CN106251012A - The path calculation method of a kind of band weak rock mass logistics transportation scheduling and device - Google Patents

Abstract

nullThe invention provides path calculation method and the device of a kind of band weak rock mass logistics transportation scheduling，It is applied to multiple from server，By parallel receive master server send come go out to send instructions and described in other from the pheromone of server time，The first path optimizing information that multiple described each Formica fuscas obtained the most successively corresponding sub-ant colony from server produce，It is achieved thereby that search mission to be allocated to multiple sub-ant colony from server simultaneously，Search can be completed concurrently，Improve the speed of search，Afterwards，Multiple described a plurality of first path optimizing information each comparing acquisition from server，Obtain respective first optimal path information，And the pheromone in the first optimal path information described in each self refresh，Afterwards，Respective first optimal path information and corresponding informance element are sent jointly to master server，And then shorten master server and compare the time of routing information bar number，Can quickly obtain the optimum Distribution path of logistics distribution.

Description

The path calculation method of a kind of band weak rock mass logistics transportation scheduling and device

Technical field

The present invention relates to technical field of intelligent traffic, in particular, relate to a kind of band weak rock mass logistics transportation and adjust The path calculation method of degree and device.

Background technology

Along with the development of economic globalization, drive the development of logistics so that it is seem more and more important；And logistics distribution In occupation of the very big proportion in logistics transportation, but cannot be able to ensure in client institute due to the reason such as traffic, weather in reality Dispensing is completed in the time zone formulated, therefore, it is proposed to the Logistic Scheduling method of band weak rock mass, i.e. at basic vehicle route Time windows constraints condition is increased, it is allowed to transport truck arrives or late in advance, it is achieved that at cost and Gu in problem An ideal Tradeoff point is have found between full meaning.

At present, solve the algorithm of the Logistics Scheduling Problem of band weak rock mass, mainly by simulating each dispensing personnel Carry out the Distribution path that this goods delivery is formed, and after comparing a plurality of Distribution path of formation, therefrom obtain optimum Distribution path, meets client's formulation to the goods time of advent as far as possible；But existing computational methods, due to must be in simulation After complete dispensing personnel reach a dispensing point, the path that next one dispensing personnel could arrive a dispensing point carries out mould Intend, until after these all dispensing personnel comprised have traveled through whole dispensing points, could pass through to compare the dispensing drawing optimum Path, and then made the speed of a searching route task more slowly, cause cannot obtaining vehicle transport mistake rapidly Distribution path optimum in journey.

Summary of the invention

In view of this, the invention provides path calculation method and the device of a kind of band weak rock mass logistics transportation scheduling, And then the search speed of raising algorithm, quickly obtain optimum Distribution path.

For achieving the above object, the present invention provides following technical scheme:

The path calculation method of a kind of band weak rock mass logistics transportation scheduling, is applied to multiple from server, including:

Multiple described receive parallel from server master server send come go out to send instructions and described in other from server Pheromone；

Updating each self-corresponding sub-ant colony parameter respectively from server described in multiple, arbitrary described sub-ant colony parameter includes joining Sending location point information, the heuristic greedy method of Path selection, taboo list and enabled node list, wherein, described delivery location point is believed Breath includes the initial information element value on the number of delivery location point, present position and each delivery location point, described taboo list Preserving the delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, described enabled node list preserves in described sub-ant colony Arbitrary Formica fusca without delivery location point；

From server respectively according to the described path choosing of each Formica fusca each self-corresponding described sub-ant colony described in multiple The heuristic greedy method selected, taboo list, enabled node list, and other are described from the pheromone of server, obtain institute successively right The first path optimizing information that each Formica fusca in the described sub-ant colony answered produces；

Multiple described obtain a plurality of described first excellent of being respectively compared from each self-corresponding described sub-ant colony from server Change routing information, it is thus achieved that each minimum routing information, as multiple described from server respectively since corresponding described sub-ant colony First optimal path information of middle acquisition；

Update the optimum road of described first respectively obtained since corresponding described sub-ant colony respectively from server described in multiple Pheromone in the information of footpath；

From server respectively by described first optimal path information each obtained and pheromone described in multiple, it is sent to institute State master server, obtain the foundation of current optimal path as described master server.

Preferably, the plurality of described from server respectively according to each Formica fusca each self-corresponding described sub-ant colony The heuristic greedy method of described Path selection, taboo list, enabled node list, and other are described from the pheromone of server, depend on The first path optimizing information that each Formica fusca in described sub-ant colony corresponding to secondary acquisition produces, including:

Make each Formica fusca each self-corresponding described sub-ant colony successively from start bit respectively from server described in multiple Put and a little set out；

Multiple described from server be utilized respectively the heuristic greedy method of respective described Path selection and receive other The described pheromone from server, calculates current next step dispensing position allowed of dispensing Formica fusca in each self-corresponding described sub-ant colony Putting a little, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, wherein, described next step is permitted The delivery location point permitted is the arbitrary delivery location point in the enabled node list of described current dispensing Formica fusca；

Current the second of Formica fusca of providing and delivering described in each self-corresponding described sub-ant colony is updated respectively from server described in multiple Path optimizing information, and return the plurality of described heuristic greedy method being utilized respectively respective described Path selection from server And receive other are described from the pheromone of server, calculate in each self-corresponding described sub-ant colony current dispensing Formica fusca next The delivery location point that step allows, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, directly To multiple described from each self-corresponding described sub-ant colony of server each Formica fusca all move a described delivery location point；

Multiple described from each self-corresponding described sub-ant colony of server each Formica fusca all move a described dispensing During location point, return the plurality of described heuristic greedy method and the reception being utilized respectively respective described Path selection from server Other arrived are described from the pheromone of server, calculate current next step permission of dispensing Formica fusca in each self-corresponding described sub-ant colony Delivery location point, and described delivery location point is saved in and described current dispensing taboo list corresponding to Formica fusca, until multiple Described from each self-corresponding described sub-ant colony of server each Formica fusca all traveled through in described delivery location dot information and comprised Whole delivery location points；

The first of multiple described each Formica fuscas generations obtained the most successively from server the described sub-ant colony of correspondence Path optimizing information, wherein, described first path optimizing information is made up of multiple described second path optimizing information.

Preferably, the plurality of described update respectively from server respectively obtain since corresponding described sub-ant colony described in Pheromone in first optimal path information, including:

Multiple described from server foundation Pheromone update formula respectively:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in each self-corresponding described first optimal path information, wherein, m is Formica fusca in arbitrary ant colony Number, n is iterations, τ_ijT () is the pheromone amount that t remains between delivery location point ij on ij line, τ_ij(t+n) For the pheromone amount remained between delivery location point ij on ij line after delivery location point nth iteration, d_ijFor delivery location point ij Between distance, Δ τ_ij(t) be arbitrary Formica fusca k be pheromone in t from the pheromone residual quantity of delivery location point i to j, ρ Degree of volatility, L_kBy arbitrary Formica fusca k length in walked path in once circulation.

The path calculation method of a kind of band weak rock mass logistics transportation scheduling, is applied to master server, including:

Send out and send instructions to multiple from server, wherein, multiple described arbitrary corresponding from server from server In a sub-ant colony, and described in go out to send instructions and described obtain from each self-corresponding described sub-ant colony respectively from server for multiple First optimal path information and the foundation of pheromone；

Receive multiple described described first optimal path information sent respectively from server and pheromone；

By receive each described in from server send come described pheromone be sent to described in other from server Sub-ant colony corresponding in；

By multiple described described first optimal path information sent respectively from server received and current optimum Length compares, it is thus achieved that current optimal path, as the current optimization length of next iteration, and returns described transmission and sets out Instruction from server, until current iteration number of times reaches default maximum iteration time, exports optimal path to multiple.

