CN104636871A - Data-based single-stage multi-product scheduling control method - Google Patents
Data-based single-stage multi-product scheduling control method Download PDFInfo
- Publication number
- CN104636871A CN104636871A CN201510070318.1A CN201510070318A CN104636871A CN 104636871 A CN104636871 A CN 104636871A CN 201510070318 A CN201510070318 A CN 201510070318A CN 104636871 A CN104636871 A CN 104636871A
- Authority
- CN
- China
- Prior art keywords
- order
- processing
- data
- scheduling
- ratio
- 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.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a data-based single-stage multi-product scheduling control method. The method comprises the following steps of for different target batch processing scenes, extracting corresponding historic production scheduling data, and according to different production scheduling tasks during a production scheduling period, respectively establishing SMSP (single-stage multi-product scheduling problem) scheduling cases; giving a heuristic machining equipment distribution rule, and using a genetic algorithm to solve the optimal order processing series of the SMSP scheduling cases; using the SMSP scheduling cases and the optimal order processing series to build a training data set, and training to obtain an order processing series prediction model; using the order processing series prediction model to process a batch processing order to obtain an order processing arrangement series; using the heuristic machining equipment distribution rule and the order processing arrangement series to distribute batch processing orders to different equipment to produce and process. According to the data-based single-stage multi-product batch processing control method, the production scheduling cases during different production scheduling periods can be quickly customized, and the method has better scheduling optimizing effect.
Description
Technical field
The present invention relates to process industry control field, be specifically related to a kind of control method of the single phase multi-product batch processing based on data.
Background technology
For meeting customer need better, Process Industry production and processing task needs to carry out usually in batches, therefore, also just faces the problem such as size and quantity, date issued and delivery date how to determine batch.
Batch process is the important component part in Process Industry production run, have product batch little, multi items, high added value, equipment investment little, produce the advantages such as updating decision, therefore, be widely used in the middle of the Process Industries such as petrochemical complex, pharmacy, food, effectively scheduling is realized to batch process, the flexibility of production run can not only be strengthened, can also economize on resources, reduce costs, make enterprise obtain better economic benefit.
Typical batch process is often made up of a series of production process, and each operation can be processed on one or more production equipment, and the work capacity of each production equipment is different, can process the product of different cultivars, different batch.At present, the batch processing production scheduling process in Process Industry belongs to single phase multi-product batch process (single-stage multi-product scheduling problem, SMSP) mostly.
Existing SMSP optimizing scheduling method for solving adopts the complex optimization such as mathematical programming approach or genetic algorithm method usually, these methods often realize comparatively complicated, need to carry out a large amount of parameter tuning work, and its Optimization Solution time can be explosive growth with the increase of scheduling problem scale, when scheduling problem is larger, these methods often cannot meet the requirement of real-time of process of manufacture because solving overlong time.
For solving the larger deficiency of existing single phase multi-product batch processing optimizing scheduling method for solving limitation in actual application, need to be optimized the scheduling scheme of multi-product batch processing.
Summary of the invention
The invention provides a kind of control method of the single phase multi-product batch processing based on data, for single phase multi-product batch processing production and processing scene, can production scheduling scheme in the rapid development different production cycle, and there is good optimizing scheduling effect.
Based on a control method for the single phase multi-product batch processing of data, comprise the following steps:
(1) for different target batch process scenes, from historical production data, take out corresponding history operation and dispatching information, according to the difference of production scheduling task in the production scheduling cycle, set up SMSP respectively and dispatch case.
Relevant information and the corresponding dispatch environment information of each order adopt four attribute descriptions, are respectively: actual process time, order completion date, Order Processing time ratio and order completion ratio.
(2) the heuristic process equipment allocation rule of given one, then utilizes SMSP in genetic algorithm for solving step (1) to dispatch the optimum Order Processing sequence of case.
Described heuristic process equipment allocation rule is obtained by artificial setting, utilizes selected heuristic process equipment allocation rule, all Fabrication Orders can be assigned to different equipment and process.
