CN115619007A - Intelligent manufacturing scheduling method and device, electronic equipment and medium - Google Patents
Intelligent manufacturing scheduling method and device, electronic equipment and medium Download PDFInfo
- Publication number
- CN115619007A CN115619007A CN202211167448.3A CN202211167448A CN115619007A CN 115619007 A CN115619007 A CN 115619007A CN 202211167448 A CN202211167448 A CN 202211167448A CN 115619007 A CN115619007 A CN 115619007A
- Authority
- CN
- China
- Prior art keywords
- product
- cost
- production
- target
- order information
- 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
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 259
- 238000000034 method Methods 0.000 title claims abstract description 92
- 230000002787 reinforcement Effects 0.000 claims abstract description 75
- 238000012545 processing Methods 0.000 claims abstract description 66
- 239000013598 vector Substances 0.000 claims description 90
- 230000006870 function Effects 0.000 claims description 26
- 230000015654 memory Effects 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 230000005284 excitation Effects 0.000 claims description 9
- 230000002441 reversible effect Effects 0.000 claims description 8
- 230000010365 information processing Effects 0.000 claims description 4
- 238000005457 optimization Methods 0.000 abstract description 14
- 238000005516 engineering process Methods 0.000 abstract description 9
- 206010012186 Delayed delivery Diseases 0.000 abstract description 8
- 230000007547 defect Effects 0.000 abstract description 7
- 239000000047 product Substances 0.000 description 173
- 230000008569 process Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 8
- 239000011265 semifinished product Substances 0.000 description 8
- 230000003111 delayed effect Effects 0.000 description 6
- 238000013459 approach Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an intelligent manufacturing scheduling method and device, electronic equipment and a medium. By applying the technical scheme of the application, the total cost including the product processing cost, the staff overtime cost, the product delayed delivery cost and the product inventory cost can be taken as an optimization target. And outputting a scheduling plan corresponding to the processing order information at the lowest cost for the user by combining with a deep reinforcement learning algorithm. Therefore, on one hand, the scheduling scheme of the manufacturer in a period of time can be automatically given in real time. On the other hand, the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost of staff in the production process is not considered in the production scheduling scheme in the related technology is also avoided.
Description
Technical Field
The present application relates to a product planning generation technology, and in particular, to an intelligent manufacturing scheduling method, apparatus, electronic device, and medium.
Background
Along with the rapid development of new generation information technologies such as industrial internet, big data, artificial intelligence, etc., the theory of intelligent manufacturing is also continuously advancing, and intelligent manufacturing has become one of the core technologies for enhancing the overall competitiveness of the manufacturing industry at present.
In the intelligent manufacturing industry in the related art, planning and scheduling orders are an important part of intelligent manufacturing. For example, the optimized scheduling scheme in a period can be given by integrating materials, product production processes, machines, staff and the like into an intelligent scheduling system and using various optimization algorithms by taking a customer order in the period as input.
However, in the existing production process according to product planning, the production cost of staff in the production process is not considered, which also causes the output scheduling scheme to be inaccurate, and further affects the business processing progress.
Disclosure of Invention
The embodiment of the application provides an intelligent manufacturing scheduling method and device, electronic equipment and a medium. The method is used for solving the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost problem of staff in the production process is not considered in the production scheduling scheme in the related technology.
According to an aspect of the embodiments of the present application, there is provided an intelligent manufacturing scheduling method, wherein:
acquiring manufacturer basic information and order information in a target time period, wherein the manufacturer basic information comprises staff production information used for reflecting staff overtime cost, and the order information comprises order quantity, product delivery date and product delay cost;
obtaining a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of a production machine or the production state of a product;
taking the plurality of feature vectors as the input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, wherein the target scheduling plan is used for reflecting a target product range required to be produced for processing the order information and a corresponding target production machine under the lowest cost;
performing scheduling manufacturing on the order information based on the target scheduling plan;
wherein the target scheduling plan comprises one of:
selecting a product range with minimum delay time and a scheduling plan of a corresponding production machine, selecting a product range with minimum time ratio of residual delivery and residual production and a scheduling plan of a corresponding production machine, selecting a product range with maximum delay time and a scheduling plan of a corresponding production machine, randomly selecting a product range and a scheduling plan of a corresponding production machine, and selecting a product range with maximum estimated delay time and a scheduling plan of a corresponding production machine.
Optionally, in another embodiment based on the foregoing method of the present application, the obtaining a plurality of state feature vectors based on the vendor basic information and the order information includes:
inputting the manufacturer basic information and the order information into an intelligent manufacturing and scheduling system to obtain a plurality of state feature vectors matched with the manufacturer basic information and the order information;
wherein the vendor basis information comprises:
the processing time of the production machine, the attendance information of staff, the number of the production machines on different production lines, the production cost and the overtime cost of the staff on different production lines, and the cost information of products.
Optionally, in another embodiment based on the foregoing method of the present application, the state feature vector includes:
the order information processing system comprises an average utilization rate vector used for reflecting the resource utilization rate of the production machines, a utilization rate variance vector used for reflecting the load balance state of each production machine, a total procedure completion rate vector used for reflecting the processing amount progress of the order information and vectors of different product completion rates, and an estimated delay rate vector used for reflecting the delay progress of the order information and an actual delay rate vector.
Optionally, in another embodiment of the method according to the present application, after performing scheduling manufacturing on the order information based on the target scheduling plan, the method further includes:
determining the production duration consumed by the target production machine to produce the products with the target product quantity based on the target product range reflected by the target scheduling plan and the corresponding target production machine;
calculating overtime cost corresponding to the target scheduling plan based on the initial processing time point and the production duration of the target production machine; and obtaining a product delay cost included in the order information;
and optimizing the reinforcement learning model based on the magnitude relation between the overtime cost and the product postponing cost.