A kind of server cluster, including:

N from server；

Described server set group control n is described receive parallel from server master server send come go out to send instructions and He is described from the pheromone of server；

Wherein, each is described from server, for receive master server send come go out to send instructions and other n-1 institute State the pheromone from server, update corresponding sub-ant colony parameter, according to the institute of each Formica fusca in corresponding described sub-ant colony State the heuristic greedy method of Path selection in sub-ant colony parameter, taboo list, enabled node list, and other n-1 described from The pheromone of server, obtains the first path optimizing letter that each Formica fusca in corresponding described sub-ant colony produces successively Breath, a plurality of described first path optimizing information obtained in the most corresponding described sub-ant colony, it is thus achieved that corresponding minimum path Information, as the first optimal path information obtained from corresponding described sub-ant colony, updates from corresponding described sub-ant colony The pheromone in described first optimal path information obtained, by described first optimal path information obtained and pheromone, sends out Give described master server；

Wherein, described sub-ant colony parameter includes delivery location dot information, the heuristic greedy method of Path selection, taboo list and can With node listing, wherein, described delivery location dot information includes the number of delivery location point, present position and each dispensing position Putting the initial information element value on a little, described taboo list preserves the delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, Described enabled node list preserve arbitrary Formica fusca in described sub-ant colony without delivery location point.

Preferably, the described son of each Formica fusca each described described sub-ant colony corresponding from server foundation described The heuristic greedy method of the Path selection in ant colony parameter, taboo list, enabled node list, and other n-1 described from service The pheromone of device, obtains the first path optimizing information that each Formica fusca in corresponding described sub-ant colony produces, tool successively Body is used for:

Make each Formica fusca in the described sub-ant colony of correspondence successively from original position point；

The heuristic greedy method utilizing the described Path selection in corresponding described sub-ant colony parameter and other n-1 received Individual described from the pheromone of server, calculate current next step delivery location allowed of dispensing Formica fusca in corresponding described sub-ant colony Point, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, wherein, described next step allows Delivery location point be described current dispensing Formica fusca enabled node list in arbitrary delivery location point；

Update the second path optimizing information of current Formica fusca of providing and delivering described in corresponding described sub-ant colony, and return described profit Described from clothes with the heuristic greedy method of the described Path selection in corresponding described sub-ant colony parameter and receive other n-1 The pheromone of business device, calculates current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and by institute State delivery location point and be saved in the taboo list corresponding with described current dispensing Formica fusca, until in corresponding described sub-ant colony each Formica fusca all moves a described delivery location point；

When each Formica fusca all moves a described delivery location point in corresponding described sub-ant colony, return described utilization The heuristic greedy method of the corresponding described Path selection in described sub-ant colony parameter and receive other n-1 are described from service The pheromone of device, calculates current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and by described Delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, until each ant in corresponding described sub-ant colony Ant has all traveled through the whole delivery location points comprised in described delivery location dot information；

Obtain the described first path optimizing information that each Formica fusca in the described sub-ant colony of correspondence produces successively, its In, described first path optimizing information is made up of multiple described second path optimizing information.

Preferably, described each described from server update obtain from corresponding described sub-ant colony described first Pheromone in shortest path information, specifically for:

According to Pheromone update formula:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in corresponding described first optimal path information, wherein, m is the number of Formica fusca in arbitrary ant colony Mesh, n is iterations, τ_ijT () is the pheromone amount that t remains between delivery location point ij on ij line, τ_ij(t+n) it is The pheromone amount remained between delivery location point ij on ij line after delivery location point nth iteration, d_ijFor delivery location point ij it Between distance, Δ τ_ij(t) be arbitrary Formica fusca k be pheromone in t from the pheromone residual quantity of delivery location point i to j, ρ Degree of volatility, L_kBy arbitrary Formica fusca k length in walked path in once circulation.

A kind of master server, including:

Sending module, is used for sending out and sends instructions to n from server, by receive each described in send out from server The described pheromone sent here is sent to other n-1 described sub-ant colony corresponding from server, and wherein, n described from clothes Arbitrary in business device from server corresponding to a sub-ant colony, and described in go out to send instructions for n described from server respectively from respectively Self-corresponding described sub-ant colony obtains the foundation of the first optimal path information and pheromone；

Receiver module, for receiving n described described first optimal path information sent respectively from server and letter Breath element；

Comparison module, is used for n described described first optimal path information sent respectively from server with current Optimization length compares, it is thus achieved that current optimal path, as the current optimization length of next iteration；

Judge module, for judging that whether current iteration number of times is equal to the maximum iteration time preset；

Sending module, is additionally operable to, when current iteration number of times is less than described default maximum iteration time, send out and bristle with anger Order is individual from server to n；

Output module, for when current iteration number of times is equal to described default maximum iteration time, output is current optimum Path, the optimal path provided and delivered as this.

Understand via above-mentioned technical scheme, compared with prior art, the invention provides a kind of band weak rock mass logistics The path calculation method of transportation dispatching and device, be applied to multiple from server, by sending at the parallel master server that receives Go out to send instructions and other are described when the pheromone of server, multiple described from server respectively according to each self-corresponding institute State the heuristic greedy method of the Path selection of each Formica fusca in sub-ant colony, taboo list, enabled node list, and other described from The pheromone of server, obtains the first path optimizing letter that each Formica fusca in corresponding described sub-ant colony produces successively Breath, it is achieved thereby that be allocated to multiple sub-ant colony from server simultaneously so that it is can complete concurrently to search by search mission Rope, improves the speed of search, afterwards, multiple described is respectively compared from each self-corresponding described sub-ant colony acquisition from server A plurality of described first path optimizing information, it is thus achieved that respective first optimal path information, and first optimum described in each self refresh Pheromone on routing information, afterwards, sends jointly to master server by respective first optimal path information and corresponding informance element, And then shorten master server and compare the time of optimal path information bar number so that it is the optimum that can quickly obtain logistics distribution is joined Send path.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to The accompanying drawing provided obtains other accompanying drawing.

Fig. 1 a kind of be applied to multiple band weak rock mass logistics transportation from server for what the embodiment of the present invention provided and dispatch Path calculation method flow chart；

Fig. 2 is applied to multiple band weak rock mass logistics transportation tune from server for the another kind that the embodiment of the present invention provides The path calculation method flow chart of degree；

The road of a kind of band weak rock mass logistics transportation scheduling being applied to master server that Fig. 3 provides for the embodiment of the present invention Footpath computational methods flow chart；

The structural representation of a kind of server cluster that Fig. 4 provides for the embodiment of the present invention；

The structural representation of a kind of master server that Fig. 5 provides for the embodiment of the present invention.

Detailed description of the invention

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise Embodiment, broadly falls into the scope of protection of the invention.

Embodiments provide the path calculation method of a kind of band weak rock mass logistics transportation scheduling, refer to accompanying drawing 1, it is applied to multiple from server, wherein, carries out the execution step of an iteration search unanimously from server described in each, institute The method of stating specifically includes following steps:

Step 101: multiple described receive parallel from server master server send come go out to send instructions and described in other from The pheromone of server；

Concrete, when calculating the optimal path of band weak rock mass logistics transportation scheduling, pre-build band weak rock mass Logistics transportation scheduling mathematic model:

$\min z=\underset{k=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\left(C_v+C_d\right)d_{ij}{x}_{ijk}+c_1\underset{i=1}{\overset{N}{\Σ}}\max \{({\mathrm{et}}_i-t_i),0\}+c_2\underset{i=1}{\overset{N}{\Σ}}\max \{(t_i-{\mathrm{lt}}_i),0\}$

$m=\[\underset{i=1}{\overset{N}{\Σ}}{g}_i/\mathrm{\α}G\]+1$

And above-mentioned model meets following constraints:

$\underset{i=1}{\overset{N}{\Σ}}{g}_i{y}_{ik}\≤G,k=1,2,...,m$

$\underset{i=0}{\overset{N}{\Σ}}\underset{j=0}{\overset{N}{\Σ}}{c}_{ij}{x}_{ijk}\≤L,k=1,2,...,m$