The end condition of described genetic algorithm is: if after continuous T time iteration, current obtained optimum solution improves not yet, then algorithm stops, and T gets the half of all order numbers.
(3) utilize SMSP to dispatch case and corresponding optimum Order Processing sequence, build training dataset, training obtains Order Processing sequential forecasting models.
The quantity building the SMSP scheduling case of training dataset is more, then train the Order Processing sequential forecasting models obtained more reliable.
Described Order Processing sequential forecasting models adopts Single hidden layer feedforward neural networks to express.Extreme learning machine algorithm is adopted to carry out the training of Order Processing sequential forecasting models.
Described training data concentrates the input feature vector of each training sample to comprise five kinds, is respectively: the ratio of the order numbers that the ratio of actual process time, the ratio of completion date, the ratio of Order Processing time ratio, the ratio of dilivery date and current production scheduling stage have arranged and total orders.
(4) utilize Order Processing sequential forecasting models to process pending batch processing order, obtain Order Processing and arrange sequence;
(5) the heuristic process equipment allocation rule of step (2) and the Order Processing of step (4) is utilized to arrange sequence, by pending batch processing Order splitting to different equipment, determine processing start time and end time, carry out production and processing.
The present invention is based on the control method of the single phase multi-product batch processing of data, mainly contain following beneficial effect:
(1) can rapid development for the production scheduling scheme of single phase multi-product batch processes, meet the requirement of real-time of actual production process;
(2) in the application on site stage, without the need to carrying out complicated modeling and parameter tuning work, it is convenient to implement;
(3) there is good optimizing scheduling ability.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the control method of the single phase multi-product batch processing that the present invention is based on data;
Fig. 2 is the scheduling scheme structure process flow diagram based on heuristic process equipment allocation rule;
Fig. 3 is genetic algorithm block diagram;
Fig. 4 is Order Processing sequential forecasting models schematic diagram;
Fig. 5 is the SMSP production scheduling scheme schematic diagram (horizontal ordinate is the time) built based on heuristic process equipment allocation rule;
Fig. 6 is the Optimal Production scheduling scheme schematic diagram (horizontal ordinate is the time) built based on genetic algorithm.
Embodiment
Below in conjunction with accompanying drawing, the control method of the single phase multi-product batch processing that the present invention is based on data is described in detail.
As shown in Figure 1, a kind of control method of the single phase multi-product batch processing based on data, comprises the steps:
(1) for different target batch process scenes, from historical production data, take out corresponding history operation and dispatching information, according to the difference of production scheduling task in the production scheduling cycle, set up SMSP respectively and dispatch case.
(2) the heuristic process equipment allocation rule of given one, then utilizes SMSP in genetic algorithm for solving step (1) to dispatch the optimum Order Processing sequence of case.
Heuristic process equipment allocation rule is by artificial selected, for all Order splitting to different equipment is carried out production and processing, such as, within a production scheduling cycle, suppose total N number of Fabrication Order O={o1, o2 ... oN}, forms an Order Processing sequence π=(π by N number of Fabrication Order
1, π
1..., π
n), π
i∈ o1, o2 ..., oN}, i ∈ { 1,2 ..., N}, according to this Order Processing sequence, successively each Order splitting in this N number of order is processed to a specific process equipment, and determine start time and the end time of processing, if still have unappropriated order, then continue to distribute; If all Order splitting are complete, then calculate the performance index f (π) of scheduling scheme, obtain complete SMSP scheduling scheme, assigning process as shown in Figure 2.
According to relevant information and the corresponding dispatch environment information of order, by the comparison between two to order, all Fabrication Orders are sorted, such as, given any two orders, be respectively " order 1 " and " order 2 ", generic attribute value is 0 or 1, generic attribute value is 1, and expression should be given priority in arranging for, and " order 1 " is processed; Generic attribute value is 0, and should give priority in arranging for " order 2 " is processed.
