CN105321042B - Genetic algorithm-based advanced plan scheduling system and method - Google Patents

Abstract

The present invention relates to a genetic algorithm-based advanced plan scheduling system and method. The system comprises a basic setting module, a data management module and a scheduling module. Optimal solutions are successively generated from potential solution populations by using the characteristics of mutation and population evolution in a genetic algorithm, so as to solve the scheduling problems of machining workshops of aviation enterprises.

Description

A kind of high-level plan program system based on genetic algorithm and method

Technical field

The present invention relates to a kind of high-level plan program system based on genetic algorithm and method, belong to computer control managing Technical field.

Background technology

The military enterprises such as Aeronautics and Astronautics belong to typical discrete manufacturing business, and Workshop Production generally has product structure complexity, Parts are many, multi items, small lot, criticize and produce and scientific research Flexible production, and the impact factor of production is many, and plan arrangement difficulty is big to wait spy Point.In actual production, staff planners need, according to batch product task, scientific research mission, other fragmentary tasks, to need with reference to stock's feelings Condition, device resource situation, frock tool ensure that situation, technological preparation situation, material supply situation, personnel's situation etc. carry out workshop The arrangement of production plan and scheduling.Because practical situation is complicated, the factor of impact plan is many, and information gathering data amount is big, a lot Workshop management software some rules to pre-define mostly, such as the shortest process time are preferential, First Come First Served, date of delivery Early priority scheduling, is had search and the sequence of " emphasis " formula, person carries to be Workshop Production Management by this didactic rule For planning scheduling decision-making.But generally there is in actual production process the complexity of production factors, the randomness of impact plan factor, The binding character of working condition, and solve site problems multiple target the features such as.Therefore, using general regular computing side Method, on fast search optimal case, the aspect such as multiple-objection optimization can not solve this certified Job Shop well A NP difficult problem for type.

The method of high-level plan scheduling has a lot, mainly includes the scheduling method based on planning, constraint planning CP, branch is fixed Boundary B＆B, decomposition technique DT, TABU search and genetic algorithm.Implemented simple and convenient based on the scheduling method of planning, but rule Then single it is impossible to realize the production scheduling of multiple constraint；Constraint planning CP although it is contemplated that various constraints in reality, but it Be primarily used to rule go out infeasible scheme and cut down substantial amounts of search space, and require rule base larger it is impossible to protect Card obtains optimal solution；Branch boundary B＆B avoids and enumerates, and makes primal problem be divided into less subproblem, but calculates time-consuming Consumption internal memory；Decomposition technique DT can reduce optimization scale, obtains relatively satisfactory solution in finite time, but limitation is more, such as It is limited to length and Moving Unit of form etc..TABU search is similar with genetic algorithm, but the initial solution of TABU search is to calculation Method search performance affects larger it is more difficult to execute.Genetic algorithm has preferable robustness, can receive in np problem in finite time Hold back and obtain comparatively ideal solution.

In the research of Production Planning Scheduling, genetic algorithm also has more application.Tan Hui etc. (2005) and Zhou Xin etc. (2010) have studied the production scheduling problem under distributed production model, the result of case verification shows that genetic algorithm can be fine The scheduling that ground meets distributed enterprise requires.But the main constraints in these researchs are distributed across the work of diverse location Factory, very big with the constraints difference of discrete type manufacturing shop.Fan Yingli etc. (2002) asks with reference to genetic algorithm and TABU search The production planning problem of solution Reconfigurable Manufacturing System；Li Xiu etc. (2001) have studied in batch production genetic algorithm in plant working The application of plan.This two researchs all demonstrate the advantage that genetic algorithm solves production planning problem, but a weighting enterprise The production schedule, a weighting Production Scheduling Problem, do not inquire into the application in production scheduling for the genetic algorithm.

