for QoS-aware manufacturing service supply chain optimization", 《NEUROCOMPUTING》 * |
| 张萌 等: "基于区间的制造服务组合多目标优化方法", 《计算机集成制造系统》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766279A (en) * | 2019-09-20 | 2020-02-07 | 西安文理学院 | Design and manufacture integration method based on QoS |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109447349B (en) | 2022-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chase-Dunn et al. | World-systems analysis | |
| Bhawalkar et al. | Preventing unraveling in social networks: the anchored k-core problem | |
| CN106157155B (en) | Social media information propagation visualization analysis method and system based on map metaphor | |
| Choi et al. | The effects of network characteristics on performance of innovation clusters | |
| Vandermeer et al. | Self-organized spatial pattern determines biodiversity in spatial competition | |
| Jaradat et al. | The effect of elite pool in hybrid population-based meta-heuristics for solving combinatorial optimization problems | |
| CN104063778A (en) | Method for allocating cargo positions for cargoes in three-dimensional warehouse | |
| Lawrence | The thermodynamics of unequal exchange: energy use, CO2 emissions, and GDP in the world-system, 1975—2005 | |
| Sauterey et al. | When everything is not everywhere but species evolve: an alternative method to model adaptive properties of marine ecosystems | |
| Lee et al. | Equivalence results in the allocation of indivisible objects: A unified view | |
| Hagberg et al. | Designing threshold networks with given structural and dynamical properties | |
| Harris et al. | A memetic algorithm for the quadratic assignment problem with parallel local search | |
| Du et al. | Structural balance in fully signed networks | |
| Crawford et al. | A hybrid soft computing approach for subset problems | |
| CN109447349A (en) | A kind of manufacturing service supply chain optimization method of Based on Networked correlation perception | |
| Barter et al. | Meta-food-chains as a many-layer epidemic process on networks | |
| Pachuau et al. | An overview of crossover techniques in genetic algorithm | |
| Drake et al. | Modified choice function heuristic selection for the multidimensional knapsack problem | |
| Kaimanovich | " MÜNCHHAUSEN TRICK" AND AMENABILITY OF SELF-SIMILAR GROUPS | |
| CN107180286B (en) | Manufacturing service supply chain optimization method and system based on improved pollen algorithm | |
| CN107545323A (en) | Logistics Distribution Method, device, equipment and the storage medium of commodity | |
| Avelino et al. | Parity effects in rock-paper-scissors type models with a number of species NS≤ 12 | |
| Baidoo et al. | Solving the TSP using traditional computing approach | |
| CN115001978B (en) | Cloud tenant virtual network intelligent mapping method based on reinforcement learning model | |
| Boveiri et al. | Static homogeneous multiprocessor task graph scheduling using ant colony optimization |
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 |