Optionally, in another embodiment of the method according to the present application, the optimizing the reinforcement learning model based on a magnitude relationship between the overtime cost and the product postponement cost includes:
if the overtime cost is detected to be larger than the product delay cost, marking an excitation function corresponding to the target scheduling plan as a negative value;
and optimizing the reinforcement learning model by using the excitation function marked as a negative value.
Optionally, in another embodiment of the method according to the present application, after performing scheduling manufacturing on the order information based on the target scheduling plan, the method further includes:
selecting a first number of sample sets from a data sample pool;
and performing reverse gradient propagation training on the reinforcement learning model by using a mean square error loss function and the sample set, and updating parameters of the reinforcement learning model until the reinforcement learning model is completely trained.
Optionally, in another embodiment based on the foregoing method of the present application, the performing inverse gradient propagation training on the reinforcement learning model includes:
taking an average utilization rate vector for reflecting the resource utilization rate of the production machine, and an estimated delay rate vector and an actual delay rate vector for reflecting the delay progress of the order information as a reward function;
and carrying out reverse gradient propagation training on the reinforcement learning model by utilizing the reward function.
According to another aspect of the embodiments of the present application, there is provided an intelligent manufacturing scheduling apparatus, wherein:
the system comprises an acquisition module and a display module, wherein the acquisition module is configured to acquire manufacturer basic information and order information in a target time period, the manufacturer basic information comprises staff production information used for reflecting staff overtime cost, and the order information comprises order quantity, product delivery date and product delay cost;
the conversion module is configured to obtain a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of a production machine or the production state of a product;
an output module, configured to take the plurality of feature vectors as an input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, where the target scheduling plan is used to reflect a target product range and a corresponding target production machine required to be produced for processing the order information at a lowest cost;
a processing module configured to perform scheduling manufacturing on the order information based on the target scheduling plan;
wherein the target scheduling plan comprises one of:
selecting a product range with minimum delay time and a scheduling plan of a corresponding production machine, selecting a product range with minimum time ratio of residual delivery and residual production and a scheduling plan of a corresponding production machine, selecting a product range with maximum delay time and a scheduling plan of a corresponding production machine, randomly selecting a product range and a scheduling plan of a corresponding production machine, and selecting a product range with maximum estimated delay time and a scheduling plan of a corresponding production machine.
According to another aspect of the embodiments of the present application, there is provided an electronic device including:
a memory for storing executable instructions; and
a display for communicating with the memory to execute the executable instructions to perform the operations of any of the intelligent manufacturing scheduling methods described above.
According to yet another aspect of the embodiments of the present application, there is provided a computer-readable storage medium for storing computer-readable instructions, which when executed perform the operations of any one of the above-mentioned smart manufacturing scheduling methods.
In the application, the manufacturer basic information of the employee production information for reflecting the overtime cost of the employee and the order information comprising the order quantity, the product delivery date and the product delay cost can be obtained; obtaining a plurality of state feature vectors based on manufacturer basic information and order information; the characteristic vectors are used as the input of a reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the characteristic vectors, and the target scheduling plan is used for reflecting the quantity of target products required to be produced for processing order information and corresponding target production machines at the lowest cost; and performing scheduling manufacturing on the order information based on the target scheduling plan. By applying the technical scheme of the application, the total cost including the product processing cost, the employee overtime cost, the product delayed delivery cost and the product inventory cost can be taken as an optimization target. And the scheduling plan corresponding to the order information is processed at the lowest cost by combining with the deep reinforcement learning algorithm. Therefore, on one hand, the scheduling scheme of the manufacturer in a period of time can be automatically given in real time. On the other hand, the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost of staff in the production process is not considered in the production scheduling scheme in the related technology is avoided.
The technical solution of the present application is further described in detail by the accompanying drawings and examples.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description, serve to explain the principles of the application.
The present application may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram illustrating an intelligent manufacturing scheduling method according to an embodiment of the present application;
FIG. 2 is a schematic diagram illustrating a training process of reinforcement learning model parameters according to an embodiment of the present application;
FIG. 3 is a schematic diagram illustrating a network architecture of a reinforcement learning model provided in an embodiment of the present application;
FIG. 4 is a flow chart illustrating an actual application of a reinforcement learning model provided by an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application;
fig. 7 is a schematic diagram of a storage medium according to an embodiment of the present application.
Detailed Description
Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be discussed further in subsequent figures.
In addition, technical solutions between the various embodiments of the present application may be combined with each other, but it must be based on the realization of the technical solutions by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not within the protection scope of the present application.
It should be noted that all directional indicators (such as upper, lower, left, right, front, and rear … …) in the present embodiment are only used to explain the relative position relationship between the components, the motion situation, etc. in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indicator is changed accordingly.
A method for performing intelligent manufacturing scheduling according to an exemplary embodiment of the present application is described below in conjunction with fig. 1-4. It should be noted that the following application scenarios are merely illustrated for the convenience of understanding the spirit and principles of the present application, and the embodiments of the present application are not limited in this respect. Rather, embodiments of the present application may be applied to any scenario where applicable.
The application also provides an intelligent manufacturing scheduling method, an intelligent manufacturing scheduling device, electronic equipment and a medium.
Fig. 1 schematically shows a flow diagram of an intelligent manufacturing scheduling method according to an embodiment of the present application. As shown in fig. 1, the method includes:
s101, acquiring manufacturer basic information and order information in a target time period, wherein the manufacturer basic information comprises staff production information used for reflecting staff overtime cost, and the order information comprises order quantity, product delivery date and product delay cost.
