CN113361813A - Optimized scheduling method for scheduling system of wafer equipment - Google Patents

Optimized scheduling method for scheduling system of wafer equipment Download PDF

Info

Publication number
CN113361813A
CN113361813A CN202110752785.8A CN202110752785A CN113361813A CN 113361813 A CN113361813 A CN 113361813A CN 202110752785 A CN202110752785 A CN 202110752785A CN 113361813 A CN113361813 A CN 113361813A
Authority
CN
China
Prior art keywords
wafer
population
production
minimum
individual
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.)
Pending
Application number
CN202110752785.8A
Other languages
Chinese (zh)
Inventor
肖攸安
单海成
邓胳峰
邹鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University of Technology WUT
Original Assignee
Wuhan University of Technology WUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuhan University of Technology WUT filed Critical Wuhan University of Technology WUT
Priority to CN202110752785.8A priority Critical patent/CN113361813A/en
Publication of CN113361813A publication Critical patent/CN113361813A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/22—Design optimisation, verification or simulation using Petri net models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06316—Sequencing of tasks or work
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G06Q10/08355—Routing methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Evolutionary Computation (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Biophysics (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Physiology (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Genetics & Genomics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Educational Administration (AREA)

Abstract

The invention discloses an optimized scheduling method for a scheduling system of wafer equipment, which comprises the following steps: coding production elements in the wafer production equipment, acquiring coding information of the production elements in a single wafer box processing path, and setting the coding information as a single chromosome coding model; constructing a Petri network structure, wherein the Petri network takes a chromosome coding model as an input parameter and takes the completion time corresponding to the corresponding chromosome coding model as an output parameter; establishing an initial population based on a chromosome coding model, carrying out iterative update on the initial population, replacing the worst individual of the initial population with an elite individual in the updated population, updating a Petri network, and obtaining the minimum and maximum completion time from the updated population; and extracting coding information from the result and scheduling the wafer production path.