$\underset{k=1}{\overset{m}{\Σ}}{y}_{ik}=1,i=1,2,...,N$

$\underset{k=1}{\overset{m}{\Σ}}{y}_{0k}=m$

t₀=T₀=0

t_i>=0 i=1,2 ..., N

T_i>=0 i=1,2 ..., N

et_i＜ t_i＜ lt_iI=1,2 ..., N

$\underset{i=1}{\overset{N}{\Σ}}{x}_{ijk}=y_{jk},j=1,2,...,N;K=1,2,...,m$

$\underset{j=1}{\overset{N}{\Σ}}{x}_{ijk}=y_{ik},i=1,2,...,N;K=1,2,...,m$

Wherein, N is dispensing target number, and N is positive integer；M is the vehicle number that home-delivery center sends, and m is positive integer； g_iThe demand of target of providing and delivering for i-th, and g_i=1,2 ..., N；[] represents the maximum positive integer of the numeral being not more than in bracket； α is the degree of difficulty and the estimation of constraint to entrucking and unloaded, and α is more than zero, less than 1；G be each car maximum load about Bundle；C_vFor unit distance cost of vehicle operation use, C_dFor driver's unit costs；d_ijThe road of dispensing target i to j has been travelled for vehicle k Journey；c₁For dispatching buses early to the penalty coefficient of dispensing objective, c₂For the late punishment system to dispensing objective that dispatches buses Number；et_iFor the permission earliest start time of task i, lt_iFor the permission late start time of task i, t_iFor vehicle point of arrival i's Time；

Afterwards, use ant colony parallel algorithm that described model is solved, first several Formica fuscas are averagely allocated to multiple From server, each described mistake that can control sub-ant colony this route searching of complete independently contained in it from server Journey, and multiple described separate between server, but receive parallel master server send come go out to send instructions and except itself Other outer described pheromones come from server transmission, wherein, and if only if controls multiple described clothes at described master server Business device performs the most independently for the first time iterative search time, multiple described in receive from server other described in send out from server The pheromone sent here is sky.

Step 102: update each self-corresponding sub-ant colony parameter described in multiple from server respectively, arbitrary described sub-ant colony is joined Number includes delivery location dot information, the heuristic greedy method of Path selection, taboo list and enabled node list, wherein, described dispensing Location point information includes the initial information element value on the number of delivery location point, present position and each delivery location point, institute Stating taboo list and preserve the delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, described enabled node list preserves described In sub-ant colony arbitrary Formica fusca without delivery location point；

Concrete, multiple described receive parallel from server master server send come go out to send instructions and described in other After server sends the pheromone come, multiple described in server each starts to control it contained sub-ant colony carry out road The first step of path search task, updates each self-corresponding sub-ant colony parameter the most respectively, and then it is multiple in control to realize master server Described when server carries out successive ignition search, multiple described can each obtain different route searching knots from server Really, meanwhile, the heuristic greedy method of the Path selection multiple described described sub-ant colony parameters arranged according to self from server And other the described pheromones come from server transmission received, decrease multiple route result obtained between server Identical probability.

Step 103: from server respectively according to the institute of each Formica fusca each self-corresponding described sub-ant colony described in multiple State the heuristic greedy method of Path selection, taboo list, enabled node list, and other are described from the pheromone of server, successively The first path optimizing information that each Formica fusca in described sub-ant colony corresponding to acquisition produces；

Concrete, when performing an iteration search, multiple each from server described can be successively from server The delivery location point that in sub-ant colony contained in acquiring it, each Formica fusca is whole in described delivery location dot information After, produced first path optimizing information.

Step 104: multiple described a plurality of institutes being respectively compared from each self-corresponding described sub-ant colony acquisition from server State the first path optimizing information, it is thus achieved that each minimum routing information, as multiple described from server respectively since corresponding institute State the first optimal path information obtained in sub-ant colony；

Concrete, described in multiple, obtain each ant the sub-ant colony of each control the most independently from server After the first path optimizing information that ant produces, multiple described can each compare oneself corresponding sub-ant colony from server and obtained A plurality of described first path optimizing information, and therefrom obtain one of shortest path as the first optimal path information.

Step 105: multiple described update described the respectively obtained since corresponding described sub-ant colony respectively from server Pheromone in one optimal path information；

Concrete, owing to updating the dispensing position the sub-ant colony parameter of correspondence independently of one another from server described in multiple When putting dot information so that be each assigned with initial information element value each described delivery location point in server, therefore, Described in each after server is by comparing and having obtained respective first optimal path information, each in again updating it Pheromone on individual delivery location point, and then utilize the first optimal path information that described pheromone record obtains, wherein, each Individual described from the initial information element value that server is its interior delivery location point distribution, can be with uniform distribution, it is also possible to uneven Distribution.

Step 106: from server respectively by described first optimal path information each obtained and pheromone described in multiple, It is sent to described master server, obtains the foundation of current optimal path as described master server；

Concrete, multiple described obtain the first optimal path information and pheromone independently of one another from server time, Need each the first corresponding optimal path information obtained and pheromone to be sent to described master server, in order to master server It is analyzed collecting to the Search Results returned, it is thus achieved that final optimal path.

Disclosed in the embodiment of the present invention band weak rock mass logistics transportation scheduling path calculation method in, be applied to multiple from Server, by multiple from server have received parallel master server send come go out to send instructions and described in other from service During the pheromone of device, independently of each other according to the Path selection of each Formica fusca heuristic in each self-corresponding described sub-ant colony The factor, taboo list, enabled node list, and other are described from the pheromone of server, obtain corresponding described son successively The first path optimizing information that each Formica fusca in ant colony produces, it is achieved thereby that search mission is allocated to simultaneously multiple from Sub-ant colony in server so that it is can complete route searching concurrently, improves the speed of search, afterwards, multiple described from The a plurality of described first path optimizing information that server more each obtains, obtains respective first optimal path information, and respectively Pheromone in first optimal path information described in self refresh, afterwards, by respective first optimal path information and corresponding informance Element sends jointly to master server, and then shortens master server and compare the time of path optimizing information bar number so that it is can be fast Speed obtains the optimum Distribution path of logistics distribution.

Embodiments provide the path calculation method of another kind of band weak rock mass logistics transportation scheduling, refer to attached Fig. 2, is applied to multiple from server, wherein, carries out the execution step of an iteration search unanimously from server described in each, Described method specifically includes following steps:

Step 201: multiple described receive parallel from server master server send come go out to send instructions and described in other from The pheromone of server.

Step 202: update each self-corresponding sub-ant colony parameter described in multiple from server respectively, arbitrary described sub-ant colony is joined Number includes delivery location dot information, the heuristic greedy method of Path selection, taboo list and enabled node list, wherein, described dispensing Location point information includes the initial information element value on the number of delivery location point, present position and each delivery location point, institute Stating taboo list and preserve the delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, described enabled node list preserves described In sub-ant colony arbitrary Formica fusca without delivery location point.

Step 203: make each Formica fusca each self-corresponding described sub-ant colony respectively successively from server described in multiple From original position point；

Concrete, multiple described carry out route searching the most independently from server time, each is described from service all Make each Formica fusca in sub-ant colony contained in it the most sequentially from respective original position point, in order to follow-up successively Obtain each Formica fusca to complete to search for produced routing information.

Step 204: multiple described heuristic greedy method and receptions being utilized respectively respective described Path selection from server Other arrived are described from the pheromone of server, calculate current next step permission of dispensing Formica fusca in each self-corresponding described sub-ant colony Delivery location point, and described delivery location point is saved in and described current dispensing taboo list corresponding to Formica fusca, wherein, described Next step delivery location point allowed is the arbitrary delivery location point in the enabled node list of described current dispensing Formica fusca；

Concrete, multiple described arbitrary from server are contained already at original position in server is according to it The sub-ant colony of point, therefrom one Formica fusca of selection is as Formica fusca of currently providing and delivering, and according to delivery location point more new formula:

And if only if j ∈ Allowed_k, otherwise

Obtain current next step the delivery location point allowed of dispensing Formica fusca in the described sub-ant colony of self correspondence, wherein, i Being currently located delivery location point for current dispensing Formica fusca k, j is delivery location point of currently next step permission of dispensing Formica fusca k, and α is The pheromone that current dispensing Formica fusca k accumulates in motor process, β is the heuristic greedy method of Path selection, d_ijRepresent delivery location Distance between some i to j,For current dispensing Formica fusca k in t from the probability of delivery location point i to j；Q₁And Q₂It is One constant；Allowedk is enabled node list, and Allowed_k=1,2 ..., n}-Tabu_k；Tabu_kFor current dispensing The taboo list of the delivery location point that Formica fusca k selected before next step；For β at current dispensing Formica fusca k by delivery location Point i transfers to the effect degree of j；

