CN115562198B - Workshop rescheduling method considering rescheduling execution cost - Google Patents
Workshop rescheduling method considering rescheduling execution cost Download PDFInfo
- Publication number
- CN115562198B CN115562198B CN202211201305.XA CN202211201305A CN115562198B CN 115562198 B CN115562198 B CN 115562198B CN 202211201305 A CN202211201305 A CN 202211201305A CN 115562198 B CN115562198 B CN 115562198B
- Authority
- CN
- China
- Prior art keywords
- rescheduling
- population
- time
- child
- workpiece
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 104
- 230000008569 process Effects 0.000 claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 36
- 230000002068 genetic effect Effects 0.000 claims abstract description 11
- 238000005457 optimization Methods 0.000 claims abstract description 9
- 210000000349 chromosome Anatomy 0.000 claims description 57
- 108090000623 proteins and genes Proteins 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 claims description 5
- 238000003754 machining Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 abstract description 3
- 238000011160 research Methods 0.000 abstract 1
- 238000004088 simulation Methods 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 9
- 230000002028 premature Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32339—Object oriented modeling, design, analysis, implementation, simulation language
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Manufacturing & Machinery (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Automation & Control Theory (AREA)
- Evolutionary Computation (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Physiology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- General Factory Administration (AREA)
Abstract
The invention provides a workshop rescheduling method considering rescheduling execution cost, which starts from two layers of a workpiece and equipment, integrates the processing time deviation cost of the same working procedure before and after rescheduling, the processing time deviation cost of the waiting working procedure of the adjacent working procedure, the equipment working time deviation cost and the processing time deviation cost of the waiting working procedure of the equipment, establishes a pre-scheduling loss function to calculate the loss in the rescheduling execution process, and provides a multi-objective optimization scheme based on the execution cost and the maximum finishing time, thereby solving the problem that the loss cost is neglected in many multi-scheduling researches at present. Then, an experiment is carried out on the simulation case by utilizing an improved genetic algorithm, the effectiveness of the invention is verified, a job shop rescheduling system is developed based on Matlab GUI, the probability of generating a pseudo-optimal scheduling scheme is greatly reduced, and the economic benefit of enterprises is ensured.
Description
Technical Field
The invention relates to the field of workshop scheduling, in particular to a workshop rescheduling method considering rescheduling execution cost.
Background
The Job Shop scheduling problem of Job Shop is a classical NP-hard problem. When the production scale is small and the production environment is simple, the optimal solution can be conveniently found through some algorithms. However, when the production scale is enlarged or there is an uncertainty in the production environment, the number of combinations of processes is increased explosively, resulting in that the general optimization method is not applicable any more. Therefore, in a practical dynamic production manufacturing shop, the influence of uncertainty factors on the scheduling quality is a difficulty to be solved in searching for an optimal rescheduling method.
In a Job Shop in an uncertain environment, there are many uncertain factors (e.g., changes in order demand, changes in processing resources, changes in equipment status) that can affect the quality and stability of the scheduling scheme. In order to timely adjust the workshop resource allocation when disturbance occurs, the implementation of rescheduling becomes important, and the scheduling scheme is ensured to be suitable for the current working condition. Rescheduling, however, is an adjustment and modification to the original schedule, so it must incur cost consumption, such as reallocation of manufacturing resources. Many studies ignore the implementation costs of rescheduling, which can seriously affect the effectiveness of rescheduling schemes, resulting in their inability to effectively cope with dynamic disturbances and no longer adapt to current operating conditions.
Considering the loss cost of rescheduling, the rescheduling scheme can be started at the optimal time, the stability of the rescheduling scheme and the coupling with the dynamic production environment can be ensured, and the probability of generating the pseudo-optimal scheduling scheme is reduced. Therefore, the invention provides dynamic scheduling considering the rescheduling execution cost, and develops a system platform in a matalb environment based on a genetic algorithm.