Description

Optimized scheduling method for scheduling system of wafer equipment
Technical Field
The invention belongs to the field of intelligent production scheduling of semiconductor equipment, and particularly relates to an optimized scheduling method for a production scheduling system of wafer equipment.
Background
Semiconductor manufacturing is an important basic industry in the information age, and wafer manufacturing is the most important and complex part of the semiconductor manufacturing process, and the main process flows include heat treatment, photolithography, etching, ion implantation, film deposition, chemical mechanical polishing and cleaning, etc., which involve a plurality of flows and apparatuses. With the development of technology, these processes are often integrated in semiconductor equipment in the form of process modules together with warehousing and material transportation systems to improve production efficiency. However, the structure of the semiconductor equipment and the process flow of the wafer are complicated, which directly results in the competition of resources for the wafer in the equipment and the requirements of changing the state of some modules and replacing process modules. Therefore, it is critical to design an intelligent scheduling system that not only meets complex requirements but also improves productivity for semiconductor devices.
Disclosure of Invention
In view of the above requirements, the invention provides a method for optimizing the route planning of a transport trolley in semiconductor equipment and the comprehensive scheduling of wafer production sequencing based on a genetic simulated annealing algorithm, which takes the minimized maximum completion time as an optimization target, thereby improving the efficiency and increasing the yield under the condition of meeting the complex constraints of different process modules.
The technical purpose is achieved, the technical effect is achieved, and the invention is realized through the following technical scheme:
an optimized scheduling method for a wafer equipment scheduling system comprises the following steps:
coding production elements in the wafer production equipment, acquiring coding information of the production elements in a single wafer box processing path, and setting the coding information as a single chromosome coding model;
constructing a Petri network structure, wherein the Petri network takes a chromosome coding model as an input parameter and takes the completion time corresponding to the corresponding chromosome coding model as an output parameter;
establishing an initial population based on a chromosome coding model, carrying out iterative update on the initial population, replacing the worst individual of the initial population with an elite individual in the updated population, updating a Petri network, and obtaining the minimum and maximum completion time from the updated population; and extracting coding information from the result and scheduling the wafer production path.
As a further improvement of the present invention, the production elements include respective processing stations, wafer cassettes, and transport carts provided in the wafer production facility;
the chromosome coding model comprises a circular crystal box code, a code corresponding to a processing station where the circular crystal box passes, and a code of a transport vehicle used for the migration of the circular crystal box among the stations.
As a further improvement of the invention, the structure of the constructed Petri net comprises
Token, each processing station in the wafer production facility;
the production state of the wafer in each wafer box is in the warehouse;
and (4) transition, namely change of the wafer production state in each wafer box.
As a further improvement of the invention, the method also comprises a formulated transportation trolley scheduling strategy, and a time parameter obtained based on the transportation trolley scheduling strategy is added into the Petri network to obtain an improved Petri network comprising a time attribute;
the trolley dispatching strategy is as follows:
judging whether the transport trolley is in an occupied state, if so, executing according to the time sequence of the Petri network structure;
if not, the user can not select the specific application,
firstly, determining whether a target station has priority setting, if so, determining to the target station according to the priority;
if one or more available target stations need to use the transport trolley after the current state is finished, the target stations with the minimum remaining time in the current state are moved to;
if the remaining time of the plurality of available target stations is the minimum, selecting the target station with the shortest distance to go to;
and if the remaining time and the distance of a plurality of available target stations are the minimum, one of the target stations is randomly selected to go to.
As a further improvement of the invention, based on the improved Petri net, the round crystal box is scheduled to be in a non-waiting state with the time length capable of being determined and the transition of the round crystal state is executed in the production process with the aim of realizing the minimum completion time.
As a further improvement of the present invention, the iterative updating of the initial population comprises updating the initial population using a tournament selection method by selecting the individual with the smallest completion time of the two chromosomes to be compared to enter the next generation population until the size of the new population reaches the size of the initial population.
As a further improvement of the invention, annealing-type mutation is carried out by combining two-point mutation in a genetic algorithm and Metropolis acceptance criterion in a simulated annealing algorithm: the method comprises the steps of randomly selecting any position of two chromosomes to carry out segmentation and exchange, then comparing the minimum completion time corresponding to the child chromosomes after the exchange with the minimum completion time corresponding to the parent chromosomes before the exchange, and determining whether new and old individuals are replaced or not by comparing the minimum completion times of the chromosomes before and after the exchange.
As a further improvement of the present invention, said preservation of elite comprises
Obtaining the individual with the minimum completion time in all chromosomes in the current population as the optimal individual s, and recording the minimum completion time value fs；
Calculating the optimal minimum completion time f of the updated populationk；
Comparison fsAnd fkIf f is a magnitude relation ofk>fsReplacing the worst individual of the current population with the best individual s in the population before updating;
as a further improvement of the invention, whether the number of updating iterations reaches the set number is judged,
if so, outputting the optimal individual corresponding to the optimal scheduling production scheme;
if not, continuing to iteratively execute the championship selection method and update the population by combining the genetic algorithm until the iteration upper limit is reached.
