A multi-source supply smoothing scheduling decision method, system, device and storage medium based on inter-period allocation operators

Abstract

The application provides a multi-source supply smoothing scheduling decision method, system, device and storage medium based on cross-period allocation operators, and relates to the technical field of supply chain planning and operation optimization. The method comprises the following steps: acquiring basic business data of a multi-source supply network; using a preset demand distribution weight vector in a week to discretely decompose the total demand of a sales area in each logical week into a daily demand sequence; performing time sequence backtracking on each demand day in the daily demand sequence based on a transportation cycle to determine the corresponding production factory delivery day and the production logical week to which the demand day belongs; generating a cross-period allocation operator according to the attribution relationship; constructing a linear programming model containing the operator and solving the model to generate an optimal supply plan. The application solves the problem of capacity planning distortion caused by non-integer week transportation cycles.

Description

Technical Field

[0001] This application relates to the field of supply chain planning and operations optimization technology, and in particular to a multi-source supply smoothing scheduling decision-making method, system, device and storage medium based on inter-period allocation operators. Background Technology

[0002] In modern large-scale supply chain management, production scheduling decisions face a fundamental contradiction between decision granularity and computational scale. Existing mainstream scheduling models typically use "week" as the basic time granularity. The underlying logic of this approach is to artificially aggregate the demand for the entire week into a fixed point in time (e.g., Monday or Friday) and make a coarse weekly offset based on the average transportation cycle. While this approach greatly reduces the scale of decision variables and makes the model mathematically easier to solve, it obscures the true temporal continuity in the physical world.

[0003] However, this coarse-grained discretization of time leads to severe boundary effects, meaning that in actual business operations, transportation cycles often cross weekly boundaries. For example, demand generated on Monday, with a transportation cycle of 3 days, must be produced last week; while demand generated on Friday of the same week, with the same transportation cycle, can be produced this week. Existing weekly models cannot perceive these daily-level differences, often forcibly categorizing all demand within a week into either production this week or production last week. This one-size-fits-all categorization leads to drastic fluctuations in production plans. The model may incorrectly instruct factories to concentrate production in a particular week, leaving them idle the following week, causing instantaneous capacity overflow or waste, severely deviating from the actual need for balanced production. While refining the model granularity to the "day" can solve the accuracy problem, this leads to an exponential explosion in the scale of decision variables, making it infeasible to obtain the optimal solution across the entire network within a finite time. Summary of the Invention

[0004] Firstly, this application provides a multi-source supply smoothing scheduling decision-making method based on inter-period allocation operators. This method aims to solve the technical problems in the prior art caused by the use of weekly granular decision-making, such as coarse inter-week transportation processing, distorted production rhythm, and inaccurate capacity utilization assessment.

[0005] The method is implemented through the following steps: First, acquire basic configuration data and business demand data of the supply chain network. The basic configuration data includes several sets of production plants, several sets of sales regions, a production calendar, and transportation cycle data from each production plant to each sales region. The business demand data includes the total weekly demand of each sales region in each demand logical week within a preset decision period. Second, based on a preset intra-week demand distribution weight vector, discretize the total weekly demand of each sales region in each demand logical week to generate daily-level demand components. Each daily-level demand component corresponds to a specific date within the demand logical week. Then, perform time-series offset calculation on the specific dates corresponding to the daily-level demand components based on the transportation cycle to determine the production and delivery dates required to produce that daily-level demand component. Map the production and delivery dates to the corresponding production logical week and generate an inter-period allocation operator based on the production logical week and the daily-level demand component. Finally, solve a multi-source supply scheduling linear programming model with the objective of minimizing the global comprehensive cost to obtain the optimal supply allocation plan.

[0006] Secondly, this application provides a multi-source supply smoothing scheduling decision-making system based on inter-period allocation operators. The system includes a data acquisition module, a demand decomposition module, an operator generation module, and a decision-solving module. The system is configured to execute the steps described in the first aspect above, achieving fully automated decision-making from data input to optimal scheduling plan output through the collaborative work of each module.

[0007] Thirdly, this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect above.