Secondly, by current dispensing next step delivery location point to be passed through of Formica fusca of choosing from described available section point range Table is deleted, and is added in described taboo list, in order to follow-up current dispensing Formica fusca can update according to delivery location point Formula has traveled through whole delivery location points that correspondence comprises from server；

Due to the heuristic greedy method difference of multiple described Path selection arranged from server self, and Formica fusca of currently providing and delivering In motor process, on the delivery location point of process, the pheromone of accumulation is different so that from server the most parallel described in multiple Independently during executive path search task, multiple road obtained an iteration search procedure from server can be prevented effectively from The situation that footpath result is consistent occurs, meanwhile, because Formica fusca can trend towards the path selecting have higher pheromone intensity level, so Other in addition to self received send, from server, the pheromones of coming, can according to other from server obtain first Information in shortest path information usually adjusts each next step delivery location point selected of Formica fusca in body ant colony, and then adjusts The whole selection in path；Wherein, the pheromone bag of accumulation on the delivery location point of current dispensing Formica fusca process in motor process Include on corresponding each delivery location point from the delivery location dot information included by server update ant colony parameter is initial Pheromone value, and other pheromones come from server transmission in addition to self received；

Additionally, only control multiple described servers at described master server to perform iterative search for the first time the most independently Time, multiple described receive from server other described in send, from server, the pheromone come be sky, Formica fusca of the most currently providing and delivering In motor process, on the delivery location point of process, the pheromone of accumulation only includes corresponding from server update ant colony parameter institute Including delivery location dot information in each delivery location point on initial information element value.

Step 205: update, from server, ant of currently providing and delivering described in each self-corresponding described sub-ant colony respectively described in multiple The second path optimizing information of ant；

Concrete, multiple described arbitrary current dispensing Formica fuscas comprised from server from server from current institute After the delivery location point at place moves to the delivery location point of next permission, corresponding needs joining of being presently in from server The distance record between location point, the next delivery location point allowed and the two dispensing point is sent to preserve, as one second Path optimizing information, in order to the corresponding fullpath result obtained after server completes an iteration search of follow-up composition.

Step 206: judge multiple described from each self-corresponding described sub-ant colony of server each Formica fusca the most all move A dynamic described delivery location point, the most then perform step 207, performs step 204 if it is not, then return；

Concrete, due to multiple described from server arbitrary from each Formica fusca contained by server need successively to Before all move a delivery location point after, could again make each Formica fusca in sub-ant colony successively from the dispensing position being presently in Put and a little move to the respective next delivery location point allowed, accordingly, it would be desirable to determining arbitrary from server each After Formica fusca all moves a delivery location point, just can carry out follow-up mobile operation.

Step 207: multiple described heuristic greedy method and receptions being utilized respectively respective described Path selection from server Other arrived are described from the pheromone of server, calculate current next step permission of dispensing Formica fusca in each self-corresponding described sub-ant colony Delivery location point, and described delivery location point is saved in and described current dispensing taboo list corresponding to Formica fusca；

Concrete, after determining that arbitrary each Formica fusca from server all moves a delivery location point, continue From the sub-ant colony contained by server, a Formica fusca is reselected as Formica fusca of currently providing and delivering so that it is according to delivery location from described Point more new formula, it is thus achieved that current next step the delivery location point allowed of dispensing Formica fusca in the described sub-ant colony of self correspondence, it After, current dispensing next step delivery location point to be passed through of Formica fusca chosen is deleted from described enabled node list, And be added in described taboo list.

Step 208: update, from server, ant of currently providing and delivering described in each self-corresponding described sub-ant colony respectively described in multiple The second path optimizing information of ant；

Concrete, record preserves multiple described arbitrary delivery location being presently in from server from server Distance record between point, the next delivery location point allowed and the two dispensing point preserves, and optimizes road as one second Footpath information.

Step 209: judge multiple described the most equal time of each Formica fusca from each self-corresponding described sub-ant colony of server Go through the whole delivery location points comprised in described delivery location dot information, the most then perform step 2010, if it is not, then return Perform step 207.

Step 2010: multiple described each Formica fuscas obtained the most successively the described sub-ant colony of correspondence from server produce The first raw path optimizing information, wherein, described first path optimizing information is made up of multiple described second path optimizing information；

Concrete, multiple described from each self-corresponding described sub-ant colony of server each Formica fusca all traveled through described After the whole delivery location points comprised in delivery location dot information, multiple described from server all record save the most right Should in sub-ant colony each Formica fusca through the first path optimizing information produced by whole delivery location points.

Step 2011: multiple described a plurality of institutes being respectively compared from each self-corresponding described sub-ant colony acquisition from server State the first path optimizing information, it is thus achieved that each minimum routing information, as multiple described from server respectively since corresponding institute State the first optimal path information obtained in sub-ant colony.

Step 2012: multiple described update described the respectively obtained since corresponding described sub-ant colony respectively from server Pheromone in one optimal path information；

Concrete, multiple described arbitrary from server from server according to Pheromone update formula:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in each self-corresponding described first optimal path information, wherein, m is arbitrary from contained by server The number of Formica fusca in sub-ant colony, n is iterations, τ_ijT () is the information that t remains between delivery location point ij on ij line Element amount, τ_ij(t+n) it is the pheromone amount of residual, d between delivery location point ij on ij line after delivery location point nth iteration_ijFor Distance between delivery location point ij, Δ τ_ij(t) be current dispensing Formica fusca k in t from the pheromone of delivery location point i to j Residual quantity, ρ is the degree of volatility of pheromone, L_kBy current dispensing Formica fusca k length in walked path in an iteration search circulation Degree.

Step 2013: from server respectively by described first optimal path information each obtained and information described in multiple Element, is sent to described master server, obtains the foundation of current optimal path as described master server.

Disclosed in the embodiment of the present invention, the path calculation method of band weak rock mass logistics transportation scheduling, is applied to multiple from clothes Business device, by multiple described arbitrary from server heuristic greedy method from server and receive other described in from service The pheromone of device, calculates current next step delivery location point allowed of dispensing Formica fusca in each self-corresponding described sub-ant colony, effectively The situation avoiding multiple route result obtained an iteration search procedure from server consistent occurs, and afterwards, will select Current dispensing next step delivery location point to be passed through of Formica fusca out is deleted from described enabled node list, and is added In described taboo list, in order to follow-up current dispensing Formica fusca can travel through correspondence from service according to delivery location point more new formula The whole delivery location points comprised in device.

Embodiments provide the path calculation method of another kind of band weak rock mass logistics transportation scheduling, refer to attached Fig. 3, is applied to master server, and described method specifically includes following steps:

Step 301: send out and send instructions to multiple from server, wherein, multiple described arbitrary from clothes from server Business device corresponding to a sub-ant colony, and described in go out to send instructions for multiple described from server respectively from each self-corresponding described sub-ant The foundation of the first optimal path information and pheromone is obtained in Qun；

Concrete, pre-building the logistics transportation scheduling model of band weak rock mass, described master server is described by sending Go out to send instructions to multiple described from server, control multiple described can carry out current iteration search the most independently i.e. from server The logistics transportation scheduling model of described band weak rock mass is solved.

Step 302: receive multiple described described first optimal path information sent respectively from server and pheromone；

Concrete, multiple described during server carries out current iteration search the most independently, multiple described Arbitrary from server can obtain respective described first optimal path information and at described first optimal path from server The pheromone value of distribution in information, afterwards, will send each self-corresponding described first optimal path information and pheromone respectively To described master server, in order to the route result that current iteration is searched for is analyzed collecting by master server.

Step 303: by receive each described in from server send come described pheromone be sent to described in other Sub-ant colony corresponding from server；

Concrete, in order to avoid multiple described from server when carrying out an iteration search, it is thus achieved that route result phase With, master server needs multiple described pheromone values of distribution from described first optimal path information that server transmission comes Again it is sent to other from server except sending this pheromone value itself multiple from server, in order to other are multiple from clothes The pheromone that business device can be distributed according to other local optimum path obtained from server, adjusts from changing in next time During generation search, each Formica fusca can be selected differently from the delivery location point of the route result that an iterative search is formed, and then Obtain the route result that more last iterative search is more excellent.