Patent application number 202011417993.4 proposes a dynamic scheduling method for solving a machine fault in a flexible job shop based on an empire competition algorithm. The rescheduling strategy is to reschedule the unprocessed working procedure and the processing working procedure on the failed machine after the machine fails, and ensure that the maximum finishing time of rescheduling and the delay time of initial scheduling are minimum. This patent only targets rescheduling for maximum completion time, without taking into account the lost costs of rescheduling. Time variances for each process during the entire process can create lost costs due to the delivery requirements of the work pieces, maintenance costs of the machine, etc., which should also be taken into account in the rescheduling scheme.
There is no clear and complete quantization method for the scheduling cost after rescheduling is implemented. Therefore, the invention focuses on the aspects of workpieces and equipment, considers the offset of the processing starting time of the same process before and after rescheduling, the offset of the waiting processing time of the adjacent process of the workpieces, the offset of the total processing time of the machine, the offset of the idle time of the machine and discusses the influence of rescheduling cost on a scheduling scheme.
Disclosure of Invention
Aiming at the defects related to the background technology, the invention provides a workshop rescheduling method considering rescheduling execution cost.
The invention adopts the following technical scheme for solving the technical problems:
a shop rescheduling method considering rescheduling execution cost, comprising the steps of:
step 1), setting parameters, namely setting related parameters of a genetic algorithm for solving JSP rescheduling problem considering rescheduling execution cost, wherein the related parameters comprise population scale popSize, evolution algebra changes and cross probability p c Probability of variation p m The current iteration number j, where j is initialized to 1, 0.9.ltoreq.p c ≤0.97，0.001≤p m ≤0.02；
Step 2), initializing a population according to the popSize set in the step 1, and obtaining an initial population with popSize chromosomes after the population is initialized, wherein each chromosome represents a sequence of procedures;
step 3), calculating fitness values of all chromosomes in the initial population generated in the step 2;
step 4), selecting the initial population with the calculated fitness value in the step 3 by adopting a roulette method, selecting (popSize-1)/2 chromosomes as Parent population, namely Parent population, and performing subsequent steps;
step 5), crossing chromosomes in the Parent population selected in the step 4 to increase population diversity, and generating a child population 1, namely child_1;
step 6), finishing mutation operation on the chromosome in the population child_1 generated in the step 5 by adopting an interchange method to generate a child population 2, namely child_2;
step 7), recording the optimal chromosome and the fitness value thereof in the population Children_2;
step 8), judging whether j is larger than the projects, if j is larger than the projects, drawing an iteration process diagram and a workpiece process Gantt chart according to the data recorded in the step 7, and ending the algorithm; if j is less than or equal to the operations, j+1 is then transferred to step 3;
step 9), if the opportunity device for executing the scheduling scheme fails, setting a failure machine, failure time and repair time according to working conditions, adopting local correction type rescheduling, finding out an unprocessed procedure after a failure point, and if no failure occurs, continuing to execute the scheduling scheme;
step 10), rescheduling the unprocessed working procedure after the fault point and the working procedure in process on the fault machine from the step 2 to generate a rescheduled new population.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
1. the invention provides an optimization index based on rescheduling execution cost by describing the loss cost of two layers of workpieces and equipment generated by rescheduling and comprehensively summarizing the cost; the rescheduling execution cost is researched and quantized, and is brought into the optimization target of a genetic algorithm, a multi-target optimization scheme is provided, and the maximum finishing time and the execution cost are comprehensively considered, so that the loss caused by the rescheduling on the resource reallocation is reduced, and the production cost is reduced.
2. The invention enlarges the population quantity to increase population diversity during initialization, uses POX crossover operator to complete crossover operation, and the offspring generated by the method can inherit excellent characteristics of the parent well and is always feasible. The invention solves the problems of premature convergence, poor solution stability and difficult determination of genetic parameters in the process of solving the scheduling problem of the traditional job shop by using the traditional genetic algorithm to a certain extent.