For a kth order, it can preferentially be arranged to process relative to kth+1 ~ n order, if using a kth order as " order 1 ", generic attribute value is 1; If using a kth order as " order 2 ", then generic attribute value is 0.
As shown in Figure 3, adopt the coded system based on arrangement, chromosome is expressed as π=(π to the flow process of genetic algorithm
1, π
2..., π
n), wherein π
i∈ o1, o2 ..., oN}, i ∈ { 1,2 ..., N}, decode procedure is, utilize given heuristic process equipment allocation rule to arrange successively to process to order wherein, thus determine complete production scheduling scheme, and the performance of each scheduling scheme is assessed, to obtain adaptation value.
The central genetic operation adopted comprises:
Select operation: adopt roulette selection method to complete selection operation;
Interlace operation: adopt partial mapped crossover method to complete interlace operation;
Mutation operation: adopt backward method to carry out mutation operation, namely the fragment between chromosome two different random positions is reversed.
Genetic algorithm end condition is: if after continuous T time iteration, current obtained optimum solution improves not yet, then algorithm stops.T gets the half of all Fabrication Order numbers.
(3) utilize SMSP to dispatch case and corresponding optimum Order Processing sequence, build training dataset, training obtains Order Processing sequential forecasting models.
For a kth order, it can preferentially be arranged to process relative to kth+1 ~ n order, therefore, a kth order and any one follow-up order are for constructing a training sample, and the relevant information of corresponding order and real-time dispatch environment information are for constructing input feature vector.
Adopt the relevant information of following four each orders of attribute description and corresponding dispatch environment information:
I, actual process time (Real_Proc): not yet arranging to carry out in the order processed, if select order i preferentially to carry out production and processing, then determine corresponding process equipment according to given heuristic process equipment allocation rule, thus the equipment switching time determined needed for this Order Processing and process time, the computing formula of actual process time is as follows:
Real_Proc
i＝PT
i(mach)+CT
ji(1)
Wherein, mach represents that according to heuristic process equipment allocation rule be the process equipment that it distributes; J represents the upper order processed on the device.If order i is first order carrying out on the device processing, then switching time is 0.
II, order completion date (Order_Comp): not yet arranging to carry out in the order processed, if select order i preferentially to carry out production and processing, then the computing formula of this order completion date is as follows:
Order_Comp
i＝Order_Comp
j+Real_Proc
i(2)
Wherein, j is the upper order processed on relevant device.If order i is first order carrying out on the device processing, then Order_Comp
jbe 0.
III, Order Processing time ratio (Proc_Ratio): the process time that same order carries out machining need on different process equipment is different.In the current production scheduling moment, if scheduling order i processes, according to selected heuristic process equipment allocation rule, determine the process equipment of its correspondence and required switching time and process time.
In current time arrangement, carried out to order i the process time needed for production and processing, with the correlative value arranged in other moment the process time needed for order i production and processing, adopt following formula to calculate:
Wherein,
represent that order i carries out at distinct device the mean value that machining needs process time;
represent the mean value being switched to the switching time needed for order i from different order.
Proc_Ratio
iless, represent the current production scheduling stage to order i arrange production machining need process time shorter, otherwise, then illustrate that the process time of needs is longer.
IV, order completion is than (Order_Num_R): order completion has arranged the ratio of order numbers and the total orders of processing in the current production scheduling stage than expression, this attribute is for describing the real-time status of current production and processing environment.
In addition, the static attribute describing order can be used as the dilivery date (DD) of order equally.
The related symbol definition adopted in I ~ IV 4 attribute is as shown in table 1.
