Fast moving consumer goods delivery plan dynamic decomposition regulation and control method and system based on deep learning

Abstract

The invention discloses a deep learning-based fast moving consumer goods delivery plan dynamic decomposition regulation and control method and system, belongs to the technical field of fast moving consumer goods sales plan decomposition and adjustment in the new retail industry, and aims to solve the technical problem of how to complete the sales task of an enterprise without affecting the consumption of consumers. According to the technical scheme, the method comprises the following steps: data acquisition and data processing: acquiring fast moving consumer goods category information, sales data and target data in a single city, and carrying out preliminary summary calculation; obtaining a target value: summarizing and calculating a quarter target total value according to monthly category target performance; acquiring an accumulated value: summarizing and calculating an annual actual sales accumulated value according to the actual performance after putting; dynamic adjustment of a quarter putting plan: dynamically adjusting a quarter target value which does not start putting; dynamically adjusting and decomposing the category putting plan; outputting a category putting plan result; and cyclically adjusting.

Description

Technical Field

[0001] The present invention relates to the technical field of decomposition and adjustment of fast-moving consumer goods sales plans in the new retail industry, and specifically to a method and system for dynamically decomposing and regulating fast-moving consumer goods launch plans based on deep learning. Background Art

[0002] Fast-moving consumer goods refer to goods that are consumed quickly in consumers' daily lives, such as food, beverages, and household items. For fast-moving consumer goods companies, it is crucial to accurately decompose sales forecasts and make intelligent dynamic adjustments based on sales conditions, which can help companies rationally arrange production and supply chains, reduce inventory and operating costs, and improve market competitiveness. For e-commerce platforms, there are many types of fast-moving consumer goods. The sales launch plans of each category not only determine the production rhythm and supply conditions of upstream manufacturers, but also affect the inventory and capital turnover of e-commerce platforms, and also affect the shopping experience of end consumers on the platform.

[0003] The study believes that the dynamic decomposition of FMCG launch plans is a complex and sophisticated process involving multiple aspects, including market trends, consumer behavior, product characteristics, and marketing channels. For example, the FMCG market is highly competitive and changes rapidly, so the launch plan needs to pay close attention to market dynamics, including the impact of macroeconomic environment, industry policies, changes in consumer demand and other factors on the FMCG market. Understanding the unique selling point (USP) and competitive advantages of the product is the basis for formulating a launch plan. By emphasizing the differentiated characteristics of the product, it can stand out in the fierce market competition. Most of the historical research focuses on the impact of marketing strategy differentiation and the macro market environment on the launch plan, and there is a lack of research on the decomposition and dynamic regulation of the launch plan.

[0004] Dynamic decomposition and control of FMCG launch plans are very important for any FMCG company or e-commerce platform, because the FMCG market is highly competitive and consumer demand is changeable. Dynamic decomposition of launch plans can ensure that companies can quickly respond to market changes and adjust marketing strategies and launch plans in a timely manner. At the same time, dynamic decomposition of launch plans can be used to formulate effective promotional activities and sales strategies based on consumer purchasing behavior and preferences, thereby promoting sales growth. In the long run, it is conducive to long-term market planning and brand building, and through continuous investment and optimized decomposition and control, the sustainable development of enterprises can be achieved.

[0005] For fast-moving consumer goods companies or e-commerce platforms, there are a wide variety of product categories, many upstream suppliers, and a long procurement time. How to complete the company's sales tasks without affecting consumer consumption while ensuring that there is no negative impact on the company's capital turnover is a technical problem that needs to be solved urgently. Summary of the invention

[0006] The technical task of the present invention is to provide a method and system for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning, so as to solve the problem of how to complete the company's sales tasks without affecting consumer consumption, while ensuring that there will be no negative impact on the company's capital turnover.

[0007] The technical task of the present invention is achieved in the following way: a method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning, the method is as follows:

[0008] Data collection and data processing: collect FMCG category information, sales data, and target data in a single city, and perform preliminary summary calculations;

[0009] Get target value: Based on the monthly category target performance, calculate the total quarterly target value;

[0010] Get cumulative value: Based on the actual performance after the launch, calculate the actual annual sales cumulative value;

[0011] Dynamic adjustment of quarterly delivery plans: Dynamic adjustment of quarterly target values ​​that have not yet started delivery;

[0012] Dynamic adjustment and decomposition of product category launch plans: Dynamically adjust and decompose the sales share of each category in the upcoming quarter;

[0013] Category launch plan result output: Output the dynamic decomposition results of the sales target values ​​of each category in the quarter;

[0014] Cyclic adjustment: Dynamically decompose and adjust the delivery plans for subsequent quarters that have not yet started.

