CN105528675A - Production distribution scheduling method based on ant colony algorithm - Google Patents

Abstract

The present invention discloses a production distribution scheduling method based on ant colony algorithm. The modeling is carried out for a production combined distribution scheduling problem with volume and production time differences, then the solution is carried out through an improved ant colony algorithm, thus a set of optimized scheme of production distribution scheduling is obtained, the total operation cost of a target position manufacturing enterprise is reduced greatly, and the enterprise operational efficiency is improved. At the same time, for a specific problem, a practical application development solution is given, and thus an enterprise can efficiently obtain a suitable optimal scheduling scheme.

Description

A kind of production distribution scheduling method based on ant group algorithm

Technical field

The invention belongs to supply chain field, specifically a kind of production distribution scheduling method based on ant group optimization.

Background technology

Under Vehicles Collected from Market environment, the core competitiveness of manufacturing enterprise is no longer simple manufacturing capacity, but supply chain operating capability, manufacturing enterprise needs that buying, production and dispensing link are carried out the overall situation and controls, production system and logistics system are carried out combined optimization, the maximization of overall economic benefit could be obtained, the competitive power of enterprise.

Combined dispatching is the optimization method of a class towards supply chain, adopts the normal form of accurately scheduling, and in design supply chain, the combined dispatching scheme of each link, realizes the optimization of enterprise-wide economic benefit, thus the service level of enterprise.

Under at present traditional mode of production pattern all being concentrated on to the research of associating scheduling problem, in this kind of production model, an equipment can process the operation of an operation or a collection of fixed qty simultaneously, but in real industry, difference batch production pattern has the character of these two kinds of production models concurrently, and more more complicated than this two class, apply more extensive.And traditional dispatching method only considers how to reduce production cost mostly simply, do not consider how to make the total cost of production dispensing combined dispatching minimum, therefore can not adapt to Production requirement instantly.

Summary of the invention

The present invention is the weak point existed to overcome prior art, provides a kind of production distribution scheduling method based on ant group optimization, to realizing the optimization of enterprise-wide economic benefit, thus can reduce enterprise cost, the efficiency of service of enterprise.

The present invention is that technical solution problem adopts following technical scheme:

The feature of a kind of production distribution scheduling method based on ant group optimization of the present invention is carried out as follows:

Step 1, suppose to exist n batch needs and carry out producing and providing and delivering, the volume of equipment of producing each batch is designated as B; The vehicle volume of each batch of providing and delivering is designated as V; Form one batch of set by described n batch, be designated as U={b ₁, b ₂..., b _k..., b _n, b _krepresent that kth is criticized; And by a kth crowd b _ksize be designated as S _k; By a kth crowd b _kproduction time be designated as T _k; A group is designated as by adding all batches of carrying out providing and delivering in same car; 1≤k≤n;

Step 2, measure-alike batch in described crowd of set U is divided into a class, thus obtains a classification; Be designated as W={w ₁, w ₂..., w _z..., w _a, w _zrepresent z classification; Described z classification w _zin batch total be designated as f _z;

The parameters of step 3, initialization ant group algorithm, comprising: m represents m ant, and initialization m=1; M represents ant sum, L represents iterations, and initialization L=1; L _maxrepresent maximum iteration time;

Step 4, defining variable are l, and initialization l=1; A definition kth crowd b _kidentifier be flag _k, and initialization flag _k=0;

Step 5, create l the group of m ant of the L time iteration and l the candidate list corresponding with it and described n batch can be assigned in a different group by m ant of the L time iteration provide and deliver; And the vehicle fleet that m ant of the L time iteration completes grouping use all batches

Step 6, by m+1 assignment to m, and return step 5 and perform, until m=M, thus the vehicle fleet set that all M the ants obtaining the L time iteration complete grouping use all batches

Step 7, from described vehicle fleet set in choose the locally optimal solution of minimum value as the L time iteration, be designated as π _l;

Step 8, i-th candidate utilizing formula (1) to upgrade the L time iteration criticize b _i' join l group with jth in candidate criticize b _j' between pheromones thus obtain the pheromones of the L+1 time iteration