Chinese patent application 200710200861.4, denomination of invention：Production schedule automatic arrangement program system and method, application People：Great Fujin accurate industry (Shenzhen) Co., Ltd.Which disclose a kind of production schedule automatic arrangement program system and side Method, there is provided the main frame database of system, the information such as the related order of plan scheduling, Work station capacity in main frame reading database Afterwards, automatically calculate and export Production Planning Scheduling result.But this patent does not refer to how realizing Production Planning Scheduling, does not have yet There is the algorithm of the scheduling that clarifies a plan.

Chinese patent application 201410506043.7, denomination of invention：Material production scheduling simulator, applicant：Wuhan Iron and steel (group) company.Which disclose a kind of material production scheduling simulator, there is provided unified data-interface, realize Being wirelessly transferred between Scheduling Algorithm server and data base, terminal unit.What this patent solved is hardware integration problem, does not have Relate to the algorithm of plan scheduling.

Content of the invention

The technology of the present invention solve problem：For the Workshop Production feature of the military enterprises such as Aeronautics and Astronautics, for plan scheduling The big problem of difficulty, overcomes the deficiencies in the prior art, provides a kind of high-level plan program system based on genetic algorithm and method, Using variation, Evolution of Population characteristic in genetic algorithm, potential solution population gradually produces optimal solution plan, solve The scheduling problem of Aviation Enterprise workshop.

The technology of the present invention solution：Based on the high-level plan program system of genetic algorithm, including basis instrument module, number According to management module and plan scheduling module three part, wherein：

Basis instrument module：Carry out material essential information, personal information, equipment and equipment group information, the working day in workshop Go through input and maintenance, the setting of information；Operating right setting is carried out to the application personnel in system；This module is data management mould Block and the support of plan scheduling module offer Back ground Information and system operation information；

Data management module：Incoming task and its precedence information, processing technique information, operation resource information and stock's letter Breath.Wherein processing technique information include process process information, operation man-hour information, working procedure processing unit information, operation setting Information, and operation material requested information；Operation resource information includes the work needed for equipment and equipment group information, working procedure processing Dress, tool information；

Plan scheduling module：Read the work calendar information in system-based setup module, read in data management module Task and its precedence information, processing technique information, operation resource information and inventory information.Input schduling parameter N, T, Pc, Pm, N represent Population Size, and T represents iterationses, and Pc represents crossover probability, and Pm represents mutation probability, calls Scheduling Algorithm, defeated Go out scheduling result, i.e. the Gantt chart of device resource dimension and task dimension.

Based on the high-level plan scheduling method of genetic algorithm, adjust two parts including preliminary scheduling and constraint, wherein：

Preliminary scheduling is：

(1) schduling parameter N, T, Pc, Pm are inputtedRead the work calendar information in basis instrument module, read data pipe Task in reason module and its precedence information, processing technique information, operation resource information and inventory information, are task encoding, Random generation scale is initial population A of N；

(2) calculate the Proper treatment of every chromosome in initial population A, using the shortest total elapsed time as evaluation criterion, The result obtaining after will be inverted for the total elapsed time of every chromosome is as the appropriate value of this chromosome；

(3) for the different select probability Ps of the appropriateness value imparting of every chromosome, (computing formula is as follows, wherein i Represent certain chromosome, its appropriate value is designated as fi), select N bar chromosome from initial population A by the way of roulette, raw Population A ＇ of Cheng Xin；

(4) the crossover probability Pc according to input, makes the parent chromosome in population A ＇ be intersected two-by-two, the dye after intersection Colour solid needs to carry out legitimacy inspection, and legal chromosome retains, and illegal chromosome abandons；

(5) the mutation probability Pm according to input, carries out gene section rearrangement to parent chromosome, same after the new chromosome of generation Sample needs to carry out legitimacy inspection；

(6) repeat this process T time, obtain the optimal solution of preliminary scheduling；

Constraint is adjusted to：

There is inserting single, delay or when material supply changes, need correspondingly to increase task, adjustment task priority letter Breath, processing technique information, operation resource information and inventory information, scheduling obtaining again meets the scheduling knot of current production status Really.