[0015] As a preferred embodiment, data collection and data processing are specifically as follows:

[0016] Collect sales product information, sales record data and sales target data of enterprises and cities;

[0017] Collect the data of indicators required for building the model, including month (SUMM_MONTH), primary category ID (PARENT_SORT_ID), primary category name (PARENT_SORT_NAME), secondary category ID (SORT_ID), secondary category name (SORT_NAME), sales (SALE_MNY), sales quantity (SALE_AMT), and sales target (SALE_MNY_TARGET);

[0018] The collected data is summarized and calculated according to the dimensions of month, primary category and secondary category.

[0019] Preferably, obtaining the target value means aggregating the initial monthly sales target into quarterly target values, quarterly target value proportions and annual target total values ​​based on the collected initial monthly sales target, and the formula is as follows:

[0020]

[0021]

[0022] Obtaining the cumulative value means calculating the annual sales cumulative value and the remaining annual target sales value by dividing the collected monthly sales by the annual target value and the quarterly target value. The formula is as follows:

[0023]

[0024] Annual remaining target sales = annual target total value - annual sales cumulative value;

[0025] Dynamic adjustment of the quarterly delivery plan refers to dynamically adjusting the quarterly target value that has not yet been delivered based on the quarterly target proportion and the remaining annual target sales. The formula is as follows:

[0026]

[0027] Preferably, the dynamic adjustment and decomposition of the category launch plan refers to dynamically adjusting and decomposing the sales target value proportion of each category in the upcoming T quarter by dynamically adjusting the quarterly target value based on the collected category sales data, as follows:

[0028] Statistically summarize the actual sales of each category and the total sales of each quarter in each quarter from T-4 to T-1;

[0029] The actual sales percentage of each category in each quarter from T-4 to T-1 is summarized and the formula is as follows:

[0030]

[0031] Weights are assigned to quarters T-4 to T-1 based on their impact and importance on quarter T, as shown in the following table:

[0032] Quarter T-4 T-3 T-2 T-1 Weight 10％ 20％ 30％ 40％

[0033] By referring to the moving average algorithm, the sales target proportion of each category in quarter T is dynamically adjusted. The formula is as follows:

[0034] Dynamically adjusted target sales share of each category in quarter T = actual sales share of the category in quarter T-4 * 10% + actual sales share of the category in quarter T-3 * 20% + actual sales share of the category in quarter T-2 * 30% + actual sales share of the category in quarter T-1 * 40%.

[0035] Preferably, the category launch plan result output refers to the dynamic decomposition of the sales target value of each category in the T quarter by using the dynamically adjusted T quarter target value and the dynamically adjusted T quarter target sales ratio of each category, using the proportion extrapolation method, as follows:

[0036] Calculate the sales target percentage of each category adjusted dynamically in quarter T;

[0037] Calculate the total target value of dynamic adjustment in quarter T;

[0038] The weighted extrapolation method is used to dynamically calculate the sales target for each category in the T quarter, and the launch plan is adjusted based on the extrapolation results. The formula is as follows:

[0039] Target sales of each category in quarter T = dynamically adjusted proportion of target sales of each category in quarter T * dynamically adjusted target value in quarter T.

[0040] A dynamic decomposition and control system for fast-moving consumer goods launch plan based on deep learning, the system comprising:

[0041] The data collection and processing module is used to collect the category information, sales data, target data, etc. of fast-moving consumer goods in a single city, and perform preliminary summary calculations;

[0042] The target value acquisition module is used to calculate the total quarterly target value based on the monthly category target performance;

[0043] The cumulative value acquisition module is used to calculate the actual annual sales cumulative value based on the actual performance after the launch;

[0044] The quarterly delivery plan dynamic adjustment module is used to dynamically adjust the quarterly target values ​​that have not yet started delivery;

[0045] The module for dynamic adjustment and decomposition of category launch plans is used to dynamically adjust and decompose the sales proportion of each category in the upcoming quarter;

[0046] Category launch plan result output module, used to output the dynamic decomposition results of the sales target values ​​of each category in the quarter;

[0047] The cyclic adjustment module is used to dynamically decompose and regulate the delivery plans for subsequent quarters that have not yet started.