Step 304: by multiple described described first optimal path information sent respectively from server received and Current optimization length compares, it is thus achieved that current optimal path, as the current optimization length of next iteration；

Concrete, multiple described the first optimal path information sent respectively from server and master server are prestored Current optimization length carries out size and compares, if there have length to be less than in multiple described first optimal path information to be described current optimum long The path of degree, then the path replacing described current optimization length is current optimal path, is made by described current optimal path meanwhile Current optimization length for next iteration, in order to check whether to there is optimal path in the route result of next iteration；If Multiple described first optimal path information have the path that length is more than or equal to described current optimization length, then keeps described working as The path of front optimization length is constant, and still as the standard of comparison of next iteration search；

Additionally, and if only if, described master server controls multiple described when server carries out first time iterative search, institute Stating the current optimization length that master server prestores is sky, and now, described master server described is entered from server by the most multiple Row the first optimal path information that iterative search is obtained for the first time, obtains the initial value of current optimization length.

Step 305: judge that current iteration number of times, whether equal to the maximum iteration time preset, if being equal to, then performs step 306, if being less than, then return and perform step 301；

Concrete, it is optimal path to ensure the route result of dispensing, needs the iterative search carried out repeatedly, i.e. described Each search mission it is iterated independently parallel from server, it is thus achieved that respective local described in the many secondary control of master server are multiple Optimal path result and distribution pheromone value on the path.

Step 306: export current optimal path；

Concrete, when current iteration number of times is equal to the maximum iteration time preset, just will pass through to compare obtain current Optimal path exports as final result.

In the path calculation method of band weak rock mass logistics transportation scheduling disclosed in the embodiment of the present invention, it is mainly used in master Server, after pre-building the logistics transportation scheduling model of band weak rock mass, is sent simultaneously to multiple by going out to send instructions From server, it is possible to obtain the first optimal path information and the letter of multiple described many sub-ant colony each corresponding from server Breath element, it is achieved thereby that be allocated to multiple many sub-ant colony from server simultaneously so that it is can be concurrently by search mission Complete search, improve the speed of search so that it is can quickly obtain the optimum Distribution path of logistics distribution.

Embodiments providing a kind of server cluster, refer to accompanying drawing 4, described server cluster includes:

N from server；

Described server cluster 401 control n described receive parallel from server master server send come go out to send instructions And other are described from the pheromone of server；

Wherein, each is described from server 401i, for receive master server send come go out to send instructions and other n-1 Individual described from the pheromone of server, update corresponding sub-ant colony parameter, each Formica fusca in the described sub-ant colony that foundation is corresponding Described sub-ant colony parameter in the heuristic greedy method of Path selection, taboo list, enabled node list, and other n-1 institute State the pheromone from server, obtain the first path optimizing that each Formica fusca in corresponding described sub-ant colony produces successively Information, a plurality of described first path optimizing information obtained in the most corresponding described sub-ant colony, it is thus achieved that corresponding minimum road Footpath information, as the first optimal path information obtained from corresponding described sub-ant colony, updates from corresponding described sub-ant colony Pheromone in described first optimal path information of middle acquisition, by described first optimal path information obtained and pheromone, It is sent to described master server；

Wherein, described sub-ant colony parameter includes delivery location dot information, the heuristic greedy method of Path selection, taboo list and can With node listing, wherein, described delivery location dot information includes the number of delivery location point, present position and each dispensing position Putting the initial information element value on a little, described taboo list preserves the delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, Described enabled node list preserve arbitrary Formica fusca in described sub-ant colony without delivery location point；

Concrete, constitute server cluster 401 from server described in n, send going out receiving master server When sending instructions, described server cluster 401 can control to receive this parallel from server described in its interior n and go out to send instructions, and then Start to perform this iterative search each, independently of the other, simultaneously can be described from service by n received from master server Other n-1 that on the local optimum path of device, the pheromone information of distribution is sent respectively in addition to self described from server；

Secondly, described receive this at n from server parallel and go out to send instructions and described in except self in addition to other n-1 from clothes After the pheromone of business device, each is described can control the Formica fusca in each self-corresponding sub-ant colony from server 401i and carry out this repeatedly Generation search, and then obtain respective local optimum path, and return to described master server, in order to described master server is to reception To a plurality of local optimum path be analyzed, obtain optimal path result；Wherein, each is described from server 401i control Formica fusca in body ant colony carries out the detailed process of path iterative search: update corresponding sub-ant colony parameter, i.e. determines this Distribution on the number of delivery location point involved by secondary iterative search, delivery location point present position, each delivery location point In initial information element value, the heuristic greedy method of Path selection and described sub-ant colony each each self-corresponding taboo list of Formica fusca with Enabled node list, afterwards, according to described from the information of server from described sub-ant colony parameter and other n-1 of receiving Element, the first path optimizing letter that in obtaining it successively, in sub-ant colony, each Formica fusca produces after whole delivery location points Breath, carries out length by a plurality of described first path optimizing information obtained and compares, therefrom select a shortest paths conduct The first optimal path information that should obtain from server, afterwards, again updating should be from the first optimal path letter described in server The pheromone value of distribution on breath, and this first optimal path information and the corresponding pheromone produced all are sent to described main service Device, and receive described described first optimal path information each acquired from server of n and right at described master server After the pheromone that should produce, just carry out subsequent analysis operation, so as to quickly obtain optimal path.

In server cluster disclosed in the embodiment of the present invention, receive main clothes parallel from server by controlling its interior n What business device transmission came goes out to send instructions and described in other after the pheromone of server, completes current iteration search independently of one another and appoints Business, it is achieved thereby that be allocated to multiple sub-ant colony from server simultaneously so that it is can complete road concurrently by search mission Path search, improves the speed of search, afterwards, by n described Formica fusca institute from server the most corresponding sub-ant colony The a plurality of local path result obtained, it is thus achieved that the pheromone of distribution in first optimal path information and this path, and by it It is sent to described master server, compares the time of path optimizing information bar number shortening master server so that it is can quickly obtain Obtain the optimum Distribution path of logistics distribution.

The work process of the unit that the embodiment of the present invention provides, refer to the flow chart corresponding to accompanying drawing 1, concrete work Repeat no more as process.

From server 401i described in n involved in the above-described embodiments each from server described, depend on According to the heuristic greedy method of Path selection in the described sub-ant colony parameter of each Formica fusca in corresponding described sub-ant colony, taboo Table, enabled node list, and other n-1 described from the pheromone of server, obtains corresponding described sub-ant colony successively In the first path optimizing information of producing of each Formica fusca, the process of implementing is:

Make each Formica fusca in the described sub-ant colony of correspondence successively from original position point；

The heuristic greedy method utilizing the described Path selection in corresponding described sub-ant colony parameter and other n-1 received Individual described from the pheromone of server, calculate current next step delivery location allowed of dispensing Formica fusca in corresponding described sub-ant colony Point, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, wherein, described next step allows Delivery location point be described current dispensing Formica fusca enabled node list in arbitrary delivery location point；

Update the second path optimizing information of current Formica fusca of providing and delivering described in corresponding described sub-ant colony, and return described profit Described from clothes with the heuristic greedy method of the described Path selection in corresponding described sub-ant colony parameter and receive other n-1 The pheromone of business device, calculates current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and by institute State delivery location point and be saved in the taboo list corresponding with described current dispensing Formica fusca, until in corresponding described sub-ant colony each Formica fusca all moves a described delivery location point；

When each Formica fusca all moves a described delivery location point in corresponding described sub-ant colony, return described utilization The heuristic greedy method of the corresponding described Path selection in described sub-ant colony parameter and receive other n-1 are described from service The pheromone of device, calculates current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and by described Delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, until each ant in corresponding described sub-ant colony Ant has all traveled through the whole delivery location points comprised in described delivery location dot information；

Obtain the described first path optimizing information that each Formica fusca in the described sub-ant colony of correspondence produces successively, its In, described first path optimizing information is made up of multiple described second path optimizing information；