In formula (1), ρ represents the evaporation rate of pheromones; m _ij(L) represent that in the L time iteration, i-th candidate criticizes b _i' join l group with jth in candidate criticize b _j' be assigned to the number of times of same group; And have:

Step 9, formula (3) and formula (4) is utilized to judge whether meet pheromone concentration and limit interval [τ _min, τ _max], if meet, then retain the pheromones of the L+1 time iteration and perform step 11; Otherwise, perform step 10:

${\mathrm{\τ}}_{max}\≤\frac{1}{(1-\mathrm{\ρ})f\left({\mathrm{\π}}^{*}\right)}---\left(3\right)$

${\mathrm{\τ}}_{\min }=\frac{{\mathrm{\τ}}_{max}(1-\sqrt[a]{0.05})}{(a/2-1)\sqrt[a]{0.05}}---\left(4\right)$

In formula (3) and formula (4), π ^*represent the minimum value in current acquired all locally optimal solutions;

If step 10 then will assignment is given if then will assignment is given

Step 11; By L+1 assignment to L, judge L < L _maxwhether set up, if set up, return step 4 and perform, otherwise complete L _maxsecondary iteration, and obtain globally optimal solution π _best, be L _maxthe minimum value of all locally optimal solutions in secondary iteration; With globally optimal solution π _bestcorresponding distribution project is as optimum distribution project;

Step 12, batch will carry out descending sort according to the production time in each group in optimum distribution project, the ranking results of acquisition as series-produced order, thus obtains Optimal Production and dispensing combined dispatching scheme.

The feature of the production scheduling method of the porcelain calcine technology based on ant group optimization of the present invention is also, in described step 5, m ant of the L time iteration is assigned to described n batch in different groups as follows to provide and deliver:

Step 5.1, defining variable f;

Step 5.2, initialization z=1;

Step 5.3, initialization f=1;

Step 5.4, judge described z classification w _zin select the identifier flag of f crowd _fwhether be 1; If 1, then represent that f batch has completed grouping, and perform step 5.5; Otherwise, by z classification w _zin select f batch and join described l candidate list in, and perform step 5.6;

Step 5.5, by f+1 assignment to f, and return step 5.4 and perform, until f=f _ztill after, perform step 5.6;

Step 5.6, by z+1 assignment to z, and return step 5.3 and perform; Until after z=a, thus obtain l candidate list to be updated remember described l candidate list to be updated in candidate criticize as { b ₁', b ₂' ..., b ' _i..., b _a'; 1≤i≤a;

Step 5.7, from described candidate list to be updated crowd b that middle selection size is maximum _key' joining described l group is criticized as key and key is criticized b _key' identifier flag ' _keybe set to 1; Then from candidate list, crucial crowd b is deleted _key';

Step 5.8, formula (5) is utilized to obtain m ant by described l candidate list to be updated in i-th candidate criticize b _i' join l group candidate probability thus obtain m ant and all candidates are criticized join l group candidate probability set

$P_{i1}^m=\left\{\begin{array}{cc}\frac{{\mathrm{\&theta;}}_{i1}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{i1}^{\mathrm{\&beta;}}}{{\mathrm{\&Sigma;}}_{b_{i^{\&prime;}}\&Element;X_1^{\&prime;\left(L\right)\left(m\right)}}{\mathrm{\&theta;}}_{i1}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{i1}^{\mathrm{\&beta;}}}& {b_i}^{\&prime;}\&Element;X_1^{\&prime;\left(L\right)\left(m\right)}\\ 0& else\end{array}\right.---\left(5\right)$

In formula (5), α is the weight of pheromones, and β is the weight of heuristic information; θ _ilrepresent that i-th candidate criticizes b _i' l group can be joined the expecting degree of middle dispensing; η _ilrepresent and i-th candidate is criticized b _i' l group can be joined the heuristic information of dispensing; And have:

${\mathrm{\&theta;}}_{i1}=\frac{1}{\left|G_1^{\left(L\right)\left(m\right)}\right|}{\mathrm{\&Sigma;}}_{j\&Element;G_1^{\left(L\right)\left(m\right)}}{\mathrm{\&tau;}}_{ij}\left(L\right)---\left(6\right)$

${\mathrm{\&eta;}}_{i1}=\frac{S_i}{{\mathrm{\&Sigma;}}_{i=1}^n{S}_i}---\left(7\right)$

In formula (6), τ _ijrepresent that i-th candidate of the L time iteration criticizes b _i' join l group with jth in candidate criticize b _j' between pheromones, 1≤i ≠ j≤a; S _irepresent that i-th candidate criticizes b ' _isize;

Step 5.9, from described candidate probability set in select Maximum alternative probability corresponding Maximum alternative is criticized, and is designated as b ' _max; Then described Maximum alternative criticizes b ' _maxsize be designated as s _max;

Step 5.10, described Maximum alternative is criticized b ' _maxjoin described l group and Maximum alternative is criticized b ' _maxidentifier flag ' _maxbe set to 1;

Step 5.11, described vehicle volume V is deducted described Maximum alternative criticize b ' _maxsize s _max, obtain residue vehicle volume, be designated as

Step 5.12, according to described residue vehicle volume from described l candidate list to be updated middle deletion size is greater than described residue vehicle volume candidate criticize; Thus obtain l the candidate list upgraded

Step 5.13, obtain described Maximum alternative and criticize b ' _maxcorresponding classification, is designated as w _max;

Step 5.14, judge that described Maximum alternative criticizes b ' _maxat corresponding classification w _maxin whether be f _zindividual batch; If so, then from l the candidate list upgraded the described Maximum alternative of middle deletion criticizes b ' _max; Otherwise, from l the candidate list upgraded the described Maximum alternative of middle deletion criticizes b ' _max, and described Maximum alternative is criticized b ' _maxcorresponding classification w _maxin f _maxcriticize l the candidate list joining described renewal for+1 in, thus obtain l the candidate list again upgraded

Step 5.15, with described l the candidate list again upgraded as l candidate list to be updated and return the execution of step 5.8 order, until l candidate list to be updated till sky, by l the group of m ant of the L time iteration fill it up with; Thus obtain l the group of m ant of the L time iteration distribution project;

Whether step 5.16, the identifier judging n batch are all 1, if be all 1, have then represented that m ant of the L time iteration is to the individual distribution criticized of described n; And l assignment is given otherwise, by l+1 assignment to l, and return the execution of step 5 order.

Compared with the prior art, beneficial effect of the present invention is embodied in:

1, the present invention is under typical difference is produced in batches pattern, the production of research manufacturing enterprise, dispensing two benches combined dispatching problem, by adopting the ant group algorithm improved, under difference is produced in batches pattern, first classify for product in batches, then propose, based on the candidate list of batch size classes, by setting up the method for candidate list for each ant, to draw the grouping scheme of each ant; Recycling pheromone updating rule, lastest imformation element, achieves successive ignition, finally obtains optimum solution; Compared to other method; Ant group algorithm, on the degree of optimization of available time and result, is that a kind of performance better optimizes the approximate data manufacturing span; Solve the problem that in real industry, job batching is produced, grouping is provided and delivered, make obtained optimum solution achieve the optimization of enterprise-wide economic benefit, reduce energy consumption, provide cost savings, improve the service level of enterprise.

2, the present invention is by carrying out Pheromone update to the optimum solution that iteration obtains each time, solves the problem focused on around this solution that search procedure is too fast; If the optimum solution obtained since first time iteration is upgraded, probably cause algorithm convergence too early; And the optimum solution of iteration has very big-difference after iteration each time each time, so upgrade the optimum solution of iteration each time, the pheromones of more solutions can be made to strengthen, thus obtain more optimal solution.