[0008] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect above.

[0009] The beneficial effects of this application are as follows:

[0010] 1. By introducing an inter-period amortization operator, this application creatively embeds day-level physical time-series logic into a week-level linear programming model. This enables the model to maintain the computational advantages of week-level decision-making while being able to perceive and handle inter-period transportation offsets like a day-level model, eliminating planning errors caused by time discretization, and achieving a "dimensionality reduction attack" in decision granularity while maintaining "dimensionality increase" in accuracy.

[0011] 2. This application can intelligently "spread" demand concentrated in a certain week to multiple adjacent production weeks based on the actual distribution of daily demand and accurate transportation cycle backtracking. This mechanism effectively eliminates the false capacity peaks and troughs common in traditional models, supports factories to achieve smoother and more economical balanced production, and realizes automatic balancing of production load.

[0012] 3. By accurately calculating the different "time windows" caused by the differences in transportation cycles of different factories, the model of this application can automatically utilize the early idle capacity of remote factories to meet the urgent needs that cannot be covered by near factories, thereby reducing the system's stockout risk and delivery default rate without increasing the total capacity investment, and improving the delivery reliability of complex supply chain networks. Attached Figure Description

[0013] Figure 1 A flowchart of a multi-source supply smoothing scheduling decision method based on inter-period allocation operators provided in an embodiment of this application;

[0014] Figure 2 This is a structural block diagram of a multi-source supply smoothing scheduling decision system based on inter-period allocation operators provided in an embodiment of this application. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0016] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0017] This embodiment provides a multi-source supply smoothing scheduling decision-making method based on inter-period allocation operators. Specifically, it pre-compresses and encodes the fine-grained daily-level transportation offset logic in the physical world into a set of static weekly-level proportional coefficients. This allows for accurate simulation and handling of complex supply and demand relationships spanning weekly boundaries even when using traditional weekly-granularity linear programming models. This method solves the technical problem in existing technologies where sacrificing time accuracy for computational speed leads to a disconnect between production plans and actual logistics, and false fluctuations in capacity planning. It achieves the beneficial effect of significantly improving the executability of scheduling decisions and resource utilization without increasing the computational complexity of the model.

[0018] Figure 1 A flowchart of the method provided in this embodiment is shown. Figure 1 As shown, the method includes the following steps:

[0019] S100, acquire basic configuration data and business demand data of the supply chain network. The basic configuration data includes multiple sets of production plants, multiple sets of sales regions, production calendar, and transportation cycle data from each production plant to each sales region. The business demand data includes the total weekly demand of each sales region in each demand logical week within a preset decision cycle.

[0020] In this embodiment, the data acquisition module first establishes a data interface with the Enterprise Resource Planning (ERP) system or the Supply Chain Management System to read and parse structured business data.

[0021] Specifically, a collection of production plants Includes all available supply source nodes; for each factory in the set The system obtains its attribute information, including (but not limited to) the factory's geographic location code, production line type, and cost structure parameters. For example, the factory... Unit production cost (Unit: Yuan / piece).

[0022] Sales Area Collection This includes all demand convergence points, for each sales region within the set. The system obtains its unique identifier.

[0023] Business requirement data includes a set of time series data. ,in This index represents the logical week of demand; the data specifies the demand within each particular logical week. Sales area The total number of products required. It should be noted that the "total weekly demand" here is the input standard of the traditional weekly model, which masks the demand fluctuations of each specific day within the week.

[0024] Transportation cycle data Including from the factory to the sales area The required physical time is measured in days.

[0025] For example, suppose the supply chain network includes a factory. and a sales area The basic configuration data read by the system may be as follows: Factory The unit production cost is 10 yuan / piece; from arrive Transportation cycle Day. Business demand data indicates that in week 49 ( Sales area Weekly total demand These data are loaded into memory and constructed into corresponding data objects for subsequent steps to call.

[0026] S200, based on a preset weekly demand distribution weight vector, discretizes the total weekly demand of each sales region in each demand logical week, generating a daily-level demand component corresponding to each specific date within that demand logical week.