Concrete, for each in described server cluster 401 from server 401i ant in obtaining it successively During the first path optimizing information that each Formica fusca in Qun produces, each Formica fusca in corresponding sub-ant colony can be made all from same Individual original position point sets out, and first selects a Formica fusca from corresponding sub-ant colony as Formica fusca of currently providing and delivering so that it is according to pre- The delivery location point more new formula first stored:

And if only if j ∈ Allowed_k, otherwise

Obtaining current next step delivery location point allowed of dispensing Formica fusca, wherein, i is currently located for current dispensing Formica fusca k Delivery location point, j is delivery location point of currently next step permission of dispensing Formica fusca k, and α is that current dispensing Formica fusca k is in motor process The pheromone of middle accumulation, β is the heuristic greedy method of Path selection, d_ijRepresent the distance between delivery location point i to j,For Current dispensing Formica fusca k in t from the probability of delivery location point i to j；Q₁And Q₂It is a constant；Allowedk is available section Point list, and Allowed_k=1,2 ..., n}-Tabu_k；Tabu_kThe dispensing selected before next step for current dispensing Formica fusca k The taboo list of location point；Transferred to the effect degree of j by delivery location point i at current dispensing Formica fusca k for β；

Afterwards, by next step delivery location point allowed of obtaining from described current dispensing available section point range corresponding to Formica fusca Delete in table, and this point is added in the taboo list of correspondence, then make described current dispensing Formica fusca from the location point being presently in Move to next step the delivery location point allowed calculated, and record this mobile route information, as one second Path optimizing information, now, completes and this one Formica fusca from the sub-ant colony of server is moved forward a delivery location point Operation, carry out the shifting of delivery location point successively according to remaining Formica fusca in the mobile step antithetical phrase ant colony respectively of this Formica fusca afterwards Dynamic, until after in described sub-ant colony, each Formica fusca all moves a delivery location point, then these Formica fuscas are carried out next round Mobile operation, until whole Formica fuscas travel throughs whole delivery location point positions, could obtain in the sub-ant colony of this server often The first path optimizing information that one Formica fusca is formed after all dispensing points.

In the embodiment of the present invention, each in described server cluster 401 can the most independently from server 401i Control in respective sub-ant colony the heuristic greedy method of Path selection that obtains after each Formica fusca updates according to self and receive Other n-1 described from the pheromone of server, can obtain each Formica fusca successively and be produced after whole delivery location points The first raw path optimizing information, effectively prevent again n from the path knot that server obtains an iteration search procedure The most consistent situation occurs.

Update from server 401i described in n involved in the above-described embodiments each from server described The pheromone in described first optimal path information obtained from corresponding described sub-ant colony, the process of implementing is:

Described in each from server 401i after obtaining respective first optimal path information, can according to pheromone more New formula:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in each self-corresponding described first optimal path information, wherein, m is Formica fusca in arbitrary ant colony Number, n is iterations, τ_ijT () is the pheromone amount that t remains between delivery location point ij on ij line, τ_ij(t+n) For the pheromone amount remained between delivery location point ij on ij line after delivery location point nth iteration, d_ijFor delivery location point ij Between distance, Δ τ_ij(t) be arbitrary Formica fusca k be pheromone in t from the pheromone residual quantity of delivery location point i to j, ρ Degree of volatility, L_kBy arbitrary Formica fusca k length in walked path in once circulation.

In the embodiment of the present invention, each in described server cluster 401 passes through Pheromone update from server 401i Formula, it is achieved that the process that the pheromone of distribution in the first optimal path information obtained is updated.

Embodiments provide a kind of master server, refer to accompanying drawing 5, including:

Sending module 501, is used for sending out and sends instructions to n from server, by receive each described in from service Device sends the described pheromone come and is sent to other n-1 described sub-ant colony corresponding from server, and wherein, n described Arbitrary from server from server corresponding to a sub-ant colony, and described in go out to send instructions for n described from server respectively The foundation of the first optimal path information and pheromone is obtained from each self-corresponding described sub-ant colony；

Concrete, when carrying out iterative search for the first time, described master server is to n the described information sent from server Element is empty, and afterwards in the successive ignition search procedure carried out, described master server all can send institute to described in n from server State out and send instructions, be sent in addition to self from the pheromone of server described in n simultaneously last iterative search obtained Other n-1 from server, in order to n described individual from clothes according to other n-1 when carrying out next iteration search from server The pheromone of business device suitably adjusts the local path result of oneself.

Receiver module 502, for receive described described first optimal path information sent respectively from server of n and Pheromone；

Concrete, at described master server in time going out to send instructions from server transmitted in parallel described in n, n is described from clothes Business device can be separate carry out current iteration search mission, and by the first optimal path information each obtained and this path The pheromone of distribution all returns to described master server, is analyzed collecting for master server.

Comparison module 503, is used for n described described first optimal path information sent respectively from server and works as Front optimization length compares, it is thus achieved that current optimal path, as the current optimization length of next iteration；

Concrete, when described master server controls to carry out iterative search for the first time from server described in n, described main clothes The current optimization length that business device prestores is sky, and now, described master server is carried out for the first time from server described in n by comparing The first optimal path information that iterative search is obtained, obtains the initial value of current optimization length；

Afterwards, iteration each time relatively in, if n is described sends the first optimal path described in the n bar come from server When information having length be less than described current optimization length, then replace the first optimal path that current optimization length is length minimum Information, if being all higher than described current optimization length, then will described current optimization length as the standard value that compares of iteration next time.

Judge module 504, for judging that whether current iteration number of times is equal to the maximum iteration time preset；

Sending module 501, is additionally operable to when current iteration number of times is less than described default maximum iteration time, and transmission is set out Instruction to n is described from server；

Concrete, when current iteration number of times is less than described default maximum iteration time, sending out in described master server Send module 501 to again send out to send instructions to n from server, afterwards, the most again by the reception mould in described master server Block 502, comparison module 503 carry out iterative computation next time, until described judge module 504 judges that current iteration number of times is equal to During the maximum iteration time preset, just stop iterative computation.

Output module 505, for when current iteration number of times is equal to described default maximum iteration time, output is Shortest path, the optimal path provided and delivered as this；

Concrete, when current iteration number of times is equal to described default maximum iteration time, just ought by obtain more afterwards The most final result of front optimal path exports, and terminates the calculating of this Distribution path.

In master server disclosed in the embodiment of the present invention, by going out to send instructions to be sent simultaneously to n from server, it is possible to Obtain the first optimal path information and the pheromone of n described many sub-ant colony each corresponding from server, thus realize Search mission is allocated to simultaneously multiple many sub-ant colony from server so that it is search can be completed concurrently, improve The speed of search so that it is can quickly obtain the optimum Distribution path of logistics distribution.

The work process of the unit that the embodiment of the present invention provides, refer to the flow chart corresponding to accompanying drawing 3, concrete work Repeat no more as process.

Some possible application scenarios that technical scheme to the embodiment of the present invention relates to below carry out citing and introduce.