Drawings
FIG. 1 is a path diagram of an overall implementation of the present invention;
FIG. 2 is a flow chart of a modified genetic algorithm;
FIG. 3 is a schematic illustration of roulette;
FIG. 4 is a process information interface of the development system;
FIG. 5 is a parameter setting interface of the system;
FIG. 6 is a Gantt chart of an original scheduling scheme;
FIG. 7 is a Gantt chart of rescheduling scheme 1;
fig. 8 is a gante diagram of rescheduling scheme 2.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings:
the invention provides a workshop rescheduling method considering rescheduling execution cost, which comprises the following steps:
step 1), setting parameters, namely setting related parameters of a genetic algorithm for solving JSP rescheduling problem considering rescheduling execution cost, wherein the related parameters comprise population scale popSize, evolution algebra changes and cross probability p c Probability of variation p m The current iteration number j, where j is initialized to 1, 0.9.ltoreq.p c ≤0.97，0.001≤p m ≤0.02；
Step 2), initializing a population according to the popSize set in the step 1, and obtaining an initial population with popSize chromosomes after the population is initialized, wherein each chromosome represents a sequence of procedures;
the specific steps of the step 2 are as follows:
(1) expanding the population scale popSize in the step 1, specifically popsize=2×popsize+1, so as to enrich population diversity;
(2) randomly generating a chromosome population comprising n×m genes, the chromosomes being based on process codes, the occurrence of the kth time from left to right of the same workpiece number representing the kth process for that workpiece, such as chromosome [3 3 21 21 ], wherein the first 3 represents the first process for workpiece number 3, the second 3 represents the second process for workpiece number 3, and so on;
step 3), calculating fitness values of all chromosomes in the initial population generated in the step 2; the fitness value of the chromosome is used for distinguishing the quality degree among the chromosomes, and the better the fitness value of the chromosome is, the larger the probability of being selected is. The calculation of fitness is mainly to calculate the maximum completion time of each chromosome, but if the initial population is a newly generated population by rescheduling, additional calculation of rescheduling execution cost of each chromosome is required.
The step 3 is specifically implemented according to the following steps:
(1) calculating fitness function values f (i) of all chromosomes in the initial population: i is any chromosome in the initial population, and the value range is [1, popsize ]. f (i) represents the fitness function of chromosome i, specifically as shown in formula (1):
f(i)＝1/C(i) (1)
in the formula (1), C (i) is the maximum completion time of the chromosome i, f (i) is the inverse of the maximum completion time, and the larger the value of f (i) is, the better the scheduling scheme is;
(2) when the initial population is newly generated by rescheduling, the fitness function is shown in formula (2):
f(i)＝1/[C(i)+C'(i)] (2)
where C '(i) represents i's rescheduling execution cost, calculated from the prescheduling loss function, as shown in equation (3),
C′＝ω 1 ·(C DT +C DW )+ω 2 ·(C DO +C DI ) (3)
ω 1 ,ω 2 and the weight coefficient of the workpiece and the equipment is represented. C (C) DT 、C DW 、C DO 、C DI The shift of the start processing time of the same process, the shift of the waiting processing time of the adjacent process of the workpiece, the shift of the total processing time of the machine and the shift of the idle time of the machine are respectively shown. The description thereof is as follows.
● The same procedure starts with a process time shift: workpiece J j Step J of (2) ij The start machining time deviation before and after rescheduling is shown in formula (4):
DT ij ＝T' ij,m -T ij,m (4)
in the above, DT ij Not less than 0, if DT ij < 0, let DT ij ＝0。
The total start machining time deviation cost for all the workpieces is shown in equation (5):
● Adjacent processes of the workpiece wait for a processing time offset: in the pre-dispatching process, the workpiece J j Step J of (2) ij The +.bias of waiting for processing is shown in equation (6)
DW ij ＝W' ij -W ij (6)
DW ij Not less than 0, if DW ij < 0, let DW ij =0. Wherein W is ij ＝T ij,m -T (i-1)j,(m-1) -P (i-1)j,(m-1) 。
When i=1, W ij ＝W' ij =0, i.e. the first pass of the workpiece is considered to be without waiting.
Therefore, the waiting process time deviation costs of all the processes are shown in the formula (7):
● Machine total process time offset: the total operating time of the machine may change if and only if the machine is working at the time of failure, scheduling may produce an operating time bias cost. Assume that the fault time is t wr The failure machine is M q The last working procedure of the fault machine processing before the fault moment is J kp Namely, the workpiece J p Wherein p=1, 2, …, N, k=1, 2, … N p 。
The total operating time deviation cost of the equipment is shown in the formula (8):
● Machine idle time offset: device M m The total idle time in the original schedule and pre-schedule schemes is shown in equations (9) and (10), respectively:
wherein k is m Representing device M m Is used for the total processing steps.
