AI optimization-based enterprise production and marketing management regulation and control method and system

Abstract

The invention relates to the field of production and marketing regulation and control, and particularly discloses an enterprise production and marketing management regulation and control method and system based on AI optimization. Comprising the steps of obtaining enterprise production and marketing data, performing analysis by utilizing historical production and marketing data and economic operation data, determining a current completion value and a future plan amount, constructing a regulation and control algorithm model based on the current completion value and the future plan amount to perform prediction calculation to obtain a whole-year completion value, and performing prediction calculation on the whole-year completion value based on a regulation and control management standard and a whole-year target value. And regulating and calculating by utilizing an optimal evaluation algorithm to obtain a predicted regulation and control result, and regulating and controlling the predicted regulation and control result by utilizing an optimal search algorithm based on the annual target value and the annual finished value according to regulation and control requirements and regulation and control targets to obtain an optimal regulation and control evaluation result. The method aims at coordinating and balancing enterprise taxation, production and sales relations and guiding quarterly and monthly production and operation activities to be developed around enterprise strategies and annual targets.

Description

Technical Field

[0001] The present invention relates to the field of production and marketing regulation, and in particular to a method and system for enterprise production and marketing management regulation based on AI optimization. Background Art

[0002] In the tobacco industry, in order to ensure the balanced development of corporate economic operations and match annual goals with corporate development strategies, every month, based on the requirements of the National Bureau, local government requirements, changes in market demand and current economic conditions, the company strives to achieve the best benefits for its annual goals by adjusting structure, inventory, output, taxation and cooperative production.

[0003] Adjusting the structure means increasing or decreasing the sales volume of different price categories and specifications while keeping the total sales volume unchanged. Adjusting inventory means satisfying the market sales supply by selling more inventory while keeping the total output unchanged, so as to affect the total tax and profit of the enterprise while keeping the total output value unchanged. Adjusting output means affecting the total output value and total tax and profit of the enterprise by increasing or decreasing the total production volume. Adjusting taxes means adjusting the economic operation status by delaying the tax payment time of some specifications, so as to adjust the tax and profit situation of the enterprise. Adjusting cooperative production volume means increasing or decreasing cooperative production volume to adjust the economic operation status and affect the total output value and tax and profit of the enterprise.

[0004] Therefore, according to management requirements, enterprises need to select one or more control measures every month to calculate and forecast the balance between corporate tax, profit, sales and production, calculate total tax and profit and total output value to evaluate the expected effect, find the best solution, and form a final resolution to issue control instructions to the production and sales links, adjust the production and sales plan, and ensure that material support, cigarette production, warehousing process and market sales can be coordinated in place within the specified time and region through planning, and ensure that the production and sales quantity of the enterprise meets the annual production and operation goals. It involves the comprehensive planning department, finance department, production department and marketing center related departments.

[0005] Currently, relevant data is mainly collected manually, imported into Excel by the finance department, and calculated using Excel tools. The calculation process is complicated and time-consuming. Summary of the invention

[0006] In response to the above problems, one object of the present invention is to provide an enterprise production and sales management and control method based on AI optimization.

[0007] The second object of the present invention is to provide an enterprise production and sales management and control system based on AI optimization.

[0008] The first technical solution adopted by the present invention is: S1, obtaining the production and sales data of the enterprise, using the historical production and sales data and economic operation data for analysis, and determining the current completion value and future planned quantity;

[0009] S2. Based on the current completion value and the future planned amount, a control algorithm model is constructed to perform prediction calculations to obtain the annual completion value;

[0010] S3. Based on the control management standards and annual target values, the optimal evaluation algorithm is used to perform control calculations to obtain predicted control results;

[0011] S4. Based on the annual target value and the annual completion value, and in accordance with the control requirements and control objectives, the predicted control result is controlled using an optimal search algorithm to obtain an optimal control evaluation result.