Table 1
| Parameter | Definition |
| m | Process equipment number |
| n | Fabrication Order number |
| PT i(mach) | The process time of order i on process equipment mach |
| CT ij | From the processing of order i being switched to the switching time needed the machining of order j |
| DD i | The dilivery date of order i |
| C i | The completion date of order i |
Training data concentrates the input feature vector of each training sample to comprise five kinds, as listed in table 2:
Table 2
| Feature | Definition |
| Real_Proc_R | The ratio of the actual process time of " order 1 " and " order 2 " |
| Order_Comp_R | The ratio of the completion date of " order 1 " and " order 2 " |
| Proc_Ratio_R | The ratio of the Order Processing time ratio of " order 1 " and " order 2 " |
| DD_R | The ratio of the dilivery date of " order 1 " and " order 2 " |
| Order_Num_R | The ratio of the order numbers that the current production scheduling stage has arranged and total orders |
As shown in Figure 4, Single hidden layer feedforward neural networks (single-hidden layer feedforward neural network, SLFN) is adopted to represent target dispatch pattern.SLFN model has 5 input nodes, respectively 5 input feature vectors of corresponding training sample, and have 2 output nodes, wherein, if the value of node 1 is higher than the value of node 0, then " order 1 " is given priority in arranging for and carried out production and processing; Otherwise " order 2 " is given priority in arranging for and is carried out production and processing.Extreme learning machine algorithm is adopted to be used for the training of SLFN model.
(4) utilize Order Processing sequential forecasting models to process pending batch processing order, obtain Order Processing and arrange sequence;
(5) the heuristic process equipment allocation rule of step (2) and the Order Processing of step (4) is utilized to arrange sequence, by pending batch processing Order splitting to different equipment, determine processing start time and end time, carry out production and processing.
Have the single phase multi-product batch processing production scene of 3 parallel fabrication production equipments below for one, the invention will be further described.
This production scene is always delayed the deadline to minimize
as optimizing scheduling performance index, deadline computing formula of always delaying is as follows:
First from historical production data, take out the history operation and dispatching information under target batch process for producing scene, for the production scheduling task in the different production scheduling cycle, set up SMSP respectively and dispatch case.
Such as, have 10 Fabrication Order O={o1 in some history production scheduling cycles, o2 ..., o10}, associated production data dispatching is as shown in table 3, and namely the production scheduling task in this production scheduling cycle is established as a SMSP and dispatches case.
Table 3
Select rule on earliest finish time (i.e. heuristic process equipment allocation rule) to be processed to different equipment by Order splitting, and determine start time and the end time of Order Processing.
For the SMSP case shown in table 3, suppose that the Order Processing sequence arranged is π=(7,6,8,10,4,3,9,1,5,2), can according to rule construct on earliest finish time production scheduling scheme as shown in Figure 5.
Dispatch case for the SMSP constructed by the data dispatching by the different production scheduling cycle, utilize genetic algorithm optimization to solve and obtain optimum Order Processing sequence.
The chromosome of genetic algorithm adopts the coding method based on arrangement, and chromosome is expressed as π=(π
1, π
2..., π
10), wherein π
i∈ o1, o2 ... o10}, i ∈ 1,2 ... 10}, decode procedure is, utilizes given rule on earliest finish time, is from left to right arranged successively by the order corresponding to each for chromosome gene to process, thus determine complete production scheduling scheme, and ask for and always delay the deadline to assess this chromosome accordingly.
The Population Size of genetic algorithm is set to 200, initial population stochastic generation, and crossover probability is set to 0.6, and mutation probability is set to 0.1.Algorithm end condition is: after continuous T time iteration, if optimum solution improves not yet, then algorithm stops, and T gets the half of all order numbers.
Dispatch case for the SMSP shown in table 3, the optimum Order Processing sequence of the one utilizing genetic algorithm to ask for is π=(1,9,10,2,8,3,7,6,5,4), and as shown in Figure 6, its completion date of always delaying is 30.76 to corresponding production scheduling scheme.
According to the relevant data dispatching in historical production data and the optimum Order Processing sequence being solved unit acquisition by genetic algorithm optimization, structure training dataset, to train Order Processing sequential forecasting models.
Case is dispatched, for optimum Order Processing sequence π=(1,9,10,2 for the SMSP shown in table 3,8,3,7,6,5,4), wherein a kth order (k ∈ 1,2 ..., 9}) can be given priority in arranging for relative to kth+1 ~ 10 orders and to be carried out processing.Therefore, namely this order and any one follow-up order can be used for structure training sample.If using a kth order as " order 1 ", the generic attribute value of training sample is 1; If using a kth order as " order 2 ", the generic attribute value of training sample is 0.