Thus apparatus M m The total idle deviation of (2) is shown in formula (11):
DI m ＝I' m -I m (11)
the total cost deviation formula is shown as formula (12), wherein mu m The calculation formula is shown in formula (13);
step 4), selecting the initial population with the calculated fitness value in the step 3 by adopting a roulette method, selecting (popSize-1)/2 chromosomes as Parent population, namely Parent population, and performing subsequent steps;
step 5), crossing chromosomes in the Parent population selected in the step 4 to increase population diversity, and generating a child population 1, namely child_1;
the step 5 is specifically implemented according to the following steps:
(1) selecting a chromosome in sequence in the population Parent as a Parent1, namely Parent1, and selecting a chromosome with a corresponding sequence number in the Parent population as a Parent2, namely Parent2 according to random integers generated between (1, (popSize-1)/2) by a random function;
(2) if the crossover probability value is greater than the value randomly generated by the random function between (0, 1), the crossover operation is performed using the POX crossover operator:
● Workpiece sets {1,2,3,.. N } were randomly divided into two non-empty subsets J1 and J2
● The workpiece numbers containing J1 in Parent1 and Parent2 are respectively copied into child 1, namely child 1, and child 2, namely child 2 according to the positions of the workpiece numbers in the chromosomes, the workpiece numbers containing J2 in Parent1 and Parent2 are respectively copied into child 2 and child 1 according to the sequences of the workpiece numbers in the chromosomes, and the child 1 and child 2 are the crossed chromosomes.
For example, parent 1= {3,2,2,3,1,1}, parent 2= {1,1,3,2,2,3}, a workpiece set {1,2,3} is randomly divided to generate J1= {2}, J2= {3,1}, and Children 1= {1,2,2,1,3,3}, children 2= {3,3,1,2,2,1}, which are obtained after POX crossing.
(3) If the cross probability value is less than or equal to the value randomly generated by the random function between (0, 1), the cross operation is not performed, and Parent1 and Parent2 are directly used as Children1 and Children2;
(4) children1 and Children2 are added to Children_1.
Step 6), finishing mutation operation on the chromosome in the child_1 population generated in the step 5 by adopting an interchange method to generate child population child_2;
the specific implementation of the step 6 is as follows:
(1) selecting a chromosome Children1 in the population Children_1 according to the sequence of the chromosomes;
(2) if the variation probability value is larger than the value randomly generated between (0, 1) by the rand function, performing variation operation by adopting a gene exchange method:
● Randomly generating two gene numbers pos1 and pos2 within the chromosome length;
● Exchanging genes on pos1 and pos2 in chromosome child 1, repeating 4 times;
● The exchanged child 1 is added to child_2.
(3) If the variation probability value is less than or equal to the value randomly generated by the rand function between (0, 1), the variation operation is not performed, and Children1 is directly added to Children_2.
Step 7), recording optimal chromosomes and fitness values thereof in the population Children_2, wherein the specific steps are the same as the step 3;
step 8), judging whether j is larger than the projects, if j is larger than the projects, drawing an iteration process diagram and a workpiece process Gantt chart according to the data recorded in the step 7, and ending the algorithm; otherwise, j+1 is switched to step 3;
step 9), if the opportunity device for executing the scheduling scheme fails, setting a failure machine, failure time and repair time according to working conditions, adopting local correction type rescheduling, finding out an unprocessed procedure after a failure point, and if no failure occurs, continuing to execute the scheduling scheme;
step 10), rescheduling the unprocessed working procedure after the fault point and the working procedure in process on the fault machine from the step 2 to generate a rescheduled new population.
The specific implementation of step 10 is as follows:
● Recording each procedure J in the original static scheduling scheme ij Is a start processing time of (1);
● If step J ij If the start processing time of (1) is greater than or equal to the failure time of the equipment, then J ij Adding an unfinished procedure set unfin_lists;
● Finding out whether a faulty machine is being processed at the faulty moment, and if the faulty machine is being processed, adding the processing procedure into unfin_lists;
● The process in unfin_lists is rearranged, placed after the processed process, and a new population is constructed. The processed portions of the chromosomes in the new population are all identical and the portions that need to be rearranged are different.
