Production simulation method, production optimization result visualization method and related device

Abstract

The invention provides a production simulation method, a production optimization result visualization method and a related device.The production simulation method provided in the first aspect of the application comprises the steps of obtaining a plurality of possible production line distributions according to order information and existing equipment information; simulating each possible production line distribution to produce orders in the order information to obtain a plurality of production target values, wherein each production target value comprises at least one sub-target type; and obtaining a visual view according to the plurality of production target values, wherein the dimensions of the visual view are the same as the type number of the sub-target types. Due to the fact that the visual views have multiple dimensions, a decision maker can select the visual views of different dimensions according to decision preferences and select order production schemes meeting the decision preferences.

Description

Technical Field

[0001] The present invention relates to the field of production optimization, and in particular to a production simulation method, a production optimization result visualization method and related devices. Background Art

[0002] In flexible production lines, flexible factories and new generation smart factories, it is necessary to comprehensively consider multiple factors such as orders, production line status, dynamic production raw material costs, labor costs, etc. For decision makers, it is necessary to seek to minimize the cost of order production, maximize production speed, and maximize efficiency. Obviously, when a flexible production line is producing a specific order, it is difficult to simultaneously minimize costs while maximizing production speed and maximizing efficiency.

[0003] Existing technologies provide planning and scheduling software, which establishes resource optimization intelligent scheduling algorithms, comprehensively considers multiple objectives such as delivery date, priority, and capacity to perform scheduling operations, and maximize capacity utilization. However, the calculation results are all single results, for example, the capacity arrangement when the output cost is minimized, which cannot meet the diverse needs of decision makers. Summary of the invention

[0004] The purpose of this application is to solve the problem that the production line scheduling method in the prior art only outputs a fixed arrangement plan for the preset input, and cannot meet the diverse needs of decision makers in flexible production lines. This application provides a production simulation method, a production optimization result visualization method and related devices, which can provide decision makers with a visualization view corresponding to multiple production target values, thereby providing multiple decision-making solutions with different target orientations.

[0005] To solve the above technical problems, in the first aspect, an implementation method of the present application provides a production simulation method, including: obtaining multiple possible production line distributions based on order information and existing equipment information; simulating each of the possible production line distributions to produce the orders in the order information to obtain multiple production target values, wherein the production target values ​​include at least one sub-target type; and obtaining a visualization view based on the multiple production target values, wherein the dimension of the visualization view is the same as the number of types of the sub-target types.

[0006] Since visualization views have multiple dimensions, decision makers can select visualization views of different dimensions based on their decision preferences and select order production plans that meet their decision preferences.

[0007] In a possible implementation of the first aspect above, obtaining multiple possible production line distributions based on order and existing equipment information includes: obtaining order information and existing equipment information; obtaining multiple possible production line distributions based on the order information and the existing equipment information in a manner that satisfies a first constraint, wherein the first constraint reduces the number of possible production line distributions.

[0008] In a possible implementation of the first aspect above, the order information includes importance information corresponding to each of the orders, and the first constraint also includes a second constraint, and the second constraint includes: for any production unit, and another production unit whose importance information is less than or equal to the production unit, the quantity of the production units is not less than the other production unit, and the production unit and the other production unit respectively represent a set of minimum production equipment that satisfies the production of the order corresponding to the production unit.

[0009] In a possible implementation of the first aspect above, the simulation of each of the possible production line distributions producing the orders in the order information to obtain multiple production target values ​​includes: for one of the possible production line distributions, according to a preset allocation rule, each order in the order information is allocated to the corresponding production line in the possible production line distribution; optimizing the order of the orders in each of the production lines; simulating each of the production lines to produce the orders corresponding to the production line to obtain production target values, and the sub-target types include: one or more of the overall order fulfillment rate, the fulfillment rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, and total production cost.

[0010] In a possible implementation of the first aspect above, a set of minimum production equipment that can meet the production of any order is taken as a production unit, and the preset allocation rules include: when an existing production line includes a production unit that can meet the production of any order, the order is allocated to the production line; when all the existing production lines do not include a production unit that can meet the production of the order, the gap between all the production units in the existing production lines and the production units that can meet the production of the order is calculated, and the order is allocated to the existing production line where the production unit with the smallest gap is located.

[0011] In a possible implementation of the first aspect above, the optimizing the sequence of the orders in each of the production lines includes: dividing the orders assigned to any production line in the possible production line distribution into units that can move freely on the production line in a preset manner; and using a preset algorithm to move the units on the production line to optimize the production sequence of the orders.

[0012] In a possible implementation of the first aspect above, the preset algorithm includes: obtaining a sequential optimization target; taking at least one of the units as an ant, using an ant colony algorithm to optimize the order of the units on the production line to obtain an optimized production line, and updating pheromones; repeating the above steps until, when the optimized production line meets the sequential optimization target and / or the number of iterations of the ant colony algorithm is equal to a preset maximum iteration value.

[0013] In a possible implementation of the first aspect above, before optimizing the order of the orders in each of the production lines, it also includes: dividing the orders allocated to all production lines in the possible production line distribution into units with the ability to move freely between different production lines in a preset manner; calculating the achievement rate of all production lines, and moving at least one of the units on the production line with the smallest achievement rate to another production line with a preset probability according to the achievement rate of each production line.

[0014] In a possible implementation of the first aspect above, in obtaining a visualization view based on the multiple production target values, wherein the dimension of the visualization view is the same as the number of sub-target types, it includes: selecting multiple sub-target types as variables to establish an objective function; using the multiple sub-target types as coordinate axes, calculating the Pareto frontiers of the multiple possible production line distributions; counting the production time of each order in each production line corresponding to any of the Pareto fronts, and visualizing it in the form of a Gantt chart.

[0015] In order to solve the above technical problems, the present application further proposes a second aspect. The embodiment of the present application provides a production optimization result visualization method, which is used for any of the above production simulation methods, and the production optimization result visualization method comprises: displaying a first curve graph with a first indicator as the horizontal coordinate and a second indicator as the vertical coordinate, wherein the curve graph comprises a plurality of data points, wherein the data points represent the Pareto front value of the objective function with the first indicator and the second indicator as variables, and different data points represent different possible production line distributions;

[0016] In a possible implementation of the second aspect above, the method also includes: in response to a user's first input for the first curve graph, displaying a second curve graph, the first input including a first preset range and a second preset range, the second curve graph including a plurality of data points where the first indicator satisfies the first preset range and the second indicator satisfies the second preset range.

[0017] In a possible implementation of the second aspect, the method includes: in response to a first enablement of any of the data points by a user, displaying at least one first Gantt chart; the first Gantt chart represents the production time of each order in a production line of the possible production line distribution. In response to a second enablement of any of the data points by a user, displaying a second Gantt chart; the second Gantt chart represents the production time in each production line of all orders in the possible production line distribution.

[0018] In order to solve the above technical problems, the present application also proposes a third aspect. The embodiment of the present application provides an electronic device comprising: one or more processors; one or more memories; the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes any one of the above production simulation methods.

[0019] In order to solve the above technical problems, the present application also proposes a fourth aspect. The implementation mode of the present application provides a computer-readable storage medium, on which instructions are stored. When the instructions are executed on a computer, the computer executes any one of the production simulation methods described above.

[0020] The beneficial effects achieved by the embodiments of the second to fourth aspects are the same as those of the first aspect. Please refer to the description of the first aspect for details, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 According to some embodiments of the present application, a schematic flow chart of a production simulation method is shown.

[0022] Figure 2A According to some embodiments of the present application, a chromosome schematic diagram of possible production line distribution is shown.

[0023] Figure 2B According to some embodiments of the present application, a chromosome schematic diagram of another possible production line distribution is shown.

[0024] Figure 2C According to some embodiments of the present application, a chromosome schematic diagram of another possible production line distribution is shown.

[0025] Figure 3 According to some embodiments of the present application, a schematically visualized Pareto front solution set is shown.

[0026] Figure 4A According to some embodiments of the present application, it is shown Figure 3 A possible production line distribution in the production plan Gantt chart from the production line perspective.

[0027] Figure 4B According to some embodiments of the present application, it is shown Figure 3 A Gantt chart showing the production plan of a possible production line from the perspective of another production line.

[0028] Figure 4C According to some embodiments of the present application, it is shown Figure 3 A Gantt chart showing the production plan of a possible production line from the perspective of another production line.

[0029] Figure 5 According to some embodiments of the present application, it is shown Figure 3 Gantt chart of production plan for a possible production line distribution from the order perspective.

[0030] Figure 6 According to some embodiments of the present application, it is shown Figure 3 A possible production line is distributed in the production cost proportion chart.

[0031] Figure 7 According to some embodiments of the present application, a schematic diagram of an electronic device is shown. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0033] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0035] Illustrative embodiments of the present application include, but are not limited to, a production simulation method.

[0036] For better explanation, some nouns that may appear in the specific implementation methods are explained.

[0037] Flexible production line: refers to a production line composed of multiple adjustable production equipment connected together and equipped with automatic conveying devices. Flexible production line can configure different production lines according to different order requirements to meet order production.

[0038] Order: A request for the production of a certain product. When an order contains production requirements for multiple products, it is split into multiple orders for the convenience of calculation, each order corresponding to a product. For example, an order includes the following information: product type, product quantity, and delivery time. Obviously, the information that an order can contain is not limited to this.

[0039] Production unit: The minimum set of production equipment that meets the production requirements of a specific order.

[0040] The production simulation method of the present application is applicable to production units with flexible production lines, such as a single factory or a production consortium consisting of multiple factories. For such production units, since the production line is flexibly configurable, the production cost can be minimized, the production speed can be maximized, and the production efficiency can be maximized by configuring various indicators of the production line.

[0041] In an illustrative example, the above-mentioned production unit is, for example, configured with multiple production equipment and receives multiple orders. For the production of any order, multiple production equipment is required to complete different process steps to obtain the product required by the order. In order to simplify the description, in each subsequent embodiment of the present application, the production equipment is matched one-to-one with the process, that is, one production equipment can only complete one process. One or more production lines can be configured in the production unit at the same time. When multiple production lines are configured, the production line can also have multiple types to meet the production requirements of different orders, but in the embodiments described later, one production line corresponds to one order (product) uniquely. In addition, when an order has production requirements for multiple products, the order is split into multiple ones so that the order and the product type to be produced correspond one-to-one. It should be noted that different products may need to use the same process (that is, one production equipment may be used by multiple orders).

[0042] For easier understanding, the composition relationship of each component in a production unit is explained. The production unit (factory) receives orders and arranges production lines (production lines) to produce products according to the orders. A production unit can include one or more production lines. A production line can be dedicated to the production of one product or multiple products. The production line includes production units, which are a collection of production equipment used to meet the production of specific products.

[0043] In the embodiments described later, obtaining or acquiring may be acquiring from external input, for example, in S100 described later, order information and existing equipment information may be acquired through an external input table, or certain data may be acquired through the previous step or the previous steps. Or it may represent obtaining certain data or icons through a preset calculation method or drawing method. For example, in S300, data is converted into a visual view through a preset image processing method. The specific method may be introduced in the embodiments described later, but in order to more accurately and appropriately express the core content of the present application, some parts are omitted in some cases. At this time, it should be understood that the method of obtaining or acquiring is a method well known to those skilled in the art, so it will not be repeated.

[0044] Some known technologies provide a flexible flow shop production scheduling system based on Plant simulation. By designing an intermediate interaction layer, the existing production simulation system and production scheduling system are isolated from each other and cannot be interconnected intelligently and autonomously, resulting in low adaptability to random changes in the production site and slow response speed. The problem with this known technology is that the modeling cost is high. Although it is suitable for flexible scenarios, it does not contribute to the improvement of flexibility and has a single optimization goal.

[0045] In other known technologies, a factory, production line and logistics simulation software is provided, which can simulate and optimize production systems such as workshop layout, production logistics design, and production capacity, and output the optimal production line layout. By simulating the equipment information, connection method and spatial layout of the physical space, combined with the production time distribution function and equipment failure probability model constructed by historical data, the production capacity of the production line can be calculated relatively accurately and the bottleneck of the production line can be found. The allocation of processes to workstations is further optimized through genetic algorithms to maximize the line balance rate. The problem with this known technology is that the modeling cost is high, the threshold for use is high, there is a lack of flexibility for different orders, and the optimization goal is single.

[0046] In other known technologies, a production system simulation software is provided that combines the features of Plant simulation and Preactor (a name for a production scheduling system), supports multi-objective analysis of production system capacity and efficiency, including production efficiency, bottleneck analysis, time / cost / consumption, etc., and uses optimization methods such as genetic algorithms and neural networks, with strong optimization capabilities. The problem with this known technology is that the modeling cost is high, the usage threshold is high, and the reliability of plan optimization is not high.

[0047] In other known technologies, a production planning system that provides scheduling plans is provided. It integrates material and equipment resources, production calendars, orders, inventory and other rule constraints to complete order expansion, order dispatch and plan evaluation, and has high usability at the planning level. This known technology has a high threshold for use, low flexibility, simple optimization logic and lacks selective production line simulation.

[0048] The following embodiments of the present application can solve at least one problem in the above-mentioned known technologies.

[0049] The production simulation method of the first embodiment of the present application is described based on the above production scenario.

[0050] refer to Figure 1 A production simulation method of the present application includes:

[0051] S100: obtaining multiple possible production line distributions according to order information and existing equipment information;

[0052] S200: simulating each of the possible production lines to produce the orders in the order information to obtain a plurality of production target values;

[0053] S300: Obtain a visualization view according to the multiple production target values.

[0054] Specifically, in S100, the order information may include product type, product quantity, processes required for the product, and production deadline (order closing date); the existing equipment information may include multiple existing equipment names (or unique identification numbers), and processes corresponding to different existing equipment. In some embodiments, the power corresponding to the existing equipment and the output capacity of the equipment for the corresponding process may also be included.

[0055] A possible production line distribution is, for example, a distribution of production equipment that uses production equipment to meet the conditions for forming a production line and enables the production line to meet the production requirements of the order. For order A and another order B included in the order information, production unit A for producing order A and production unit B for producing order B are selected. An exemplary production line distribution is: the first production line has 2 production units A, and the second production line has 1 production unit B. Another exemplary production line distribution is: the first production line has 1 production unit A, the second production line has 1 production unit B, and the third production line has 1 production unit A. Another exemplary production line distribution is: the first production line has 1 production unit C for producing order A, and the second production line has 1 production unit B. Production unit C is also used to produce order A but its efficiency is twice that of the production unit.

[0056] In S200, the production target value includes one or more sub-target types. Specific sub-target types can be the overall fulfillment rate of one or more orders, the fulfillment rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, and total production cost. For a production line distribution, after determining the processes it includes, the above-mentioned sub-target types of the production target value can be calculated through simulated production. The reason why this application needs to perform simulated production is that it is necessary to consider many variables that change over time, such as electricity prices and labor prices.

[0057] In S300, the dimension of the visualization view is the same as the number of sub-goal types. When the dimension of the visualization view is two-dimensional, its view is a plane. When the dimension is three-dimensional, its view is a three-dimensional space. Decision makers can select the sub-goal type and the number of sub-goals according to their own needs to obtain the required visualization view.

[0058] In the first embodiment, different visualization views can be generated according to different production line distributions. The decision maker can select visualization views of different dimensions (sub-goal types) according to decision preferences, and select order production plans that meet the decision preferences from the visualization views.

[0059] The present application also has a second embodiment. In the second embodiment, S100 obtains multiple possible production line distributions according to order information and existing equipment information, including:

[0060] S110: Obtain order information and existing equipment information;

[0061] S120: Obtain multiple possible production line distributions according to the order information and the existing equipment information in a manner that satisfies a first constraint.

[0062] The first embodiment has the following problem: there are too many possible production line distributions that use production equipment to meet the condition of forming a production line and making the production line meet the order production requirements, resulting in excessive computing resources being used in the simulation process described later. Therefore, in the second embodiment, a first constraint is imposed based on the order information and the existing equipment information to reduce the possible production line distributions obtained.

[0063] An example of a first constraint is:

[0064] (1) The total number of production equipment owned by the possible production line distribution is less than the total number of existing equipment in the existing equipment information;

[0065] (2) The number of production equipment per process in all production lines in the possible production line distribution should not exceed the sum of the number of production equipment per process at present;

[0066] (3) The number of production lines in the possible production line distribution does not exceed the preset maximum number of production lines;

[0067] (4) Each production unit has no more than the preset upper limit of production quantity.

[0068] In some embodiments, production equipment can be added by investment instead of being limited to existing production equipment. Therefore, the first constraint in this embodiment is, for example:

[0069] (3) The number of production lines in the possible production line distribution does not exceed the preset maximum number of production lines;

[0070] (4) Each production unit has no more than the preset upper limit of production quantity;

[0071] (5) Equipment investment should not exceed the preset maximum investment value.

[0072] In the second embodiment, the total number of possible production line distributions is reduced through the first constraint, so that the amount of calculation in the subsequent simulation process is reduced, thereby saving computing resources.

[0073] The present application also has a third embodiment. In the third embodiment, S100 obtains multiple possible production line distributions according to order information and existing equipment information, including:

[0074] The second constraint includes: for any production unit, and another production unit whose importance information is less than or equal to the production unit, the number of the production unit is not less than the other production unit, and the production unit and the other production unit respectively represent the set of minimum production equipment that meets the production order corresponding to the production unit.

[0075] The second constraint mentioned above can be part of the first constraint. The so-called importance information can be obtained by voting. For a production unit that performs multiple orders, any order i has a product production quantity ON i , the average number of products that need to be produced for each order in this production unit is ON AVE , then the reusability information Voteweight i for:

[0076] Voteweight i =ON i / ON AVE (Formula 1)

[0077] So the second constraint is, for example:

[0078] (6) The number of production units with higher reusability information should not be less than the number of production units with lower reusability information scores.

[0079] The role of the above reusability information is not limited to this. There is a third constraint:

[0080] (7) In the same production line, the production unit with higher reusability information should be ranked no lower than the production unit with the lower reusability information score.

[0081] The third embodiment mentioned above can be combined with the second embodiment, for example, that is, the second constraint is taken as a part of the first constraint.

[0082] In the third embodiment, the total number of possible production line distributions is further reduced by the second constraint, thereby reducing the amount of calculation in the subsequent simulation process and saving computing resources.

[0083] The present application also has a fourth embodiment. In the fourth embodiment, S200: simulating each of the possible production lines to produce the orders in the order information to obtain multiple production target values, including:

[0084] S210: for a possible production line distribution, allocating each order in the order information to a corresponding production line in the possible production line distribution according to a preset allocation rule;

[0085] S220: Optimizing the order of orders in each production line;

[0086] S230: Simulate each production line to produce the order corresponding to the production line to obtain the production target value.

[0087] Specifically, in a certain possible production line distribution, it is also necessary to sort the various production units (corresponding orders) in each production line. The purpose of sorting is to optimize the order of orders to achieve the optimization of one or more sub-goal types of production target values. Sub-goal types include, for example, one or more of: the overall order fulfillment rate, the fulfillment rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, and total production cost.

[0088] The optimization method in S220 is, for example, through a plurality of preset decision trees. Each decision tree has different preset branches, and through the random forest model integration, the decision tree inputs a possible production line distribution, order information, and existing equipment information into the model to obtain the order sorting optimization of each production line in the possible production line distribution.

[0089] The order information is input into a trained model, which includes at least one convolutional layer, a joint layer, and a normalization layer. The order information and existing equipment information of a possible production line distribution are input into the model to obtain the order sorting optimization of each production line in the possible production line distribution. During the model training process, at least one relevant loss function among the preset overall order fulfillment rate, each order fulfillment rate, production line process efficiency, overall production line efficiency, electricity cost, labor cost, and total production cost is used as a discriminator to manually screen the output results and complete the training of the model.

[0090] The fourth embodiment of the present application optimizes the order of orders for each production line and takes into account the impact of the order within the production line on the production target value, so that the visualization view obtained according to multiple production target values ​​in S300 can more intuitively present a better solution.

[0091] According to the fifth embodiment of the present application, the initial allocation is further optimized based on the fourth embodiment.

[0092] Specifically, in the fifth embodiment, S210: for a possible production line distribution, allocating each order in the order information to a corresponding production line in the possible production line distribution according to a preset allocation rule, further comprising:

[0093] S211: When an existing production line includes a production unit that satisfies the production of any order, the order is allocated to the production line; when all existing production lines do not include a production unit that satisfies the production of the order, the gap between all production units in the existing production lines and the production units that satisfy the production of the order is calculated, and the order is allocated to the existing production line where the production unit with the smallest gap is located.

[0094] That is, according to S211, in a certain production line distribution, the initial distribution for any order i can be completed in the following way:

[0095] (1) When there is a production unit corresponding to the order in the existing production line, the order is allocated to the production line;

[0096] (2) Otherwise, calculate all production units in all existing production lines and the production unit Cell corresponding to the order i The minimum Euclidean distance L.

[0097] L n =min(|Cell i -Cell j |2) (Formula 2)

[0098] In formula 2, n is the number of existing production lines. j Represents an existing production line L n The jth production unit in . Where Cell=(C1,C2,…,C m ), C m Represents the mth equipment of the production unit.

[0099] Through the fifth embodiment of the present application, allocating each order can reduce the conversion time and conversion cost of the production line while taking into account the existing production line distribution.

[0100] According to the sixth embodiment of the present application, S220 is further optimized based on the fourth embodiment.

[0101] In the sixth embodiment of the present application, S220 optimizes the order of orders in each production line including:

[0102] S221: Divide the orders allocated to any production line in the possible production line distribution into units that can move freely on the production line in a preset manner;

[0103] S222: Use a preset algorithm to move the unit on the production line to optimize the production sequence of the order.

[0104] Specifically, a genetic algorithm can be used to divide an order into multiple parts according to preset rules. For example, in a specific implementation, a production line is regarded as a chromosome, and the order is equally divided into granules that can move freely in the chromosome. Each granule is regarded as a gene, and the size of each granule is Grainsize (granule length) × the number of orders. The granule length can be preset, for example, in a specific embodiment, Grainsize is equal to 4.

[0105] The gene code uses a 3-bit integer, which represents the order number. The 1-bit "0 / 1" code represents whether the grain size is distributed on the current chromosome. Theoretically, the maximum coding length of the chromosome is equal to 4×1000×(1 / Grainsize).

[0106] The collection of multiple chromosomes is regarded as an individual (chromosome set), and an individual represents a possible production line distribution.

[0107] refer to Figure 2A , Figure 2B , Figure 2C . Figure 2A According to some embodiments of the present application, a chromosome schematic diagram of possible production line distribution is shown. Figure 2B According to some embodiments of the present application, a chromosome schematic diagram of another possible production line distribution is shown. Figure 2C According to some embodiments of the present application, another possible chromosome distribution diagram of the production line is shown. Figure 2A A chromosome set 1 in the figure represents a possible production line distribution. The chromosome set 1 includes three chromosomes, namely the first chromosome 100, the second chromosome 200, and the third chromosome 300. Therefore, the possible production line distribution includes three production lines, and the grain size 101 in each chromosome is 4.

[0108] Then, in S222, a preset algorithm is used to move each gene on the chromosome to optimize the order sequence on a production line.

[0109] For example, the preset algorithm first selects at least one preset sub-goal type. Based on the sub-goal type, multiple environmental pressure parameters are generated. Then, each gene is moved on the chromosome with a random factor. After completing the preset round of movement, the production target value of the production line corresponding to the chromosome is calculated and compared with the environmental pressure parameter. When the production target value is less than the preset range given by the environmental pressure parameter, the chromosome combination is eliminated and the next iteration is performed. When the production target value is within the preset range given by the environmental pressure parameter, a chromosome mutation is performed based on the chromosome, and the taboo table is updated.

[0110] Thus, the optimized chromosomes are obtained and the optimized sequence of the production line is obtained.

[0111] The sixth embodiment of the present application realizes order sequence optimization on a single production line through a genetic algorithm.

[0112] According to the seventh embodiment of the present application, S222 is further optimized based on the sixth embodiment. Specifically, it includes:

[0113] S222A: Obtaining sequential optimization targets;

[0114] S222B: Using at least one of the units as an ant, use the ant colony algorithm to optimize the order of the units on the production line to obtain an optimized production line, and update the pheromone; repeat the above steps until the optimized production line meets the sequence optimization goal and / or the number of iterations of the ant colony algorithm is equal to a preset maximum iteration value.

[0115] Furthermore, in S222A, the objective function is set according to each sub-objective type of the optimization target value that is ultimately desired to be obtained. Specifically, for the optimization of two sub-objective types including electricity cost and overall efficiency of the production line, the objective value can be set to calculate the objective function of the electricity cost and the overall efficiency of the production line through the production line.

[0116] In S222B, the ant colony algorithm is used to sort the granularity on the production line. Specifically:

[0117] For the chromosome corresponding to any production line that has completed the initial allocation.

[0118] (1) Initialize the number of ants, the maximum number of iterations, the pheromone importance parameter α, the heuristic factor importance parameter β, the evaporation coefficient rho, and the pheromone growth intensity Q;

[0119] (2) Randomly place the ants on the particle size and calculate the probability of the particle size to be selected according to the preset calculation formula;

[0120] (3) Calculate the target optimization value of each chromosome according to the objective function, record the chromosome with the best target optimization value, update the pheromone, and clear the taboo table;

[0121] (4) Determine whether the optimal chromosome target optimization value meets the preset sequential optimization target. If so, the chromosome with the optimal value is used as the chromosome of the current production line; otherwise, continue iterative optimization until the preset maximum number of iterations is reached.

[0122] refer to Figure 2A and Figure 2C , Figure 2A In the first chromosome 100, one granule 101 is exchanged with another granule 102. The positions of the granules in the second chromosome 200 and the third chromosome 300 are also changed.

[0123] The above calculation formula may be obtained by combining the known calculation method of ant colony algorithm with the specific implementation environment.

[0124] Based on the seventh embodiment of the present application, the optimization of production line sorting is better achieved by combining the genetic algorithm and the ant colony algorithm.

[0125] According to the eighth embodiment of the present application, before optimizing the order of orders in each production line in S222, the following steps are also included:

[0126] S221a: Divide the orders allocated to all production lines in the possible production line distribution into units with the ability to move freely between different production lines in a preset manner;

[0127] S221b: Calculate the achievement rates of all production lines, and according to the achievement rates of the production lines, move at least one unit on the production line with the lowest achievement rate to another production line with a preset probability.

[0128] Specifically, refer to Figure 2A , 2B , optimize the orders included in different production lines in a possible production line distribution, and calculate the achievement rate of each chromosome corresponding to the production line in a possible production line distribution. When a chromosome has the lowest achievement rate among all chromosomes, for a granularity located on this chromosome, there is a probability p that it will move right to another chromosome. Figure 2A , 2B In the figure, a particle 301 of chromosome 300 is moved to the end of chromosome 100 through the above steps.

[0129] In conjunction with the seventh embodiment of the present application, the preset calculation formula of step (2) in the seventh embodiment can be expressed as:

[0130]

[0131] In formula 3, α is the pheromone importance parameter, β is the heuristic factor importance parameter, τ ij represents the pheromone content of particle size i to j, η ij Represents the heuristic factor, where the process switching cost from granularity i to j is selected and the ant selects the next granularity, and the taboo table is updated to complete a tour.

[0132] That is, the Figure 2A , Figure 2B , Figure 2C There is coherence between them. For example, S221a and S221b are executed first, and then S222A and S222B are executed using the chromosomes with exchanged granularity, or the order is reversed.

[0133] In some further embodiments, combined with the importance information in the third embodiment, the production lines in a possible production line are sorted, and the production lines corresponding to the orders with high importance are arranged further to the left, and the production lines corresponding to the orders with low importance are arranged further to the right. It is easy to understand that the above left and right are just concepts created for the purpose of clarity, and do not refer to left and right in a physical or real sense.

[0134] Based on the above production line arrangement, S221b can further calculate the achievement rate of all production lines, and when the achievement rate of any production line is lower than a preset threshold, at least one unit on the production line is moved to another production line on the right, and the unit is placed at the end of the other production line. Through the eighth embodiment of the present application, order optimization between different production lines in a possible production line distribution is taken into account so that the target optimization value can be better obtained.

[0135] Through the sixth, seventh and eighth embodiments described above, the present application completes the sorting problem in a possible production line distribution by combining the genetic algorithm and the ant colony algorithm, and the combination of these two algorithms can adapt to the optimization of multiple production lines and multiple objectives.

[0136] According to a ninth embodiment of the present application, in S300, obtaining a visualization view according to the multiple production target values ​​includes:

[0137] S310: Select multiple sub-goal types as variables to establish a target function;

[0138] S320: Using the multiple sub-goal types as coordinate axes, calculating a Pareto frontier solution set of multiple possible production line distributions;

[0139] S330: Count the production time of each order in each production line corresponding to any of the Pareto fronts, and visualize it in the form of a Gantt chart.

[0140] In the ninth embodiment, an objective function is established based on at least one different sub-objective type, such as the overall order fulfillment rate, the fulfillment rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, total production cost, etc.

[0141] The objective function of the embodiment is introduced below.

[0142] For the overall order fulfillment rate and the fulfillment rate of each order. i The maximum working time TPH of the order production line every day is: TPH = Parameter (daytime working time) + Parameter (nighttime working time). Then the deadline hours of the order can be expressed as DueHoursi = TPH * Due i Based on the start and end time of each workpiece, compare the end time of the last product and the last process of the order with the deadline of the order. i , the overall fulfillment rate of all orders and the fulfillment rate of each order can be calculated.

[0143] For electricity cost. Convert the start and end time matrix of each production line into the form of (days, hours), and calculate the electricity cost of processing time and idling time for batches and processes respectively. Get the total power of the process, first calculate the processing cost of the whole day, then calculate the remaining cost distributed in each hour according to the time-of-use electricity price and power, and finally add up the sum of the electricity costs of the process and batch. The processing electricity cost DaysCost of a certain order batch process of a production line can be calculated as:

[0144]

[0145]

[0146] In Formula 4, for the processing power cost for a full day, EP(t) represents the average electricity price at the tth hour, power represents the total power of the current process, and T represents the last hour of processing each day.

[0147] Formula 5 calculates the electricity cost for the remaining hours after deducting the entire process time. EP(t) represents the average electricity price at the tth hour, ps and pe represent the hours when the process starts and ends, and T represents the last hour of the process each day.

[0148] Idle power is similar to processing power cost for batches and processes and is calculated by average electricity price × idling time.

[0149] Regarding labor costs. The calculation of labor costs is similar to that of electricity costs, except that the labor costs per hour are fixed. Therefore, the specific calculation can refer to Formula 4 and Formula 5, except that EP(t) is set as a function corresponding to the day shift wages and night shift wages.

[0150] For the process efficiency of the production line: for each process of each production line, the production efficiency is calculated using the method of effective time / total processing time, and the production line-process efficiency matrix is ​​returned.

[0151] For the overall efficiency of the production line: production line efficiency (the sum of the effective processing time of all processes / the total time of all processes).

[0152] In addition, when the equipment investment is not 0, the sub-goal type may also include the sub-goal type of equipment investment, which is completed by the first constraint or the second constraint when forming multiple possible production line distributions.

[0153] Based on the sixth, seventh, and eighth embodiments and the above objective functions, multiple curves can be obtained for different chromosomes in different possible production line distributions, and the Pareto fronts between different objective functions can be calculated based on these curves, thereby obtaining the Pareto front solution set and visualizing it.

[0154] Therefore, the ninth embodiment of the present application also includes S320: using the multiple sub-target types as coordinate axes, calculating the Pareto front solution set of the multiple possible production line distributions.

[0155] Figure 3 According to some embodiments of the present application, a schematic visualization of the Pareto front solution set is shown. Figure 3 The objective function (sub-goal type) is shown in Figure 1. The Pareto frontier of the exemplary output when the order fulfillment rate and the total production cost are selected. Each data point represents a Pareto frontier solution. Decision makers can choose a solution that meets their decision preferences based on the selected sub-goal type. It is easy to understand that three sub-goal types can be added to obtain a Pareto solution set space to assist better decision-making, or more sub-goal types can be added to obtain new dimensions.

[0156] Therefore, S330 is executed: the production time of each order in each production line corresponding to any of the Pareto fronts is counted, and it is visualized in the form of a Gantt chart.

[0157] Refer to Figure 4 to Figure 6 . Figure 4A According to some embodiments of the present application, it is shown Figure 3 A possible production line distribution in the production plan Gantt chart from the production line perspective. Figure 4B According to some embodiments of the present application, it is shown Figure 3 A Gantt chart showing the production plan of a possible production line from the perspective of another production line. Figure 4C According to some embodiments of the present application, it is shown Figure 3 A Gantt chart showing the production plan of a possible production line from the perspective of another production line. Figure 5 According to some embodiments of the present application, it is shown Figure 3 Gantt chart of production plan for a possible production line distribution from the order perspective. Figure 6 According to some embodiments of the present application, it is shown Figure 3 A possible production line is distributed in the production cost proportion chart.

[0158] Figure 4 to Figure 6 is calculated based on one or more of the above objective functions. Figure 6 The decision maker can see in more detail the situation of each production line in a Pareto front solution of the possible production line distribution. Obviously, similar visualizations can be made for other objective functions or subtypes of target production values.

[0159] According to the first to ninth embodiments described above, in some embodiments, it is necessary to consider order production-a production model based on flow / batch / changeover.

[0160] Specifically, the following assumptions are made for the flow production of the production line:

[0161] (1) Parallel flow production. Each production line can be composed of one or more identical production units. On each production line, all production units should start and end production of the same order of products at the same time. The actual processing time of the order is proportional to the number of production units on the production line. The same product instance should complete all processes on the same production line, and the process steps between different production lines are interleaved.

[0162] (2) Production time logic for different batches: The production time logic determines whether there is time overlap between different instances (workpieces) in the process, whether there is a cost of changing the equipment to process different orders, and whether there is batch management between processes. Based on the comparison between the equipment availability time and the product arrival time, dynamic programming is used to calculate the start and end time of each process for each product instance. There are three specific situations:

[0163] First batch: The first process of the first product instance of the first batch: The start time is equal to the end time of the first process of the previous batch of products plus the switching time to the first process of the current order. The subsequent process of the first product instance of the first batch: The start time is equal to the sum of the switching time to the current order, the end time of the process of the previous batch, and the transit time, plus the maximum value of the end time of the previous process of the current batch of products;

[0164] Starting from the second batch, the process start time of the first instance of the batch does not need to consider the changeover time;

[0165] There is no need to consider changeover time and transportation time between processing instances within the same batch;

[0166] The simulation production assumes that once the product is processed in a certain process, it will not stop until it is completed. The flow process of orders and production instances records a corresponding start time and end time for each workpiece.

[0167] According to actual conditions, those skilled in the art may also adopt other hypothetical models.

[0168] According to the production simulation method of the first aspect of the present application. First, through the first constraint and even the second constraint, find the possible production line distribution that meets the conditions, and then find the order in each possible production line distribution The better arrangement position, finally based on the above arrangement and possible production line distribution, simulate production and output multiple target production target values. That is, the present application (1) simultaneously considers the allocation of production line equipment and the allocation of orders in the production line, and makes corresponding optimizations to their allocation methods, so that it fits the situation of flexible production lines producing multiple orders; (2) Because different orders / decision makers have different requirements for production indicators, it is difficult to find an optimal solution that satisfies all parties for a specific production line distribution / order arrangement. Therefore, the result output by the present application is a visualization view based on different sub-goal types, which facilitates decision makers to better select production line distribution / order arrangement that meets the requirements; (3) Furthermore, in some embodiments of the present application, the Pareto front of multiple production target values ​​is calculated and output, and a visualization view of the Pareto front is output, and a screening method and a method for observing the order / possible production line distribution corresponding to a certain Pareto front solution from different perspectives are provided, so that decision makers can make selections more easily.

[0169] The second aspect of the present application also provides a production optimization result visualization method, the method comprising:

[0170] A first curve graph is displayed with the first indicator as the horizontal axis and the second indicator as the vertical axis, wherein the curve graph includes multiple data points, wherein the data points represent a solution of the Pareto front value of the objective function with the first indicator and the second indicator as variables, and different data points represent different possible production line distributions or different chromosome combinations.

[0171] refer to Figure 3 The method of the second aspect of the present application is used, for example, in a human-computer interaction facility with a display device, such as a computer monitor, a tablet, a smart phone, a television, a projector, a wearable device with a display device, an AR eye based on 3D light, or a display device based on an optical waveguide. And input devices such as a mouse, a keyboard, a touch screen, a stylus, a brain-computer interface, a brain wave receiving instrument, an eyeball mouse, a motion recognition device, etc., which are equipped for the above display devices. The above input devices are used for users to trigger enabling actions.

[0172] The above-mentioned first indicator and second indicator, for example, are one of the overall achievement rate, the achievement rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, and total production cost. The calculation method thereof refers to the objective function part in the embodiment of the first aspect.

[0173] In some embodiments, the production optimization result visualization method may further add a third indicator to convert the curve graph into a three-dimensional space.

[0174] In some embodiments, the method of the second aspect further includes:

[0175] In response to a user's first input to the first curve graph, a second curve graph is displayed, the first input includes a first preset range and a second preset range, and the second curve graph includes a plurality of data points where the first indicator satisfies the first preset range and the second indicator satisfies the second preset range.

[0176] Specifically, decision makers can reduce the difficulty of decision-making by screening the Pareto frontier solution set by setting the upper limit / lower limit / value range of a certain objective function.

[0177] The method of the second aspect further includes:

[0178] In response to a first enabling of any of the data points by the user, at least one first Gantt chart is displayed; the first Gantt chart represents the production time of each order in a production line of the possible production line distribution;

[0179] In response to a second enabling of any of the data points by the user, a second Gantt chart is displayed; the second Gantt chart represents the production time in each production line for all orders of the possible production line distribution.

[0180] Specifically, in the above curve chart, each data point has a push button. When the user clicks the button representing the data point with the left mouse button, or touches the data point with the touch screen, the first enable is triggered, and the first Gantt chart is displayed based on the Pareto front value of the possible production line distribution corresponding to the data point. The first Gantt chart, for example, displays Figure 4A , Figure 4B , Figure 4C . The above Figure 4A , Figure 4B , Figure 4C It can be displayed on the same interface, or based on a third enablement given by the user, such as sliding a touch screen or a mouse wheel, the image displayed by the display device can be displayed on the same interface. Figure 4A , Figure 4B , Figure 4C Switch between.

[0181] The second enablement refers to an enabling behavior for a data point that is different from the first enablement, such as clicking the right button of a mouse, long pressing the touch screen, and other enabling operations. Figure 5 Gantt chart shown.

[0182] Based on the second aspect above, a multi-objective visual display suitable for flexible production line management is realized. The known technologies are unable to display different production line configurations for different decision-making objectives.

[0183] The present application also has a third aspect that proposes an electronic device comprising: one or more processors; one or more memories; the one or more memories storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes any one of the above-mentioned production simulation methods. Figure 7 The electronic device 700 includes a memory 701, a processor 702, a communication bus 703, a communication interface 704, and a computer program stored in the memory 701 and executable on the processor 702. The communication bus 703 is used to realize the connection and communication between the processor 702 and the memory 701; when the processor 702 executes the computer program, it realizes any of the above-mentioned production simulation methods or any of the above-mentioned production optimization result visualization methods. The electronic device is, for example, a consumer-grade computer system such as a desktop, a laptop, a handheld computer, a notebook computer, an ultra-mobile personal computer (UMPC), or a portable small terminal device such as a smart phone, a palm computer, a tablet computer, an industrial terminal, etc.

[0184] Further, the present application also proposes a computer-readable storage medium. The computer-readable storage medium includes a volatile or non-volatile, removable or non-removable medium implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include but are not limited to RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic box, tape, disk storage or other magnetic storage device, or any other medium that can be used to store desired information and can be accessed by a computer. The production simulation method and / or production optimization result visualization method of the present application are stored on a computer-readable storage medium.

[0185] Although the above embodiments mention that the production equipment and the process correspond one to one, that is, one production equipment can only complete one process, it is not limited to this that the production equipment and the process do not have a one-to-one correspondence. For example, when a process needs to correspond to multiple production equipment, it only needs to "package" the production equipment and regard it as one production equipment and add it to the above method without affecting the realization of the beneficial effects of the present application.

[0186] Although the calculation method of importance information is given in the third embodiment above, it is not limited thereto. Importance information can be a fixed value directly assigned by an order or obtained by other well-known scoring and weighting methods.

[0187] Although the specific second constraint (6) is mentioned in the third embodiment above, it is not limited to this. Any method that uses importance to reduce the number of possible production line distributions can be adopted. However, the second constraint (6) is a better embodiment of the scheme as a whole. In practice, the second constraint (6) can control the possible production line distribution to a moderate number, and will not screen out the Pareto optimal possible production line distribution that can significantly improve some sub-goal types because the number becomes too small.

[0188] Although genetic algorithms are mentioned in the sixth and eighth embodiments above, they are not limited thereto. Other decision algorithms such as decision tree algorithms, competitive decision algorithms, multi-agent decision algorithms and other existing algorithms can also be selectively used. However, in this application, by combining genetic algorithms with ant colony algorithms, the purpose of this application is better achieved under the premise of a smaller computing scale, and no pre-training is required.

[0189] Although various behaviors of the first enabling and the second enabling are given in the above-mentioned embodiment of the second aspect, it is not limited thereto. The first enabling and the second enabling can be any human-computer interaction actions collected by the input device of the corresponding display device. Changes in the settings of these actions do not affect the realization of the beneficial effects of the present application.

[0190] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0191] In addition, the methods in various embodiments of the present application may be integrated into one processing unit, or may be recorded in separate units so that each unit exists physically separately, or two or more units may be integrated into one unit.

[0192] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute limitations on the embodiments of the present application, and may include more or fewer steps than shown, or a combination of certain steps, or different steps. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0193] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A production simulation method, characterized in that: include: Obtain multiple possible production line distributions based on order information and existing equipment information; Simulate each of the possible production lines to produce the orders in the order information to obtain a plurality of production target values, wherein the production target values ​​include at least one sub-target type; A visualization view is obtained according to the multiple production target values, wherein a dimension of the visualization view is the same as the number of types of the sub-target types.

2. The production simulation method according to claim 1, characterized in that: The multiple possible production line distributions obtained based on the order and existing equipment information include: Obtain order information and existing equipment information; A plurality of possible production line distributions are obtained according to the order information and the existing equipment information in a manner that satisfies a preset first constraint.

3. The production simulation method according to claim 2, characterized in that: The order information includes importance information corresponding to each of the orders, and the first constraint also includes a second constraint. The second constraint includes: for any production unit, and another production unit whose importance information is less than or equal to the production unit, the quantity of the production unit is not less than the other production unit, and the production unit and the other production unit respectively represent the set of minimum production equipment that satisfies the production of the order corresponding to the production unit.

4. The production simulation method according to claim 1, characterized in that: The simulating each of the possible production lines to produce the orders in the order information to obtain a plurality of production target values ​​includes: For one of the possible production line distributions, allocating each order in the order information to a corresponding production line in the possible production line distribution according to a preset allocation rule; Optimizing the sequence of the orders in each of the production lines; The production target value is obtained by simulating each production line to produce the orders corresponding to the production line, and the sub-target types include: the overall order fulfillment rate, the fulfillment rate of each order, the process efficiency of the production line, the overall production line efficiency, electricity cost, labor cost, and one or more of the total production cost.

5. The production simulation method according to claim 4, characterized in that: The minimum set of production equipment that can meet any order is considered a production unit. The preset allocation rules include: When the existing production line includes the production unit that meets the production of any of the orders, the order is allocated to the production line; when all the existing production lines do not include the production unit that meets the production of the order, the gap between all the production units in the existing production lines and the production units that meet the production of the order is calculated, and the order is allocated to the existing production line where the production unit with the smallest gap is located.

6. The production simulation method according to claim 4, characterized in that: The optimizing the sequence of the orders in each of the production lines comprises: Divide the orders allocated to any production line in the possible production line distribution into units that can move freely on the production line in a preset manner; The production sequence of the orders is optimized using a preset algorithm to move the units on the production line.

7. The production simulation method according to claim 6, characterized in that: The preset algorithm includes: Get the sequential optimization target; Taking at least one of the units as an ant, the ant colony algorithm is used to optimize the order of the units on the production line to obtain an optimized production line, and the pheromone is updated; the above steps are repeated until the optimized production line meets the sequence optimization target and / or the number of iterations of the ant colony algorithm is equal to a preset maximum iteration value.

8. The production simulation method according to any one of claims 4 to 7, characterized in that: Before optimizing the order of the orders in each of the production lines, the method further includes: Dividing the orders allocated to all production lines in the possible production line distribution into units with the ability to move freely between different production lines in a preset manner; The achievement rates of all production lines are calculated, and at least one of the units on the production line with the smallest achievement rate is moved to another production line with a preset probability according to the achievement rates of the respective production lines.

9. The production simulation method according to claim 1, characterized in that: In the step of obtaining a visualization view according to the multiple production target values, wherein the dimension of the visualization view is the same as the number of types of the sub-target types, the step includes: Select multiple sub-goal types to establish the objective function for the variable; Using the multiple sub-goal types as coordinate axes, calculating the Pareto frontier solution sets of the multiple possible production line distributions; The production time of each order in each production line corresponding to any of the Pareto fronts is counted and visualized in the form of a Gantt chart.

10. A method for visualizing production optimization results, characterized in that: The method is used for the production simulation method described in any one of claims 1 to 9, and the production optimization result visualization method comprises: A first curve graph is displayed with the first indicator as the horizontal axis and the second indicator as the vertical axis. The first curve graph includes multiple data points. The data points represent the Pareto front value of the objective function with the first indicator and the second indicator as variables. Different data points represent different possible production line distributions.

11. The method for visualizing production optimization results according to claim 10, characterized in that: The method further comprises: In response to a user's first input to the first curve graph, a second curve graph is displayed, the first input includes a first preset range and a second preset range, and the second curve graph includes a plurality of data points where the first indicator satisfies the first preset range and the second indicator satisfies the second preset range.

12. The method for visualizing production optimization results according to claim 10 or 11, characterized in that: The method further comprises: In response to a first enabling of any of the data points by the user, at least one first Gantt chart is displayed; the first Gantt chart represents the production time of each order in a production line of the possible production line distribution; In response to a second enabling of any of the data points by the user, a second Gantt chart is displayed; the second Gantt chart represents the production time in each production line for all orders of the possible production line distribution.

13. An electronic device, characterized in that: include: one or more processors; One or more memories; the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes the production simulation method described in any one of claims 1 to 9.

14. A computer-readable storage medium, characterized in that: The storage medium stores instructions, which, when executed on a computer, cause the computer to execute the production simulation method according to any one of claims 1 to 9.

Images