S102, obtaining a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of a production machine or the production state of a product.
S103, taking the plurality of feature vectors as the input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, wherein the target scheduling plan is used for reflecting the quantity of target products required to be produced for processing order information and corresponding target production machines at the lowest cost.
And S104, performing scheduling manufacturing on the order information based on the target scheduling plan.
In the related art, the current generation planning algorithm tends to solve the product planning and planning problem in a short period (such as within one day), so that the maximum machine utilization rate or the minimum product delay is taken as an optimization target of the generation planning algorithm, and the problem is solved by adopting a heuristic algorithm such as an accurate algorithm or a genetic algorithm.
However, the current algorithm has certain technical defects. Firstly, only solving the planning problem in a short period requires determining the number of products to be processed each day in advance, which requires additional personnel to participate in actual production, and reduces the degree of automation.
In addition, many current studies neglect the constraint of the attendance time of the staff, do not consider the influence of the staff on the product processing sequence when the staff goes on and off duty, and do not consider the overtime cost which can be brought by the staff to carry out the process processing in the non-working time.
Moreover, although the operational research optimization algorithm has high solving precision and can ensure that the optimal solution is obtained, the solving time is extremely long, and the optimal solution can be obtained only after several hours, which is unacceptable in actual production. The genetic algorithm is one of the most common optimization algorithms, has short running time, can ensure the quality of the solution, and is widely applied to planning and planning algorithms. However, when a planning problem with a large scale is solved, the solving efficiency is not satisfactory, and the solving quality also decreases with the increase of the scale. In addition, although the heuristic programming method based on priority scheduling programming has the advantage of high solving speed, the generalization performance is poor, the rules which are well represented in a certain scene cannot guarantee the same effect in other scenes, and after all, no rule is optimal in all scenes.
Finally, it is not comprehensive to consider only machine utilization and pull-outs in optimizing objectives. In actual production, the pull-in may result in delayed delivery costs, but completing the process ahead of time may also result in inventory costs, and producing half of the product may result in semi-finished inventory costs due to being left on the production line. These costs are of great concern to the administrator.
In view of the above existing problems, the embodiment of the present application provides an intelligent manufacturing scheduling method, which takes total costs including product processing cost, employee overtime cost, product deferred delivery cost, and product inventory cost as optimization targets, and combines a scheduling plan corresponding to processing order information at the lowest cost output for a user based on a deep reinforcement learning algorithm.
As shown in fig. 2, the intelligent manufacturing scheduling method proposed in the present application is specifically explained as follows:
The manufacturer production information and the manufacturer product information comprise:
(1) The processing time of each process of each production task on different production machines.
(2) The attendance time of employees on different production lines.
(3) The capacity of different production lines is represented by the number of production machines on different production lines.
(4) Normal production costs and staff overtime costs on different production lines.
(5) Inventory costs for different types of products and semi-finished inventory costs.
It can be understood that, in actual production, the above data are generally stable and do not change frequently, so the embodiment of the present application integrates the above data as parameters into an algorithm.
And step 2, obtaining order information.
The order information may be order information within a target time period (e.g., within one day or one week), wherein the order information includes an order quantity, a product delivery date, and a product deferral cost.
And 3, obtaining a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of the production machine or the production state of the product.
Further, the method can be based on the order quantity, the product delivery date and the product delay cost which are included in the order information; and pre-collecting manufacturer production information and manufacturer product information, and inputting the manufacturer production information and the manufacturer product information into the intelligent manufacturing scheduling system to obtain a plurality of state feature vectors matched with the manufacturer basic information and the order information.
It should be noted that the following parameters are explained:
based on the above parameters, the following describes the state feature vector proposed in the present application:
(1) And the average utilization rate vector is used for reflecting the utilization rate of the resources of each production machine in the manufacturer. Is marked as Uave, wherein,
Wherein,
(2) The utilization variance vector represents the balance degree of loads on different production machines in the current system and is recorded as U std 。
Wherein,
(3) And the total process completion rate is used for describing how much the current planning of the order is carried out, wherein the approach of 0 indicates that the planning is just started, and the approach of 1 indicates that the planning is about to be finished and is marked as CRT.
Wherein,
(4) The average completion rate of each product is used for describing the completion degree of the current plan and is marked as CRJ ave . Wherein,
wherein CRJ i The process completion rate of product i is shown. And,
(5) The variance of the completion rates of different products can describe the balance degree of completion of different products, and the larger the difference of the completion degrees of different products is, the larger the difference is, the difference is recorded as CRJ std 。
Wherein,
(6) The estimated delay rate of the product means that although there is no delay in the current schedule, the current time plus the remaining processing time will exceed the delivery date, i.e., a delay will occur in the future. The estimated delay rate is recorded as Te.
Wherein,
N left indicating the number of products not yet finished, N te Indicating that a delayed delivery of the product is likely to occur.
(7) Actual delay rate of the product. By actual delay is meant that the product has been delayed at the current time. The actual delay rate is denoted as Ta.
And 4, acquiring a preset reinforcement learning model.
The reinforcement learning model in the embodiment of the application adopts a DQN structure, and DQN is the most common algorithm in reinforcement learning and is widely applied to multiple fields.
In one approach, the present application may employ a Double DQN algorithm to achieve output of a target scheduling plan. The Double DQN algorithm is optimized on the basis of DQN, the correlation between a data sample and a network before training is solved by using two identical neural networks, and the problem of overestimation caused by that a Q value is too late to approach a possible optimization target in the training process is avoided.