[0012] Optionally, S2 includes:

[0013] S21. Using the production and sales data of the enterprise, conduct tax and profit analysis to determine the tax and profit related data within the enterprise and the tax and profit related data of cooperative production;

[0014] S22. Calculate the annual total tax and profit target value model based on the enterprise tax and profit related data and the cooperative production tax and profit related data, wherein the annual total tax and profit target value model is:

[0015]

[0016] In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

[0017] Optionally, S2 further includes:

[0018] S23, using the production and sales data of the enterprise, performing output value analysis to determine the output value-related data within the enterprise;

[0019] S24. Calculate the annual total output value target value model based on the enterprise output value related data to obtain the annual total output value target value model, wherein the annual total output value target value model is:

[0020]

[0021] In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax;

[0022] S25. Based on the annual total output value target model and the annual total tax and profit target model, obtain the control algorithm model.

[0023] Optionally, the S3 includes:

[0024] S31. Determine the key indicator evaluation value and difference based on the control management standard and the annual target value;

[0025] S32, based on the key indicator evaluation value and the difference, calculate to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is:

[0026]

[0027] In the formula, S is the comprehensive evaluation value, w1 is the weight corresponding to the tax and profit evaluation value, w2 is the weight corresponding to the output value evaluation value, θ1 is the tax and profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value;

[0028] S33. Utilize the optimal evaluation algorithm to perform control calculations and obtain predicted control results.

[0029] Optionally, the optimal search algorithm of S4 includes:

[0030] S41. Determine a specification set that needs to be adjusted based on the prediction and control result, wherein the specification set includes an original value;

[0031] S42, taking out two of the original values ​​from the specification set in a preset order, adjusting them according to a preset adjustment step, and calculating an adjusted evaluation value to obtain a plurality of the adjusted evaluation values;

[0032] S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification;

[0033] S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification;

[0034] S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

[0035] The second technical solution adopted by the present invention is: a data analysis module is used to obtain the production and sales data of the enterprise, analyze the historical production and sales data and economic operation data, and determine the current completion value and future planned quantity;

[0036] A prediction and calculation module is used to construct a control algorithm model to perform prediction and calculation based on the current completion value and the future planned amount to obtain the annual completion value;

[0037] The control and measurement module is used to perform control and measurement based on the control management standards and the annual target values ​​using the optimal evaluation algorithm to obtain the predicted control results;

[0038] The control evaluation result module is used to control the predicted control result based on the annual target value and the annual completion value, and according to the control requirements and control objectives, using the optimal search algorithm to obtain the optimal control evaluation result.

[0039] Optionally, the control and measurement module includes:

[0040] A tax-profit analysis module, for performing tax-profit analysis using the production and sales data of the enterprise, and determining the tax-profit related data within the enterprise and the tax-profit related data of cooperative production;

[0041] The tax-profit model building module is used to calculate the annual total tax-profit target value model based on the tax-profit related data within the enterprise and the tax-profit related data of cooperative production, wherein the annual total tax-profit target value model is:

[0042]

[0043] In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

[0044] Optionally, the control and measurement module further includes:

[0045] An output value analysis module, used to use the production and sales data of the enterprise to perform output value analysis and determine the output value related data within the enterprise;

[0046] The output value model building module is used to calculate according to the output value related data within the enterprise to obtain the annual total output value target value model, wherein the annual total output value target value model is,

[0047]

[0048] In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax;

[0049] The control algorithm model construction module is used to obtain the control algorithm model based on the annual total output value target value model and the annual total tax and profit target value model.

[0050] Optionally, the regulation and control evaluation result module includes:

[0051] A key indicator determination module is used to determine the key indicator evaluation value and difference based on the control management standard and the annual target value;

[0052] The optimal evaluation algorithm construction module is used to calculate based on the key indicator evaluation value and the difference to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is:

[0053]

[0054] In the formula, S is the comprehensive evaluation value, w1 is the weight corresponding to the tax and profit evaluation value, w2 is the weight corresponding to the output value evaluation value, θ1 is the tax and profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value;

[0055] The control and measurement module is used to use the optimal evaluation algorithm to perform control and measurement to obtain predicted control results.