[0048] Preferably, the data collection and processing module collects information on primary and secondary categories of fast-moving consumer goods in cities, sales, sales volume and target sales, and summarizes and calculates the collected data according to the dimensions of month, primary category and secondary category;

[0049] The target value acquisition module calculates the quarterly target value, the quarterly target value ratio and the annual target value based on the monthly target value of each category;

[0050] The cumulative value acquisition module calculates the annual sales cumulative value and the annual target sales remaining to be released according to the actual monthly performance of the category after release.

[0051] More preferably, the quarterly delivery plan dynamic adjustment module dynamically adjusts the quarterly target value for which delivery has not yet started by allocating the remaining target sales in proportion according to the quarterly target proportion;

[0052] The dynamic adjustment and decomposition module of the category launch plan dynamically adjusts and decomposes the sales target value proportion of each category in the upcoming T quarter according to the actual sales proportion of each category in the T-4 to T-1 quarters by means of the weighting method and the moving average algorithm; wherein the weighting method assigns weights according to the degree of influence and importance of the T-4 to T-1 quarters on the T quarter, and the weights of the four periods of T-4 to T-1 are 10%, 20%, 30%, and 40% respectively; the moving average algorithm completes the dynamic adjustment of the target sales proportion of each category in the T quarter by multiplying and adding the respective weights of T-4 to T-1 with the actual category sales proportion of each quarter;

[0053] The category launch plan result output module intelligently outputs the dynamic decomposition results of the quarterly sales target values ​​of each category based on the dynamically adjusted sales target proportion of each category in the T quarter and the adjusted total target value of the T quarter according to the proportion deduction algorithm;

[0054] The cyclic adjustment module dynamically decomposes and regulates the process according to the systematic delivery plan, and dynamically decomposes and regulates the delivery plan for each subsequent quarter that has not yet started.

[0055] An electronic device comprising: a memory and at least one processor;

[0056] Wherein, the memory stores a computer program;

[0057] The at least one processor executes the computer program stored in the memory, so that the at least one processor executes the above-mentioned method for dynamic decomposition and control of fast-moving consumer goods release plan based on deep learning.

[0058] A computer-readable storage medium having a computer program stored therein, wherein the computer program can be executed by a processor to implement the above-mentioned method for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning.

[0059] The method and system for dynamic decomposition and control of fast-moving consumer goods delivery plan based on deep learning of the present invention have the following advantages:

[0060] (1) The present invention adopts a systematic dynamic decomposition process of the delivery plan: according to the overall delivery plan and historical actual performance, the quarterly delivery plan adjustment and the decomposition and adjustment of the delivery plan for subdivided categories are completed in sequence; the dynamic decomposition of the delivery plan is systematized and streamlined, and the work efficiency of the delivery work is improved;

[0061] (ii) The present invention adopts an intelligent dynamic decomposition algorithm: according to the actual historical delivery results, the moving average algorithm is used to intelligently adjust the quarterly delivery plan, and the weighted derivation algorithm is used to complete the intelligent decomposition of the delivery plan for each category, thereby realizing the intelligent regulation of the delivery plan and improving the accuracy of the delivery of fast-moving consumer goods;

[0062] (III) The present invention can enable enterprises to intelligently and dynamically decompose and regulate the release plans of fast-moving consumer goods, complete sales target tasks, and dynamically regulate the quarterly or monthly sales release plans of each category of fast-moving consumer goods according to the actual historical sales progress, thereby improving the accurate decomposition and intelligent adjustment of sales plans for various commodities, reducing inventory backlogs, and maximizing revenue. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The present invention is further described below in conjunction with the accompanying drawings.