Present invention advantage compared with prior art is：

(1) the most of problem all np problems in scheduling field, current method is mainly based upon planning and exhaustive method, Small-scale scheduling problem can only be solved, and converge to that optimal solution is slower, also have larger problem apart from practical application.The present invention By the high-level plan scheduling technology based on genetic algorithm, using variation, Evolution of Population characteristic in genetic algorithm, solve potential Certainly gradually produce optimal solution plan in scheme population, solve the scheduling problem of Aviation Enterprise workshop.

(2) present invention is taking into full account various resource contents (equipment in as basis instrument module and equipment group, working day Go through, order, processing technique, operation resource, stock etc. in data management module) situation, and combine related parameter configuration, fortune With genetic algorithm, the production task automatic arrangement program to workshop, provide science for the daily production plan decision-making of workshop plan personnel Foundation.In the present invention, the high-level plan scheduling technology based on genetic algorithm, overcomes the local of the specific priority rule of conventional utilization Search or sequence form so-called " optimal case " shortcoming, and this optimal case is the non-overall situationization obtained under given conditions Best answer.And utilize genetic algorithm, by colony between interaction, keep the information that searched, this is to be based on The optimization method of single search procedure is incomparable, in addition, genetic algorithm use selection, intersect, make a variation these three calculation Son is all random operation, rather than specify really establish rules then.

Brief description

Fig. 1 is the composition frame chart of the present invention；

Fig. 2 is the flowchart of basis instrument module in Fig. 1；

Fig. 3 is the flowchart of data management module in Fig. 1；

Fig. 4 is the flowchart planning arranging module in Fig. 1；

Fig. 5 is the flowchart using genetic algorithm scheduling；

Fig. 6 is the convergence graph using genetic algorithm scheduling；

Fig. 7 is the output Gantt chart (equipment dimension) of specific embodiment.

Specific embodiment

As shown in figure 1, the present invention includes basis instrument module, data management module, plan scheduling module.

As shown in Fig. 2 being implemented as follows of basis instrument module in the present invention：

1) the material setting to workshop or plant produced processing, including COM code, title, Zhu Zhi production division, and Other attribute informations；

2) the information setting to personnel in system, include personnel encode, personnel's attribute, and be subordinate to the information such as department and set Put；

3) to the teams and groups' information setting included in production division, including teams and groups' coding, teams and groups' title, and the genus of teams and groups Property setting；

4) the work calendar of workshop or factory is arranged, include the specified of working day and nonworkdays, different departments can Different work calendars are set；

5) operating right setting is carried out to systematic difference personnel, including system administration authority, data manipulation authority, business Function privilege.

As shown in figure 3, being implemented as follows of data management module in the present invention：

1) order in input workshop or task, or these orders are obtained from ERP system or MES system by interface or appoints Business information；

2) technique information of input workshop process component, adds including the process information of processing, operation man-hour information, operation Work order metamessage, operation configuration information, and operation material requested information；

3) input operation resource information, including the frock needed for equipment and equipment group information, working procedure processing, tool information；

4) the production inventory information in input workshop or the stock's letter obtaining workshop by interface from ERP system or MES system Breath.

As shown in figure 4, planning being implemented as follows of arranging module in the present invention：

1) read the work calendar information of production division in basis instrument module, as shown in table 3 equipment group break time schedule；

2) read mission bit stream and task priority information in data management module, as shown in table 8；

3) the processing technique information in data management module, such as table 1 task-operation-equipment group table, table 2 tasks-work are read Sequence-time, table 5 operation raw material use shown in table and table 6 operations-raw material quantity table；

4) read the operation resource information in workshop in data management module, as contained by table 4 equipment group shown in device data table；

5) read the inventory information in workshop in data management module, as shown in table 7 materials procurement table；