3. the present invention adopts candidate list strategy to find optimum solution; In traditional ant group algorithm, as long as volume is less than the operation of vehicle volume, can select into candidate list, so the substantial amounts of feasible solution; But in actual entrucking process, a lot of batch size is the same, if identical for volume size criticizing is classified as a class, the categorical measure criticized can never exceed the total quantity criticized, the porcelain base of same size is criticized more, batch number of candidate list is fewer, which reduces the quantity of feasible solution, effectively reduces the time of solving.

Accompanying drawing explanation

Fig. 1 is that the present invention produces dispensing span schematic diagram.

Embodiment

In the present embodiment, a kind of production distribution scheduling method based on ant group optimization, carry out modeling for volume and production time discrepant production dispensing combined dispatching problem, then solved by a kind of ant group algorithm that improves, thus obtain the prioritization scheme of a set of production distribution scheduling, greatly reduce total operating cost of target bit manufacturing enterprise with this, improve the efficiency of operation of enterprises; Simultaneously for particular problem, give practical program development scheme, make enterprise's high efficiency acquisition of energy be applicable to its optimal scheduling scheme.Specifically, be carry out as follows:

Step 1, suppose to exist n batch needs and carry out producing and providing and delivering, the volume of equipment of producing each batch is designated as B; The vehicle volume of each batch of providing and delivering is designated as V; Form one batch of set by n batch, be designated as U={b ₁, b ₂..., b _k..., b _n, b _krepresent that kth is criticized; And by a kth crowd b _ksize be designated as S _k; By a kth crowd b _kproduction time be designated as T _k; A group is designated as by adding all batches of carrying out providing and delivering in same car; 1≤k≤n;

Step 2, measure-alike batch in crowd set U is divided into a class, thus obtains a classification; Be designated as W={w ₁, w ₂..., w _z..., w _a, w _zrepresent z classification; Z classification w _zin batch total be designated as f _z; In traditional ant group algorithm, as long as volume is less than the operation of place capacity, can select into candidate list, so the substantial amounts of feasible solution.But in actual ceramic roasting process, the volume size of a lot of pottery is the same, if porcelain base identical for volume size is classified as a class, the categorical measure of porcelain base can never exceed the total quantity of porcelain base, the porcelain base of same size is more, the porcelain base number of candidate list is fewer, so just can reduce the quantity of feasible solution, effectively can reduce the time of solving.

The parameters of step 3, initialization ant group algorithm, comprising: m represents m ant, and initialization m=1; M represents ant sum, L represents iterations, and initialization L=1; L _maxrepresent maximum iteration time;

Step 4, defining variable are l, and initialization l=1; A definition kth crowd b _kidentifier be flag _k, and initialization flag _k=0;

Step 5, create l the group of m ant of the L time iteration and l the candidate list corresponding with it and n batch can be assigned in a different group by m ant of the L time iteration provide and deliver; And the vehicle fleet that m ant of the L time iteration completes grouping use all batches

Step 5.1, defining variable f;

Step 5.2, initialization z=1;

Step 5.3, initialization f=1;

Step 5.4, judge z classification w _zin select the identifier flag of f crowd _fwhether be 1; If 1, then represent that f batch has completed grouping, and perform step 5.5; Otherwise, by z classification w _zin select f batch and join l candidate list in, and perform step 5.6;

Step 5.5, by f+1 assignment to f, and return step 5.4 and perform, until f=f _ztill after, perform step 5.6;

Step 5.6, by z+1 assignment to z, and return step 5.3 and perform; Until after z=a, thus obtain l candidate list to be updated remember l candidate list to be updated in candidate criticize as { b ₁', b ₂' ..., b ' _i..., b _a'; 1≤i≤a;

Step 5.7, from candidate list to be updated crowd b that middle selection size is maximum _key' joining l group is criticized as key and key is criticized b _key' identifier flag ' _keybe set to 1; Then from candidate list, crucial crowd b is deleted _key';

Step 5.8, formula (1) is utilized to obtain m ant by l candidate list to be updated in i-th candidate criticize b _i' join l group candidate probability thus obtain m ant and all candidates are criticized join l group candidate probability set