For order " o3 " and " o4 ", the constitution step of the input feature vector of training sample is described:
In the current production scheduling stage, existing 5 orders (o1, o9, o10, o2, o8) are arranged to carry out production and processing.According to rule on earliest finish time, order " o3 " and " o4 " all process being arranged to process equipment u3.Accordingly, the property value of relevant information for describing each order and real-time production scheduling environmental information is calculated:
(1) formula (1) is utilized to calculate the attribute actual process time of " o3 " and " o4 ".The actual process time of " o3 " is 18.93, and the actual process time of " o4 " is 20.82.
(2) formula (2) is utilized to calculate the order completion date attribute of " o3 " and " o4 ".The order completion date of " o3 " is 29.63, and the order completion date of " o4 " is 31.52.
(3) the Order Processing time utilizing formula (3) to calculate " o3 " and " o4 " compares attribute.The Order Processing time ratio of " o3 " is 1.06, and the Order Processing time ratio of " o4 " is 1.10.
(4) before order " o3 " and " o4 ", existing 5 orders are arranged to carry out processing, and therefore, current order completion is than being 0.5.
(5) attribute at delivery date of " o3 " is 31; The attribute at delivery date of " o4 " is 38.
For prediction should give priority in arranging in two orders which carry out production and processing, to need on the basis of existing attribute structure input feature vector further with the comparative information both describing.Input feature vector is as follows:
(1) feature Real_Proc_R is the ratio of the actual process time of two orders.
(2) feature Order_Comp_R is the ratio of the order completion date of two orders.
(3) feature Proc_Ratio_R is the ratio of the Order Processing time ratio of two orders.
(4) feature DD_R is the ratio of the dilivery date of two orders.
(5) feature Order_Num_R is current order completion ratio.
Two training samples of final structure are as shown in table 4, and class is generic attribute.
Table 4
| No. | Real_Proc_R | Order_Comp_R | Proc_Ratio_R | DD_R | Order_Num_R | class |
| 1 | 0.91 | 0.94 | 0.96 | 0.82 | 0.5 | 1 |
| 2 | 1.10 | 1.06 | 1.04 | 1.23 | 0.5 | 0 |
Wherein, training sample 1 is using " o3 " as " order 1 "; Training sample 2 is using " o4 " as " order 1 ".
The Order Processing sequential forecasting models obtained can regard a kind of Order Sorting for SMSP problem rule as, for verifying the scheduling performance of this model, by itself and genetic algorithm, the earliest delivery date rule (Earliest Due Date, EDD) and random Order Sorting rule (Random Rule, RR) contrast.
For the SMSP problem minimizing the deadline of always delaying, delivery date, rule can be used for the processing arrangement sequence generating order the earliest, and had good dispatching effect.Select the SMSP case corresponding to 10 production scheduling cycles as test cases.Comprise the order that 10 need to carry out production and processing in each SMSP case, the scheduling result of four kinds of methods is as shown in table 5.
Table 5
| Test cases | Genetic algorithm | SLFN rule | EDD rule | Random rule |
| 1 | 8.94 | 16.7 | 30.1 | 31.6 |
| 2 | 6.31 | 8.47 | 10.6 | 25.8 |
| 3 | 33.9 | 41.4 | 46.7 | 68.9 |
| 4 | 9.32 | 12.3 | 12.2 | 17.3 |
| 5 | 17.2 | 20 | 37.5 | 44.3 |
| 6 | 5.53 | 9.48 | 18.7 | 26 |
| 7 | 13.1 | 18.3 | 39.9 | 43.3 |
| 8 | 18.4 | 25.3 | 29 | 75.6 |
| 9 | 10 | 12.1 | 14.1 | 44.5 |
| 10 | 14 | 25.7 | 32 | 44.8 |
As shown in Table 5, on all test cases, SLFN rule all can reappear the Optimized Operation ability of genetic algorithm preferably, and is obviously better than EDD rule and random rule.