6) automatic arrangement program is carried out by genetic algorithm, including generation initial population, calculating moderate function, selection operation, friendship Fork operation and five steps of mutation operation, as shown in Figure 5.First, it is that task is encoded, generate the first of scale N=50 at random Beginning population A；Second, calculate the Proper treatment of every chromosome in population A, using the shortest total elapsed time as evaluation criterion, will The result obtaining after the total elapsed time of every chromosome is inverted is as the appropriate value of this chromosome；3rd, contaminate for every The appropriateness value of colour solid gives different probability functions, selects 50 chromosomes by the way of roulette from population A, generates new Population A ＇；For the different select probability Ps of the appropriateness value imparting of every chromosome, (computing formula is as follows, wherein i generation Certain chromosome of table, its appropriate value is designated as fi), select N bar chromosome from initial population A by the way of roulette, generate New population A ＇；4th, the crossover probability with 0.6 (is adjusted according to actual production, crossover probability value in present case 0.6) parent chromosome in population A ＇, is made to be intersected two-by-two, the chromosome after intersection needs to carry out legitimacy inspection, legal Chromosome retain, illegal chromosome abandons；5th, with 0.01 as mutation probability (it is adjusted according to actual production, this Mutation probability value 0.01 in case), parent chromosome is carried out with gene section rearrangement, need also exist for after generating new chromosome into Row legitimacy is checked；6th, repeat this process 50 times, finally obtain one group of optimum scheduling result.

7) scheduling result exports Gantt chart from the dimension of device resource, can view every equipment in each period daily Plan arrangement, as shown in fig. 7, black block therein represents the unavailable time section of machine.Scheduling result is：

Chromosome (coding)=[3,4,4,1,2,5,2,3,3,3,5,1,6,4,1,2,2,4,1,6,6,3,2,5,1], T= 54.

Wherein numeral 1,2,3,4,5 and 6 represents different tasks, and task i ＆ lt goes out to represent i-th work of this task Sequence, decoded result is：

Chromosome (decoding)=[(J3,1), (J4,1), (J4,2), (J1,1), (J2,1), (J5,1), (J2,2), (J3, 2),(J3,3),(J3,4),(J5,2),(J1,2),(J6,1),(J4,3),(J1,3),(J2,3),(J2,4),(J4,4),(J6, 2), (J6,3), (J3,5), (J2,5), (J5,3), (J1,5)], T=54.

From scheduling convergence graph (as shown in Figure 6) as can be seen that during T=40, scheduling result restrains, that is, obtain optimal solution.

Task quantity is 6, operation maximum 5 needed for every single task, and equipment group number is 5, and unit of time is 1.

Table 1 task-operation-equipment group table

Table 2 tasks-operation-timetable

Table 3 equipment group break time schedule

Group_num	Start_time	End_time
			Q21-4(M1)	0	16
Q21-4(M1)	40	45
			R5(M2)	0	10
X51(M3)	35	40
			X51(M3)	80	200
X51(M3)	250	450
			X51(M3)	900	950
SCX-73(M4)	30	40

Number of devices scale contained by table 4 equipment group

Group_num	Number
		Q21-4(M1)	2
R5(M2)	2
		X51(M3)	2

SCX-73(M4)	1
		CM6140(M5)	1

Table 5 operations-raw material uses table

Table 6 operations-raw material quantity table

Table 7 materials procurement table

Material	Quantity	Arrival Time
			R28966	65	0
R22269	75	0
			R19910	70	0
R21883	80	0
			R28966	100	10
R21883	25	20
			R22269	30	35

Table 8 task priority list

Task_num	Priority
		LJ586962(1)	1

LJ586961(2)	1
		LJ586949(3)	2
LJ586998(4)	2
		LJ586943(5)	2
LJ586922(6)	2

Claims (2)

1. the high-level plan program system based on genetic algorithm, including basis instrument module, data management module and plan scheduling Module three part, wherein：

Basis instrument module：Carry out material essential information, personal information, equipment and equipment group information, the work calendar letter in workshop The input of breath and maintenance, setting；Operating right setting is carried out to the application personnel in system；This module be data management module and Plan scheduling module provides the support of Back ground Information and system operation information；