$P_{i1}^m=\left\{\begin{array}{cc}\frac{{\mathrm{\&theta;}}_{i1}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{i1}^{\mathrm{\&beta;}}}{{\mathrm{\&Sigma;}}_{b_{i^{\&prime;}}\&Element;X_1^{\&prime;\left(L\right)\left(m\right)}}{\mathrm{\&theta;}}_{i1}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{i1}^{\mathrm{\&beta;}}}& {b_i}^{\&prime;}\&Element;X_1^{\&prime;\left(L\right)\left(m\right)}\\ 0& else\end{array}\right.---\left(1\right)$

In formula (1), α is the weight of pheromones, and β is the weight of heuristic information; θ _ilrepresent that i-th candidate criticizes b _i' l group can be joined the expecting degree of middle dispensing, pheromones τ _ij(L) represent that i-th candidate criticizes b _i' criticize b with a jth candidate _j' the expecting degree being arranged in the same set.Due to lot number order and operation place group number uncertain, therefore directly can not use pheromones τ _ij, therefore defining variable θ (L) _ilindirect utilization is carried out to pheromones; η _ilrepresent and i-th candidate is criticized b _i' l group can be joined the heuristic information of dispensing; And have:

${\mathrm{\&theta;}}_{i1}=\frac{1}{\left|G_1^{\left(L\right)\left(m\right)}\right|}{\mathrm{\&Sigma;}}_{j\&Element;G_1^{\left(L\right)\left(m\right)}}{\mathrm{\&tau;}}_{ij}\left(L\right)---\left(2\right)$

${\mathrm{\&eta;}}_{i1}=\frac{S_i}{{\mathrm{\&Sigma;}}_{i=1}^n{S}_i}---\left(3\right)$

In formula (2), τ _ijrepresent that i-th candidate of the L time iteration criticizes b _i' join l group with jth in candidate criticize b _j' between pheromones, 1≤i ≠ j≤a; S _irepresent that i-th candidate criticizes b ' _isize;

Step 5.9, from candidate probability set in select Maximum alternative probability corresponding Maximum alternative is criticized, and is designated as b ' _max; Then Maximum alternative criticizes b ' _maxsize be designated as s _max;

Step 5.10, Maximum alternative is criticized b ' _maxjoin l group and Maximum alternative is criticized b ' _maxidentifier flag ' _maxbe set to 1;

Step 5.11, vehicle volume V is deducted Maximum alternative criticize b ' _maxsize s _max, obtain residue vehicle volume, be designated as

Step 5.12, according to residue vehicle volume from l candidate list to be updated middle deletion size is greater than residue vehicle volume candidate criticize; Thus obtain l the candidate list upgraded

Step 5.13, acquisition Maximum alternative criticize b ' _maxcorresponding classification, is designated as w _max;

Step 5.14, judge that Maximum alternative criticizes b ' _maxat corresponding classification w _maxin whether be f _zindividual batch; If so, then from l the candidate list upgraded middle deletion Maximum alternative criticizes b ' _max; Otherwise, from l the candidate list upgraded middle deletion Maximum alternative criticizes b ' _max, and Maximum alternative is criticized b ' _maxcorresponding classification w _maxin f _maxcriticize l the candidate list joining renewal for+1 in, thus obtain l the candidate list again upgraded

Step 5.15, with l the candidate list again upgraded as l candidate list to be updated and return the execution of step 5.8 order, until l candidate list to be updated till sky, by l the group of m ant of the L time iteration fill it up with; Thus obtain l the group of m ant of the L time iteration distribution project;

Whether step 5.16, the identifier judging n batch are all 1, if be all 1, have then represented that m ant of the L time iteration is to the individual distribution criticized of n; And l assignment is given otherwise, by l+1 assignment to l, and return the execution of step 5 order.

L the candidate list upgraded in, if i-th candidate criticizes b _i' criticize b with a jth candidate _j' meet S _i< S _jand S _it _i> S _jt _j, select i-th candidate to criticize b _i' add current group to criticize b than selecting a jth candidate _j' can more free space be reduced, and select i-th candidate to criticize b _i' the residual capacity of latter current group is larger, l candidate list in optional lot number also more, therefore select i-th candidate criticize b _i' add current group more excellent.