[0064] Attached Figure 1 This is a schematic diagram of a dynamic decomposition and control method for fast-moving consumer goods launch plans based on deep learning;

[0065] Attached Figure 2 This is a schematic diagram of the adjusted category launch plan;

[0066] Attached Figure 3 This is a schematic diagram of the product launch plan. DETAILED DESCRIPTION

[0067] The method and system for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning of the present invention are described in detail below with reference to the accompanying drawings and specific embodiments of the specification.

[0068] Embodiment 1:

[0069] As attached Figure 1 , 2 As shown in FIG. 3 , this embodiment provides a method for dynamically decomposing and controlling a fast-moving consumer goods delivery plan based on deep learning. The method is specifically as follows:

[0070] S1. Collect the sales product information, sales record data, and sales target data of the enterprise city. Collect the data of the indicators required for building the model, including month (SUMM_MONTH), primary category ID (PARENT_SORT_ID), primary category name (PARENT_SORT_NAME), secondary category ID (SORT_ID), secondary category name (SORT_NAME), sales (SALE_MNY), sales quantity (SALE_AMT), and sales target (SALE_MNY_TARGET); and summarize and calculate the collected data according to the dimensions of month, primary category, and secondary category. All data in the case are explained in detail using a single city as an example.

[0071] S2. Based on the monthly sales target collected in step S1, summarize it into quarterly target value, quarterly target value proportion, and annual target total value. As shown in Table 2, the total annual investment plan of the city is 3,000,000,000 yuan, and the investment plan proportions of each quarter are 24.81%, 25.32%, 22.97%, and 26.90% respectively.

[0072] Total annual target value = summary of monthly sales target values;

[0073] Quarterly target value = summary of sales target values ​​for each month in the current quarter;

[0074]

[0075] Table 2 Summary of quarterly target values

[0076] Quarter 2023.Q1 2023.Q2 2023.Q3 2023.Q4 2023 Quarterly target value 744,163,116 759,688,877 689,179,250 806,968,757 3,000,000,000 Quarterly target value ratio 24.81％ 25.32％ 22.97％ 26.90％ 100％

[0077] S3. Calculate the annual sales cumulative value and the annual remaining target sales based on the monthly sales collected in step S1 and the annual target value and quarterly target value calculated in step 2. As shown in Table 3, as of the second quarter of 2023, the annual actual sales cumulative value is 1,411,241,837, and the annual remaining target sales is 1,588,758,163.

[0078] Annual sales cumulative value = the sum of all monthly sales up to now;

[0079] Annual remaining target sales = annual target total value - annual sales cumulative value;

[0080] Table 3 Actual sales progress in the year

[0081] Quarter 2023.Q1 2023.Q2 2023.Q3 2023.Q4 2023 Cumulative sales value 698,336,091 1,411,241,837 Annual remaining sales target 1,588,758,163

[0082] S4. Based on the quarterly target proportion in step S2 and the remaining annual target sales in step 3, dynamically adjust the target value for the T quarter that has not yet been launched. As shown in Table 4, the target value for the third quarter after dynamic adjustment based on historical actual sales is 731,838,798.14, and the target value for the fourth quarter is 856,919,364.92.

[0083]

[0084] Table 4 Adjustment of quarterly launch plan

[0085]

[0086]

[0087] S5. Based on the category sales data collected in steps S1-S4 and the dynamically adjusted quarterly target values, dynamically adjust and decompose the sales target value proportion of each category in the upcoming T quarter.

[0088] S501. Statistically summarize the actual sales of each category and the total sales of each quarter from T-4 to T-1, as shown in Table 5:

[0089] Table 5 Actual sales of each category in the quarter

[0090]

[0091] S502. Statistics summarize the actual sales ratio of each category in each quarter from T-4 to T-1, as shown in Table 6:

[0092]

[0093] Table 6 Actual sales share of each category in the quarter

[0094]

[0095] S503. Weights are assigned to quarters T-4 to T-1 based on their impact and importance on quarter T. The farther away from quarter T, the smaller the impact. The weights of each quarter are set based on this principle:

[0096] Table 7 Quarterly Weights

[0097] Quarter 2022.Q3 2022.Q4 2023.Q1 2023.Q2 Weight 10％ 20％ 30％ 40％

[0098] S504. Using the moving average algorithm, dynamically adjust the sales target proportion of each category in quarter T. The launch plan of each category after dynamic adjustment is shown in Table 8 and Figure 2As shown, among them, casual snacks accounted for 36.15%, convenient fast food accounted for 18.68%, candy and chocolate accounted for 11.96%, kitchen seasoning accounted for 10.46%, biscuits and pastries accounted for 10.08%, grains, oils and dry goods accounted for 8.62%, and nuts, candied fruits and dried meat accounted for 4.04%.