Data management module：Incoming task and its precedence information, processing technique information, operation resource information and inventory information, Wherein processing technique information includes the process information of part, operation man-hour information, working procedure processing unit information, operation setting letter Breath, and operation material requested information；Operation resource information include frock needed for equipment and equipment group information, working procedure processing, Tool information；

Plan scheduling module：Read the work calendar information in system-based setup module, read appointing in data management module Business and its precedence information, processing technique information, operation resource information and inventory information；Input schduling parameter N, T, Pc, Pm, N Represent Population Size, T represents iterationses, Pc represents crossover probability, Pm represents mutation probability, call Scheduling Algorithm, output row Journey result, i.e. the Gantt chart of device resource dimension and task dimension；

Plan scheduling module implements process：

(1) input schduling parameter N, T, Pc, Pm, read the work calendar information in basis instrument module, read data management mould Task in block and its precedence information, processing technique information, operation resource information and inventory information, are task encoding, at random Generation scale is initial population A of N；

(2) calculate the Proper treatment of every chromosome in initial population A, using the shortest total elapsed time as evaluation criterion, will be every The result obtaining after the total elapsed time of bar chromosome is inverted is as the appropriate value of this chromosome；

(3) the appropriateness value being directed to every chromosome gives different select probability Ps, and computing formula is as follows, and wherein i represents Certain chromosome, its appropriate value is designated as fi, selects N bar chromosome, generate new by the way of roulette from initial population A Population A ＇；

$Ps=f_i/\underset{i=1}{\overset{N}{\Σ}}{f}_i$

(4) the crossover probability Pc according to input, makes the parent chromosome in population A ＇ be intersected two-by-two, the chromosome after intersection Need to carry out legitimacy inspection, legal chromosome retains, and illegal chromosome abandons；

(5) the mutation probability Pm according to input, carries out gene section rearrangement to parent chromosome, equally needs after generating new chromosome Carry out legitimacy inspection；

(6) repeat this process T time, obtain the optimal solution of preliminary scheduling；

Constraint is adjusted to：

There is inserting single, delay or when material supply changes, need correspondingly to increase task, adjustment task priority information, Processing technique information, operation resource information and inventory information, scheduling obtain the scheduling result meeting current production status again.

2. the high-level plan scheduling method based on genetic algorithm it is characterised in that：Including preliminary scheduling and constraint adjustment two parts, Wherein：Preliminary scheduling is：

(1) input schduling parameter N, T, Pc, Pm, read the work calendar information in basis instrument module, read data management mould Task in block and its precedence information, processing technique information, operation resource information and inventory information, are task encoding, at random Generation scale is initial population A of N；

(2) calculate the Proper treatment of every chromosome in initial population A, using the shortest total elapsed time as evaluation criterion, will be every The result obtaining after the total elapsed time of bar chromosome is inverted is as the appropriate value of this chromosome；

(3) the appropriateness value being directed to every chromosome gives different select probability Ps, and computing formula is as follows, and wherein i represents Certain chromosome, its appropriate value is designated as fi, selects N bar chromosome, generate new by the way of roulette from initial population A Population A ＇；

$Ps=f_i/\underset{i=1}{\overset{N}{\Σ}}{f}_i$

(4) the crossover probability Pc according to input, makes the parent chromosome in population A ＇ be intersected two-by-two, the chromosome after intersection Need to carry out legitimacy inspection, legal chromosome retains, and illegal chromosome abandons；

(5) the mutation probability Pm according to input, carries out gene section rearrangement to parent chromosome, equally needs after generating new chromosome Carry out legitimacy inspection；

(6) repeat this process T time, obtain the optimal solution of preliminary scheduling；

Constraint is adjusted to：

There is inserting single, delay or when material supply changes, need correspondingly to increase task, adjustment task priority information, Processing technique information, operation resource information and inventory information, scheduling obtain the scheduling result meeting current production status again.