Step 6, by m+1 assignment to m, and return step 5 and perform, until m=M, thus the vehicle fleet set that all M the ants obtaining the L time iteration complete grouping use all batches

Step 7, from vehicle fleet set in choose the locally optimal solution of minimum value as the L time iteration, be designated as π _l;

Step 8, i-th candidate utilizing formula (4) to upgrade the L time iteration criticize b _i' join l group with jth in candidate criticize b _j' between pheromones thus obtain the pheromones of the L+1 time iteration

In formula (4), ρ represents the evaporation rate of pheromones; m _ij(L) represent that in the L time iteration, i-th candidate criticizes b _i' join l group with jth in candidate criticize b _j' be assigned to the number of times of same group; And have:

If the optimum solution obtained since first time iteration is upgraded, probably cause too fast the focusing on this and separate around of search procedure, make algorithm convergence too early; And the optimum solution of iteration has very big-difference after iteration each time each time, so upgrade the optimum solution of iteration each time, the pheromones of more solutions can be made to strengthen.Also rotation strategy can be adopted to upgrade, namely Pheromone update is carried out to the optimum solution that iteration obtains each time, then often through corresponding iteration, to the renewal of the optimum solution obtained since first time through row primary information element.

Step 9, formula (6) and formula (7) is utilized to judge whether meet pheromone concentration and limit interval [τ _min, τ _max], if meet, then retain the pheromones of the L+1 time iteration and perform step 11; Otherwise, perform step 10:

${\mathrm{\&tau;}}_{max}\&le;\frac{1}{(1-\mathrm{\&rho;})f\left({\mathrm{\&pi;}}^{*}\right)}---\left(6\right)$

${\mathrm{\&tau;}}_{\min }=\frac{{\mathrm{\&tau;}}_{max}(1-\sqrt[a]{0.05})}{(a/2-1)\sqrt[a]{0.05}}---\left(7\right)$

In formula (6) and formula (7), π ^*represent the minimum value in current acquired all locally optimal solutions;

If step 10 then will assignment is given if then will assignment is given the concentration limits of pheromones fixes on by ant group algorithm between, with the difference between the pheromones reducing feasible solution.

Step 11; By L+1 assignment to L, judge L < L _maxwhether set up, if set up, return step 4 and perform, otherwise complete L _maxsecondary iteration, and obtain globally optimal solution π _best, be L _maxthe minimum value of all locally optimal solutions in secondary iteration; With globally optimal solution π _bestcorresponding distribution project as optimum distribution project, in batches with delivery process as shown in Figure 1;

Step 12, batch will carry out descending sort according to the production time in each group in optimum distribution project, the ranking results of acquisition as series-produced order, thus obtains Optimal Production and dispensing combined dispatching scheme.

Claims (2)

1., based on a production distribution scheduling method for ant group optimization, it is characterized in that carrying out as follows:

Step 1, suppose to exist n batch needs and carry out producing and providing and delivering, the volume of equipment of producing each batch is designated as B; The vehicle volume of each batch of providing and delivering is designated as V; Form one batch of set by described n batch, be designated as U={b ₁, b ₂..., b _k..., b _n, b _krepresent that kth is criticized; And by a kth crowd b _ksize be designated as S _k; By a kth crowd b _kproduction time be designated as T _k; A group is designated as by adding all batches of carrying out providing and delivering in same car; 1≤k≤n;

Step 2, measure-alike batch in described crowd of set U is divided into a class, thus obtains a classification; Be designated as W={w ₁, w ₂..., w _z..., w _a, w _zrepresent z classification; Described z classification w _zin batch total be designated as f _z;

The parameters of step 3, initialization ant group algorithm, comprising: m represents m ant, and initialization m=1; M represents ant sum, L represents iterations, and initialization L=1; L _maxrepresent maximum iteration time;

Step 4, defining variable are l, and initialization l=1; A definition kth crowd b _kidentifier be flag _k, and initialization flag _k=0;