[0099] Dynamically adjusted target sales share of each category in quarter T = actual sales share of the category in quarter T-4 * 10% + actual sales share of the category in quarter T-3 * 20% + actual sales share of the category in quarter T-2 * 30% + actual sales share of the category in quarter T-1 * 40%.

[0100] Table 8 Adjustment of product category proportion in the third quarter

[0101]

[0102] S6. Based on the dynamically adjusted target value for quarter T in step S4 and the target sales ratio of categories for quarter T in step 5, the dynamic decomposition of the sales target value of each category for quarter T is completed by referring to the weighting method, as follows:

[0103] S601. Calculate the sales target percentage of each category adjusted in quarter T;

[0104] S602. Calculate the total target value adjusted in quarter T;

[0105] S603. Using the weighted extrapolation method, dynamically calculate the sales target of each category in the T quarter, and adjust the launch plan accordingly. The launch plan of each category after dynamic adjustment is shown in Table 9, including 264,556,656 casual snacks, 136,720,953, 87,529,742, 76,515,833 instant food, 73,796,122 biscuits and pastries, 63,118,516 grains and oils, and 29,600,976 nuts, candied fruits, and dried meat.

[0106] Target sales of each category in quarter T = dynamically adjusted proportion of target sales of each category in quarter T * dynamically adjusted target value in quarter T.

[0107] Table 9 Adjustment of product category launch plan in the third quarter

[0108]

[0109] S7. Based on steps S1-S6, on the basis of actual sales in each quarter of each year, dynamically decompose and regulate the launch plan for each subsequent quarter that has not yet started.

[0110] Based on the dynamic decomposition and regulation completed in the above steps S1-S7, a release plan for each category in the next cycle is obtained. This method combines the release target value with the actual market consumption demand and continuously and dynamically adjusts the market release plan. According to this method, the company can be better guided to complete the release of goods and the annual total goal.

[0111] Embodiment 2:

[0112] This embodiment provides a system for dynamically decomposing and controlling a fast-moving consumer goods launch plan based on deep learning, the system comprising:

[0113] The data collection and processing module is used to collect the category information, sales data, target data, etc. of fast-moving consumer goods in a single city, and perform preliminary summary calculations;

[0114] The target value acquisition module is used to calculate the total quarterly target value based on the monthly category target performance;

[0115] The cumulative value acquisition module is used to calculate the actual annual sales cumulative value based on the actual performance after the launch;

[0116] The quarterly delivery plan dynamic adjustment module is used to dynamically adjust the quarterly target values ​​that have not yet started delivery;

[0117] The module for dynamic adjustment and decomposition of category launch plans is used to dynamically adjust and decompose the sales proportion of each category in the upcoming quarter;

[0118] Category launch plan result output module, used to output the dynamic decomposition results of the sales target values ​​of each category in the quarter;

[0119] The cyclic adjustment module is used to dynamically decompose and regulate the delivery plans for subsequent quarters that have not yet started.

[0120] The data collection and processing module in this embodiment collects information on first-level and second-level categories of fast-moving consumer goods in cities, sales, sales volume and target sales, and summarizes and calculates the collected data according to the dimensions of month, first-level category and second-level category.

[0121] The target value acquisition module in this embodiment calculates the quarterly target total value, the quarterly target value ratio and the annual target total value based on the monthly target value of each category.

[0122] The cumulative value acquisition module in this embodiment calculates the annual sales cumulative value and the annual target sales remaining to be released based on the actual monthly performance of the category after release.

[0123] The quarterly delivery plan dynamic adjustment module in this embodiment dynamically adjusts the quarterly target value for which delivery has not yet started by allocating the remaining target sales in proportion according to the quarterly target proportion.