The basic parameter of the band weak rock mass logistics transportation scheduling by getting: the position of client, the delivery car in parking lot Maximum carrying capacity is 12 tons, and the max mileage of trip is 150 every time, and parking lot coordinate is (0,0), vehicle time of departure the earliest It is 8:00, pre-builds band weak rock mass logistics transportation scheduling mathematic model:

$\min z=\underset{k=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\left(C_v+C_d\right)d_{ij}{x}_{ijk}+c_1\underset{i=1}{\overset{N}{\Σ}}\max \{({\mathrm{et}}_i-t_i),0\}+c_2\underset{i=1}{\overset{N}{\Σ}}\max \{(t_i-{\mathrm{lt}}_i),0\}$

$m=\[\underset{i=1}{\overset{N}{\Σ}}{g}_i/\mathrm{\α}G\]+1$

And above-mentioned model meets following constraints:

$\underset{i=1}{\overset{N}{\Σ}}{g}_i{y}_{ik}\≤G,k=1,2,...,m$

$\underset{i=0}{\overset{N}{\Σ}}\underset{j=0}{\overset{N}{\Σ}}{c}_{ij}{x}_{ijk}\≤L,k=1,2,...,m$

$\underset{k=1}{\overset{m}{\Σ}}{y}_{ik}=1,i=1,2,...,N$

$\underset{k=1}{\overset{m}{\Σ}}{y}_{0k}=m$

t₀=T₀=0

t_i>=0 i=1,2 ..., N

T_i>=0 i=1,2 ..., N

et_i＜ t_i＜ lt_iI=1,2 ..., N

$\underset{i=1}{\overset{N}{\Σ}}{x}_{ijk}=y_{jk},j=1,2,...,N;K=1,2,...,m$

$\underset{j=1}{\overset{N}{\Σ}}{x}_{ijk}=y_{ik},i=1,2,...,N;K=1,2,...,m$

Wherein, client's number to be serviced is 16, and m represents the vehicle number (m is positive integer) that home-delivery center sends, i-th The demand of client is g_i(g_i=1,2 ..., N), in formula, [] represents the maximum positive integer of the numeral being not more than in bracket, α (0 ＜ α ＜ 1) it is the degree of difficulty and the estimation of constraint to entrucking and unloaded, unit distance cost of vehicle operation is 50 yuan/km, driver 0.5 yuan/km of unit costs, d_ijRepresent that vehicle k has travelled the distance of client i to j, wherein waiting cost c₁It is 10 yuan/h, vehicle Deferred charges c₂It is 100 yuan/h, wherein et_iFor the permission earliest start time of task i, lt_iFor task i allow start at the latest Time, t_iTime for vehicle point of arrival i；

Described band weak rock mass logistics transportation scheduling mathematic model represents the cost consumption during logistics transportation, including thing Flow cost of vehicle operation use in transporting, driver's emolument expense, do sth. in advance or delay to reach a wait for generation of providing and delivering and lose and delay Fine.

It is 6 that master server arranges the maximum iteration time value of this path computing, makes initial iterations=0, and often Carry out an iteration and just add 1, and by total ant colony that size is 200 Formica fuscas, be divided into four sub-ant colony, the number of every sub-ant colony Amount is 50 Formica fuscas, and transfers to four each to control from server these four sub-ant colony；

When master server send out send instructions simultaneously to four from server time, i.e. started for the first time iterative search, four Individual can receive this parallel from server and go out to send instructions, and be iterated search mission independently of one another；And each is from server Control its interior sub-ant colony and be iterated the process of search unanimously, but when each iterative search starts, four from server meeting It is respectively provided with the control parameter of each Formica fusca in body ant colony: the number of dispensing client, the position at dispensing client place, each Individual dispensing the pheromone value of client's initial distribution, the heuristic greedy method of Path selection, record the dispensing that arbitrary Formica fusca passes through The taboo list of client and record arbitrary Formica fusca without the enabled node list of dispensing client, wherein, due to each from clothes The heuristic greedy method of Path selection that business device is arranged can affect this server in sub-ant colony Formica fusca select different dispensing clients, Therefore, it can make four heuristic greedy method arranging different Path selection from server, to avoid four to carry out from server Identical local path result is obtained after an iteration search；

Afterwards, each making its 50 interior Formica fuscas be both placed in logistics distribution center i.e. parking lot from server, its coordinate sets For (0,0), initial time t=0, select from 50 Formica fuscas one out as Formica fusca of currently providing and delivering, and from its correspondence can With node listing being searched its unbeaten dispensing client, according to formula:

And if only if j ∈ Allowed_k, otherwise

Select current dispensing next step dispensing client to be passed through of Formica fusca, and by this dispensing client from enabled node list Middle deletion, adds in taboo list, also records current dispensing Formica fusca simultaneously and moves to first from logistics distribution center at present The part Distribution path that dispensing client is formed；

Afterwards so that the process that 49 Formica fuscas of the residue in this sub-ant colony are all advanced according to first Formica fusca, order completes The respective first step moves；Then, move the process of a dispensing client according still further to 50 Formica fuscas in sub-ant colony, circulation make this 50 Formica fusca again moves into another dispensing client, until they have moved 16 dispensing clients that this server is arranged, now, This server obtains 50 the local optimum paths corresponding with 50 Formica fuscas, and this server first oneself can optimize road by these 50 Footpath compares, and one that therefrom selects minimum is sent to master server, to reduce the time that main service is compared；Meanwhile, from Pheromone on this minimum local optimum path can be updated by server, to highlight the path of this local optimum；

Main service is receiving four 4 minimum local optimum paths sent respectively from server and is being distributed in this After pheromone on a little paths, the length in these 4 minimum local optimum paths can be compared, select minimum one as current Optimization length, in order to when carrying out next iteration search, as standard of comparison；

Afterwards, master server can again send out issue a command to four from server, control its carry out second time iteration search Rope, meanwhile, the letter also first time iterative search four being distributed from 4 minimum local optimum paths that server is corresponding Breath element be sent respectively to except self in addition to other three from server, changes carrying out second time from server reducing four The probability that the local optimum route result that generation search obtains is identical with the local optimum route result obtained for the first time, and also can make Four usually adjust next second son ant colony from server by the information receiving other three the local optimum paths from server The route result that Formica fusca produces, until master server judges that the most accumulative iterative search number of times has reached 6 times, the most directly ought Front optimal path exports, the optimal path result calculated as this Distribution path, and then improves the search speed of algorithm, makes It can quickly obtain optimum Distribution path.

Described above to the disclosed embodiments, makes professional and technical personnel in the field be capable of or uses the present invention. Multiple amendment to these embodiments will be apparent from for those skilled in the art, as defined herein General Principle can realize without departing from the spirit or scope of the present invention in other embodiments.Therefore, the present invention It is not intended to be limited to the embodiments shown herein, and is to fit to and principles disclosed herein and features of novelty phase one The widest scope caused.

Claims (8)

1. the path calculation method of a band weak rock mass logistics transportation scheduling, it is characterised in that be applied to multiple from server, Described method includes:

Multiple described receive parallel from server master server send come go out to send instructions and described in other from the information of server Element；

Update each self-corresponding sub-ant colony parameter described in multiple respectively from server, arbitrary described sub-ant colony parameter includes position of providing and delivering Put dot information, the heuristic greedy method of Path selection, taboo list and enabled node list, wherein, described delivery location dot information bag Including the initial information element value on the number of delivery location point, present position and each delivery location point, described taboo list preserves The delivery location point that in described sub-ant colony, arbitrary Formica fusca passes through, described enabled node list preserves in described sub-ant colony arbitrary Formica fusca without delivery location point；

From server respectively according to the described Path selection of each Formica fusca each self-corresponding described sub-ant colony described in multiple Heuristic greedy method, taboo list, enabled node list, and other are described from the pheromone of server, successively corresponding to acquisition The first path optimizing information that each Formica fusca in described sub-ant colony produces；

Multiple described a plurality of described first optimization roads being respectively compared from each self-corresponding described sub-ant colony acquisition from server Footpath information, it is thus achieved that the most minimum routing information, respectively obtains from server described in multiple since corresponding described sub-ant colony The first optimal path information taken；

Multiple described described first optimal paths respectively obtained since corresponding described sub-ant colony that update respectively from server are believed Pheromone on breath；

From server respectively by described first optimal path information each obtained and pheromone described in multiple, it is sent to described master Server, obtains the foundation of current optimal path as described master server.