● For example chromosome 3,2,2,3,1,1, where J 13 ，J 12 ，J 11 The start processing time in the original schedule is less than the failure time. Then only the J is rearranged 23 ，J 22 ，J 21 The new population is {3,2,1, x, y, z }. { x, y, z } is J 23 ， J 22 ，J 21 Is a random arrangement of (a) to (b).
● The method of the invention performs experiments on FT06 calculation examples on Matlab 2017. Firstly, specific processing information is read from an excel table edited according to working conditions and displayed on a system interface, as shown in fig. 4. Then setting the iteration times 100, the population scale 100, the crossover probability 0.9 and the variation probability 0.05 according to experience in a parameter editing interface, as shown in fig. 5.
And when no fault occurs, performing static scheduling, namely original scheduling, wherein a Gantt chart of a scheduling result is shown in fig. 6. When equipment fails, setting the failed equipment as 2, the failure time as 3, the repair time as 10, and taking the maximum finishing time as an optimization target, wherein a rescheduling scheme Gantt chart is shown in fig. 7, and a multi-target optimization scheme Gantt chart provided by the invention is shown in fig. 8.
To verify the advantages of the present invention over the current state, the costs of the two rescheduling schemes (rescheduling scheme 1 and rescheduling scheme 2) of fig. 7 and 8 are calculated, wherein ω 1 ＝0.7，ω 2 The results of =0.3 are shown in table 1:
table 1 results comparison table
| Rescheduling scheme 1 | Rescheduling scheme 2 | |
| C DT | 129 | 106 |
| C DW | 62 | 49 |
| C DO | 5 | 5 |
| C DI | 29.192 | 37.305 |
| C | 65 | 67 |
| C' | 143.9588 | 121.1903 |
| sumC | 208.9588 | 188.1903 |
As can be seen from table 1, the implementation costs incurred when the system employs different rescheduling schemes can vary greatly. The maximum time to completion (c=67) for rescheduling scheme 2 is slightly greater than the maximum time to completion (c=65) for rescheduling scheme 1, but from an implementation cost perspective, the cost of loss (C '= 121.1903) for rescheduling scheme 2 is much less than the cost of loss (C' = 143.9588) for rescheduling scheme 1, and from the sum of the maximum time to completion and the cost of loss, scheduling scheme 2 (sumc= 188.1903) is much less than scheduling scheme 1 (sumc= 208.9588).
In order to embody the effectiveness of the algorithm, the convergence process of three schemes is recorded, the original scheme and the rescheduling scheme 1 are iterated for about 50 times, the convergence speed is high, the rescheduling scheme 2 converges for about 80 times, the convergence speed is low, the algorithm can be improved subsequently, and the convergence speed is increased.
The invention enlarges the population size when initializing the population, increases the diversity of individuals and the competitiveness among individuals, calculates the individual fitness value by adopting a new fitness function, increases the degree of distinction among individuals, solves the problems of difficult parameter determination, premature convergence and unstable solution in the traditional genetic algorithm by adopting the self-adaptive crossover and variation probability, improves the optimizing capability and convergence speed of the algorithm, and tests the performance of the invention by using a reference case, thereby proving that the method is applicable.
The invention provides a workshop rescheduling method considering rescheduling execution cost, which integrates rescheduling execution cost of two layers of workpieces and equipment, provides a multi-objective optimization scheme comprehensively considering maximum finishing time and execution cost, and uses an optimized genetic algorithm to experiment FT06 cases, so that the effectiveness of the invention is proved.