[0124] The category launch plan dynamic adjustment and decomposition module in this embodiment dynamically adjusts and decomposes the sales target value proportion of each category in the upcoming T quarter according to the actual sales proportion of each category in the T-4 to T-1 quarters through the weighting method and the moving average algorithm; wherein, the weighting method assigns weights according to the degree of influence and importance of the T-4 to T-1 quarters on the T quarter, and the weights of the four periods of T-4 to T-1 are 10%, 20%, 30%, and 40% respectively; the moving average algorithm completes the dynamic adjustment of the target sales proportion of each category in the T quarter by multiplying and adding the respective weights of T-4 to T-1 with the actual category sales proportion of each quarter.

[0125] The category launch plan result output module in this embodiment intelligently outputs the dynamic decomposition results of the quarterly sales target values ​​of each category according to the dynamically adjusted target sales proportion of each category in the T quarter and the adjusted total target value of the T quarter based on the proportion inference algorithm.

[0126] The cyclic adjustment module in this embodiment performs dynamic decomposition and regulation of the delivery plan for each subsequent quarter that has not yet started according to the systematic delivery plan dynamic decomposition and regulation process.

[0127] Embodiment 3:

[0128] This embodiment also provides an electronic device, including: a memory and a processor;

[0129] Wherein, the memory stores computer-executable instructions;

[0130] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the method for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning in any embodiment of the present invention.

[0131] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or any conventional processor, etc.

[0132] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory can also include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash memory card, at least one disk storage period, a flash memory device, or other volatile solid-state storage devices.

[0133] Embodiment 4:

[0134] This embodiment also provides a computer-readable storage medium, which stores a plurality of instructions, which are loaded by a processor, so that the processor executes the method for dynamic decomposition and control of fast-moving consumer goods delivery plans based on deep learning in any embodiment of the present invention. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code for implementing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0135] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0136] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0137] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0138] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning, characterized in that: The method is as follows: Data collection and data processing: collect FMCG category information, sales data, and target data in a single city, and perform preliminary summary calculations; Get target value: Based on the monthly category target performance, calculate the total quarterly target value; Get cumulative value: Based on the actual performance after the launch, calculate the actual annual sales cumulative value; Dynamic adjustment of quarterly delivery plan: Dynamic adjustment of quarterly target values ​​that have not yet started delivery; Dynamic adjustment and decomposition of product category launch plans: Dynamically adjust and decompose the sales share of each category in the upcoming quarter; Category launch plan result output: Output the dynamic decomposition results of the sales target values ​​of each category in the quarter; Cyclic adjustment: Dynamically decompose and adjust the delivery plans for subsequent quarters that have not yet started.

2. The method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning according to claim 1, characterized in that: The details of data collection and data processing are as follows: Collect sales product information, sales record data and sales target data of enterprises and cities; Collect data on indicators needed to build the model, including month, first-level category ID, first-level category name, second-level category ID, second-level category name, sales volume, sales quantity, and sales target; The collected data is summarized and calculated according to the dimensions of month, primary category and secondary category.

3. The method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning according to claim 1 or 2, characterized in that: Obtaining the target value means summarizing the initial monthly sales target into quarterly target values, quarterly target value proportions, and annual target totals based on the collected initial monthly sales target. The formula is as follows: Obtaining the cumulative value means calculating the annual sales cumulative value and the remaining annual target sales value by dividing the collected monthly sales by the annual target value and the quarterly target value. The formula is as follows: Among them, j represents the current month; Annual remaining target sales = annual target total value - annual sales cumulative value; Dynamic adjustment of the quarterly delivery plan refers to dynamically adjusting the quarterly target value that has not yet been delivered based on the quarterly target proportion and the remaining annual target sales. The formula is as follows:

4. The method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning according to claim 3 is characterized in that: Dynamic adjustment and decomposition of category launch plans refers to dynamically adjusting and decomposing the sales target value proportion of each category in the upcoming T quarter through the quarterly target value dynamically adjusted by the collected category sales data domain. The details are as follows: Statistically summarize the actual sales of each category and the total sales of each quarter in each quarter from T-4 to T-1; The actual sales percentage of each category in each quarter from T-4 to T-1 is summarized and the formula is as follows: Weights are assigned to quarters T-4 to T-1 based on their impact and importance on quarter T, as shown in the following table: By referring to the moving average algorithm, the sales target proportion of each category in quarter T is dynamically adjusted. The formula is as follows: Dynamically adjusted target sales ratio of each category in quarter T = actual sales ratio of the category in quarter T-4 * 10% + actual sales ratio of the category in quarter T-3 * 20% + T- The actual sales proportion of this category in the 2nd quarter*30% + the actual sales proportion of this category in the T-1st quarter*40%.