Step 5, create l the group of m ant of the L time iteration and l the candidate list corresponding with it and described n batch can be assigned in a different group by m ant of the L time iteration provide and deliver; And the vehicle fleet that m ant of the L time iteration completes grouping use all batches

Step 6, by m+1 assignment to m, and return step 5 and perform, until m=M, thus the vehicle fleet set that all M the ants obtaining the L time iteration complete grouping use all batches $\{N_{um}^{\left(L\right)\left(1\right)},N_{um}^{\left(L\right)\left(2\right)},...,N_{um}^{\left(L\right)\left(m\right)},...,N_{um}^{\left(L\right)\left(M\right)}\};$

Step 7, from described vehicle fleet set in choose the locally optimal solution of minimum value as the L time iteration, be designated as π _l;

Step 8, i-th candidate utilizing formula (1) to upgrade the L time iteration criticize b _i' join l group with jth in candidate criticize b _j' between pheromones τ _ij, thus obtain the pheromones τ of the L+1 time iteration (L) _ij(L+1):

${\mathrm{\&tau;}}_{ij}\left(L+1\right)=(1-\mathrm{\&rho;}){\mathrm{\&tau;}}_{ij}\left(L\right)+m_{ij}\left(L\right)\&dtri;{\mathrm{\&tau;}}_{ij}---\left(1\right)$

In formula (1), ρ represents the evaporation rate of pheromones; m _ij(L) represent that in the L time iteration, i-th candidate criticizes b _i' join l group with jth in candidate criticize b _j' be assigned to the number of times of same group; And have:

Step 9, formula (3) and formula (4) is utilized to judge τ _ij(L+1) whether meet pheromone concentration and limit interval [τ _min, τ _max], if meet, then retain the pheromones τ of the L+1 time iteration _ij, and perform step 11 (L+1); Otherwise, perform step 10:

${\mathrm{\&tau;}}_{max}\&le;\frac{1}{(1-\mathrm{\&rho;})f\left({\mathrm{\&pi;}}^{*}\right)}---\left(3\right)$

${\mathrm{\&tau;}}_{min}=\frac{{\mathrm{\&tau;}}_{max}(1-\sqrt[a]{0.05})}{(a/2-1)\sqrt[a]{0.05}}---\left(4\right)$

In formula (3) and formula (4), π ^*represent the minimum value in current acquired all locally optimal solutions;

If step 10 τ _ij(L+1)>=τ _max, then by τ _maxassignment is to τ _ij(L+1); If τ _ij(L+1)≤τ _min, then by τ _minassignment is to τ _ij(L+1);

Step 11; By L+1 assignment to L, judge L < L _maxwhether set up, if set up, return step 4 and perform, otherwise complete L _maxsecondary iteration, and obtain globally optimal solution π _best, be L _maxthe minimum value of all locally optimal solutions in secondary iteration; With globally optimal solution π _bestcorresponding distribution project is as optimum distribution project;

Step 12, batch will carry out descending sort according to the production time in each group in optimum distribution project, the ranking results of acquisition as series-produced order, thus obtains Optimal Production and dispensing combined dispatching scheme.

2. the production scheduling method of the porcelain calcine technology based on ant group optimization according to claim 1, is characterized in that, in described step 5, m ant of the L time iteration is assigned to described n batch in different groups as follows to provide and deliver:

Step 5.1, defining variable f;

Step 5.2, initialization z=1;

Step 5.3, initialization f=1;

Step 5.4, judge described z classification w _zin select the identifier flag of f crowd _fwhether be 1; If 1, then represent that f batch has completed grouping, and perform step 5.5; Otherwise, by z classification w _zin select f batch and join described l candidate list in, and perform step 5.6;

Step 5.5, by f+1 assignment to f, and return step 5.4 and perform, until f=f _ztill after, perform step 5.6;

Step 5.6, by z+1 assignment to z, and return step 5.3 and perform; Until after z=a, thus obtain l candidate list to be updated remember described l candidate list to be updated in candidate criticize as { b ₁', b ₂' ..., b _i' ..., b _a'; 1≤i≤a;