[0027] In this embodiment, the coarse-grained weekly demand is deconstructed to restore its microscopic distribution in the time dimension.

[0028] Among them, the weekly demand distribution weight vector It is a vector of length equal to the number of days in a week (usually 7), which describes the proportion of demand for each day from Monday to Sunday to the total demand for the whole week. The vector satisfies the normalization condition, that is, the sum of all elements in the vector is equal to 1. The setting of this vector can be based on statistical analysis of historical sales data, or on specific business scenario assumptions (such as uniform distribution or weekend peak distribution).

[0029] The system iterates through each requirement logic cycle. For each day of that week ( For a specific calendar date, the system calculates the day-level demand component for that day. :

[0030]

[0031] in, Indicates date Indexes within the week (e.g., index 1 for Monday and index 7 for Sunday).

[0032] For example, for sales regions Demand in week 49 Item.

[0033] Scene A (Uniform Distribution): Setting The system calculated the daily demand for each day from Monday to Sunday in week 49 as follows: Item.

[0034] Scene B (Non-uniform distribution): Setting The system calculates the daily demand as follows: Monday to Thursday, and Saturday and Sunday. Items; while the daily demand on Friday was Item.

[0035] This parametric decomposition method allows for the flexible simulation of various real-world market demand patterns.

[0036] S300, for each combination of sales region and each production plant, performs time-series offset calculation on the specific date corresponding to the daily-level demand component based on the transportation cycle, determines the production and delivery date required to produce the daily-level demand component; maps the production and delivery date to the corresponding production logical week, and calculates the sum of the weights of the daily-level demand components from the specific demand logical week carried by each production logical week, generating an inter-period allocation operator. The inter-period allocation operator is used to indicate the proportion coefficient for allocating the demand of a specific sales region in a specific demand logical week to the production of a specific production plant in a specific production logical week.

[0037] In this embodiment, the connection requirement week is generated through the logic of "daily backtracking and weekly aggregation". With production week Inter-period apportionment operator .

[0038] Specifically, it includes three tightly coupled sub-processes: time-series offset, week-based attribution determination, and weight aggregation.

[0039] First, perform timing offset calculation.

[0040] For a given factory and sales area and the demand logic cycle One day The system calculates the theoretical delivery date corresponding to the demand for that day. This means that in order to meet the date The latest date by which the factory must complete production and ship the goods, according to the demand:

[0041]

[0042] in, This refers to the number of days in the transportation cycle.

[0043] Secondly, the week-end attribution determination is performed.

[0044] The system utilizes calendar mapping functions Identify and calculate the shipping date In which production logic cycle does it actually fall? In, that is Specifically, this belongs to the demand cycle. The demand for production tasks may fall to [the target]. This week, it may also fall into Zhou, or even earlier week.

[0045] Finally, the weight aggregation operator is generated.

[0046] For a fixed pair and fixed demand cycle The system counts all items that fall into the production cycle. Date The corresponding weight And sum them up to get the operator. :

[0047]

[0048] The operator The physical meaning is: sales area In the Of the total weekly demand, a proportion is The part that needs to be in the first Zhouyou Factory Production completed.

[0049] For example, suppose a factory to the sales area Transportation cycle Day, demand week Zhou, using uniformly distributed weights .

[0050] Therefore, the dates included in week 49 and their backtracking calculations are as follows:

[0051] 1. Monday (Day 1): Shipping date = Day 1 - 3 days = last Friday; last Friday belongs to week 48, weight contribution = 1 / 7.

[0052] 2. Tuesday (Day 2): Shipping date = Day 2 - 3 days = last Saturday; last Saturday belongs to week 48, weight contribution = 1 / 7.

[0053] 3. Wednesday (Day 3): Shipping date = Day 3 - 3 days = last Sunday; last Sunday belongs to week 48, weight contribution = 1 / 7.

[0054] 4. Thursday (Day 4): Shipping date = Day 4 - 3 days = Monday of this week; Monday of this week belongs to week 49, weight contribution = 1 / 7.

[0055] 5. Friday (Day 5): Shipping date = Day 5 - 3 days = Tuesday of this week; Tuesday of this week belongs to week 49, weight contribution = 1 / 7.

[0056] 6. Saturday (Day 6): Shipping date = Day 6 - 3 days = Wednesday of this week; Wednesday of this week belongs to week 49, weight contribution = 1 / 7.

[0057] 7. Sunday (Day 7): Shipping date = Day 7 - 3 days = this Thursday; this Thursday belongs to week 49, weight contribution = 1 / 7.

[0058] Aggregation results:

[0059] For production week : .

[0060] For production week : .

[0061] For other production weeks: .

[0062] This set of operators The meaning is: In order to meet the demand of week 49, the factory Approximately 42.8% of the production task needs to be completed in week 48, and the remaining 57.2% needs to be completed in week 49; this is fundamentally different from the traditional weekly model that assigns 100% of the task to a single week.

[0063] S400 constructs a multi-source supply scheduling linear programming model. The linear programming model uses the supply volume of each production plant to each sales region in each demand logic week as the decision variable, and includes capacity constraints. The capacity constraints are based on the inter-period allocation operator, which limits the total capacity load of any production plant in any production logic week to no less than the sum of the converted demand of all demand logic weeks undertaken by the production plant in that production logic week.

[0064] In this embodiment, the generated operators are used to construct an optimized model that operates on the weekly dimension but has the logical precision of the daily dimension.

[0065] The decision variables include:

[0066] : indicates factory Determine to supply to the sales region Used to satisfy its first Weekly demand supply (unit: pieces).

[0067] : Indicates the sales area In the Weekly stockout (unit: pieces).

[0068] Construct the objective function:

[0069] The goal of this model is to minimize the global comprehensive cost. :

[0070]

[0071] in:

[0072] It is a factory The unit production cost;

[0073] From the factory To the area The unit transportation cost;

[0074] This is the unit's penalty cost for stockouts.

[0075] In addition, core constraints need to be constructed:

[0076] 1. Capacity constraints:

[0077] Traditional capacity constraints are usually However, in this model, the production cycle... The production capacity not only serves the demand cycle The needs of, and may also serve Or demand in the more distant future (if transportation cycles are long). Therefore, the constraint transforms into:

[0078]

[0079] The left side of the inequality represents the factory. In the The actual production load of a week is determined by iterating through all future demand weeks. Each decision variable Multiply by the corresponding amortization operator This allows future demand to be factored back into the current production week and summed up.

[0080] For example, for a factory The model will automatically take into account the capacity constraints in week 48. This item; this means that, although This was to meet the demand of week 49, but due to The existence of this requirement forces factories to reserve 3 / 7 of their capacity in week 48 to produce this part of the goods.

[0081] 2. Demand balance constraint:

[0082]

[0083] This constraint ensures that for each sales region and each demand week, the total supply plus the stockout at each factory is exactly equal to the total demand.

[0084] S500 solves a multi-source supply scheduling linear programming model with the objective of minimizing the overall global cost, and outputs the optimal supply allocation plan.

[0085] In this embodiment, the system will construct a pre-defined object containing the objective function, constraints, and input parameters. The mathematical model of the matrix is ​​submitted to commercial mathematical programming solvers such as COPT, Gurobi, and Cplex; the solvers then use the simplex method or interior point method to search within the feasible region for a solution that... smallest matrix.

[0086] The output not only tells the sales staff how much goods each factory should send to each region ( More importantly, combining Operators allow business personnel to derive actual production plans. For example, in a factory... The actual production schedule recommendation for week 48 is And so on.

[0087] Example 2

[0088] Based on the same inventive concept, this embodiment also provides a multi-source supply smoothing scheduling decision system based on inter-period allocation operators; such as Figure 2 As shown, the system includes:

[0089] 1. Data Acquisition Module: Used to establish data channels and extract factory capacity tables, logistics route tables containing LT, and weekly sales forecast tables from the database.

[0090] 2. Demand decomposition module: It has built-in multiple weekly distribution curves (such as uniform distribution and normal distribution approximation) to break down the acquired weekly demand into daily demand sequences.

[0091] 3. Operator generation module: It maintains a dynamic date-week mapping table and calculates a specific inter-period allocation operator matrix for each logistics route ODPair; this module can handle complex cases of non-integer weeks, such as LT=10 days (spanning two weeks).

[0092] 4. Model Building Module: Responsible for translating business logic into mathematical language; it automatically generates large-scale sparse matrices, including decision variables. and The index range, and then the output of the operator generation module. The values ​​are filled into the corresponding coefficient positions in the capacity constraint matrix.

[0093] 5. Decision Solving Module: This module integrates a solver interface; it is responsible not only for calling the solver but also for analyzing the solution results and converting the mathematical solution... Restore the data to a business-readable weekly shipping plan report and production load report.

[0094] Example 3

[0095] Building upon Example 1, this example further incorporates more complex real-world production constraints, specifically including:

[0096] Non-continuous production calendar: This calendar corrects for time series offset calculations, i.e., when the latest calculated shipping date is... Located in the factory Non-working days (according to) When making a judgment, We must go back to the most recent working day.

[0097]

[0098] This adjustment will directly affect The result generates an inter-period allocation operator 'a' that reflects the actual production rhythm. For example, a task that was originally scheduled for production on Sunday will now be forced to be moved to Friday, and its capacity occupancy will be shifted from the following week to this week.

[0099] Minimum production batch constraint:

[0100] In the construction of the S500 model, new decision variables and constraints are added, specifically including:

[0101] Binary variables: If the factory During the production week If production has been carried out, the value is 1; otherwise, it is 0.

[0102] Total output variable: , representing the total planned output of factory i in production week w:

[0103]

[0104] Add constraints:

[0105] a)

[0106] b)

[0107] These two constraints ensure that as long as production occurs ( = 1), the output must be greater than or equal to the minimum batch size. Furthermore, the capacity cannot be exceeded, which allows the model to make more realistic start-up and shutdown decisions.

[0108] Example 4

[0109] This embodiment extends the original deterministic model to scenarios capable of handling uncertainty, thereby improving the robustness of the plan. Specifically, it includes:

[0110] Security library: Business requirement data includes not only predicted mean It also includes the standard deviation of the forecast, and calculates the target safety stock based on this. .

[0111] Modify demand balance constraints: (Allowing oversupply to build up inventory).

[0112] Introducing inventory balance constraints:

[0113] In planning models, predicted demand can be used instead of actual demand.

[0114] Modify the objective function: Add an inventory holding cost term. .

[0115] Application of inter-period allocation operators: The establishment of safety stock also requires the use of production capacity, so a safety stock allocation operator can be created. The logic is similar to that of the demand allocation operator. It is used to calculate how much production is needed in week w to ensure that the safety stock level is met at the beginning of week W.

[0116] Managing uncertainty in transportation lead times:

[0117] Transportation cycle It is no longer a single value, but a distribution, for example... .

[0118] When calculating the inter-period allocation operator, a more conservative lead time is used, for example... Here, k is the service level coefficient (e.g., k=1.65 for a 95% confidence level), meaning that production must be scheduled earlier to address potential shipping delays. By adjusting the value of k, companies can make trade-offs between cost and delivery reliability.

[0119] Example 5

[0120] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it can implement the above-described S100 to S500 steps and all their sub-steps. This medium records the core algorithm code for generating the inter-period amortization operator, ensuring the portability and reproducibility of the method on different computing devices.

[0121] Example 6

[0122] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0123] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0124] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware.

[0125] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A multi-source supply smoothing scheduling decision-making method based on inter-period allocation operators, characterized in that, The method includes: Acquire basic configuration data and business demand data of the supply chain network. The basic configuration data includes a set of several production plants, a set of several sales regions, a production calendar, and transportation cycle data from each production plant to each sales region. The business demand data includes the total weekly demand of each sales region in each demand logical week within a preset decision cycle. Based on the preset weekly demand distribution weight vector, the total weekly demand of each sales region in each demand logical week is discretized to generate daily-level demand components, which correspond to each specific date within the demand logical week. Based on the transportation cycle, a time-series offset calculation is performed on the specific date corresponding to the daily-level demand component to determine the production and delivery date required to produce the daily-level demand component; The production and delivery dates are mapped to the corresponding production logical weeks, and an inter-period allocation operator is generated based on the production logical weeks and the daily-level demand components. The goal is to solve a multi-source supply scheduling linear programming model to minimize the overall global cost, so as to obtain the optimal supply allocation plan.

2. The method according to claim 1, characterized in that, The discretization of the total weekly demand for each sales region in each logical demand week, based on a preset weekly demand distribution weight vector, includes: Determine the number of days contained in the demand logic week, and obtain a demand distribution weight vector within the week with a length equal to the number of days, wherein the sum of the elements in the vector is equal to one; Based on the total weekly demand and the weight element of the first date in the weekly demand distribution weight vector, the daily-level demand component of the first date is determined.

3. The method according to claim 1, characterized in that, The calculation of the time-series offset for the specific dates corresponding to the daily-level demand components based on the transportation cycle includes: Get the specific date and shipping time values; Based on the specific date corresponding to the daily-level demand component and the transportation cycle value, the production and delivery date is determined, wherein the production and delivery date is the forward offset date of the specific date on the time axis.

4. The method according to claim 1, characterized in that, The method of generating inter-period allocation operators based on the production logic week and the daily-level demand components includes: Iterate through all specific dates within a given logical week for a specific requirement; Determine whether the production and delivery date obtained after calculating the time offset for each specific date falls within the time range of a specific production logical week; If the judgment result is yes, then the intra-week demand distribution weight corresponding to the specific date is accumulated into the inter-period allocation operator corresponding to the combination of the specific production logic week and the specific demand logic week; If the judgment result is negative, then the contribution of that specific date to the inter-period allocation operator for that particular production logic week is zero.

5. The method according to claim 1, characterized in that, The method further includes: For any given production plant and any given production logic week, calculate the sum of the products of the supply decision variables for all sales regions in all future demand logic weeks and the current demand logic week with the corresponding inter-period allocation operators, and make the sum of the products less than or equal to the maximum available capacity limit of the given production plant in the given production logic week.

6. The method according to claim 1, characterized in that, The method further includes: applying demand balance constraints when solving the multi-source supply scheduling linear programming model, the demand balance constraints including: For any given sales region and any given demand logic week, the sum of the supply decision variables provided by all production plants to that sales region during that demand logic week, plus the stockout variable for that sales region during that demand logic week, is equal to the total weekly demand for that sales region during that demand logic week.

7. A multi-source supply smoothing scheduling decision system based on inter-period allocation operators, characterized in that, The system includes: The data acquisition module is used to acquire basic configuration data and business demand data of the supply chain network. The basic configuration data includes a set of several production plants, a set of several sales regions, a production calendar, and transportation cycle data from each production plant to each sales region. The business demand data includes the total weekly demand of each sales region in each demand logical week within a preset decision cycle. The demand decomposition module is used to discretize the total weekly demand of each sales region in each demand logical week based on a preset weekly demand distribution weight vector to generate daily-level demand components, which correspond to each specific date in the demand logical week. The operator generation module is used to perform time-series offset calculation on the specific date corresponding to the daily-level demand component based on the transportation cycle, determine the production and delivery date required to produce the daily-level demand component, map the production and delivery date to the corresponding production logical week, and generate an inter-period allocation operator based on the production logical week and the daily-level demand component. The decision-solving module is used to solve the multi-source supply scheduling linear programming model with the goal of minimizing the overall global cost, so as to obtain the optimal supply allocation plan.

8. The system according to claim 7, characterized in that, The discretization of the total weekly demand for each sales region in each logical demand week, based on a preset weekly demand distribution weight vector, includes: Determine the number of days contained in the demand logic week, and obtain a demand distribution weight vector within the week with a length equal to the number of days, wherein the sum of the elements in the vector is equal to one; Based on the total weekly demand and the weight element of the first date in the weekly demand distribution weight vector, the daily-level demand component of the first date is determined.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images