The scheduling performance of further employing following performance index checking SLFN model:
Wherein,
represent the average behavior desired value of genetic algorithm on test cases;
represent a kind of Order Sorting rule (SLFN rule, regular, the random rule of EDD) the average behavior desired value on test cases;
η is for representing that this Order Sorting rule is measured relative to the Optimal performance of this complex optimization algorithm of genetic algorithm.
The η result of calculation of three kinds of Order Sorting rules is as shown in table 6, and the η value of SLFN rule much smaller than the η value of EDD rule and random rule, thus further demonstrates the scheduling performance of constructed SLFN forecast model.
Table 6
| (SLFN) | (EDD) | (RR) |
| 36.8 | 110.3 | 309.8 |
On 10 test cases, employing SLFN rule, EDD are regular, random rule three kinds of methods build orders and arrange all can ignore the computing time required for sequence, and the computing time of genetic algorithm is as shown in table 7, the sequence number of the first behavior test cases in table 7, the second behavior computing time (s).
Table 7
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| 1.2168 | 1.6848 | 1.7316 | 3.3072 | 1.1700 | 2.3088 | 1.2012 | 2.6208 | 1.6068 | 1.8408 |
Be longer than the computing time of SLFN rule the computing time of genetic algorithm, when scheduling problem is larger, genetic algorithm often cannot meet the requirement of real-time of process of manufacture because solving overlong time, and constructed forecast model only needs few computing time and calculated amount can obtain preferably scheduling result, therefore have more practicality.
Claims (6)
1., based on a control method for the single phase multi-product batch processing of data, it is characterized in that, comprise the following steps:
(1) for different target batch process scenes, from historical production data, take out corresponding history operation and dispatching information, according to the difference of production scheduling task in the production scheduling cycle, set up SMSP respectively and dispatch case;
(2) the heuristic process equipment allocation rule of given one, then utilizes SMSP in genetic algorithm for solving step (1) to dispatch the optimum Order Processing sequence of case;
(3) utilize SMSP to dispatch case and corresponding optimum Order Processing sequence, build training dataset, training obtains Order Processing sequential forecasting models;
(4) utilize Order Processing sequential forecasting models to process pending batch processing order, obtain Order Processing and arrange sequence;
(5) the heuristic process equipment allocation rule of step (2) and the Order Processing of step (4) is utilized to arrange sequence, by pending batch processing Order splitting to different equipment, determine processing start time and end time, carry out production and processing.
2. as claimed in claim 1 based on the control method of the single phase multi-product batch processing of data, it is characterized in that, described Order Processing sequential forecasting models adopts Single hidden layer feedforward neural networks to express.
3. as claimed in claim 2 based on the control method of the single phase multi-product batch processing of data, it is characterized in that, the end condition of described genetic algorithm is: if after continuous T time iteration, current obtained optimum solution improves not yet, then algorithm stops, and T gets the half of all order numbers.
4. as claimed in claim 3 based on the control method of the single phase multi-product batch processing of data, it is characterized in that, relevant information and the corresponding dispatch environment information of each order adopt four attribute descriptions, are respectively: actual process time, order completion date, Order Processing time ratio and order completion ratio.
5. as claimed in claim 4 based on the control method of the single phase multi-product batch processing of data, it is characterized in that, described training data concentrates the input feature vector of each training sample to comprise five kinds, is respectively: the ratio of the order numbers that the ratio of actual process time, the ratio of completion date, the ratio of Order Processing time ratio, the ratio of dilivery date and current production scheduling stage have arranged and total orders.