[0056] Optionally, the optimal search algorithm of the regulation and evaluation result module includes:

[0057] S41. Determine a specification set that needs to be adjusted based on the prediction and control result, wherein the specification set includes an original value;

[0058] S42, taking out two of the original values ​​from the specification set in a preset order, adjusting them according to a preset adjustment step, and calculating an adjusted evaluation value to obtain a plurality of the adjusted evaluation values;

[0059] S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification;

[0060] S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification;

[0061] S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

[0062] Beneficial effects of the above technical solution:

[0063] This technology systematically collects data on the production and operation processes of enterprises, uses data modeling and AI optimization algorithms to automatically adjust the core economic operation indicators of tobacco enterprises, coordinate and balance the relationship between corporate taxes and profits and production and sales, and guide quarterly and monthly production and operation activities around corporate strategies and annual goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flowchart of the enterprise production and marketing management and control method based on AI optimization provided by the present invention;

[0065] Figure 2 A schematic diagram of the structure of the enterprise production and marketing management and control system based on AI optimization provided by the present invention;

[0066] Figure 3 It is a schematic diagram of the control process flow;

[0067] Figure 4 Schematic diagram of the control algorithm structure. DETAILED DESCRIPTION

[0068] The following detailed description of the embodiments of the present invention is further described in detail in conjunction with the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are used to exemplarily illustrate the principles of the present invention, but cannot be used to limit the scope of the present invention, that is, the present invention is not limited to the preferred embodiments described, and the scope of the present invention is defined by the claims.

[0069] In the description of the present invention, it should be noted that, unless otherwise specified, “plurality” means two or more than two; the terms “first”, “second”, etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance; for ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0070] Embodiment 1

[0071] An embodiment of the present invention provides an enterprise production and marketing management and control method based on AI optimization, such as Figure 1 As shown,

[0072] like Figure 3 As shown in the figure, in the whole regulation process, the core economic benefit indicators are tax and output value, and the two indicators are supported by sales volume and output. The structure, inventory, output and tax of cigarettes are the specific adjustment means and adjustment values ​​of these two core indicators. Changes in the adjustment means will affect the final sales volume, output, and the final tax and profit value.

[0073] At the beginning of the year, each specification will determine the sales target and production target, which is the reference point for specification adjustment. The sum of the production of each specification equals the total production, and the total production can be regarded as a fixed value (if the total production changes, only this fixed value needs to be adjusted). Therefore, in the regulation of the Nth month, it is only necessary to remove the completed sales and production of each specification to obtain the remaining sales and production that can be adjusted. The actual tax profit and actual output value are calculated through sales and production.

[0074] The basic requirement of the control algorithm is to find the best path by adjusting the sales volume and output of different specifications while keeping the total output relatively fixed, so that the actual tax profit and actual output value are close to the annual tax profit target and output value target.

[0075] In this regulation process, assume that the annual tax and profit target is Y, and the annual output value target is P. The ultimate goal is to maximize the comprehensive quality and efficiency of tax and output value (set as S), so the formula can be simply expressed as S = fun (Y, P), which means that Y and P are comprehensively evaluated through the fun algorithm. The larger the S value, the better the comprehensive effect of regulation. The fun algorithm is actually composed of four parts: tax and profit regulation algorithm, output value regulation algorithm, optimal evaluation algorithm and optimal search algorithm.

[0076] The system control model needs to obtain the current completion value of all indicators based on measurement and analysis, and use the tax and profit control algorithm and the output value control algorithm to calculate the annual completion value in combination with the future planned amount. Then, combined with the annual target value, the target deviation degree is calculated, and the measurement and control results are output according to the optimal evaluation algorithm. The system will automatically control the production and sales volume of each specification according to the optimal search algorithm, and output the optimal evaluation and control results under the current conditions to the user. The user confirms whether the results given by the system meet the control requirements. If not, other control measures will be added. If the control requirements are met, the control plan will be output, and the adjustment notice of the goals, plans and budgets will be issued to the relevant business departments.

[0077] like Figure 4 As shown in the figure, the main modules of the system control model are composed of five parts: control target, control means, control content, measurement parameters and control algorithm:

[0078] Regulation and control target: refers to the reference object of regulation and control evaluation, which includes the production and operation targets of the enterprise, including tax and profit targets, production targets and sales targets.