5. The method for dynamic decomposition and control of fast-moving consumer goods launch plan based on deep learning according to claim 4, characterized in that: The category launch plan result output refers to the dynamic decomposition of the sales target value of each category in the T quarter by using the dynamically adjusted T quarter target value and the dynamically adjusted T quarter target sales ratio of each category, using the proportion extrapolation method, as follows: Calculate the sales target percentage of each category adjusted dynamically in quarter T; Calculate the total target value of dynamic adjustment in quarter T; The weighted extrapolation method is used to dynamically calculate the sales target for each category in the T quarter, and the launch plan is adjusted based on the extrapolation results. The formula is as follows: Target sales of each category in quarter T = dynamically adjusted proportion of target sales of each category in quarter T * dynamically adjusted target value in quarter T.

6. A dynamic decomposition and control system for fast-moving consumer goods launch plan based on deep learning, characterized in that: The system includes: The data collection and processing module is used to collect the category information, sales data, target data, etc. of fast-moving consumer goods in a single city, and perform preliminary summary calculations; The target value acquisition module is used to calculate the total quarterly target value based on the monthly category target performance; The cumulative value acquisition module is used to calculate the actual annual sales cumulative value based on the actual performance after the launch; The quarterly delivery plan dynamic adjustment module is used to dynamically adjust the quarterly target values ​​that have not yet started delivery; The module for dynamic adjustment and decomposition of category launch plans is used to dynamically adjust and decompose the sales proportion of each category in the upcoming quarter; Category launch plan result output module, used to output the dynamic decomposition results of the sales target values ​​of each category in the quarter; The cyclic adjustment module is used to dynamically decompose and regulate the delivery plans for subsequent quarters that have not yet started.

7. The system for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning according to claim 6, characterized in that: The data collection and processing module collects information on first-level and second-level categories of fast-moving consumer goods in cities, sales, sales volume and target sales, and summarizes and calculates the collected data according to the dimensions of month, first-level category and second-level category; The target value acquisition module calculates the quarterly target value, the quarterly target value ratio and the annual target value based on the monthly target value of each category; The cumulative value acquisition module calculates the annual sales cumulative value and the annual target sales remaining to be released based on the actual monthly performance of the category after release.

8. The system for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning according to claim 6 or 7, characterized in that: The quarterly delivery plan dynamic adjustment module dynamically adjusts the quarterly target value that has not yet been delivered by allocating the remaining target sales in proportion according to the quarterly target proportion; The dynamic adjustment and decomposition module of the category launch plan dynamically adjusts and decomposes the sales target value proportion of each category in the upcoming T quarter according to the actual sales proportion of each category in the T-4 to T-1 quarters by means of the weighting method and the moving average algorithm; wherein the weighting method assigns weights according to the degree of influence and importance of the T-4 to T-1 quarters on the T quarter, and the weights of the four periods of T-4 to T-1 are 10%, 20%, 30%, and 40% respectively; the moving average algorithm completes the dynamic adjustment of the target sales proportion of each category in the T quarter by multiplying and adding the respective weights of T-4 to T-1 with the actual category sales proportion of each quarter; The category launch plan result output module intelligently outputs the dynamic decomposition results of the quarterly sales target values ​​of each category based on the dynamically adjusted sales target proportion of each category in the T quarter and the adjusted total target value of the T quarter according to the proportion deduction algorithm; The cyclic adjustment module dynamically decomposes and regulates the process according to the systematic delivery plan, and dynamically decomposes and regulates the delivery plan for each subsequent quarter that has not yet started.

9. An electronic device, characterized in that: include: memory and at least one processor; Wherein, the memory stores a computer program; The at least one processor executes the computer program stored in the memory, so that the at least one processor executes the method for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which can be executed by a processor to implement the method for dynamic decomposition and control of fast-moving consumer goods launch plans based on deep learning as described in any one of claims 1 to 5.

Images