As a further improvement of the present invention, the iterative process further includes setting a temperature parameter, and the current temperature is updated according to the following formula:
T＝T0*αn
in the formula: t is0And alpha is a set temperature reduction coefficient and n is the current iteration number.
The invention has the beneficial effects that:
(1) the optimal scheduling of the wafer scheduling system is solved by adopting a genetic simulation annealing algorithm, the model has strong adaptability, is not easy to fall into a local optimal solution, and has strong solution space searching capability.
(2) A simulation model of the wafer scheduling system is constructed based on the Petri network to realize complete simulation of the system, and the simulation model is easily combined with various group optimization algorithms through a path table. Meanwhile, the time complexity of optimization by using a group optimization algorithm is greatly reduced by a variable step size simulation method.
Drawings
FIG. 1 is a flow chart illustrating a scheduling method according to the present invention;
FIG. 2 is a schematic flow chart of step 3 in the scheduling method of the present invention;
FIG. 3 is a diagram showing a distribution of semiconductor internal production resources in one embodiment;
FIG. 4 is a schematic representation of a chromosome coding model in an embodiment of the invention;
FIG. 5 is a schematic diagram of the first half of the basic Petri net model of the complete production chain of the No. 1 and No. 2 wafer boxes in the embodiment of the present invention;
FIG. 6 is a schematic diagram of the second half of a basic Petri net model of a complete production chain of the No. 1 and No. 2 wafer boxes according to the embodiment of the present invention;
FIG. 7 is an iteration curve obtained after use of the present invention in an example;
FIG. 8 is a Gantt chart of wafer processing solved in an embodiment of the present invention;
wherein: 1-warehouse, 2-transport vehicle, 3-loading and unloading station, 4-I type wafer transport vehicle, 5-merging station, 6-II type wafer transport vehicle, 7-buffer zone and 8-processing station.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
As shown in fig. 3, the semiconductor device includes 3 warehouses 1,2 wafer cassette transport carts 2,3 wafer loading/unloading stations 3, 2I-type wafer transport carts 4, 2 wafer merging stations 5, 1 II-type wafer transport cart 6, 1 buffer 7, and 8 wafer processing stations 8.
Description of the function: the warehouse 1 is used for storing wafer boxes, 4 wafer boxes are stored in each warehouse 1, each wafer box is loaded with 5 wafers to be processed, and the wafer processes in the same wafer box are consistent. The wafer loading and unloading stations 3 are used for unloading/loading wafers from/to cassettes, and each wafer loading and unloading station 3 can only load one cassette at a time, and cannot directly load wafers. The wafer merging stations 5 are used for merging unprocessed wafers in the same process path and splitting processed wafers of the same type, and each wafer merging station 5 can only load 10 wafers at the same time and cannot load a wafer box. The buffer zone 7 is used for temporary storage before wafer processing, and only 10 wafers can be loaded at the same time, and a wafer box cannot be loaded. The wafer processing stations 8 are used for processing wafers in a single process, and each wafer processing station 8 can only process 10 wafers at the same time and cannot load a wafer box. The wafer box transport vehicle 2 is used for transporting wafer boxes between the stations of the warehouse 1 and the wafer loading and unloading station 3, and only one wafer box can be transported. The type I wafer transport cart 4 is used for transporting wafers between the stations of the wafer loading/unloading station 3 and the wafer merging station 5, and only 5 wafers can be transported at the same time. The type II wafer transport vehicle 6 is used for transporting wafers among the stations of the wafer merging station 5, the buffer area 7 and the wafer processing station 8, and only 10 wafers can be transported at the same time.
The production requirements are as follows: 3 total 12 wafer boxes in warehouse 1, each wafer box all fills pending processing wafer, and the wafer all has appointed technology route in each wafer box. The complete production process requires that a wafer box transport vehicle 2 transports a wafer box to a wafer loading and unloading station 3 and returns an empty wafer box to a warehouse 1 after the wafer is unloaded, a type I wafer transport vehicle 4 transports the wafer in the wafer box from the loading and unloading station 3 to a wafer merging station 5, the wafer merging station 5 merges 10 unprocessed wafers in the same process route into a whole, a type II wafer transport vehicle 6 transports the whole wafer from the wafer merging station 5 to a buffer area 7 for temporary storage, and then sequentially transports the whole wafer to each processing station 8 for processing according to the process route, and the whole wafer needs to be taken away immediately after the processing is completed in one wafer processing station 8. After the processing is finished, the II type wafer transport vehicle 6 sends the whole to the wafer merging station 5 to split two 5 processed wafers, the I type wafer transport vehicle 4 sends the processed wafers to the wafer loading and unloading station 3 from the wafer merging station 5, when an empty wafer box is loaded in the wafer loading and unloading station 3, the processed wafers are loaded into the empty box, and finally the wafer box transport vehicle 2 sends the wafer box loaded with the processed wafers back to the original warehouse 1. The operations of loading and unloading the transport cart, merging/splitting the wafer merging station 5, processing the wafer processing station 8, and the like are all clearly time-consuming.
As shown in fig. 1, the scheduling method is implemented as follows:
step 1, numbering production resources: uniformly numbering a warehouse 1, a wafer loading and unloading station 3, a wafer merging station 5, a wafer processing station 8, a buffer zone 7, a transport trolley and the like; numbering production original sheets: all the wafer boxes stored in the warehouse 1 are numbered uniformly, and all the wafers in the wafer boxes share one serial number with the wafer boxes as a whole because the wafer boxes are processed in batch in the production process.
And 2, setting a single chromosome coding model, and coding the parameters of all wafer boxes into the same chromosome by taking five parameters, namely the time of taking out the wafer box 1 of each wafer box (namely the time of starting the wafer in the wafer box to enter production), the serial number of a loading and unloading station 3 passing before processing, the serial number of a merging station 5 passing after processing and the serial number of a loading and unloading station 3 passing after processing and returning as coding objects. Thus, as shown in FIG. 4, the single chromosome set has 60 codes, describing 12 numbered cassettes.