Step 5.7, from described candidate list to be updated crowd b that middle selection size is maximum _key' joining described l group is criticized as key and key is criticized b _key' identifier flag ' _keybe set to 1; Then from candidate list, crucial crowd b is deleted _key';

Step 5.8, formula (5) is utilized to obtain m ant by described l candidate list to be updated in i-th candidate criticize b _i' join l group candidate probability thus obtain m ant and all candidates are criticized join l group candidate probability set

$P_{il}^m=\left\{\begin{array}{cc}\frac{{\mathrm{\&theta;}}_{il}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{il}^{\mathrm{\&beta;}}}{{\mathrm{\&Sigma;}}_{b_i^{\&prime;}\&Element;X_l^{\&prime;\left(L\right)\left(m\right)}}{\mathrm{\&theta;}}_{il}^{\mathrm{\&alpha;}}{\mathrm{\&eta;}}_{il}^{\mathrm{\&beta;}}}& b_i^{\&prime;}\&Element;X_l^{\&prime;\left(L\right)\left(m\right)}\\ 0& else\end{array}\right.---\left(5\right)$

In formula (5), α is the weight of pheromones, and β is the weight of heuristic information; θ _ilrepresent that i-th candidate criticizes b _i' l group can be joined the expecting degree of middle dispensing; η _ilrepresent and i-th candidate is criticized b _i' l group can be joined the heuristic information of dispensing; And have:

${\mathrm{\&theta;}}_{il}=\frac{1}{\left|G_l^{\left(L\right)\left(m\right)}\right|}{\&Sigma;}_{j\&Element;G_l^{\left(L\right)\left(m\right)}}{\mathrm{\&tau;}}_{ij}\left(L\right)---\left(6\right)$

${\mathrm{\&eta;}}_{il}=\frac{S_i}{{\&Sigma;}_{i=1}^n{S}_i}---\left(7\right)$

In formula (6), τ _ijrepresent that i-th candidate of the L time iteration criticizes b _i' join l group with jth in candidate criticize b _j' between pheromones, 1≤i ≠ j≤a; S _irepresent that i-th candidate criticizes b ' _isize;

Step 5.9, from described candidate probability set in select Maximum alternative probability corresponding Maximum alternative is criticized, and is designated as b ' _max; Then described Maximum alternative criticizes b ' _maxsize be designated as s _max;

Step 5.10, described Maximum alternative is criticized b ' _maxjoin described l group and Maximum alternative is criticized b ' _maxidentifier flag ' _maxbe set to 1;

Step 5.11, described vehicle volume V is deducted described Maximum alternative criticize b ' _maxsize s _max, obtain residue vehicle volume, be designated as

Step 5.12, according to described residue vehicle volume from described l candidate list to be updated middle deletion size is greater than described residue vehicle volume candidate criticize; Thus obtain l the candidate list upgraded

Step 5.13, obtain described Maximum alternative and criticize b ' _maxcorresponding classification, is designated as w _max;

Step 5.14, judge that described Maximum alternative criticizes b ' _maxat corresponding classification w _maxin whether be f _zindividual batch; If so, then from l the candidate list upgraded the described Maximum alternative of middle deletion criticizes b ' _max; Otherwise, from l the candidate list upgraded the described Maximum alternative of middle deletion criticizes b ' _max, and described Maximum alternative is criticized b ' _maxcorresponding classification w _maxin f _maxcriticize l the candidate list joining described renewal for+1 in, thus obtain l the candidate list again upgraded

Step 5.15, with described l the candidate list again upgraded as l candidate list to be updated and return the execution of step 5.8 order, until l candidate list to be updated till sky, by l the group of m ant of the L time iteration fill it up with; Thus obtain l the group of m ant of the L time iteration distribution project;

Whether step 5.16, the identifier judging n batch are all 1, if be all 1, have then represented that m ant of the L time iteration is to the individual distribution criticized of described n; And l assignment is given otherwise, by l+1 assignment to l, and return the execution of step 5 order.