[0079] Regulation content: refers to the most fine-grained measurement object, which includes the production, sales volume and tax and profit of various cigarette specifications on sale.

[0080] Regulatory measures: refers to a collection of various regulatory measures that have been proven feasible in practice to achieve regulatory objectives, including taxation, structure, output, inventory, cooperative production, etc.

[0081] Calculation parameters: refers to the business parameters used in the process of calculating tax, profit and output value, including unit price, inventory-to-sales ratio, tax rate, etc., which are generally obtained through historical data and treated as operation constants.

[0082] Regulation algorithm: The core of regulation is to maximize the comprehensive quality and efficiency S of tax, profit and output value. The calculation process consists of four parts: tax and profit regulation algorithm, output value regulation algorithm, optimal evaluation algorithm and optimal search algorithm.

[0083] S1. Obtain the production and sales data of the enterprise, analyze the historical production and sales data and economic operation data, and determine the current completion value and future planned volume;

[0084] S2. Based on the current completion value and the future planned amount, a control algorithm model is constructed to perform prediction calculations to obtain the annual completion value;

[0085] S21. Using the production and sales data of the enterprise, conduct tax and profit analysis to determine the tax and profit related data within the enterprise and the tax and profit related data of cooperative production;

[0086] S22. Calculate the annual total tax and profit target value model based on the enterprise tax and profit related data and the cooperative production tax and profit related data, wherein the annual total tax and profit target value model is:

[0087]

[0088] In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

[0089] When the enterprise runs to the Nth month, the annual tax and profit target value consists of the completed tax and profit (which can be regarded as a constant value, set as A) and the remaining adjustable tax and profit (variable value). The main purpose of economic operation regulation is to calculate the remaining adjustable variable value.

[0090] The remaining adjustable tax and profit is composed of the remaining adjustable tax and profit of each specification, which is obtained by multiplying the single box tax value of this specification by the remaining adjustable allocation of this specification. The single box tax value is mainly obtained based on the historical financial volume principal and profit, which can be regarded as a relatively fixed coefficient value, set as β. The real variable is the adjustable allocation value of the remaining months (set as x). The calculation formula for the remaining tax and profit of a single specification is expressed as xβ.

[0091] Assume that there are m specifications involved in the calculation of annual tax and profit. Among these m specifications, some specifications need to be limited according to brand development requirements and market demand, that is, these limited specifications cannot adjust the allocation amount, and the remaining tax and profit generated by these specifications can be regarded as constants (set to B). Then, after excluding the specifications that need to be limited, the remaining specifications are allowed to be adjusted, and the set of these specifications is set to N. Then the control algorithm is to find the path of the best control method that can be satisfied in the set N = {x1...xn}.

[0092] Since cooperative production is not included in the company's tax and profit, the tax and profit of cooperative production must be removed in the calculation. Assuming that there are k specifications involved in cooperative production, the calculation method is also obtained by multiplying the cooperative production allocation of each specification (set as x') by the cooperative production per box tax and profit of this specification (set as β'). The calculation formula is expressed as x'β', and finally the total tax and profit of cooperative production of k specifications is summed up. Let the set of these specifications be K, then the regulation algorithm of cooperative production is to find the path that can satisfy the best regulation method in the set K = {x'1...x'n}.

[0093] Now the whole algorithm is to adjust the values ​​of {x1...xn} in the set N so that the y value is close to or equal to the annual target value Y (denoted as y→Y).

[0094] S23, using the production and sales data of the enterprise, performing output value analysis to determine the output value-related data within the enterprise;

[0095] S24. Calculate the annual total output value target value model based on the enterprise output value related data to obtain the annual total output value target value model, wherein the annual total output value target value model is:

[0096]

[0097] In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax;

[0098] S25. Based on the annual total output value target model and the annual total tax and profit target model, obtain the control algorithm model.

[0099] When the enterprise runs to the Nth month, the annual output value target is composed of the completed output value (which can be regarded as a constant value, set as C) and the remaining adjustable output value (variable value). The main purpose of operation control is to calculate the remaining adjustable output value.