And 3, establishing a Petri network-based multi-type wafer transportation path planning and processing sequencing hybrid model, wherein the input parameter of the model is a single chromosome, and the return parameter is the completion time under the condition of the chromosome coding model.
The specific process of step 3 is as follows:
and 3.1, reading the input chromosome, and extracting the time for the wafer in each wafer box to start to enter production and a complete production route, namely station numbers of a wafer loading and unloading station 3, a wafer merging station 5, a wafer processing station 8 and the like which pass through in sequence.
And 3.2, constructing a basic Petri net structure according to the complete production route of the wafers in each wafer box, converting the state-transition pair constraint condition in the wafer production route into time sequence description, wherein the specific construction process comprises the steps of taking production resources such as a warehouse 1, a wafer loading and unloading station 3, a wafer merging station 5, a wafer processing station 8, various transport vehicles 2 and the like as tokens, taking the production state of the wafers in each wafer box as a library, and taking the change of the production state of the wafers in each wafer box as transition, so that the basic Petri net graphic model is used for describing all wafer production chains. Taking wafers in the wafer cassettes of nos. 1 and 2 as an example, the whole production process is shown in fig. 5-6.
3.3, improving the basic Petri network structure, and formulating a route selection strategy of the transport trolley: if the transport trolley is in an occupied state, executing according to the time sequence of the basic Petri network structure; if the trolley is in a non-occupied state, the method is carried out according to the following rules: if the available target sites have priority setting, the target sites are forwarded according to the priority; if the remaining time of the available target stations is the minimum, selecting the target station with the shortest distance to go to, and if the remaining time of the available target stations is the minimum and the distance is the shortest, randomly selecting one of the target stations to go to; otherwise, if the transport trolley is not needed after the current states of all the available target stations are finished, the transport trolley stays in place.
And 3.4, continuously improving the Petri network model on the basis of the step 3.3, adding a time attribute to a library in the Petri network, namely converting time constraint in the wafer production route into production state residual time description, and then simulating the improved Petri network model by adopting a variable step size simulation strategy. The specific process is as follows:
and 3.4.1, establishing an empty table schedule for storing the simulation process of the input chromosome. Establishing a wafer current state remaining time table: each row corresponds to a wafer in a wafer box, and the parameters in each row are the residual time, the occupied production resource number and the wafer box number from top to bottom in sequence. All production states of the wafer can be divided into two types, the first is an operation state with determined duration, and the second is a waiting state with undetermined duration. Therefore, for the first state, the remaining time is a determined value, and for the second state, the remaining time is set to be the minimum value of the remaining time of the wafer currently in the first state. The current state remaining time table is set up as shown in FIG. 7, which shows a wafer MiAt position piThe time remaining for the transition above, i ═ {1,2,3, K,11,12} represents the wafer number. The constructed remaining time table is shown in Table 1
Table 1: remaining time table
Figure BDA0003145591770000061
Step 3.4.2, select and execute transition: and taking out the minimum remaining time from the current state remaining time table, and executing state transition on the wafers in the wafer boxes with one or more numbers meeting the minimum remaining time, wherein the transition rule is the time sequence determined by the improved Petri network structure. Since the waiting state always has the minimum remaining time, the wafer in the waiting state inevitably performs state transition in each selection until the relevant timing condition is met so as to jump out of the waiting state in the transition.
Step 3.4.3, updating the remaining time table: and subtracting the minimum remaining time from all the time in the remaining time table of the current state, and updating the remaining time and the number of the occupied production resources of the wafer which is transferred to the new state according to the new state. And storing the updated remaining time of the wafer, the number of the occupied production resource and the number parameter of the wafer box into a schedule table.
Step 3.4.4, illegal solution treatment: illegal solutions are mainly classified into deadlock situations and unsatisfied time constraints. Deadlock refers to the situation that the contention between two parties is continuously in a waiting state due to the simultaneous contention of limited resources, and the contention can only occur between transport vehicles, so that a resource reservation method is adopted for preventing, after the transport vehicle determines a destination station of the next stage, the station resource is marked to be reserved, and then a series of operations of transportation, loading/unloading are carried out. If deadlock occurs in the simulation operation, if a certain time limit is exceeded, the current chromosome is judged to be an invalid solution, and the returned completion time is set to be a maximum value. And if the time constraint condition is not met, the current chromosome is also judged to be an invalid solution, and the returned completion time is set to be a maximum value.
Step 3.4.5: judging whether the simulation can be finished or not, if all the wafers are finished or production errors occur, finishing the simulation, and returning to the finishing time; otherwise go to step 3.4.2 for execution.
Step 4, setting initial parameters of the genetic simulated annealing algorithm, setting the initial population size Num to be 200, the upper limit value Lim of the iteration times to be 300 and the initial temperature T 01000, cooling coefficient alpha 0.95, current temperature T0. Then setting an initial population, wherein the specific process is as follows:
and 4.1, calculating an overall finishing time upper limit value MaxTime, wherein the corresponding condition of the value is that each numbered wafer finishes complete processing and then the next numbered wafer starts to be processed, the contention of the wafers with different numbers on resources during production does not happen at the moment, and the shortest finishing time is necessarily less than MaxTime.