6. as claimed in claim 5 based on the control method of the single phase multi-product batch processing of data, it is characterized in that, adopt extreme learning machine algorithm to carry out the training of Order Processing sequential forecasting models.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510070318.1A CN104636871B (en) | 2015-02-10 | 2015-02-10 | A kind of control method of the single phase multi-product batch processing based on data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510070318.1A CN104636871B (en) | 2015-02-10 | 2015-02-10 | A kind of control method of the single phase multi-product batch processing based on data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636871A true CN104636871A (en) | 2015-05-20 |
| CN104636871B CN104636871B (en) | 2018-02-23 |
Family
ID=53215590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510070318.1A Expired - Fee Related CN104636871B (en) | 2015-02-10 | 2015-02-10 | A kind of control method of the single phase multi-product batch processing based on data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636871B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105321042A (en) * | 2015-10-19 | 2016-02-10 | 金航数码科技有限责任公司 | Genetic algorithm-based advanced plan scheduling system and method |
| CN108122055A (en) * | 2016-11-28 | 2018-06-05 | 北京理工大学 | The resource regulating method and device of a kind of Flow Shop |
| CN108985617A (en) * | 2018-07-11 | 2018-12-11 | 广东人励智能工程有限公司 | A kind of product manufacturing process dispatching method and system based on intelligence manufacture |
| CN109154809A (en) * | 2016-03-16 | 2019-01-04 | 通快机床两合公司 | Production programming system and method |
| CN110059886A (en) * | 2019-04-25 | 2019-07-26 | 哈尔滨理工大学 | Consider the integrated dispatch method that the single group process of equipment batch processing terminates simultaneously |
| WO2021037025A1 (en) * | 2019-08-30 | 2021-03-04 | 京东数字科技控股股份有限公司 | Method and apparatus for predicting product scheduling time |
| CN112907068A (en) * | 2021-02-09 | 2021-06-04 | 刘连英 | Method for the batch production of fastener groups |
| CN113031543A (en) * | 2021-02-24 | 2021-06-25 | 同济大学 | Control scheduling method and device for semiconductor production line |
| CN116523151A (en) * | 2023-07-05 | 2023-08-01 | 深圳市鑫冠亚科技有限公司 | Electrode production management method, system and storage medium based on artificial intelligence |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271543A (en) * | 2008-04-23 | 2008-09-24 | 永凯软件技术（上海）有限公司 | Production scheduling system and method using genetic algorithm based on elite solution pool |
| EP2610696A1 (en) * | 2010-08-27 | 2013-07-03 | Hitachi, Ltd. | Process design/production planning device |
| CN103309316A (en) * | 2013-05-28 | 2013-09-18 | 北京理工大学 | Scheduling method of multi-stage variation hybrid flow shop with batch processor |
-
2015
- 2015-02-10 CN CN201510070318.1A patent/CN104636871B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271543A (en) * | 2008-04-23 | 2008-09-24 | 永凯软件技术（上海）有限公司 | Production scheduling system and method using genetic algorithm based on elite solution pool |
| EP2610696A1 (en) * | 2010-08-27 | 2013-07-03 | Hitachi, Ltd. | Process design/production planning device |
| CN103309316A (en) * | 2013-05-28 | 2013-09-18 | 北京理工大学 | Scheduling method of multi-stage variation hybrid flow shop with batch processor |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105321042A (en) * | 2015-10-19 | 2016-02-10 | 金航数码科技有限责任公司 | Genetic algorithm-based advanced plan scheduling system and method |
| CN105321042B (en) * | 2015-10-19 | 2017-02-22 | 金航数码科技有限责任公司 | Genetic algorithm-based advanced plan scheduling system and method |
| CN109154809B (en) * | 2016-03-16 | 2021-12-31 | 通快机床两合公司 | Production planning system and method |
| CN109154809A (en) * | 2016-03-16 | 2019-01-04 | 通快机床两合公司 | Production programming system and method |
| CN108122055A (en) * | 2016-11-28 | 2018-06-05 | 北京理工大学 | The resource regulating method and device of a kind of Flow Shop |
| CN108985617B (en) * | 2018-07-11 | 2021-07-13 | 广东人励智能工程有限公司 | Product production flow scheduling method and system based on intelligent manufacturing |