Claims (6)
1. A shop rescheduling method considering rescheduling execution cost, comprising the steps of:
step 1), setting parameters, namely setting related parameters of JSP rescheduling problem considering rescheduling execution cost by a genetic algorithm, wherein the related parameters comprise population scale popSize, initial processing sequence scale which is randomly generated, evolution algebra changes, the number of processing sequence optimization iteration, and cross probability p c The probability of the exchange of two processing sequences and the variation probability p are shown m The probability of exchanging two procedures in a certain processing sequence is represented, and the current iteration number j represents the current iteration number of the processing sequence;
step 2), initializing a population according to the popSize set in the step 1), and obtaining an initial population with popSize chromosomes after the population is initialized, wherein each chromosome represents a sequence of procedures;
step 3), calculating fitness values of all chromosomes in the initial population generated in the step 2);
step 4), selecting the initial population with the fitness value calculated in the step 3) by adopting a roulette method, selecting (popSize-1)/2 chromosomes as Parent population, namely Parent population, and performing subsequent steps;
step 5), crossing chromosomes in the Parent population selected in the step 4) to increase population diversity and generate child population 1, namely child_1;
step 6), finishing mutation operation on the chromosome in the population child_1 generated in the step 5) by adopting an interchange method to generate a child population 2, namely child_2;
step 7), recording the optimal chromosome and the fitness value thereof in the population Children_2;
step 8), judging whether the previous iteration times j is larger than the evolution algebra changes, if j is larger than the changes, drawing an iteration process diagram and a workpiece procedure Gantt chart according to the data recorded in the step 7), and ending the algorithm; if j is less than or equal to the Items, j+1 is then transferred to step 3);
step 9), if the opportunity device for executing the scheduling scheme fails, setting a failure machine, failure time and repair time according to working conditions, adopting local correction type rescheduling, finding out an unprocessed procedure after a failure point, and if no failure occurs, continuing to execute the scheduling scheme; the local modified rescheduling is determined by:
the same procedure starts with a process time shift: workpiece J j Step J of (2) ij The start machining time deviation before and after rescheduling is shown in formula (4):
DT ij ＝T′ ij,m -T ij,m (4)
in the above, DT ij Not less than 0, if DT ij < 0, let DT ij ＝0；
The total start machining time deviation cost for all the workpieces is shown in equation (5):
adjacent processes of the workpiece wait for a processing time offset: in the pre-dispatching process, the workpiece J j Step J of (2) ij The time deviation of waiting for processing is as shown in formula (6) Shown is
DW ij ＝W′ ij -W ij (6)
DW ij Not less than 0, if DW ij < 0, then set DW ij =0; wherein W is ij ＝T ij,m -T (i-1)j,(m-1) -P (i-1)j,(m-1) ；
When i=1, W ij ＝W′ ij =0, i.e. consider that the first process of the workpiece does not need to wait;
therefore, the waiting process time deviation costs of all the processes are shown in the formula (7):
machine total process time offset: if and only if a faulty machine is processing at the moment of failure, the total working time of the machine will change, and scheduling will produce a working time deviation cost; assume that the fault time is t wr The failure machine is M q The last working procedure of the fault machine processing before the fault moment is J kp Namely, the workpiece J p Wherein p=1, 2, …, N, k=1, 2, … N p ；
The total operating time deviation cost of the equipment is shown in the formula (8):
machine idle time offset: device M m The total idle time in the original schedule and pre-schedule schemes is shown in equations (9) and (10), respectively:
wherein k is m Representing device M m Is a total number of processing steps;
thus apparatus M m The total idle deviation of (2) is shown in formula (11):
DI m ＝I′ m -I m (11)
the total cost deviation formula is shown as formula (12), wherein mu m The calculation formula is shown in formula (13);
step 10), rescheduling the unprocessed working procedure after the fault point and the working procedure in process on the fault machine from the step 2) to generate a rescheduled new population.
2. The shop rescheduling method considering the rescheduling execution cost according to claim 1, wherein the step 2) specifically comprises the steps of:
step 2.1, expanding the population scale popSize in the step 1);
step 2.2, randomly generating a chromosome population comprising n×m genes, the chromosomes being encoded based on the process, and from left to right being the kth process of the workpiece.