Method the most according to claim 1, it is characterised in that the plurality of described the most corresponding from server foundation respectively Described sub-ant colony in the heuristic greedy method of described Path selection of each Formica fusca, taboo list, enabled node list, Yi Jiqi He is described from the pheromone of server, obtains the first optimization that each Formica fusca in corresponding described sub-ant colony produces successively Routing information, including:

Make each Formica fusca each self-corresponding described sub-ant colony successively from original position point respectively from server described in multiple Set out；

Multiple described from server be utilized respectively the heuristic greedy method of respective described Path selection and receive other described in From the pheromone of server, calculate current next step delivery location allowed of dispensing Formica fusca in each self-corresponding described sub-ant colony Point, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, wherein, described next step allows Delivery location point be described current dispensing Formica fusca enabled node list in arbitrary delivery location point；

Multiple described the second optimizations updating currently dispensing Formica fusca described in each self-corresponding described sub-ant colony from server respectively Routing information, and return and the plurality of described be utilized respectively the heuristic greedy method of respective described Path selection from server and connect Other received are described from the pheromone of server, and next step is permitted to calculate in each self-corresponding described sub-ant colony current dispensing Formica fusca The delivery location point permitted, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, until many Described in individual, from each self-corresponding described sub-ant colony of server, each Formica fusca all moves a described delivery location point；

Multiple described from each self-corresponding described sub-ant colony of server each Formica fusca all move a described delivery location During point, return and the plurality of described be utilized respectively the heuristic greedy method of respective described Path selection from server and receive Other are described from the pheromone of server, calculate joining of next step permission of Formica fusca of currently providing and delivering in each self-corresponding described sub-ant colony Send location point, and described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, until multiple described From each self-corresponding described sub-ant colony of server, each Formica fusca has all traveled through comprise in described delivery location dot information complete Portion's delivery location point；

The first optimization that multiple described each Formica fuscas obtained the most successively the described sub-ant colony of correspondence from server produce Routing information, wherein, described first path optimizing information is made up of multiple described second path optimizing information.

Method the most according to claim 1, it is characterised in that update each since right described in the plurality of from server respectively The pheromone in described first optimal path information obtained in the described sub-ant colony answered, including:

Multiple described from server foundation Pheromone update formula respectively:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in each self-corresponding described first optimal path information, wherein, m is the number of Formica fusca in arbitrary ant colony Mesh, n is iterations, τ_ijT () is the pheromone amount that t remains between delivery location point ij on ij line, τ_ij(t+n) it is The pheromone amount remained between delivery location point ij on ij line after delivery location point nth iteration, d_ijFor delivery location point ij it Between distance, Δ τ_ij(t) be arbitrary Formica fusca k be pheromone in t from the pheromone residual quantity of delivery location point i to j, ρ Degree of volatility, L_kBy arbitrary Formica fusca k length in walked path in once circulation.

4. the path calculation method of a band weak rock mass logistics transportation scheduling, it is characterised in that be applied to master server, described Method includes:

Send out and send instructions to multiple from server, wherein, multiple described arbitrary from server from server corresponding to one Individual sub-ant colony, and described in go out to send instructions and described from each self-corresponding described sub-ant colony, obtain first respectively from server for multiple Optimal path information and the foundation of pheromone；

Receive multiple described described first optimal path information sent respectively from server and pheromone；

By receive each described in send, from server, the described pheromone come and be sent to described in other institute from server Corresponding sub-ant colony；

By multiple described described first optimal path information sent respectively from server received and current optimization length Compare, it is thus achieved that current optimal path, as the current optimization length of next iteration, and send out described in returning and send instructions To multiple from server, until current iteration number of times reaches default maximum iteration time, export optimal path.

5. a server cluster, it is characterised in that including:

N from server；

Described server set group control n is described receive parallel from server master server send come go out to send instructions and other institutes State the pheromone from server；

Wherein, each is described from server, for receive master server send come go out to send instructions and other n-1 described from The pheromone of server, updates corresponding sub-ant colony parameter, according to the described son of each Formica fusca in corresponding described sub-ant colony The heuristic greedy method of the Path selection in ant colony parameter, taboo list, enabled node list, and other n-1 described from service The pheromone of device, obtains the first path optimizing information that each Formica fusca in corresponding described sub-ant colony produces successively, than The a plurality of described first path optimizing information obtained in more corresponding described sub-ant colony, it is thus achieved that corresponding minimum routing information, As the first optimal path information obtained from corresponding described sub-ant colony, update acquisition from corresponding described sub-ant colony Pheromone in described first optimal path information, by described first optimal path information obtained and pheromone, is sent to institute State master server；

Wherein, described sub-ant colony parameter includes delivery location dot information, the heuristic greedy method of Path selection, taboo list and available section Point list, wherein, described delivery location dot information includes the number of delivery location point, present position and each delivery location point On initial information element value, described taboo list preserves the delivery location point that arbitrary Formica fusca passes through in described sub-ant colony, described Enabled node list preserve arbitrary Formica fusca in described sub-ant colony without delivery location point.

Server cluster the most according to claim 5, it is characterised in that described each described from server according to corresponding Described sub-ant colony in each Formica fusca described sub-ant colony parameter in the heuristic greedy method of Path selection, taboo list, available Node listing, and other n-1 are described from the pheromone of server, obtain in corresponding described sub-ant colony successively is each The first path optimizing information that Formica fusca produces, specifically for:

Make each Formica fusca in the described sub-ant colony of correspondence successively from original position point；

The heuristic greedy method utilizing the described Path selection in corresponding described sub-ant colony parameter and other n-1 the institute received State the pheromone from server, calculate current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, And described delivery location point is saved in the taboo list corresponding with described current dispensing Formica fusca, wherein, next step permission described Delivery location point is the arbitrary delivery location point in the enabled node list of described current dispensing Formica fusca；

Update the second path optimizing information of current Formica fusca of providing and delivering described in corresponding described sub-ant colony, and return described utilize right The heuristic greedy method of the described Path selection in the described sub-ant colony parameter answered and receive other n-1 are described from server Pheromone, calculate current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and join described Location point is sent to be saved in the taboo list corresponding with described current dispensing Formica fusca, until each Formica fusca in corresponding described sub-ant colony All move a described delivery location point；

When each Formica fusca all moves a described delivery location point in corresponding described sub-ant colony, return described utilization correspondence Described sub-ant colony parameter in the heuristic greedy method of described Path selection and receive other n-1 described from server Pheromone, calculates current next step delivery location point allowed of dispensing Formica fusca in corresponding described sub-ant colony, and by described dispensing Location point is saved in the taboo list corresponding with described current dispensing Formica fusca, until each Formica fusca is equal in corresponding described sub-ant colony The whole delivery location points comprised in described delivery location dot information are traveled through；

Obtain the described first path optimizing information that each Formica fusca in the described sub-ant colony of correspondence produces, wherein, institute successively First path optimizing information of stating is made up of multiple described second path optimizing information.

Server cluster the most according to claim 5, it is characterised in that described each described from server update from right The pheromone in described first optimal path information obtained in the described sub-ant colony answered, specifically for:

According to Pheromone update formula:

${\mathrm{\τ}}_{ij}(t+n)=(1-\mathrm{\ρ}){\mathrm{\τ}}_{ij}+\mathrm{\ρ}\·\underset{k=1}{\overset{m}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k$

In this circulation that and if only if, Formica fusca k is through i, j, otherwise

And if only if, and this circulation i, j belong to local optimum path Lmin, otherwise

Updating the pheromone in corresponding described first optimal path information, wherein, m is the number of Formica fusca in arbitrary ant colony, and n is Iterations, τ_ijT () is the pheromone amount that t remains between delivery location point ij on ij line, τ_ij(t+n) for dispensing position The pheromone amount remained between delivery location point ij on ij line after putting a nth iteration, d_ijFor between delivery location point ij away from From, Δ τ_ij(t) be arbitrary Formica fusca k in t from the pheromone residual quantity of delivery location point i to j, ρ is the volatilization journey of pheromone Degree, L_kBy arbitrary Formica fusca k length in walked path in once circulation.

8. a master server, it is characterised in that including:

Sending module, is used for sending out and sends instructions to n from server, by receive each described in send from server Described pheromone be sent to the described sub-ant colony corresponding from server of other n-1, wherein, n is described from server In arbitrary from server corresponding to a sub-ant colony, and described in go out to send instructions for n described from server respectively from the most right The described sub-ant colony answered obtains the foundation of the first optimal path information and pheromone；

Receiver module, for receiving n described described first optimal path information sent respectively from server and pheromone；

Comparison module, for optimum with current by n described described first optimal path information sent respectively from server Length compares, it is thus achieved that current optimal path, as the current optimization length of next iteration；

Judge module, for judging that whether current iteration number of times is equal to the maximum iteration time preset；

Sending module, is additionally operable to, when current iteration number of times is less than described default maximum iteration time, send out and send instructions to n Individual from server；

Output module, for when current iteration number of times is equal to described default maximum iteration time, exporting current optimal path, The optimal path provided and delivered as this.