In step 4.2, there are 12 numbered wafers in the present embodiment, and each numbered wafer has 5 chromosome parameter code values according to step 2, so that each chromosome has 60 code values. For each numbered wafer, the parameter is set as a random number selected in a designated range, the time range of the parameter out of the warehouse 1 is [0, MaxTime ], the number of the loading and unloading station 3 passing through when the parameter goes to be processed and the number range of the loading and unloading station 3 passing through when the processing is completed and returned are all the numbers of the loading and unloading station 3 in the step 1, and the number of the merging station 5 passing through when the parameter goes to be processed and the number range of the merging station 5 passing through when the processing is completed and returned are all the numbers of the merging station 5 in the step 1. A single chromosome is generated and its encoding is initialized accordingly.
And 4.3, reading the set initial population size Num, and generating and initializing Num chromosomes according to the method in the step 4.2, wherein the Num chromosomes form the initial population.
Step 5, firstly setting an individual objective function value calculation method: inputting the chromosome coding model into the model established in the step 3 and executing the step 3, then taking the returned completion time as the objective function value of the individual, and executing the subsequent calculation of the objective function value of the individual according to the completion time. Then calculating the objective function values of all chromosome individuals in the population, and selecting the minimum objective function value f from the objective function valuessAnd the corresponding individual s, the minimum objective function value is recorded as the optimal objective function value, and the corresponding individual is recorded as the optimal individual.
Step 6, adopting a championship match selection method to update the population: and determining the number of individuals selected each time to be 2, randomly picking out 2 individuals from the population, and selecting one individual with the minimum objective function value from the population to enter the next generation of the population. This selection operation was repeated until the new population size reached the original population size, and the updated population of this step was recorded as temp 1.
And 7, combining the single-point crossing in the genetic algorithm and the Metropolis acceptance criterion in the simulated annealing algorithm to carry out annealing type crossing: randomly selecting two chromosomes from temp1 in the population updated in step 6, segmenting at a randomly selected position point and exchanging the right part to obtain two different sub-chromosomes, wherein the process is as follows:
Figure BDA0003145591770000071
then, judging whether to accept the new individual to replace the old individual according to a Metropolis acceptance criterion, namely judging whether the daughter A replaces the father A and whether the daughter B replaces the father B respectively, wherein the Metropolis acceptance criterion judgment process comprises the following steps:
calculating the probability P of accepting a new individual to replace an old individual
Figure BDA0003145591770000072
Wherein EjFor new individual objective function values, EiThe values are the old individual objective function values. According to the formula, if the objective function value of the new individual is smaller, the new individual is used for replacing the old individual; otherwise, the new individual replaces the old individual according to the calculated probability, and if the new individual is not replaced according to the probability, the new individual is discarded, namely the number of chromosomes is still 2 after the crossover is finished. The two chromosomes with completed crossover are then added to the new population temp2 and removed from the population temp1 and the above steps are repeated for the remaining chromosomes in the population temp1 until all updates are complete.
And 8, annealing type mutation is carried out by combining two-point mutation in a genetic algorithm and Metropolis acceptance criterion in a simulated annealing algorithm: randomly selecting 1 chromosome from the population temp2 updated in step 7, randomly selecting two position points on the chromosome for mutation, and replacing the current position point code value with a random number in the feasible solution range by the mutation operation of a certain position point, wherein the feasible solution range is shown in step 4.2. The process is schematically shown as follows:
individuals before mutation0 0 1 1000 → individuals after mutation1 0 1 0 0 0 0
And then judging whether to accept the new individual to replace the old individual according to Metropolis acceptance criteria, wherein the judging method is shown in step 7. This mutated chromosome is then added to the new population temp3 and removed from the population temp2, and the above steps are repeated for the remaining chromosomes in the population temp2 until all updates are complete.
Step 9, performing elite preservation: recording the minimum objective function value f in the current population temp3kThen, the minimum objective function value f before selection, intersection and variation is addedsComparison, if fkIf the value is larger, a worse solution is obtained after selection, crossing and mutation, at this time, the best individual s before selection, crossing and mutation needs to be stored, and the specific method is to replace the worst individual in the current population temp3 with s; if f iskAnd smaller, no operation is performed.
Step 10, after the elite is stored, judging whether the upper limit value of the iteration times is reached: if so, outputting the current optimal individual, the optimal objective function value and the corresponding schedule production table and transferring to the step 11 for execution; if not, the current temperature T is updated according to the following formula and the process goes to step 5.
T＝T0*αn
Wherein T isUAnd alpha is a set temperature reduction coefficient and n is the current iteration number.
And 11, the current optimal objective function value is the total minimum completion time, the code of the current optimal individual is the production path of each wafer, and the schedule table records the complete processing process of the wafer in a matrix form. And programming and visualizing the schedule table and the optimal individual code, so that two key parts, namely a vehicle path and a processing sequence, in the scheduling system are output and displayed, and the vehicle path table and the wafer processing Gantt chart are correspondingly and respectively obtained. The vehicle routing table for the type ii wafer transport vehicle 2A is shown in table 2, and the wafer processing gantt chart is shown in fig. 8.
Table 2: type II transport vehicle A path table
Figure BDA0003145591770000081
Figure BDA0003145591770000091
In conclusion, the improved Petri network structure is combined with the genetic simulation annealing optimization algorithm, accurate and efficient description and scheduling optimization of the complex multi-constraint discrete parallel system of the semiconductor equipment are achieved, and an effective scheme is provided for improving the production efficiency of the semiconductor equipment.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (10)