[0100] The remaining adjustable output value is composed of the remaining adjustable output value of each specification, which is obtained by multiplying the tax-exclusive transfer unit price of this specification by the remaining production volume (self-produced) of this specification. The tax-exclusive transfer unit price can be regarded as a relatively fixed coefficient value, set to ɡ, and the real variable is the production volume of the remaining months (set to z). The calculation formula for the remaining output value of a single specification is zɡ.

[0101] Assume that there are a total of m specifications involved in the calculation of annual output value. Among these m specifications, some specifications need to be limited according to brand development requirements and market demand, that is, these limited specifications cannot adjust the output value, and the total output value of these specifications can be regarded as a constant (set to D). Then after excluding the specifications that need to be limited, the remaining ones are the specifications that are allowed to be adjusted, and the set of these specifications is set to R. Then the control algorithm is to find the path of the best control method that can be satisfied in the set R = {z1...zr}.

[0102] In this formula, two more limiting conditions need to be added: first, the total output h of m specifications must be close to or equal to the company's annual total output (set as H), denoted as h→H; second, self-produced output + cooperative production output + surplus inventory must be greater than or equal to the annual total sales volume.

[0103] Based on the above formula and two limiting conditions, the entire algorithm now adjusts the values ​​of {x1...xr} in the set R so that the calculated p value is close to or equal to the annual target value P (denoted as p→P).

[0104] S3. Based on the control management standards and annual target values, the optimal evaluation algorithm is used to perform control calculations to obtain predicted control results;

[0105] S31. Determine the key indicator evaluation value and difference based on the control management standard and the annual target value;

[0106] S32, based on the key indicator evaluation value and the difference, calculate to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is:

[0107]

[0108] In the formula, S is the comprehensive evaluation value, w1 is the corresponding weight of the tax-profit evaluation value, w2 is the corresponding weight of the output value evaluation value, θ1 is the tax-profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value.

[0109] In terms of tax and profit, output, output value, and the final comprehensive evaluation, the issue of how to conduct optimal evaluation will be involved, that is, how to evaluate that the actual value o approaches and equals the target value O (denoted as o→O), and define the evaluated value as the θ value, that is, θ = o→O. And the process of calculating the θ value is to calculate the similarity between the two points o and O. The higher the similarity, the better.

[0110] To ensure that the final comprehensive evaluation value θ is positively distributed, and the values of different types (tax and profit, output, output value) can be calculated and compared together, it is carried out by comparing two values. The specific algorithm: if o < O, θ = o / O; if o > O, θ = O / o; if o = O, θ = 1. In this way, when o ≠ O, θ will infinitely approach 1. The larger the value, the higher the similarity, and when it equals 1, it means the best.

[0111] Through this algorithm, the θ value of tax and profit (denoted as θ1 = y→Y), the θ value of output value (denoted as θ2 = p→P), and the θ value of output (denoted as θ3 = h→H) can be calculated. Since these θ values have been normalized, only the weighted average (with the weight denoted as w) is needed to calculate the final comprehensive evaluation value.

[0112] Because the comprehensive evaluation value (denoted as S) is mainly composed of the tax and profit θ1 and the output value θ2, and the output θ3 is only used as a limiting condition. Assuming that the deviation value of the output θ3 from 1 is required to be controlled within α (0 < α < 1), then the final calculation formula.

[0113] The comprehensive evaluation value S calculated by this formula will infinitely approach 1. The larger the value, the better the effect after regulation. In this way, the optimal evaluation problem of the algorithm is solved.

[0114] S33. Use the optimal evaluation algorithm to conduct regulation measurement and obtain the predicted regulation result.

[0115] S4. Based on the annual target value and the annual completed value, and according to the regulation requirements and regulation goals, use the optimal search algorithm to regulate the predicted regulation result to obtain the optimal regulation evaluation result.