3. The shop rescheduling method considering the rescheduling execution cost according to claim 1, wherein the step 3) specifically comprises the following steps:
step 3.1, calculating fitness function values f (i) of all chromosomes in the initial population: i is any chromosome in the initial population, the value range is [1, popsize ], and f (i) represents the fitness function of the chromosome i, and the fitness function is specifically shown as a formula (1):
f(i)＝1/C(i) (1)
in the formula (1), C (i) is the maximum completion time of the chromosome i, and f (i) is the reciprocal of the maximum completion time;
step 3.2, when the initial population is newly generated by rescheduling, the fitness function is shown in formula (2):
f(i)＝1/[C(i)+C′(i)] (2)
where C '(i) represents i's rescheduling execution cost, calculated from the prescheduling loss function, as shown in equation (3),
C′＝ω 1 ·(C DT +C DW )+ω 2 ·(C DO +C DI ) (3)
ω 1 ,ω 2 representing the weight coefficient of the workpiece and the equipment, C DT 、C DW 、C DO 、C DI The shift of the start processing time of the same process, the shift of the waiting processing time of the adjacent process of the workpiece, the shift of the total processing time of the machine and the shift of the idle time of the machine are respectively shown.
4. The shop rescheduling method taking the rescheduling execution cost into consideration as set forth in claim 1, wherein the step 5) is specifically implemented as follows:
step 5.1, selecting a chromosome in sequence in the population Parent as a Parent1, namely Parent1, and selecting a chromosome with a corresponding serial number in the Parent population as a Parent2, namely Parent2 according to random integers generated between (1, (popSize-1)/2) according to a random function;
step 5.2, if the crossover probability value is greater than the value randomly generated by the random function between (0, 1), performing crossover operation by using the POX crossover operator:
randomly dividing the workpiece set {1,2, 3..n } into two non-empty subsets J1 and J2
Copying the workpiece numbers containing J1 in the Parent1 and the Parent2 into a child 1, namely child 1, and a child 2, namely child 2 according to the positions of the workpiece numbers in the chromosomes, and copying the workpiece numbers containing J2 in the Parent1 and the Parent2 into the child 2 and the child 1 according to the sequences of the workpiece numbers in the chromosomes, wherein the child 1 and the child 2 are crossed chromosomes;
step 5.3, if the cross probability value is less than or equal to the value randomly generated by the random function between (0, 1), performing no cross operation, and directly using Parent1 and Parent2 as Children1 and Children2;
step 5.4, children1 and Children2 are added to Children_1.
5. The shop rescheduling method considering the rescheduling execution cost according to claim 1, wherein the implementation of step 6) is as follows:
step 6.1, selecting a chromosome child 1 in the population child_1 according to the sequence of the chromosomes;
and 6.2, if the variation probability value is larger than the value randomly generated by the rand function between (0 and 1), performing variation operation by adopting a gene exchange method, and if the variation probability value is smaller than or equal to the value randomly generated by the rand function between (0 and 1), directly adding the Children1 into the Children_2 without performing variation operation.
6. The shop rescheduling method considering the rescheduling execution cost according to claim 1, wherein the implementation of step 10) is as follows:
step 10.1, recording each procedure J in the original static scheduling scheme ij If the processing time of step J ij If the start processing time of (1) is greater than or equal to the failure time of the equipment, then J ij Adding an unfinished procedure set unfin_lists;
step 10.2, finding out whether a faulty machine is being processed at the moment of fault, if so, adding the processing procedure into unfin_lists;
step 10.3, rearranging the procedures in unfin_lists, placing the procedures after the processed procedures, and constructing a new population.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211201305.XA CN115562198B (en) | 2022-09-29 | 2022-09-29 | Workshop rescheduling method considering rescheduling execution cost |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211201305.XA CN115562198B (en) | 2022-09-29 | 2022-09-29 | Workshop rescheduling method considering rescheduling execution cost |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115562198A CN115562198A (en) | 2023-01-03 |
| CN115562198B true CN115562198B (en) | 2023-09-29 |
Family
ID=84742593