In one mode, there are 5 hidden layers between the input layer and the output layer of the reinforcement learning model in the embodiment of the present application, the number of nodes of each hidden layer is 30, and the activation function is tanh. The neural structure of the Q network is shown in fig. 3.
And 5, taking the plurality of feature vectors as the input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors.
Further, the embodiment of the present application utilizes a reinforcement learning model to implement a target scheduling plan corresponding to the lowest cost according to different state feature vector outputs. That is, the target scheduling plan is the range of target products and corresponding target production machines that need to be produced to process the order information at the lowest cost.
In particular, the target scheduling plan is one of a plurality of scheduling plans. According to the method and the device, the plurality of feature vectors can be used as the input of the reinforcement learning model, and the target scheduling plan matched with the feature vectors is selected from the plurality of scheduling plans.
Wherein, a plurality of scheduling plans in the present application include the following:
(1): and selecting the minimum delay time product range and the scheduling plan of the corresponding production machine.
Wherein the product with the smallest average delayed relaxation time + the production machine that finishes the processing the earliest can also be selected. For example, the actual delayed product set D, the incomplete product set U, may be remembered. If D is not null, selecting the product with the largest actual delay time; if D is empty, a set S of products that have been processed is selected from the unfinished set U.
Through the strategy, the prior processing of the products processed in the existing working procedures is guaranteed, and the phenomenon that the time interval between different working procedures of the same product is too large is avoided, so that the inventory cost of semi-finished products is reduced. The product with the smallest average delay and relaxation time, i.e. the product with the most delay, is selected from the product set S, and the average delay and relaxation time of the product is equal to the remaining processing time divided by the number of the remaining processes.
In addition, when the production machine is selected, the production machine that completes the processing earliest is selected. The time for starting processing of the product on each production machine is determined by measuring the overtime cost and the delay cost in consideration of the working time of the production machine. In one mode, if the overtime cost is higher than the cost brought by the delay, the overtime is not worth, and the product is processed after being delayed to work again. If the overtime cost is lower than the delay cost, the delay cost caused by the sequential delay processing is too large, and the processing operation of the process is more cost-effective when the process is started immediately.
(2) The product range and the scheduling plan of the corresponding production machine are selected for the minimum time ratio of remaining delivery to remaining production.
That is, the product with the smallest ratio of the remaining delivery time to the remaining processing time + the production machine that completes the processing earliest is selected. For example, delay set D, incomplete product set U, may be noted. If D is not null, selecting the product with the largest estimated delay time; if D is empty, a set S of products that have been processed by a process is selected from the incomplete set U, and the product with the smallest ratio of the relaxation time to the remaining processing time is selected in S.
It will be appreciated that the smaller the ratio, the more difficult it is to deliver the product on time. If S is null, it is selected in U. When a production machine is selected, the production machine that completes the processing earliest is selected.
(3) And selecting the product range with the maximum delay time and the scheduling plan of the corresponding production machine.
That is, the product with the greatest delay time + the production machine with the lowest utilization or the greatest load is selected. Wherein, for all products which have not been processed, the possible delay time is calculated, and the product with the largest delay time is selected, preferably the product which has been processed by a part of the process. When selecting a production machine, the production machine with the lowest utilization rate is selected with a probability of 50%, and the production machine with the highest workload is selected with a certain probability (for example, 50%).
(4) And selecting the product range with the maximum delay time and the scheduling plan of the corresponding production machine.
That is, it is also the case that the product with the largest delay time is selected, but on the other hand, the production machine that completes the processing the earliest needs to be selected. Wherein, for all the products which are not finished, the possible delay time is calculated, and the product with the maximum delay time is selected, and the products which are already processed by a part of the process are preferentially selected. When a production machine is selected, the production machine that completes the processing earliest is selected.
(5) And randomly selecting a product range and a scheduling plan of a corresponding production machine.
I.e. randomly selecting a ready product. When a production machine is selected, the production machine that completes the processing earliest is selected.
(6) And selecting the product range with the maximum estimated delay time and the scheduling plan of the corresponding production machine.
That is, the product with the largest estimated delay plus the production machine that completed processing earliest is selected. The estimated delay degree is expressed by the product of the estimated delay time and the reciprocal of the product completion rate, and the larger the value is, the more serious the delay of the product is.
For example, note delay set D, unfinished product set U. If D is not null, selecting the product with the maximum estimated delay degree; if D is empty, then a set S of products that have been processed is selected from the incomplete set U, and the product with the smallest product of the remaining delivery time and the product completion rate is selected in S. And if S is empty, selecting from U. When a production machine is selected, the production machine that completes the processing earliest is selected.
From the above, the order information in a period of time is obtained currently, and according to the order quantity, the product delivery date and the product delay cost contained in the order information, the current manufacturer production information and the manufacturer product information of the manufacturer, the manufacturer can obtain the order information in the period of time. And converting the order information into the corresponding production state of a production machine which processes the order information by the manufacturer, or the production state characteristic vector of the product. And automatically outputting a scheduling plan corresponding to the state feature vector by combining a mode of a reinforcement learning model so as to complete the planning processing of the order information based on the scheduling plan subsequently.
In the application, the manufacturer basic information of the employee production information for reflecting the overtime cost of the employee and the order information comprising the order quantity, the product delivery date and the product delay cost can be obtained; obtaining a plurality of state feature vectors based on manufacturer basic information and order information; the characteristic vectors are used as input of a reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the characteristic vectors, and the target scheduling plan is used for reflecting the quantity of target products required to be produced for processing order information and corresponding target production machines at the lowest cost; and performing scheduling manufacturing on the order information based on the target scheduling plan.