1. An optimized scheduling method for a wafer equipment scheduling system is characterized by comprising the following steps:
coding production elements in the wafer production equipment, acquiring coding information of the production elements in a single wafer box processing path, and setting the coding information as a single chromosome coding model;
constructing a Petri network structure, wherein the Petri network takes a chromosome coding model as an input parameter and takes the completion time corresponding to the corresponding chromosome coding model as an output parameter;
establishing an initial population based on a chromosome coding model, carrying out iterative update on the initial population, replacing the worst individual of the initial population with an elite individual in the updated population, updating a Petri network, and obtaining the minimum and maximum completion time from the updated population; and extracting coding information from the result and scheduling the wafer production path.
2. The optimized scheduling method of the wafer equipment scheduling system of claim 1, wherein: the production elements comprise processing stations, wafer boxes and transport vehicles which are arranged in the wafer production equipment;
the chromosome coding model comprises a circular crystal box code, a code corresponding to a processing station where the circular crystal box passes, and a code of a transport vehicle used for the migration of the circular crystal box among the stations.
3. The optimized scheduling method of the wafer equipment scheduling system of claim 2, wherein: the structure of the constructed Petri net comprises
Token, each processing station in the wafer production facility;
warehouse, production state of wafer in each wafer box
And (4) transition, namely change of the wafer production state in each wafer box.
4. The optimized scheduling method of the wafer equipment scheduling system of claim 1, wherein:
the method also comprises a formulated transportation trolley scheduling strategy, and a time parameter obtained based on the transportation trolley scheduling strategy is added into the Petri network to obtain an improved Petri network comprising a time attribute;
the trolley dispatching strategy is as follows:
judging whether the transport trolley is in an occupied state, if so, executing according to the time sequence of the Petri network structure;
if not, the user can not select the specific application,
firstly, determining whether a target station has priority setting, if so, determining to the target station according to the priority;
if one or more available target stations need to use the transport trolley after the current state is finished, the target stations with the minimum remaining time in the current state are moved to;
if the remaining time of the plurality of available target stations is the minimum, selecting the target station with the shortest distance to go to;
and if the remaining time and the distance of a plurality of available target stations are the minimum, one of the target stations is randomly selected to go to.
5. The optimized scheduling method of the wafer equipment scheduling system of claim 4, wherein: based on the improved Petri network, the round crystal box is scheduled to be in a non-waiting state capable of determining time length in the production process with the aim of realizing the minimum completion time, and transition of the round crystal state is executed.
6. The optimized scheduling method of the wafer equipment scheduling system of claim 1, wherein: the iterative updating of the initial population comprises updating the initial population by a championship selection method, and selecting and comparing individuals with the minimum completion time in two chromosomes to enter a next generation population until the scale of the new population reaches the scale of the initial population.
7. The optimized scheduling method of the wafer equipment scheduling system of claim 6, wherein: further comprising annealing-type mutation using a combination of two-point mutation in genetic algorithm and Metropolis acceptance criterion in simulated annealing algorithm: the method comprises the steps of randomly selecting any position of two chromosomes to carry out segmentation and exchange, then comparing the minimum completion time corresponding to the child chromosomes after the exchange with the minimum completion time corresponding to the parent chromosomes before the exchange, and determining whether new and old individuals are replaced or not by comparing the minimum completion times of the chromosomes before and after the exchange.
8. The optimized scheduling method of the wafer equipment scheduling system of claim 7, wherein:
the essence preservation comprises
Obtaining the individual with the minimum completion time in all chromosomes in the current population as the optimal individual s, and recording the minimum completion time value fs；
Calculating the optimal minimum completion time f of the updated populationk；
Comparison fsAnd fkIf f is a magnitude relation ofk>fsAnd replacing the worst individual of the current population by the best individual s in the population before updating.
9. The optimized scheduling method of the wafer equipment scheduling system of claim 8, wherein: judging whether the number of updating iterations reaches the set number,
if so, outputting the optimal individual corresponding to the optimal scheduling production scheme;
if not, continuing to iteratively execute the championship selection method and update the population by combining the genetic algorithm until the iteration upper limit is reached.
10. The optimized scheduling method of the wafer equipment scheduling system of claim 1, wherein: the iterative process also comprises setting temperature parameters, and the current temperature is updated according to the following formula:
T＝T0*αn
in the formula: t is0And alpha is a set temperature reduction coefficient and n is the current iteration number.
CN202110752785.8A 2021-07-02 2021-07-02 Optimized scheduling method for scheduling system of wafer equipment Pending CN113361813A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110752785.8A CN113361813A (en) 2021-07-02 2021-07-02 Optimized scheduling method for scheduling system of wafer equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110752785.8A CN113361813A (en) 2021-07-02 2021-07-02 Optimized scheduling method for scheduling system of wafer equipment