| CN108985617A (en) * | 2018-07-11 | 2018-12-11 | 广东人励智能工程有限公司 | A kind of product manufacturing process dispatching method and system based on intelligence manufacture |
| CN110059886A (en) * | 2019-04-25 | 2019-07-26 | 哈尔滨理工大学 | Consider the integrated dispatch method that the single group process of equipment batch processing terminates simultaneously |
| WO2021037025A1 (en) * | 2019-08-30 | 2021-03-04 | 京东数字科技控股股份有限公司 | Method and apparatus for predicting product scheduling time |
| CN112907068A (en) * | 2021-02-09 | 2021-06-04 | 刘连英 | Method for the batch production of fastener groups |
| CN113031543A (en) * | 2021-02-24 | 2021-06-25 | 同济大学 | Control scheduling method and device for semiconductor production line |
| CN116523151A (en) * | 2023-07-05 | 2023-08-01 | 深圳市鑫冠亚科技有限公司 | Electrode production management method, system and storage medium based on artificial intelligence |
| CN116523151B (en) * | 2023-07-05 | 2024-01-09 | 深圳市鑫冠亚科技有限公司 | Electrode production management method, system and storage medium based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636871B (en) | 2018-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104636871A (en) | Data-based single-stage multi-product scheduling control method | |
| Gan et al. | Joint optimization of maintenance, buffer, and spare parts for a production system | |
| CN104035816A (en) | Cloud computing task scheduling method based on improved NSGA-II | |
| CN103927628A (en) | Order management system and order management method oriented to customer commitments | |
| CN104842564A (en) | NSGA-II-based three-dimensional printing multi-task optimal scheduling method | |
| Wei et al. | Research on cloud design resources scheduling based on genetic algorithm | |
| CN102298737A (en) | Customer commitment-oriented order management system and method thereof | |
| CN107146039A (en) | The customized type mixed-model assembly production method and device of a kind of multiple target Collaborative Control | |
| Yue et al. | Hybrid Pareto artificial bee colony algorithm for multi-objective single machine group scheduling problem with sequence-dependent setup times and learning effects | |
| Sun et al. | A teaching-learning-based optimization with feedback for LR fuzzy flexible assembly job shop scheduling problem with batch splitting | |
| Zhang et al. | Effective dispatching rules mining based on near-optimal schedules in intelligent job shop environment | |
| Więcek | Intelligent approach to inventory control in logistics under uncertainty conditions | |
| Keshavarz et al. | Efficient upper and lower bounding methods for flowshop sequence-dependent group scheduling problems | |
| CN104281917A (en) | Fuzzy job-shop scheduling method based on self-adaption inheritance and clonal selection algorithm | |
| Zeng et al. | Multi-skilled worker assignment in seru production system for the trade-off between production efficiency and workload fairness | |
| CN110675055A (en) | Automatic production line modeling and layout planning method and system | |
| Gan et al. | A multi-objective evolutionary algorithm for emergency logistics scheduling in large-scale disaster relief | |
| CN113485278B (en) | Flexible job shop scheduling multi-target distribution estimation method for optimizing two production indexes | |
| Mei et al. | A method for man hour optimisation and workforce allocation problem with discrete and non-numerical constraints in large-scale one-of-a-kind production | |
| Zhipeng et al. | Small‐World Optimization Algorithm and Its Application in a Sequencing Problem of Painted Body Storage in a Car Company | |
| Jiang et al. | Improved heuristic algorithm for modern industrial production scheduling | |
| CN114819558A (en) | Dual-target scheduling optimization method for distributed mixed flow shop | |
| Zhou et al. | A modified column generation algorithm for scheduling problem of reentrant hybrid flow shops with queue constraints | |
| CN112734286B (en) | Workshop scheduling method based on multi-strategy deep reinforcement learning | |
| Quan et al. | Multi-objective evolutionary scheduling based on collaborative virtual workflow model and adaptive rules for flexible production process with operation reworking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180223 Termination date: 20210210 |