By applying the technical scheme of the application, the total cost including the product processing cost, the employee overtime cost, the product delayed delivery cost and the product inventory cost can be taken as an optimization target. And the scheduling plan corresponding to the order information is processed at the lowest cost by combining with the deep reinforcement learning algorithm. Therefore, on one hand, the scheduling scheme of the manufacturer in a period of time can be automatically given in real time. On the other hand, the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost of staff in the production process is not considered in the production scheduling scheme in the related technology is also avoided.
Optionally, in another embodiment based on the foregoing method of the present application, obtaining a plurality of state feature vectors based on the vendor basic information and the order information includes:
inputting the manufacturer basic information and the order information into an intelligent manufacturing and scheduling system to obtain a plurality of state feature vectors matched with the manufacturer basic information and the order information;
wherein the vendor basis information comprises:
the processing time of the production machine, the attendance information of staff, the number of the production machines on different production lines, the production cost and the overtime cost of the staff on different production lines, and the cost information of products.
Optionally, in another embodiment based on the foregoing method of the present application, the state feature vector includes:
the order information processing system comprises an average utilization rate vector used for reflecting the resource utilization rate of the production machines, a utilization rate variance vector used for reflecting the load balance state of each production machine, a total procedure completion rate vector used for reflecting the processing amount progress of the order information and vectors of different product completion rates, and an estimated delay rate vector used for reflecting the delay progress of the order information and an actual delay rate vector.
Optionally, in another embodiment of the method according to the present application, after performing scheduling manufacturing on the order information based on the target scheduling plan, the method further includes:
determining the production duration consumed by the target production machine to produce the products with the target product quantity based on the target product range reflected by the target scheduling plan and the corresponding target production machine;
calculating overtime cost corresponding to the target scheduling plan based on the initial processing time point and the production duration of the target production machine; and obtaining a product delay cost included in the order information;
and optimizing the reinforcement learning model based on the magnitude relation between the overtime cost and the product postponing cost.
Optionally, in another embodiment of the method according to the present application, the optimizing the reinforcement learning model based on a magnitude relationship between the overtime cost and the product postponement cost includes:
if the overtime cost is detected to be larger than the product delay cost, marking an excitation function corresponding to the target scheduling plan as a negative value;
and optimizing the reinforcement learning model by using the excitation function marked as a negative value.
In one mode, after the target scheduling plan output by the reinforcement learning model is obtained, the overtime cost and the corresponding postponed cost of processing the target product and the corresponding target production machine reflected by the target scheduling plan can be further compared, so that the model is optimized according to the comparison result.
Specifically, if it is determined that the overtime cost required for executing the target scheduling plan is large (i.e., greater than the deferral cost), it indicates that the cost consumed by the target scheduling plan is large. Therefore, the embodiment of the application can mark the result (i.e. the target scheduling plan) output by the reinforcement learning model as a negative excitation function, so that the reinforcement learning model is prevented from outputting the target scheduling plan for the order information subsequently. And further, the purposes of reducing the delay cost, overtime cost, starting cost of a production machine and stocking cost of semi-finished products are achieved, and therefore the optimal output target of the reinforced model is ensured.
Optionally, in another embodiment of the method according to the present application, after performing scheduling manufacturing on the order information based on the target scheduling plan, the method further includes:
selecting a first number of sample sets from a data sample pool;
and performing reverse gradient propagation training on the reinforcement learning model by using a mean square error loss function and the sample set, and updating parameters of the reinforcement learning model until the reinforcement learning model is completely trained.
When the scheduling is executed each time, the system state is updated and the reward is acquired through the scheduling plan given by the execution model, and a quadruple (s, a, r, s) consisting of the original system state s, the action a, the reward r and the updated system state s _ is obtained. The quadruples are stored into a pool of data samples. So that after each subsequent scheduling, a certain number of sample sets are selected from the data sample pool and the mean square error loss function (y) is used j -Q(v j ,a j )) 2 And carrying out reverse gradient propagation on the reinforcement learning model to update the parameters of the network Q, and continuously learning until the training termination condition is met.
Further, after the trained reinforcement learning model is obtained, the reinforcement learning model can be stored. So that the intelligent manufacturing scheduling method provided by the application can be further executed by utilizing the trained reinforcement learning model. As an example, the performing step may be as shown in fig. 4.
Optionally, in another embodiment based on the above method of the present application, the training of inverse gradient propagation for the reinforcement learning model includes:
taking an average utilization rate vector for reflecting the resource utilization rate of the production machine, and an estimated delay rate vector and an actual delay rate vector for reflecting the delay progress of the order information as a reward function;
and performing inverse gradient propagation training on the reinforcement learning model by using the reward function.
In one way, the embodiment of the application can further perform optimization training on the reinforcement learning model by designing a reward function. The reward function includes three parts, namely, the actual delay rate of the product, the estimated delay rate and the average utilization rate of the production machine.
It will be appreciated that the arrangement of the process steps as compact as possible can be achieved by using these three feature vectors as reward functions to reduce the overall length of time required for processing. The machining process is as compact as possible, so that the delay cost, the overtime cost, the starting cost of a production machine and the stock cost of semi-finished products can be effectively reduced, and the optimal overall target is ensured.
In an alternative scheme, in order to further optimize the inventory cost, the embodiment of the application may further perform an operation of judging whether the start processing time of the process can be delayed or not for each process from the back to the front for the processing sequence on different production machines. And further determining whether the delay of the operation can reduce the inventory cost without influencing other costs (such as overtime cost and delay cost) according to the judgment result. If so, the start time of the procedure is postponed, otherwise no adjustment is made.