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211201305.XA Active CN115562198B (en) | 2022-09-29 | 2022-09-29 | Workshop rescheduling method considering rescheduling execution cost |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115562198B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504540A (en) * | 2015-01-13 | 2015-04-08 | 郑州航空工业管理学院 | Method for dynamic flexible job workshop scheduling control based on multistage intelligent optimization algorithm |
| CN110705844A (en) * | 2019-09-16 | 2020-01-17 | 新疆大学 | Robust optimization method of job shop scheduling scheme based on non-forced idle time |
| CN111738578A (en) * | 2020-06-11 | 2020-10-02 | 电子科技大学 | Discrete type workshop scheduling method under dynamic environment |
| WO2022000924A1 (en) * | 2020-07-01 | 2022-01-06 | 北京工业大学 | Double-resource die job shop scheduling optimization method based on ammas-ga nested algorithm |
| CN114926033A (en) * | 2022-05-24 | 2022-08-19 | 燕山大学 | Flexible job shop dynamic event scheduling method based on improved NSGAII |
-
2022
- 2022-09-29 CN CN202211201305.XA patent/CN115562198B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504540A (en) * | 2015-01-13 | 2015-04-08 | 郑州航空工业管理学院 | Method for dynamic flexible job workshop scheduling control based on multistage intelligent optimization algorithm |
| CN110705844A (en) * | 2019-09-16 | 2020-01-17 | 新疆大学 | Robust optimization method of job shop scheduling scheme based on non-forced idle time |
| CN111738578A (en) * | 2020-06-11 | 2020-10-02 | 电子科技大学 | Discrete type workshop scheduling method under dynamic environment |
| WO2022000924A1 (en) * | 2020-07-01 | 2022-01-06 | 北京工业大学 | Double-resource die job shop scheduling optimization method based on ammas-ga nested algorithm |
| CN114926033A (en) * | 2022-05-24 | 2022-08-19 | 燕山大学 | Flexible job shop dynamic event scheduling method based on improved NSGAII |
Non-Patent Citations (1)
| Title |
|---|
| 设备故障条件下柔性作业车间重调度方法;包博等;计算机工程与应用(第19期);248-253 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115562198A (en) | 2023-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zandieh et al. | A robust genetic algorithm for scheduling realistic hybrid flexible flow line problems | |
| CN113610233B (en) | Flexible job shop scheduling method based on improved genetic algorithm | |
| CN110276481B (en) | Distributed hybrid pipeline scheduling optimization method | |
| CN111667071B (en) | Traditional job shop scheduling method based on improved genetic algorithm | |
| CN108460463B (en) | High-end equipment assembly line production scheduling method based on improved genetic algorithm | |
| CN110796355A (en) | Flexible job shop scheduling method based on dynamic decoding mechanism | |
| Bozorgirad et al. | A comparison of local search algorithms with population-based algorithms in hybrid flow shop scheduling problems with realistic characteristics | |
| Pan et al. | A novel differential evolution algorithm for no-idle permutation flow-shop scheduling problems | |
| CN110245842B (en) | Production line risk scheduling method for sudden large faults of equipment | |
| CN110378583B (en) | Method for interchanging adjacent procedures of pseudo-critical path and equipment | |
| CN111382915A (en) | Flexible job shop scheduling method for co-fusion AGV | |
| CN114493337A (en) | Flexible job shop scheduling method based on improved particle swarm genetic hybrid algorithm | |
| CN114841581A (en) | Feature selection method in dynamic job shop scheduling rule based on GEP-VNS evolution | |
| CN114519455A (en) | Distributed flexible workshop scheduling method and system with transportation process | |
| CN111340345A (en) | Cutter scheduling method based on improved particle swarm optimization | |
| CN112286152A (en) | Distributed flow shop scheduling method and system with batch delivery constraint | |
| CN112990716A (en) | Dual-resource constraint flexible workshop scheduling and layout integrated optimization method and system | |
| CN117852825B (en) | Deadlock-free scheduling method of flexible manufacturing system containing central resources based on deep learning | |
| CN115562198B (en) | Workshop rescheduling method considering rescheduling execution cost | |
| Wong et al. | Optimization of manual fabric-cutting process in apparel manufacture using genetic algorithms | |
| CN117077975A (en) | Distributed heterogeneous flow shop scheduling method based on mixed initialization modular factor algorithm | |
| CN110705844B (en) | Robust optimization method for job shop scheduling scheme based on non-forced idle time | |
| CN114415615B (en) | Mixed flow assembly line balanced distribution method and device under uncertain demand | |
| CN114676987B (en) | Intelligent flexible job shop active scheduling method based on hyper-heuristic algorithm | |
| Nearchou | Flow-shop sequencing using hybrid simulated annealing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |