CN114862231A

CN114862231A - Production scheduling method and device and electronic equipment

Info

Publication number: CN114862231A
Application number: CN202210550032.3A
Authority: CN
Inventors: 陈锦华; 王晗; 易东明; 李陈赞; 林旭武
Original assignee: Guangzhou Xiaopeng Motors Technology Co Ltd
Current assignee: Zhaoqing Xiaopeng New Energy Investment Co Ltd
Priority date: 2022-05-20
Filing date: 2022-05-20
Publication date: 2022-08-05

Abstract

The application relates to a production scheduling method and device and electronic equipment. The method comprises the following steps: obtaining order data of the order to be sorted according to the product attribute of the vehicle; respectively determining production constraint conditions and a sequencing neighborhood structure according to a preset target of production, wherein the production constraint conditions comprise necessary constraint conditions and custom constraint conditions; obtaining candidate sorting schemes which accord with preset targets through at least two preset algorithms according to order data and a sorting neighborhood structure, and screening and obtaining a production sorting scheme of an order to be sorted according to a user-defined constraint condition from the candidate sorting schemes which accord with necessary constraint conditions; and determining a production plan of the order to be sorted according to the production sorting scheme. The scheme provided by the application can be used for meeting the production planning and scheduling of a complex manufacturing scene, so that the production plan meets the complex requirement of finished automobile manufacturing.

Description

Production scheduling method and device and electronic equipment

Technical Field

The present disclosure relates to the field of manufacturing technologies, and in particular, to a method and an apparatus for scheduling production, and an electronic device.

Background

Different from the traditional automobile, the intelligent electric car is not only a travel tool, but also integrates various revolutionary technologies such as the Internet, big data, artificial intelligence and the like, so that the automobile has the capability of computing power and intelligent software, can meet the more personalized requirements of users, has more intelligent performance and the like. For this reason, when the intelligent electric vehicle is produced and sequenced, in addition to considering conditions such as vehicle systems, colors, power assemblies and the like, the whole vehicle needs to be provided with various personalized intelligent modules, so that some difficulties are brought to the whole vehicle manufacturing of the intelligent electric vehicle.

Currently, the manufacture of smart electric vehicles requires the support of Advanced Planning and Scheduling (APS). The APS system is a system engineering for planning a feasible production operation plan in consideration of the limitation conditions of producer resources, and is used for managing and planning capacity resources, thereby converting extensive and manual management type production into resource intensive and intelligent planning type production. In the planning and scheduling process, the APS system considers resource and capability constraints inside and outside a producer from the whole situation, selects an optimal scheme from a large number of feasible schemes by using a complex intelligent algorithm to guide the production process of the producer, helps the producer to plan, execute, analyze, optimize and decide the resource utilization in production, arranges the operation tasks needing production into a production sequence according to a set constraint rule, and can achieve the purposes of minimizing production time, maximizing production profit and the like under the action of the constraint rule, thereby generating benefits for the producer.

The intelligent electric car is subjected to the stamping, welding, coating and final assembly process flows which are required to be subjected to car manufacturing, and the whole car manufacturing of the intelligent electric car also has the characteristics of high technological content, complex manufacturing process, variable user requirements and the like. However, the current APS system generally schedules production for a certain plant, and the introduced algorithm is single, without considering the complexity of the problem. Therefore, the production scheduling plan obtained by the related technology cannot meet the complex requirements of multi-train, less-batch, short-term delivery and personalized manufacture of the intelligent electric vehicle in the whole vehicle manufacturing.

Disclosure of Invention

In order to solve or partially solve the problems in the related art, the application provides a production scheduling method, a production scheduling device and electronic equipment, which can meet the production planning scheduling of a complex manufacturing scene, so that a production plan meets the complex requirement of finished automobile manufacturing.

A first aspect of the present application provides a production scheduling method, including:

obtaining order data of the order to be sorted according to the product attribute of the vehicle;

respectively determining a production constraint condition and an ordering neighborhood structure according to a preset target of production, wherein the production constraint condition comprises a necessary constraint condition and a custom constraint condition;

obtaining candidate sorting schemes which accord with preset targets through at least two preset algorithms according to the order data and the sorting neighborhood structure, and screening and obtaining a production sorting scheme of the order to be sorted according to user-defined constraint conditions from the candidate sorting schemes which accord with necessary constraint conditions;

and determining the production plan of the order to be sorted according to the production sorting scheme.

In one embodiment, the requisite constraints include at least one of:

the sum of the manufacturing starting time of the order to be sorted and the time offset value is greater than or equal to the current time;

each order to be sorted is sorted only once;

the plan starting time of the order to be sorted is earlier than the corresponding plan ending time;

the single-day capacity consumption value of all the orders to be sorted in a single workshop is less than or equal to the total single-day capacity of the workshop.

In one embodiment, the custom constraints include at least one of:

setting the ratio of the daily production quantity of the order of the specified product attribute to the daily total production quantity according to a custom ratio;

producing the order with the specified product attribute according to the user-defined batch;

the production quantity of the order with the specified product attribute on the specified date is greater than or equal to the preset production quantity.

In one embodiment, the ranking neighborhood structure includes at least one of:

randomly adjusting the production sequence of any order to be ordered in a known production ordering scheme;

randomly exchanging the production sequence of any two orders to be sequenced in the known production sequencing scheme.

In an embodiment, the obtaining, by at least two preset algorithms, a candidate ranking scheme that meets a preset target includes:

solving is carried out in sequence through at least two solving stages, and a candidate ordering scheme which accords with a preset target is output in the last solving stage; and the preset algorithm adopted in each solving stage is respectively and independently set.

In one embodiment, the sequentially solving through at least two solving stages and outputting the candidate ranking scheme meeting the preset target in the last solving stage includes:

in the first solving stage, a random ordering scheme is used as an initial solution, iterative solution is carried out according to a corresponding ordering neighborhood structure and a preset target by adopting a preset algorithm, and when a preset termination condition is reached, the solution is ended and an optimal ordering scheme is output;

and taking the optimal ordering scheme of the last solving stage as an initial solution, carrying out iterative solving according to a corresponding ordering neighborhood structure and a preset target by adopting a preset algorithm, and finishing the solving and outputting a candidate ordering scheme at the last solving stage when a preset termination condition is reached.

In an embodiment, the screening of the production ordering scheme for obtaining the order to be ordered according to the custom constraint condition from the candidate ordering schemes meeting the necessary constraint condition includes:

and judging the relation between the candidate sorting scheme and the self-defined constraint condition according to a preset judgment rule when the candidate sorting scheme meets the necessary constraint condition, and screening according to a corresponding judgment result to take the screened candidate sorting scheme as the production sorting scheme of the order to be sorted.