By applying the technical scheme of the application, the total cost including the product processing cost, the employee overtime cost, the product delayed delivery cost and the product inventory cost can be taken as an optimization target. And outputting a scheduling plan corresponding to the processing order information at the lowest cost for the user by combining with a deep reinforcement learning algorithm. Therefore, on one hand, the scheduling scheme of the manufacturer in a period of time can be automatically given in real time. On the other hand, the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost of staff in the production process is not considered in the production scheduling scheme in the related technology is also avoided.
It can be understood that the intelligent manufacturing scheduling method provided by the application adds constraints related to the attendance cost of the staff to the reinforcement learning model. And the total cost including the product processing cost, the staff overtime cost, the product delayed delivery cost, the product inventory cost and the semi-finished product inventory cost is taken as an optimization target. And further, more accurate output of the scheduling scheme is realized.
In addition, since the action set of the reinforcement learning algorithm proposed by the present application is expressed in the form of rules, the rules include both the rules for selecting products and the rules for selecting production machines. In the rules for selecting products, the products are classified into three categories, namely, products that have been postponed, unfinished products, and semi-finished products. When selecting products, preferentially selecting products from delay products so as to reduce delay cost as much as possible; secondly, selecting from semi-finished products to reduce the warehousing cost of the semi-finished products; when both product sets are empty, the product is selected from the unfinished product.
Secondly, when determining the starting processing time of a certain product on a certain production machine in the reinforced learning scheduling environment provided by the application, the comparison method is also adopted to select the starting processing time which can minimize the total cost.
Finally, in order to reduce the product inventory cost and the semi-finished product inventory cost in the reinforcement learning scheduling environment, a heuristic optimization method is designed, a judgment operation is executed on each procedure from back to front for processing sequences on different production machines, whether the processing starting time of the procedure can be delayed or not is judged, and whether the delay can reduce the inventory cost while not influencing other costs (such as overtime cost and delay cost) or not is judged. If so, the start time of the procedure is postponed, otherwise no adjustment is made.
Optionally, in another embodiment of the present application, as shown in fig. 5, the present application further provides an intelligent manufacturing scheduling apparatus. Which comprises the following steps:
an obtaining module 201 configured to obtain manufacturer basic information and order information in a target time period, wherein the manufacturer basic information includes staff production information used for reflecting staff overtime cost, and the order information includes order quantity, product delivery date and product delay cost;
a conversion module 202 configured to obtain a plurality of status feature vectors based on the manufacturer basic information and the order information, wherein each status feature vector is used for reflecting a production status of a production machine or a production status of a product;
an output module 203, configured to take the plurality of feature vectors as an input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, where the target scheduling plan is used to reflect a target product range and a corresponding target production machine required to be produced for processing the order information at a lowest cost;
a processing module 204 configured to perform scheduling manufacturing on the order information based on the target scheduling plan;
wherein the target scheduling plan comprises one of:
selecting a product range with minimum delay time and a scheduling plan of a corresponding production machine, selecting a product range with minimum time ratio of residual delivery and residual production and a scheduling plan of a corresponding production machine, selecting a product range with maximum delay time and a scheduling plan of a corresponding production machine, randomly selecting a product range and a scheduling plan of a corresponding production machine, and selecting a product range with maximum estimated delay time and a scheduling plan of a corresponding production machine.
By applying the technical scheme of the application, the total cost including the product processing cost, the employee overtime cost, the product delayed delivery cost and the product inventory cost can be taken as an optimization target. And the scheduling plan corresponding to the order information is processed at the lowest cost by combining with the deep reinforcement learning algorithm. Therefore, on one hand, the scheduling scheme of the manufacturer in a period of time can be automatically given in real time. On the other hand, the defect that the production scheduling scheme is not accurate enough due to the fact that the production cost of staff in the production process is not considered in the production scheduling scheme in the related technology is also avoided.
In another embodiment of the present application, the conversion module 202 is configured to perform the following steps:
inputting the manufacturer basic information and the order information into an intelligent manufacturing and scheduling system to obtain a plurality of state feature vectors matched with the manufacturer basic information and the order information;
wherein the vendor base information comprises:
the processing time of the production machine, the attendance information of staff, the number of the production machines on different production lines, the production cost and the overtime cost of the staff on different production lines, and the cost information of products.
In another embodiment of the present application, the conversion module 202 is configured to perform the steps of:
the order information processing system comprises an average utilization rate vector used for reflecting the resource utilization rate of the production machines, a utilization rate variance vector used for reflecting the load balance state of each production machine, a total procedure completion rate vector used for reflecting the processing amount progress of the order information and vectors of different product completion rates, and an estimated delay rate vector used for reflecting the delay progress of the order information and an actual delay rate vector.
In another embodiment of the present application, the conversion module 202 is configured to perform the steps of:
determining the production duration consumed by the target production machine to produce the products with the target product quantity based on the target product range reflected by the target scheduling plan and the corresponding target production machine;
calculating overtime cost corresponding to the target scheduling plan based on the initial processing time point and the production duration of the target production machine; and obtaining a product delay cost included in the order information;
and optimizing the reinforcement learning model based on the magnitude relation between the overtime cost and the product postponing cost.
In another embodiment of the present application, the conversion module 202 is configured to perform the steps of:
if the overtime cost is detected to be larger than the product delay cost, marking an excitation function corresponding to the target scheduling plan as a negative value;
and optimizing the reinforcement learning model by using the excitation function marked as a negative value.
In another embodiment of the present application, the conversion module 202 is configured to perform the steps of:
selecting a first number of sample sets from a data sample pool;
and performing reverse gradient propagation training on the reinforcement learning model by using a mean square error loss function and the sample set, and updating parameters of the reinforcement learning model until the reinforcement learning model is completely trained.