Publications (1)

Publication Number Publication Date
CN113361813A true CN113361813A (en) 2021-09-07

Family

ID=77538073

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110752785.8A Pending CN113361813A (en) 2021-07-02 2021-07-02 Optimized scheduling method for scheduling system of wafer equipment

Country Status (1)

Country Link
CN (1) CN113361813A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115600773A (en) * 2022-12-13 2023-01-13 合肥喆塔科技有限公司（Cn） Production path analysis method and system based on sequence pattern mining
WO2023142653A1 (en) * 2022-01-28 2023-08-03 弥费科技(上海)股份有限公司 Area-based management method and apparatus, computer device, and storage medium

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101493857A (en) * 2009-02-13 2009-07-29 同济大学 Semiconductor production line model building, optimizing and scheduling method based on petri net and immune arithmetic
CN101901425A (en) * 2010-07-15 2010-12-01 华中科技大学 Flexible job shop scheduling method based on multi-species coevolution
CN104820872A (en) * 2015-05-06 2015-08-05 华北电力大学 Method for optimizing project duration of engineering project based on potential anti-key working procedures
CN104835026A (en) * 2015-05-15 2015-08-12 重庆大学 Automatic stereoscopic warehouse selection operation scheduling modeling and optimizing method based on Petri network and improved genetic algorithm
CN110598943A (en) * 2019-09-19 2019-12-20 郑州航空工业管理学院 Method for solving flexible job shop scheduling with transportation time by using improved cultural genetic algorithm
CN111079987A (en) * 2019-11-28 2020-04-28 电子科技大学 Semiconductor workshop production scheduling method based on genetic algorithm
CN111563336A (en) * 2020-04-30 2020-08-21 南通大学 Deadlock-free scheduling method of flexible manufacturing system based on improved genetic algorithm
CN111967654A (en) * 2020-07-27 2020-11-20 西安工程大学 Method for solving flexible job shop scheduling based on hybrid genetic algorithm
CN111985647A (en) * 2020-07-21 2020-11-24 西安理工大学 Printing bookbinding job scheduling optimization method based on genetic algorithm
CN112001541A (en) * 2020-08-24 2020-11-27 南京理工大学 Improved genetic algorithm for path optimization
CN112966822A (en) * 2021-02-03 2021-06-15 广东工业大学 Mixed-flow manufacturing workshop scheduling method based on improved genetic algorithm