In an embodiment, the method further comprises:

and presetting the priority of each production constraint condition, wherein the self-defined constraint condition has the optional attributes of necessity and non-necessity.

The second aspect of the present application provides a production scheduling apparatus, which includes:

the data acquisition module is used for acquiring order data of the order to be sorted according to the product attribute of the vehicle;

the system comprises a configuration module, a sorting module and a display module, wherein the configuration module is used for respectively determining production constraint conditions and a sorting neighborhood structure according to a preset target of production, and the production constraint conditions comprise necessary constraint conditions and custom constraint conditions;

the processing module is used for obtaining candidate sorting schemes which accord with preset targets through at least two preset algorithms according to the order data and the sorting neighborhood structure, and screening and obtaining a production sorting scheme of the order to be sorted according to a user-defined constraint condition from the candidate sorting schemes which accord with necessary constraint conditions;

and the planning module is used for determining the production plan of the order to be sorted according to the production sorting scheme.

In one embodiment, the method comprises the following steps:

the processing module comprises at least one solving submodule, each solving submodule is used for sequentially solving through at least two solving stages, and a candidate ordering scheme which accords with a preset target is output in the last solving stage; and the preset algorithm adopted in each solving stage is respectively and independently set.

A third aspect of the present application provides an electronic device comprising:

a processor; and

a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method as described above.

A fourth aspect of the present application provides a computer-readable storage medium having stored thereon executable code, which, when executed by a processor of an electronic device, causes the processor to perform the method as described above.

The technical scheme provided by the application can comprise the following beneficial effects:

according to the technical scheme, the production ordering scheme achieving the preset target can be obtained by combining the preset algorithm according to different production constraint conditions and ordering neighborhood structures, and an optimal production plan is obtained. According to the design, the production ordering of the orders can be flexibly restricted through multi-dimensional and increasing and decreasing production restriction conditions, more potential production ordering schemes can be found through combination of ordering neighborhood structures, and the production ordering schemes which meet preset targets can be more comprehensively screened out through at least two preset algorithms, so that the production plan of the orders to be ordered can be determined by utilizing the better production ordering scheme obtained in the whole situation, and the finally determined production plan can meet the complexity requirement of finished automobile manufacturing. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The foregoing and other objects, features and advantages of the application will be apparent from the following more particular descriptions of exemplary embodiments of the application as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the application.

FIG. 1 is a schematic flow chart illustrating a method for scheduling production according to an embodiment of the present disclosure;

FIG. 2 is another flow chart of a production scheduling method according to an embodiment of the present application;

FIG. 3 is a flow chart of the production scheduling method shown in FIG. 2;

FIG. 4 is a schematic flow chart illustrating solving an objective function in the production scheduling method according to the embodiment of the present application;

FIG. 5 is a schematic flow diagram of the first solution phase shown in FIG. 4 for solving an objective function;

FIG. 6 is a schematic flow diagram of the second solution phase shown in FIG. 4 for solving an objective function;

FIG. 7 is a schematic flow diagram of the third solution phase shown in FIG. 4 for solving an objective function;

FIG. 8 is a schematic flow diagram of the fourth solution phase shown in FIG. 4 for solving an objective function;

FIG. 9 is a schematic diagram of a production scheduling apparatus according to an embodiment of the present application;

FIG. 10 is a schematic diagram of another embodiment of a production scheduling apparatus;

fig. 11 is a schematic structural diagram of an electronic device shown in an embodiment of the present application.

Detailed Description

Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While embodiments of the present application are illustrated in the accompanying drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms "first," "second," "third," etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.

In the related technology, the current APS system is difficult to meet the complex requirements of multi-train, less-batch, short-term delivery and personalized manufacture in the whole vehicle manufacturing of the intelligent electric vehicle, and the designed planning and scheduling cannot complete the production target.

In view of the above problems, embodiments of the present application provide a production scheduling method, which can quickly and reasonably deal with production planning and scheduling in a complex manufacturing scenario, and obtain a globally optimal production ordering scheme.

The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 is a flowchart illustrating a production scheduling method according to an embodiment of the present application.

Referring to fig. 1, an embodiment of the present application provides a production scheduling method, which includes:

and S110, acquiring order data of the order to be sorted according to the product attribute of the vehicle.

Since product attributes such as the model number, color, vehicle configuration, and expected delivery date of a vehicle are generally considered when an owner purchases a vehicle, at least data of the product attributes such as the model number, color, vehicle configuration, and expected delivery date of a vehicle is described in order data of the vehicle. It will be appreciated that for orders with the same product attributes, quantity consolidation may be performed so that the production ordering of the orders may be performed according to the corresponding order quantities, each at, for example, a paint shop for color or an assembly shop for vehicle configuration. And S120, respectively determining production constraint conditions and an ordering neighborhood structure according to a preset target, wherein the production constraint conditions comprise necessary constraint conditions and custom constraint conditions.

The preset target includes, but is not limited to, maximizing production profit, minimizing production time, or maximizing productivity efficiency according to the user's demand. In some embodiments, the corresponding objective function to achieve the preset objective may be set according to different preset objectives, so as to achieve the preset objective by solving the objective function.

In order to achieve the preset target, production constraint conditions can be set according to objective factors and subjective factors, wherein the production constraint conditions comprise necessary constraint conditions and custom constraint conditions, the necessary constraint conditions refer to constraint conditions which must be met by the solution result of the objective function, and if the necessary constraint conditions are violated, the solution result is unreasonable. The user-defined constraint conditions are independently set by a user according to actual production requirements, the number of the user-defined constraint conditions is not limited, the number of the user-defined constraint conditions can be increased or decreased by the selection of the user, and meanwhile, specific limited contents of each user-defined constraint condition can also be independently set by the user. The solution result may not meet all the custom constraints, that is, the solution result may meet some or all of the custom constraints or may not meet all of the custom constraints on the premise of meeting the necessary constraints.

In order to facilitate evolution and iteration of candidate ranking schemes in subsequent steps, the mode of generating different candidate ranking schemes, namely, the scheme of how to rank each order to be ranked, can be determined by setting the ranking neighborhood structure. In some embodiments, ordering the neighborhood structure includes, but is not limited to, the following: randomly adjusting the production sequence of any order to be ordered in the known production sequence; randomly transposing the production sequence of any two orders to be sorted in the known production sequence. It can be understood that each sort neighborhood structure can generate different candidate sort schemes, so that alternative candidate sort schemes can be expanded, and a subsequent algorithm can find a better candidate sort scheme as a final production sort scheme.

S130, obtaining candidate sorting schemes according with preset targets through at least two preset algorithms according to the order data and the sorting neighborhood structure, and screening the candidate sorting schemes according with necessary constraint conditions to obtain a production sorting scheme of the order to be sorted according to the user-defined constraint conditions.

The preset algorithm includes, but is not limited to, a combination of two or more of a tabu search algorithm, a simulated annealing algorithm, a delayed acceptance algorithm, and a hill climbing algorithm. Through the respective calculation of various algorithms, the local optimal solution of the sequencing scheme, which is generated by the self limitation of a single algorithm, can be avoided. Furthermore, various candidate sorting schemes provided by the sorting neighborhood structure and the constraint of the candidate sorting schemes through the production constraint condition can establish a mathematical model for achieving a preset target, namely solving an objective function. The mathematical model can generate a series of candidate sorting schemes according to the sorting neighborhood structure, and then at least two preset algorithms respectively calculate each candidate sorting scheme according to order data to determine the probability of receiving the candidate sorting scheme, so that the received candidate sorting scheme can be ensured to achieve a preset target as far as possible. And judging the acceptable candidate sorting schemes according to the production constraint conditions, wherein the candidate sorting schemes meeting the necessary constraint conditions can be reserved after the production sorting schemes not meeting the necessary constraint conditions are eliminated. Further, for example, the candidate ranking schemes may be further evaluated according to a preset evaluation rule, that is, the advantages and disadvantages of the candidate ranking schemes under all the custom constraint conditions are evaluated, so as to screen out a preferred candidate ranking scheme. Optionally, the screened best candidate ranking scheme may be used as the final production ranking scheme for each order to be ranked.

And S140, determining a production plan of the order to be sorted according to the production sorting scheme.

After the production ordering scheme of each order to be ordered is determined, the production plan of the order to be ordered is determined according to the processing sequence of each order in the production ordering scheme in different process plants. It should be understood that the production plan includes corresponding production tasks for manufacturing corresponding vehicles in the order to be ordered within a preset time at each workshop.

It can be seen from this embodiment that, according to the technical scheme of the present application, a candidate ranking scheme that achieves a preset target can be obtained by combining at least two preset algorithms according to a production constraint condition and a ranking neighborhood structure, and an optimal production ranking scheme is obtained by screening to serve as a production plan. By means of the design, the production ordering of orders can be flexibly restricted through multi-dimensional and increasing and decreasing production restriction conditions, potential ordering schemes can be quickly and effectively found through an ordering neighborhood structure and a plurality of preset algorithms, and therefore better production ordering schemes which accord with preset targets can be screened in the overall situation.

Fig. 2 and fig. 3 are another flow chart of the production scheduling method according to the embodiment of the present application.

Referring to fig. 2 and 3, a production scheduling method according to an embodiment of the present application includes:

and S210, acquiring order data of the order to be sorted according to the product attribute of the vehicle.

The product attributes of the vehicles comprise the types, colors, vehicle configurations, expected delivery dates and the like of the vehicles, and different product attributes need to be processed and produced through production workshops of different processes, so that order data corresponding to each vehicle to be produced comprise the product attributes. In order to reasonably arrange the production sequence of each order, wherein the number of the orders to be ordered generally cannot exceed the maximum order carrying capacity of a producer, namely a factory, the number of the orders to be ordered which can be loaded in the current scheduling can be determined according to the maximum order carrying capacity, the scheduling starting date and the scheduling ending date of the factory, and corresponding order data is obtained.

And S220, determining an objective function according to a preset target.

Taking the production of a whole intelligent electric car as an example, the preset target may be, for example, minimizing the production completion time of all orders. Specifically, for example, n orders to be ordered for manufacturing a finished automobile are provided, and each order needs to go through m processes, the preset target, that is, the objective function f of the minimum completion time, may be shown by the following formula (1):

f＝min(max _1≤i≤n C _ij ) (1)

therein, max _1≤i≤n C _ij And j is more than or equal to 1 and less than or equal to m, and j and m are integers. f denotes the waiting of all intelligent electric vehicle manufactureThe minimum elapsed time from the beginning to the end of the order is ordered.

It should be understood that when the preset targets are different, the target function is adjusted accordingly.

And S230, determining production constraint conditions, wherein the production constraint conditions comprise necessary constraint conditions and custom constraint conditions.

In one embodiment, the necessary constraints include at least one of:

constraint 1, the sum of the time to start manufacture of the order to be sorted and the time offset value is greater than or equal to the current time.

Constraint 2, each order to be sorted is sorted only once.

Constraint 3, the planned start time of the order to be sorted is earlier than the corresponding planned end time.

Constraint 4, the single-day capacity consumption value of all orders to be sorted in a single plant is less than or equal to the total single-day capacity of the plant.

In a specific embodiment, the constraint 1 can be expressed by the following equation (2):

StartingTime(C _ij )+FirstProcessOffset(C _ij )≥CurrentTime() (2)

wherein StartingTime (C) _ij ) Represents order C _ij The manufacturing start time of (2) may be expressed as a manufacturing start time of the entire vehicle, taking an order of the entire vehicle manufacturing as an example. firstProcessOffset (C) _ij ) Is order C _ij The offset value of the first process start time of (2), CurrentTime () represents the current time. It should be understood that in actual production, production may not be performed completely at the set time beat, so that it is necessary to set the offset value, that is, the delay time period for starting the first process. It should be understood that the constraint 1 is a sum of offset values of the manufacturing start time of the order and the first process start time of the order, and must be a time in the future, that is, a time which must be later than the current time and cannot be passed any more, so as to satisfy the objective factor.

In a specific embodiment, the constraint 2 can be expressed by the following equation (3):

wherein the content of the first and second substances,

representing the ranking results for all orders within T days,

representing all unsorted orders in T days, C _ijt The time of completion of the order i on the process j on the T day is shown, and the value of T is customized by the user. It will be appreciated that the products of each order can only be manufactured once and then only participate in the ordering once, so the number of orders before ordering must be equal to the number of orders after ordering.

In a specific embodiment, the constraint 3 can be expressed by the following formula (4):

PlanningTime(C _ij )<EndingTime(C _ij ) (4)

among them, PlanningTime (C) _ij ) Order C _ij Scheduled start time of (C), EndingTime (C) _ij ) Order C _ij The manufacturing completion time of (1). The meaning of equation (4) means that the scheduled start time of the ith order in the jth process must be earlier than the scheduled end time of the order.

In a specific embodiment, the constraint condition 4 can be expressed by the following formula (5):

∑ _st C _ijst ≤P _st (5)

wherein s denotes the individual workshops in the process line, for example in the production of a complete vehicle, such as stamping, painting, welding, etc., P _st Representing the capacity of the s workshop in t days, wherein the capacity can be the working time of workers and equipment; c _ijst Order C representing plant s on day t _ij When the time is finished. Equation (5) tableWhen planning and scheduling the capacity of a workshop, the capacity consumption of all orders to be sorted must not exceed the total daily capacity P of the workshop _st Thus conforming to objective factors.

Preferably, to rationally implement the planning schedule for each order to be ordered, in one embodiment, the production constraints comprise all necessary constraints. Optionally, the necessary constraint condition may be further expanded according to actual requirements, and is not limited herein.

Further, the production constraints also include custom constraints, which in one embodiment include at least one of:

and the constraint condition 5, the ratio of the daily production quantity of the order with the specified product attribute to the daily production total quantity is set according to a custom ratio. Wherein the product attribute includes, but is not limited to, product color, product model, etc.

And (6) producing the orders with the specified product attributes according to the custom batch according to the constraint conditions.

Constraint 7, the production quantity of an order for a specified product attribute on a specified date is greater than or equal to a preset production quantity.

In a specific embodiment, the constraint 5 can be expressed by the following equation (6):

wherein A is _kt Representing the production volume of the order with product attribute k on day t,

represents the total order throughput in t days, where p ₁ ，p ₂ Is customized for the user, p ₁ And p ₂ Are each an integer greater than or equal to 0, p ₁ ≤p ₂ . The above formula (6) represents that the ratio of the two is p ₁ :p ₂ . For ease of understanding, for example, the production volume of an order for a vehicle that requires production attributes red for the day is100, and the total production of orders in the day is 600, the ratio of the production of orders to the production of orders is 100:600, and the ratio can be independently set on different dates. It should be understood that this ratio is the actual order throughput, and is in an objective sense, and is not allowed to be approximately 1: 6.

In a specific embodiment, the constraint condition 6 can be expressed by the following formula (7):

A ₁ :A ₂ :A ₃ …:A _k ＝q ₁ :q ₂ :q ₃ …:q _k (7)

wherein A is _k Representing the production quantity with product attribute k, q _k Representing the production quantity per batch with product attribute k, q _k ∈[a,b]And a and b can be any natural numbers defined by a user. For convenience of understanding, for example, the product attributes are specified as product colors, a is 3, b is 5, and the product attributes of the vehicles in the order to be sorted respectively include white, red and black, so that the production quantity per batch of the vehicles in each color in the production sorting needs to be within 3-5, for example, a white vehicle: red vehicle: the black car is 3:4:5, that is, when the production sequence of the order is performed, 3 white cars are required to be produced in each batch, 4 red cars are required to be produced in each batch, and then 5 black cars are required to be produced in each batch, it is understood that there is no calculation of the divisor in the formula (7), that is, when the white cars: red vehicle: when a black car is 5:5:5, it cannot be equated with a white car: red vehicle: black car 1:1: 1.

In a specific embodiment, the constraint condition 7 can be expressed by the following equation (8):

A _kt ≥N (8)

wherein A is _kt The production quantity of the product with the attribute of k at the t day is represented, N represents the preset yield, t and N are positive integers, t is larger than or equal to 1, and N is larger than or equal to 0. Equation (8) shows that on the production sequence of day t, an order with product attribute k requires N or more outputs. For example, it may be constrained that on day t, a vehicle with a product attribute of white needs to produce over 100.

It can be understood that the number of the custom constraints is not limited to the above example, and the user can also expand according to the actual requirement, and autonomously increase or decrease the custom constraints each time the production sorting is performed. In some embodiments, the custom constraint conditions include necessary and non-necessary attributes to be selected, that is, a user can respectively select the attributes of each custom constraint condition according to requirements, and when the necessity is selected, the corresponding custom constraint condition can be regarded as a necessary constraint condition; when the non-necessity is selected, then the corresponding custom constraint does not have the mandatory of the necessary constraint.

Further, in some embodiments, the priority of each production constraint is set in advance. It can be understood that the priority of all necessary constraint conditions is higher than that of all self-defined constraint conditions with non-necessary attributes, on the basis, the weight of each production constraint condition is respectively set, and the priority of each production constraint condition is adjusted according to the weight, so that the guidance of a subsequent algorithm in calculation is facilitated, and the effective iteration and convergence of a solution result are facilitated. In addition, according to different priority settings, the result of the judgment according to the production constraint condition with higher priority may be used as an initial value when the judgment is performed according to the production constraint condition with lower priority, and the specific scheme is described in detail later, and is not described herein again.

S240, determining a sorting neighborhood structure of the production sorting scheme.

It is to be understood that a neighborhood is any open interval centered at point x, denoted as u (x). The interval in which δ is a neighborhood is U (x, δ) ═ x- δ, x + δ. However, when x is a solution, for example, x is a production ordering scheme of each order to be ordered, it is difficult to determine the neighborhood structure of the solution at this time, because there may be a production ordering scheme better than the current x in different types of neighborhood structures. Thus, determining the neighborhood structure of x also determines the way in which different solutions, i.e., different production ordering schemes, are generated.

In order to solve the problem of production sorting of each order, in this step, the sorting neighborhood structure may be expressed according to the following formula:

wherein insert (x) indicates that the production order of any order to be ordered in the known production ordering scheme x is randomly adjusted, so that the order is shifted forwards or backwards. For example, in the known production sequence scheme x 1 → 2 → 3 → 4 → 5 → 6 → 7, order 2 is extracted and inserted between order 4 and order 5 by means of insert (x), forming a new production sequence scheme x': 1 → 3 → 4 → 2 → 5 → 6 → 7. Due to the fact that orders in the two production sorting schemes are sorted differently, the result of solving the objective function in the subsequent algorithm is different, and meanwhile the judgment scores g (x) of the subsequent step in the screening and judging process for the different production sorting schemes are different. For example, if g (x ') > g (x), then x ' is a more optimal solution, then the iterative search may continue on the neighborhood of x ' rather than spread out on the neighborhood of x.

Wherein swap (x) means to randomly swap the production sequence of any two orders to be sorted in the known production sorting scheme, i.e. swap the production sequence of two orders in the known production sorting scheme x to generate a new production sorting scheme x ". For example, the known production sequence scheme x of the existing vehicle orders 1 to 7 is 1 → 2 → 3 → 4 → 5 → 6 → 7, and the positions of the order 2 and the order 6 are exchanged in swap (x) manner to form a new production sequence scheme x ″:1 → 6 → 3 → 4 → 5 → 2 → 7. Similarly, different production ordering schemes can be generated by adopting the ordering neighborhood structure of swap (x), so that potential solution results can be searched in different ways to obtain more production ordering schemes for screening.

It should be appreciated that the ordering neighborhood structure is not limited to the two examples described above, but is merely illustrative. The steps S230 and S240 may be executed without being separated from each other.

S250, solving an objective function in at least two solving stages in sequence according to order data, a sorting neighborhood structure and production constraint conditions to obtain a candidate sorting scheme; and according to a preset termination condition, taking the optimal ordering scheme of the last solving stage as a production ordering scheme of the order to be ordered.

In order to avoid taking the local optimal solution as the final production sequencing, a plurality of solving stages can be preset for calculation, and corresponding preset algorithms can be configured in each solving stage respectively, so that an algorithm combination is formed. In one embodiment, the solution is performed through at least two solution stages in sequence, and a candidate ordering scheme meeting a preset target is output in the last solution stage; and the preset algorithms adopted in each solving stage are respectively and independently set. In a specific embodiment, a random ordering scheme is adopted as an initial solution in a first solving stage, a preset algorithm is adopted to carry out iterative solution according to a corresponding ordering neighborhood structure and a preset target, and when a preset termination condition is reached, the solution is ended and an optimal ordering scheme is output; and taking the optimal ordering scheme of the last solving stage as an initial solution, carrying out iterative solving according to a corresponding ordering neighborhood structure and a preset target by adopting a preset algorithm, and finishing the solving and outputting a candidate ordering scheme at the last solving stage when a preset termination condition is reached.

For convenience of understanding, taking the number of the solution stages as 2 as an example, a random solution, that is, a random ordering scheme, is adopted as an initial solution in the first solution stage to perform iterative solution in the ordering neighborhood structure, then a preferred solution, that is, a preferred ordering scheme, in the 1 st solution stage is adopted as an initial solution in the 2 nd solution stage to perform iterative solution in the ordering neighborhood structure, and a candidate ordering scheme meeting a preset target is output in the last solution stage, where the number of the candidate ordering schemes may be 1 or more. In addition, the preset algorithms adopted in each solving stage are respectively and independently set. For example, when the preset algorithms include a tabu search algorithm, a simulated annealing algorithm, a delayed acceptance algorithm, a hill climbing algorithm, etc., one of the algorithms may be selected at each solution stage as the preset algorithm of the current solution stage, and the preset algorithms selected at different solution stages may be the same or different. Alternatively, the ordering neighborhood structures selected by the different solution phases may be the same or different. By selecting respective pre-set algorithms and ordering neighborhood structures at different solution stages, a set of comprehensive mathematical models can be formed to solve the objective function.

Further, in order to improve the calculation efficiency of each solution stage, each solution stage may be respectively provided with a corresponding preset termination condition, so as to control the termination of the solution process of the preset algorithm, and avoid the situation that the calculation is performed for a long time and the convergence cannot be achieved. Optionally, in an embodiment, the preset termination condition may be: when the optimal solution is obtained in the current solving stage, the solving is finished; or ending the solution when the optimal solution is not obtained in the current solution stage and the preset solution time is reached. Of course, the preset termination condition may also be other conditions, which are only exemplified and not limited herein. Preferably, in an embodiment, the preset solution time of each solution stage may be set independently, that is, the solution time of each stage may be the same or different. In order to control the total solving time length and ensure the calculation efficiency, the accumulated total solving time length of all the solving stages can be set by a user in a self-defined way or a default value can be preset. When the total solving time length is set by user self-definition, the solving time length of each solving stage can be distributed according to a preset proportion, and the method is not limited in this respect.

Further, after the solution of each solution phase is finished according to the preset termination condition, there may be more than one solution result, that is, each solution phase may obtain a plurality of candidate ranking schemes after finishing. In an embodiment, one of the preferred ranking schemes may be selected as the production ranking scheme of the order to be ranked, or the optimal solution, that is, the optimal ranking scheme, may be selected as the production ranking scheme of the order to be ranked.

In order to obtain a more reliable production sorting scheme, in an embodiment, when the candidate sorting scheme meets the necessary constraint condition, the relation between the candidate sorting scheme and the custom constraint condition is judged according to a preset judgment rule, and screening is performed according to a corresponding judgment result, so that the screened candidate sorting scheme is used as the production sorting scheme of the order to be sorted. That is, on the premise that the solution results, that is, the candidate sorting schemes, must meet the necessary constraint conditions but may not meet the user-defined constraint conditions, the relevance between the solution results and the user-defined constraint conditions may be evaluated, so that each solution result may obtain a corresponding evaluation result, for example, the evaluation results may be visually represented in the form of scores, thereby determining the merits of each solution result according to the size of the scores, and then screening out the production sorting scheme of the optimal solution as the production sorting scheme of the order to be sorted. By means of the design, judgment is performed in a quantitative mode, intuitive selection of a user is facilitated, processing efficiency is improved, and meanwhile reliability of a judgment result is guaranteed. In one embodiment, each of the defined constraints has a corresponding predetermined evaluation rule. That is, different custom constraints can adopt independent evaluation rules, so that the production sorting scheme can be evaluated more objectively.

For ease of understanding, it is assumed that this step sets 4 solving stages to solve the objective function, and the following description uses 4 solving stages to illustrate with reference to fig. 4 to 8. Of course, in other embodiments, the number of specific solution phases is not limited thereto.

S251, solving an objective function through a first algorithm in a first solving stage according to a random ordering scheme as an initial solution; and when the preset termination condition is reached, screening to obtain the optimal solution of the first solving stage.

As shown in fig. 4 and 5, after the sorting scheme of the initial solution x is randomly selected, a new solution, that is, a new sorting scheme, is generated according to two different sorting neighborhood structures insert (x) and swap (x) in sequence. For example, the ordering scheme of each order of the initial solution x is 1 → 2 → 3 → 4 → 5 → 6, and after the ordering scheme of the initial solution is reordered according to the neighborhood structure of insert (x), a new solution 1 → 3 → 4 → 2 → 5 → 6 can be obtained; the new solution ordering scheme 1 → 3 → 4 → 2 → 5 → 6 is reordered according to the neighborhood structure of swap (x), e.g., the latest solution ordering scheme 3 → 1 → 4 → 2 → 5 → 6 can be obtained. For example, the first algorithm is a tabu search algorithm, the latest solution is input into the first algorithm, and an objective function is solved according to whether the latest solution meets a preset target or not according to order data so as to calculate the acceptance probability of the latest solution. And when the acceptance probability of each latest solution is greater than a preset probability threshold, determining that the latest solution can be accepted, and forming a corresponding candidate sorting scheme by using the sorting scheme of the latest solution.

Further, when the judgment is performed according to the preset judgment rule, the preset judgment rule corresponding to the necessary constraint condition is a scheme that the candidate sorting scheme must meet the necessary constraint condition, and the candidate scheme that does not meet the necessary constraint condition is directly excluded without further judging the relationship with the user-defined constraint condition. Further, if the candidate sorting scheme is determined to belong to the optimal solution according to the preset evaluation rule corresponding to the user-defined constraint condition, the current solution stage can be ended when the preset termination condition is reached, and the optimal solution is used as the initial solution of the second solution stage. If the candidate sorting scheme determines that the candidate sorting scheme does not belong to the optimal solution according to the preset evaluation rule corresponding to the custom constraint condition and does not reach the preset solving time of the solving stage, for example, 10 minutes, the current solving stage does not reach the preset termination condition, the latest solution needs to be generated into a new solution again according to neighborhood structures of insert (x) and Swap (x), then the calculation is performed by the first algorithm and the evaluation is performed according to the preset evaluation rule, and the process is repeated until the preset termination condition is reached, the solving of the current solving stage is ended, and the optimal solution is screened according to the preset evaluation rule. That is, the ordering scheme of the optimal solution may be taken as the preferred ordering scheme.

And judging whether the candidate sorting scheme in the solution result meets the corresponding self-defined constraint condition or not according to a preset judgment rule, and obtaining a corresponding judgment result in a scoring mode so as to visually and objectively screen.

Specifically, the predefined evaluation rules corresponding to the respective defined constraints are respectively described as examples by taking the customized constraints in the constraints 5 to 7 as examples, but the predefined evaluation rules are not limited to the following examples and are only described as examples herein. For example, let g (x) be the score of candidate ranking scheme x, score be the initial score, A _kt Yield on day t for product with product attribute k, p _kt Specifying for the user the production proportion on day t of a product with product attribute k, Y _t For products of all product attributes on day tTotal yield. w is the weight of the self-defined constraint, and the weights of different constraints are independently set and can be self-defined by a user, or the default setting w is 1.

For the constraint condition 5, the candidate ranking scheme x may be evaluated according to a preset evaluation rule in the following formula (9).

For ease of understanding, assuming that the total production of all product attributes on a day is 200 entire vehicles, where the user specifies that 100 vehicles A and 100 vehicles B are to be produced, then p for vehicle A _kt 100/200-0.5. If the vehicle is ranked according to the candidate ranking scheme x, only 40 vehicles A can be produced in the candidate ranking scheme x, namely A _kt For 40, if there are 60 less vehicles than the 100 vehicles specified by the user, 40/200 is 0.2 and 0.2 is not equal to 0.5, i.e., a _kt :Y _t ≠p _kt . Assuming that score has an initial score of-30 and w is 1, f (x) -30-1 × |40-200 × 0.5| -30-60 ═ 90. If it is determined that 100 vehicles a can be produced after ranking according to the candidate ranking scheme x, the final score g (x) is-30, i.e., no discount is required. On the premise that the addend item is not preset, the candidate ranking scheme x capable of maintaining the initial score is the optimal ranking scheme, namely the optimal solution under the current constraint condition.

With respect to the above-mentioned constraint 6, the production sort plan x may be evaluated according to a preset evaluation rule in the following formula (10).

Specifically, it is known to determine q in the aforementioned formula (7) _k ∈[a,b]That is, when the candidate ranking scheme completely satisfies the constraint condition 6, it can be determined that a is less than or equal to A _k B is less than or equal to b. However, when the candidate ranking scheme does not satisfy the constraint condition 6, there are four cases of non-compliance, and different cases may adopt the above equation (10)The score value of (2) is calculated, thereby generating different deduction situations. Wherein, four specific nonconformities are as follows:

1) case 1: the last capacity is not satisfied with A _k <a. For example, the total energy of a factory of the intelligent trolley enterprise on a certain day is 13 vehicles, and a user specifies that 3-4 vehicles are required to be produced in one batch. If the plant produces three vehicles, white, red and black, the total production capacity of 13 vehicles will produce the following production sequencing scheme: white red black white red black red. And because the last order is only produced by 1 vehicle and does not meet the requirement that 3-4 vehicles are one batch, deduction is needed.

2) case 2: the initial capacity can not satisfy A _k A is less than or equal to a. Similar to case 1, for example, the candidate sorting scheme is red white, red, black, white, and black, and the first order only schedules 1 vehicle for production, and does not satisfy the requirement that 3-4 vehicles are one batch, so that the deduction is needed.

3) case 3: product quantity A of a certain product attribute _k Greater than b, i.e. b < A _k 。

4) case 4: product quantity A of a certain product attribute _k Less than a, and the product is not at the beginning capacity or the end capacity of the production of the same day.

It will be appreciated that, according to the 4 cases described above, the score g (x) may be calculated according to the corresponding calculation in the actual candidate ranking scheme selection formula (10).

Further, in formula (10), let index be the index of the last order of the target attribute product specified by the user in any candidate sorting scheme, and index is an integer greater than or equal to 0. The user can specify a product with a plurality of target product attributes, and the processing methods of the products are the same. For example, taking the product attribute as a color, there are 10 candidate ranking schemes x of vehicle products with different colors: and if the index starts from 0, the index of each product is 0-9 in sequence, and the index of the product with the red target product attribute is 6. Let maxIndex be the first different product attribute product that the product lot of the target product attribute indexes forward from index. In the candidate ranking scheme, the product with the first different product attribute before the red vehicle is a white vehicle, the index of the white vehicle is 2, and the maxIndex of the red vehicle is 2. Let minIndex be the index of the first non-product attribute product after the target attribute product batch, and in the candidate sorting scheme, if the first different product attribute product after the red vehicle batch is a black vehicle, the corresponding minIndex is 7.

Further, g (x) calculated according to the constraint condition 5 can be used as the initial score of the constraint condition 6, and the corresponding calculation mode is selected in the formula (10) according to the actual situation to calculate the corresponding score of the production sequencing scheme.

For the constraint condition 7, the candidate ranking scheme may be evaluated according to a preset evaluation rule in the following formula (11).

Wherein the constraint condition 7 requires that the order production quantity of the product with the product attribute of k at the t-th day is not lower than the preset production quantity N. Further, the constraint condition 7 can be further expanded to require that the order production quantity of the product with the product attribute k at the t-th day is not higher than the preset production quantity M, and the value of M can be set by a user in a customized manner. In the evaluation, a corresponding calculation mode is selected in formula (11) according to the actual candidate ranking scheme. Similarly, the score of g (x) calculated according to constraint 6 as described above can be used as the initial score of constraint 7.

It can be understood that in the same formula, the w values of the weights are the same, and in different formulas, the corresponding w values of the weights can be different and are set by a user in a self-defined manner. According to the weight set by the user for different self-defined constraint conditions, the priority of each constraint condition can be determined. Preferably, the larger the weight is, the higher the priority is, the order of the evaluation is preferentially performed, and then the score after preferentially performing the calculation may be taken as an initial score of the evaluation rule according to the subsequent constraint condition.

After scoring is sequentially performed according to the preset scoring rules corresponding to all the custom constraint conditions, the final scoring result, i.e. the final score, of the candidate sorting scheme can be obtained. And when the final score is equal to the initial score, determining that the candidate sorting scheme is the optimal solution, otherwise, determining that the candidate sorting scheme is not the optimal solution, and continuously iterating and calculating the score of a new candidate sorting scheme according to the first algorithm before the preset solving time is not cut off. And if the optimal solution is not obtained after the preset solving time is reached, taking the candidate sorting scheme with the highest score in all the production sorting schemes as the optimal solution of the current solving stage, namely the optimal sorting scheme, and taking the optimal solution as the initial solution of the next solving stage.

S252, solving an objective function through a second algorithm in a second solving stage according to the optimal solution in the first solving stage as an initial solution; and when the preset termination condition is reached, screening to obtain the optimal solution of the second solving stage.

As shown in fig. 6, after the sorting scheme of the optimal solution in the first solution phase is obtained, a new solution, that is, a new sorting scheme, is generated according to two different sorting neighborhood structures insert (x) and swap (x) in sequence. For example, if 3 → 1 → 4 → 2 → 5 → 6 is the optimal solution for the first solution stage, then the new solution can be obtained, e.g., 3 → 4 → 2 → 1 → 5 → 6, by reordering according to the neighborhood structure of insert (x); after the new solution ordering scheme 3 → 4 → 2 → 1 → 5 → 6 is reordered according to the neighborhood structure of swap (x), for example, the latest solution ordering scheme 4 → 3 → 2 → 1 → 5 → 6 can be obtained. For example, the second algorithm is a simulated annealing algorithm, the latest sorting scheme, i.e., the latest solution, is input into the first algorithm, and the objective function is solved according to whether the latest solution meets the preset target or not according to the order data, so as to calculate the acceptance probability of the latest solution. And when the acceptance probability of each latest solution is greater than a preset probability threshold, determining that the latest solution can be accepted, and forming a corresponding candidate sorting scheme.

Similarly, referring to the first solving stage, if the candidate sorting scheme determines to belong to the optimal solution according to the preset evaluation rule, the current solving stage can be ended when the preset termination condition is reached, and the optimal solution is used as the initial solution of the second solving stage. If the candidate sorting scheme is determined not to belong to the optimal solution according to the preset evaluation rule and the preset solving time of the solving stage is not reached, for example, 10 minutes, the candidate sorting scheme indicates that the current solving stage does not reach the preset termination condition, the latest solutions need to be sorted again according to insert (x) and Swap (x) in sequence to generate new solutions, then the new solutions are calculated by a second algorithm and evaluated according to the preset evaluation rule until the preset termination condition is reached, the solving of the current solving stage is ended, the optimal solutions are screened according to the preset evaluation rule, and the optimal solutions are used as initial solutions of the next solving stage.

S253, solving an objective function through a third algorithm in a third solving stage according to the optimal solution of the second solving stage as an initial solution; and when the preset termination condition is reached, screening to obtain the optimal solution of the third solving stage.

As shown in fig. 7, similarly, the optimal solution of the third solution phase can be obtained by referring to the related description of the first solution phase, which is not described herein again. The third algorithm may be, for example, a delayed acceptance algorithm.

S254, solving an objective function through a fourth algorithm in a fourth solving stage according to the optimal solution of the third solving stage as an initial solution; and when the preset termination condition is reached, screening to obtain the optimal solution of the fourth solving stage.

As shown in fig. 8, similarly, the optimal solution of the fourth solution phase can be obtained by referring to the related description of the first solution phase, which is not described herein again. The fourth algorithm may be, for example, a hill climbing algorithm. Different from the three solving stages, the solving stage adopts a more-time ordering neighborhood structure for ordering, and the preset solving time length adopted in the stage is longer than the preset solving time lengths of other stages, so that different solving results are obtained through more comprehensive iteration in the last solving stage, more candidate ordering schemes are obtained, the omission of potential ordering schemes is avoided, and the global optimal ordering scheme is found. It should be noted that, in order to control the calculation efficiency, the user may customize the total solution time, and the solution time of the last stage should not exceed the total solution time of all the solution stages.

It should be noted that, in the above steps S251 to S254, the preset algorithm used in each solving stage is only for example, and is not limited herein, and all the steps may also use the same or partially the same algorithm to perform the calculation, but different algorithm combinations are used in different solving stages, so that a more comprehensive candidate ranking scheme can be obtained, and the solving stages are verified with each other, thereby avoiding generating a local optimal solution. Of course, the type and the number of the sorting neighborhood structures for each solution phase can be freely combined, and the method is not limited herein.

As can be seen from this example, in the scheme of the present application, a combined configuration can be formed by using respective preset algorithms in different solution stages, and an optimal solution can be found by using different sorting neighborhood structures; meanwhile, by flexibly increasing or decreasing the user-defined constraint conditions and setting the corresponding preset evaluation rules, a more objective and reliable evaluation result can be obtained, the actual requirements of a user can be flexibly met, the method is suitable for different industrial manufacturing scenes, and the application is wide.

Corresponding to the embodiment of the application function implementation method, the application also provides a production scheduling device, electronic equipment and a corresponding embodiment.

Fig. 9 is a schematic structural diagram of a production scheduling apparatus according to an embodiment of the present application.

Referring to fig. 9, the production scheduling apparatus according to an embodiment of the present application includes a data obtaining module 910, a configuration module 920, a processing module 930, and a determining module 940. Wherein:

the data obtaining module 910 is configured to obtain order data of the order to be sorted according to the product attribute of the vehicle.

The configuration module 920 is configured to determine a production constraint condition and a sorting neighborhood structure according to a preset target of production, where the production constraint condition includes a necessary constraint condition and a custom constraint condition.

The processing module 930 is configured to obtain candidate ordering schemes meeting preset targets through at least two preset algorithms according to the order data and the ordering neighborhood structure, and obtain a production ordering scheme of the order to be ordered by screening according to the custom constraint condition from the candidate ordering schemes meeting the necessary constraint condition.

The determining module 940 is configured to determine a production plan of the order to be sorted according to the production sorting scheme.

The order data acquired by the data acquisition module 910 includes, but is not limited to, product attribute data such as the model number, color, vehicle configuration, and product attributes such as the expected delivery date of the vehicle. By acquiring order data, accurate calculations by the processing module 930 are facilitated.

The configuration module 920 may define a preset target by a user, and determine a corresponding target function, a corresponding production constraint condition, and a corresponding sorting neighborhood structure according to the corresponding preset target. Wherein the necessary constraints comprise at least one of: the sum of the manufacturing starting time of the order to be sorted and the time offset value is greater than or equal to the current time; each order to be sorted is only sorted once; the plan starting time of the order to be sorted is earlier than the corresponding plan ending time; the single-day capacity consumption value of all orders to be sorted in a single workshop is less than or equal to the total single-day capacity of the workshop. The custom constraints include at least one of: setting the ratio of the daily production quantity of the order of the specified product attribute to the daily total production quantity according to a custom ratio; producing the order with the specified product attribute according to the user-defined batch; the production quantity of the order with the specified product attribute on the specified date is greater than or equal to the preset production quantity. The ordering neighborhood structure includes at least one of: randomly adjusting the production sequence of any order to be ordered in the known production sequence; randomly transposing the production sequence of any two orders to be sorted in the known production sequence.

Alternatively, the content and number of custom constraints may be customized by the user, with alternative attributes of the custom constraints being necessary and unnecessary, and then whether or not the custom constraints are converted into necessary constraints may be customized by the user. Further, the configuration module 920 is further configured to preset priorities of the production constraints, where the custom constraints have candidate attributes of necessity and non-necessity. Configuration module 920 can configure the candidate attributes of the custom constraint.

Referring to fig. 10, fig. 10 is another schematic structural diagram of a production scheduling apparatus according to an embodiment of the present application.

As shown in fig. 10, the processing module 930 of the production scheduling apparatus according to an embodiment of the present invention may include a sorting module 931, at least one solving sub-module 932, and an evaluating module 933.

The ranking module 931 is configured to iteratively generate a candidate ranking scheme according to the ranking neighborhood structure, where the candidate ranking scheme is used by the solving submodule as an initial solution or a preferred solution to be screened in each solving stage.

Each solving submodule 932 is respectively used for sequentially solving through at least two solving stages and outputting a candidate ordering scheme which accords with a preset target in the last solving stage; and the preset algorithms adopted in each solving stage are respectively and independently set. Further, the configuration module 920 may also be configured to configure a preset algorithm used in each solution phase.

Further, the solving submodule 932 is also configured to use a random ordering scheme as an initial solution in the first solving stage, perform iterative solution according to a preset target by using a preset algorithm according to the corresponding ordering neighborhood structure, and when a preset termination condition is reached, end the solution and output an optimal ordering scheme; and taking the optimal ordering scheme of the last solving stage as an initial solution, carrying out iterative solving according to a corresponding ordering neighborhood structure and a preset target by adopting a preset algorithm, and finishing the solving and outputting a candidate ordering scheme at the last solving stage when a preset termination condition is reached.

The evaluation module 933 is configured to, when the candidate sorting scheme meets the necessary constraint condition, evaluate the relationship between the candidate sorting scheme and the custom constraint condition according to a preset evaluation rule, and perform screening according to a corresponding evaluation result, so as to use the screened candidate sorting scheme as a production sorting scheme of the order to be sorted. And when the solving result does not accord with the necessary constraint condition, directly excluding the corresponding solving result.

In summary, the production scheduling device of the application can flexibly restrict the production ordering of orders and screen better solution results through multidimensional and increasing and decreasing production restriction conditions, and can quickly and more comprehensively and effectively find out a potential ordering scheme through an ordering neighborhood structure in combination with a preset algorithm, so that an optimal ordering scheme can be obtained in the whole situation, and a production ordering scheme of orders to be ordered can be obtained.

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Referring to fig. 11, the electronic device 1000 includes a memory 1010 and a processor 1020.

The Processor 1020 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 1010 may include various types of storage units, such as system memory, Read Only Memory (ROM), and permanent storage. Wherein the ROM may store static data or instructions that are needed by the processor 1020 or other modules of the computer. The persistent storage device may be a read-write storage device. The persistent storage may be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off. In some embodiments, the persistent storage device employs a mass storage device (e.g., magnetic or optical disk, flash memory) as the persistent storage device. In other embodiments, the permanent storage may be a removable storage device (e.g., floppy disk, optical drive). The system memory may be a read-write memory device or a volatile read-write memory device, such as a dynamic random access memory. The system memory may store instructions and data that some or all of the processors require at runtime. Further, the memory 1010 may comprise any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic and/or optical disks, among others. In some embodiments, memory 1010 may include a removable storage device that is readable and/or writable, such as a Compact Disc (CD), a digital versatile disc read only (e.g., DVD-ROM, dual layer DVD-ROM), a Blu-ray disc read only, an ultra-dense disc, a flash memory card (e.g., SD card, min SD card, Micro-SD card, etc.), a magnetic floppy disk, or the like. Computer-readable storage media do not contain carrier waves or transitory electronic signals transmitted by wireless or wired means.

The memory 1010 has stored thereon executable code that, when processed by the processor 1020, may cause the processor 1020 to perform some or all of the methods described above.

Furthermore, the method according to the present application may also be implemented as a computer program or computer program product comprising computer program code instructions for performing some or all of the steps of the above-described method of the present application.

Alternatively, the present application may also be embodied as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) having executable code (or a computer program or computer instruction code) stored thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the various steps of the above-described method according to the present application.

Having described embodiments of the present application, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for production scheduling, comprising:

respectively determining a production constraint condition and a sequencing neighborhood structure according to a preset target of production, wherein the production constraint condition comprises a necessary constraint condition and a custom constraint condition;

2. The method of claim 1, wherein the requisite constraints comprise at least one of:

each order to be sorted is sorted only once;

3. The method of claim 1, wherein the custom constraints comprise at least one of:

the production quantity of the order with the specified product attribute on the specified date is larger than or equal to the preset production quantity.

4. The method of claim 1, wherein the ranking neighborhood structure comprises at least one of:

5. The method according to claim 1, wherein the obtaining the candidate ranking schemes meeting the preset goal through at least two preset algorithms comprises:

6. The method of claim 5, wherein the sequentially solving through at least two solution stages and outputting the candidate ranking scheme meeting the predetermined goal at the last solution stage comprises:

7. The method according to claim 1, wherein the screening of the production ordering scheme of the order to be ordered according to the custom constraint condition from the candidate ordering schemes meeting the necessary constraint condition comprises:

8. The method of claim 1, further comprising:

and presetting the priority of each production constraint condition, wherein the self-defined constraint condition has optional attributes of necessity and non-necessity.

9. A production scheduling apparatus, comprising:

10. The apparatus of claim 9, comprising:

11. An electronic device, comprising:

a processor; and

a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method of any one of claims 1-8.

12. A computer-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to perform the method of any one of claims 1-8.