In another embodiment of the present application, the conversion module 202 is configured to perform the following steps:
taking an average utilization rate vector for reflecting the resource utilization rate of the production machine, and an estimated delay rate vector and an actual delay rate vector for reflecting the delay progress of the order information as a reward function;
and carrying out reverse gradient propagation training on the reinforcement learning model by utilizing the reward function.
The embodiment of the application also provides electronic equipment for executing the intelligent manufacturing scheduling method. Please refer to fig. 6, which illustrates a schematic diagram of an electronic device according to some embodiments of the present application. As shown in fig. 6, the electronic apparatus 3 includes: the system comprises a processor 300, a memory 301, a bus 302 and a communication interface 303, wherein the processor 300, the communication interface 303 and the memory 301 are connected through the bus 302; the memory 301 stores a computer program that can be executed on the processor 300, and the processor 300 executes the computer program to perform the intelligent manufacturing scheduling method provided in any one of the foregoing embodiments of the present application.
The Memory 301 may include a Random Access Memory (RAM) and a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The communication connection between the network element of the apparatus and at least one other network element is realized through at least one communication interface 303 (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, etc. may be used.
In one approach, the processor 300 may also be a Graphics Processing Unit (GPU). Which may implement or perform the methods, steps, and logic blocks disclosed in the embodiments of the present application. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 301, and the processor 300 reads the information in the memory 301 and completes the steps of the method in combination with the hardware thereof.
The electronic equipment provided by the embodiment of the application and the intelligent manufacturing scheduling method provided by the embodiment of the application have the same inventive concept and have the same beneficial effects as the method adopted, operated or realized by the electronic equipment.
Referring to fig. 7, the computer readable storage medium is an optical disc 40, on which a computer program (i.e., a program product) is stored, and when the computer program is executed by a processor, the computer program executes the intelligent manufacturing scheduling method according to any of the foregoing embodiments.
It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, a phase change memory (PRAM), a Static Random Access Memory (SRAM), a Dynamic Random Access Memory (DRAM), other types of Random Access Memories (RAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a flash memory, or other optical and magnetic storage media, which are not described in detail herein.
The computer-readable storage medium provided by the above-mentioned embodiments of the present application and the method for data identification provided by the embodiments of the present application have the same beneficial effects as the method adopted, executed or implemented by the application program stored in the computer-readable storage medium.
It should be noted that:
in the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application may be practiced without these specific details. In some instances, well-known structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the application, various features of the application are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the application and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted to reflect the following schematic: this application is intended to cover such departures from the present disclosure as come within known or customary practice in the art to which this invention pertains. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this application.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the application and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An intelligent manufacturing scheduling method, wherein:
acquiring manufacturer basic information and order information in a target time period, wherein the manufacturer basic information comprises staff production information used for reflecting staff overtime cost, and the order information comprises order quantity, product delivery date and product delay cost;
obtaining a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of a production machine or the production state of a product;
taking the plurality of feature vectors as the input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, wherein the target scheduling plan is used for reflecting a target product range required to be produced for processing the order information and a corresponding target production machine under the lowest cost;
performing scheduling manufacturing on the order information based on the target scheduling plan;
wherein the target scheduling plan comprises one of:
selecting a product range with minimum delay time and a scheduling plan of a corresponding production machine, selecting a product range with minimum time ratio of residual delivery and residual production and a scheduling plan of a corresponding production machine, selecting a product range with maximum delay time and a scheduling plan of a corresponding production machine, randomly selecting a product range and a scheduling plan of a corresponding production machine, and selecting a product range with maximum estimated delay time and a scheduling plan of a corresponding production machine.
2. The method of claim 1, wherein said deriving a plurality of status feature vectors based on said vendor basis information and said order information comprises:
inputting the manufacturer basic information and the order information into an intelligent manufacturing and scheduling system to obtain a plurality of state feature vectors matched with the manufacturer basic information and the order information;
wherein the vendor basis information comprises:
the processing time of the production machine, the attendance information of staff, the number of the production machines on different production lines, the production cost and the overtime cost of the staff on different production lines, and the cost information of products.
3. The method of claim 1 or 2, wherein the state feature vector comprises:
the order information processing system comprises an average utilization rate vector used for reflecting the resource utilization rate of the production machines, a utilization rate variance vector used for reflecting the load balance state of each production machine, a total procedure completion rate vector used for reflecting the processing amount progress of the order information and vectors of different product completion rates, and an estimated delay rate vector used for reflecting the delay progress of the order information and an actual delay rate vector.
4. The method of claim 1, wherein after said scheduling manufacturing said order information based on said target scheduling plan, further comprising:
determining the production duration consumed by the target production machine to produce the products with the target product quantity based on the target product range reflected by the target scheduling plan and the corresponding target production machine;
calculating overtime cost corresponding to the target scheduling plan based on the initial processing time point and the production duration of the target production machine; and obtaining a product delay cost included in the order information;
and optimizing the reinforcement learning model based on the magnitude relation between the overtime cost and the product postponing cost.
5. The method of claim 4, wherein optimizing the reinforcement learning model based on the magnitude relationship between the overtime cost and the product deferral cost comprises:
if the overtime cost is detected to be larger than the product delay cost, marking an excitation function corresponding to the target scheduling plan as a negative value;
and optimizing the reinforcement learning model by using the excitation function marked as a negative value.
6. The method of claim 1, wherein after said scheduling manufacturing said order information based on said target scheduling plan, further comprising:
selecting a first number of sample sets from a data sample pool;
and performing reverse gradient propagation training on the reinforcement learning model by using a mean square error loss function and the sample set, and updating parameters of the reinforcement learning model until the reinforcement learning model is completely trained.