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101493857A (en) * 2009-02-13 2009-07-29 同济大学 Semiconductor production line model building, optimizing and scheduling method based on petri net and immune arithmetic
CN101901425A (en) * 2010-07-15 2010-12-01 华中科技大学 Flexible job shop scheduling method based on multi-species coevolution
CN104820872A (en) * 2015-05-06 2015-08-05 华北电力大学 Method for optimizing project duration of engineering project based on potential anti-key working procedures
CN104835026A (en) * 2015-05-15 2015-08-12 重庆大学 Automatic stereoscopic warehouse selection operation scheduling modeling and optimizing method based on Petri network and improved genetic algorithm
CN110598943A (en) * 2019-09-19 2019-12-20 郑州航空工业管理学院 Method for solving flexible job shop scheduling with transportation time by using improved cultural genetic algorithm
CN111079987A (en) * 2019-11-28 2020-04-28 电子科技大学 Semiconductor workshop production scheduling method based on genetic algorithm
CN111563336A (en) * 2020-04-30 2020-08-21 南通大学 Deadlock-free scheduling method of flexible manufacturing system based on improved genetic algorithm
CN111985647A (en) * 2020-07-21 2020-11-24 西安理工大学 Printing bookbinding job scheduling optimization method based on genetic algorithm
CN111967654A (en) * 2020-07-27 2020-11-20 西安工程大学 Method for solving flexible job shop scheduling based on hybrid genetic algorithm
CN112001541A (en) * 2020-08-24 2020-11-27 南京理工大学 Improved genetic algorithm for path optimization
CN112966822A (en) * 2021-02-03 2021-06-15 广东工业大学 Mixed-flow manufacturing workshop scheduling method based on improved genetic algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
任磊、王峰、邢科义: "基于Petri网的柔性制造系统无死锁遗传调度算法", 《控制理论与应用》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023142653A1 (en) * 2022-01-28 2023-08-03 弥费科技(上海)股份有限公司 Area-based management method and apparatus, computer device, and storage medium
CN115600773A (en) * 2022-12-13 2023-01-13 合肥喆塔科技有限公司（Cn） Production path analysis method and system based on sequence pattern mining

Similar Documents

Publication Publication Date Title
CN111507641B (en) Batch processing equipment scheduling method and device
Mönch et al. Production planning and control for semiconductor wafer fabrication facilities: modeling, analysis, and systems
CN113361813A (en) Optimized scheduling method for scheduling system of wafer equipment
CN112561194B (en) Integrated scheduling method and system for production and logistics of mixed flow shop
CN113822525B (en) Flexible job shop multi-target scheduling method and system based on improved genetic algorithm
CN114707294B (en) Multi-target scheduling method for job shop with limited transportation capacity constraint
CN114330870B (en) Vehicle path planning method with time window based on multiple swarm evolution algorithm
Huang et al. An improved genetic algorithm for job-shop scheduling problem with process sequence flexibility
CN110554673A (en) Intelligent RGV processing system scheduling method and device
CN112232545A (en) AGV task scheduling method based on simulated annealing algorithm
Alaia et al. A Comparative Study of the PSO and GA for the m-MDPDPTW
Ma et al. A hybrid particle swarm optimization and simulated annealing algorithm for job-shop scheduling
Zhang et al. A performance analytical model of automated material handling system for semiconductor wafer fabrication system
Wu et al. Scheduling semiconductor in-line steppers in new product/process introduction scenarios
CN115730799A (en) Method, system and equipment for scheduling production tasks of flexible assembly job workshop
CN112700190B (en) Improved method for distributing tray materials by scanning method and genetic simulation annealing method
Tamssaouet et al. A study on the integration of complex machines in complex job shop scheduling
Yang et al. Multiobjective lot scheduling and dynamic OHT routing in a 300-mm wafer fab
ElWakil et al. A New Hybrid Salp Swarm-simulated Annealing Algorithm for the Container Stacking Problem.
Jeong et al. Machine Learning-Based Dispatching for a Wet Clean Station in Semiconductor Manufacturing
CN111639822A (en) Express distribution method based on 0-1 knapsack problem analysis
Dang et al. Adaptive large neighborhood search for scheduling of mobile robots
CN116795054B (en) Intermediate product scheduling method in discrete manufacturing mode
CN117875189B (en) Three-dimensional warehouse space layout method based on GA optimization GRO
Gen et al. Hybrid genetic algorithm for designing logistics network, VRP and AGV problems

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210907