7. The method of claim 6, wherein the training the reinforcement learning model for inverse gradient propagation comprises:
taking an average utilization rate vector for reflecting the resource utilization rate of the production machine, and an estimated delay rate vector and an actual delay rate vector for reflecting the delay progress of the order information as a reward function;
and carrying out reverse gradient propagation training on the reinforcement learning model by utilizing the reward function.
8. The utility model provides an intelligent manufacturing scheduling device which characterized in that, wherein:
the system comprises an acquisition module and a display module, wherein the acquisition module is configured to acquire manufacturer basic information and order information in a target time period, the manufacturer basic information comprises staff production information used for reflecting staff overtime cost, and the order information comprises order quantity, product delivery date and product delay cost;
the conversion module is configured to obtain a plurality of state feature vectors based on the manufacturer basic information and the order information, wherein each state feature vector is used for reflecting the production state of a production machine or the production state of a product;
an output module, configured to take the plurality of feature vectors as an input of the reinforcement learning model, so that the reinforcement learning model outputs a target scheduling plan matched with the plurality of feature vectors, where the target scheduling plan is used to reflect a target product range and a corresponding target production machine required to be produced for processing the order information at a lowest cost;
a processing module configured to perform scheduling manufacturing on the order information based on the target scheduling plan;
wherein the target scheduling plan comprises one of:
selecting a product range with minimum delay time and a scheduling plan of a corresponding production machine, selecting a product range with minimum time ratio of residual delivery and residual production and a scheduling plan of a corresponding production machine, selecting a product range with maximum delay time and a scheduling plan of a corresponding production machine, randomly selecting a product range and a scheduling plan of a corresponding production machine, and selecting a product range with maximum estimated delay time and a scheduling plan of a corresponding production machine.
9. An electronic device, comprising:
a memory for storing executable instructions; and the number of the first and second groups,
a processor for executing the executable instructions with the memory to perform the operations of the intelligent manufacturing scheduling method of any of claims 1-7.
10. A computer-readable storage medium storing computer-readable instructions that, when executed, perform the operations of the intelligent manufacturing scheduling method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211167448.3A CN115619007A (en) | 2022-09-23 | 2022-09-23 | Intelligent manufacturing scheduling method and device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211167448.3A CN115619007A (en) | 2022-09-23 | 2022-09-23 | Intelligent manufacturing scheduling method and device, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115619007A true CN115619007A (en) | 2023-01-17 |
Family
ID=84858277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211167448.3A Pending CN115619007A (en) | 2022-09-23 | 2022-09-23 | Intelligent manufacturing scheduling method and device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115619007A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116629552A (en) * | 2023-05-26 | 2023-08-22 | 讯猫软件集团有限公司 | Intelligent industrial management regulation and control system |
| CN117132088A (en) * | 2023-10-26 | 2023-11-28 | 四川省致链数字科技有限公司 | Order scheduling method, system and readable storage medium |
-
2022
- 2022-09-23 CN CN202211167448.3A patent/CN115619007A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116629552A (en) * | 2023-05-26 | 2023-08-22 | 讯猫软件集团有限公司 | Intelligent industrial management regulation and control system |
| CN117132088A (en) * | 2023-10-26 | 2023-11-28 | 四川省致链数字科技有限公司 | Order scheduling method, system and readable storage medium |
| CN117132088B (en) * | 2023-10-26 | 2024-01-26 | 四川省致链数字科技有限公司 | Order scheduling method, system and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115619007A (en) | Intelligent manufacturing scheduling method and device, electronic equipment and medium | |
| US20190080244A1 (en) | Scheduling method and system based on improved variable neighborhood search and differential evolution algorithm | |
| Han et al. | Appointment scheduling and routing optimization of attended home delivery system with random customer behavior | |
| TW201732696A (en) | Method and device warehouse storage space planning and electronic device | |
| CN110210731A (en) | Reminding task distribution method, device, computer equipment and storage medium | |
| Lee et al. | Flexible job-shop scheduling problems with ‘AND’/‘OR’precedence constraints | |
| CN108509220B (en) | Revit engineering calculation amount parallel processing method, device, terminal and medium | |
| CN112149925B (en) | Automatic allocation method and device for warehouse tasks, warehouse management method and system | |
| CN112926808A (en) | Logistics path planning method, device, equipment and computer readable storage medium | |
| Babiceanu et al. | Real-time holonic scheduling of material handling operations in a dynamic manufacturing environment | |
| CN106874080B (en) | Data calculation method and system based on distributed server cluster | |
| Yang et al. | A versatile multi-agent reinforcement learning benchmark for inventory management | |
| CN116579586B (en) | Resource scheduling method, device and system | |
| Gan et al. | Scheduling of flexible-sequenced process plans in a mould manufacturing shop | |
| CN115271574A (en) | Goods replenishment method and device, electronic equipment and storage medium | |
| CN117057550A (en) | Material demand plan calculation method, device, network equipment and medium | |
| Leung et al. | A stochastic programming model for production planning of perishable products with postponement | |
| CN116663800A (en) | Task ship determining method and device | |
| CN108681838B (en) | Scheduling plan determining method and device | |
| JP2009126621A (en) | Proper inventory quantity determination system and its method | |
| CN108090713A (en) | Inventory demand forecasting system | |
| CN109903074A (en) | State of market division methods and device based on data analysis | |
| CN117114452B (en) | Method, processor, device and storage medium for determining construction scheme | |
| CN118536912B (en) | Automatic warehouse-in and warehouse-out scheduling method and device based on Q-Learning algorithm | |
| CN115879744A (en) | Limited production scheduling method, device, storage medium and electronic equipment |
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 |