[0116] The optimal search algorithm includes:

[0117] S41. Based on the predicted regulation result, determine the set of specifications that need to be adjusted, where the set of specifications includes the original values;

[0118] S42. In the preset order, take out two of the original values from the set of specifications and adjust them according to the preset adjustment step size, and calculate the adjustment evaluation value to obtain a number of the adjustment evaluation values;

[0119] S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification;

[0120] S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification;

[0121] S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

[0122] After determining the algorithms and rules for tax, output value and optimal evaluation, the next step is to solve the problem of optimal search. The best combination value should be searched in the set N = {x1...xn} and the set R = {z1...zr} to maximize the S value. If the optimal search algorithm is not designed and the search is done manually, it will be necessary to try and compare the elements of the sets N and R for Nn×Rr times until the maximum S value is found. This search process will be an astronomical number. If the regulation of the cooperative production set K = {x'1...x'n} is also taken into account, the search range will be even larger.

[0123] Therefore, the optimal search algorithm is also the core part of this regulation algorithm. The basic idea of ​​the search is the same for the three sets N, R, and K. Take the set N = {x1...xn} as an example for analysis.

[0124] All the original values ​​to be adjusted are stored in the set N in advance. The search algorithm is to continuously adjust the size of the values ​​in the set, calculate the θ value of the tax profit, find the largest θ through comparison, and complete the search.

[0125] First, assume that the total amount is fixed. When adjusting the x value in the set N = {x1...xn}, it is done by increasing a certain step size (set to ε, an integer), that is, the adjusted value x' = x + ε. When the total amount remains unchanged, increasing a certain specification means that other specifications need to be reduced. Therefore, this step size always appears in positive and negative pairs (i.e. ±ε). In each search step, it is necessary to take out two specifications a and b from the set N = {x1...xn} in a certain order, adjust their values ​​to x'a = xa + ε, x'b = xb - ε, and then calculate θ. Through multiple selections and comparisons, the maximum θ value under this step size (set to θ1) will be found, and the two specifications a and b that really need to be adjusted are obtained, and their values ​​xa and xb are adjusted to xa + ε and xb - ε.

[0126] After completing one step, increase the step length ε and repeat the above process to find the maximum θ value of the new step length (set as θ2). By comparing θ1 and θ2, if the value of θ2 is getting bigger, it means that the situation is developing in a good direction. Continue to increase the step length ε and continue searching; if the value of θ2 is getting smaller, it means that the situation is developing in a bad direction, and stop searching for this round of ε.

[0127] When the previous round of step length ε search is completed, the step length ε is reduced to a finer adjustment, and the new step length is set to ε' = ε / 2, where ε' is an integer. If ε' is greater than 0 or a certain control value (which can be set), the new round of step length search processing is repeated; otherwise, the optimal search algorithm is exited. The entire algorithm is a cyclic processing process, and the calculation formula is expressed as follows:

[0128]

[0129] Finally, the optimal solution adjusted at the finest granularity ε will be generated, and the x value in the set N = {x1....xn} is the new value after adjustment of each specification.

[0130] If the total amount needs to be adjusted, just adjust the original value in the set N = {x1...xn} so that the sum of the values ​​in the set is equal to the adjusted total amount, and then search according to the above algorithm.

[0131] In this search process, the step of taking out two specifications a and b from the set N = {x1...xn} in a certain order should be able to optimize performance. Since the tax and profit per box of each specification in the set N is a fixed value, but the tax and profit per box will directly affect the change of the y value, the set N needs to be sorted according to the tax and profit per box. If the tax and profit is too low, the structure needs to be improved. The selected a should be searched from high to low according to the tax and profit per box, and the selected b should be searched from low to high according to the tax and profit per box; vice versa.

[0132] This embodiment is a production and sales management and control method for enterprises based on AI optimization, including obtaining the production and sales data of cigarettes of enterprises, analyzing the historical production and sales data and economic operation data, determining the current completion value and the future planned quantity; performing forecast calculation based on the current completion value and the future planned quantity, obtaining the annual completion value, obtaining the annual target data of the enterprise, constructing a control algorithm model based on the control management standard, regulating the future planned quantity and the annual target value, obtaining the forecast control result, and evaluating the forecast control result based on the annual target value and the annual completion value according to the control requirements and goals using the evaluation algorithm to obtain the control evaluation result, if the control evaluation result is optimal, then outputting the control plan, if the control evaluation result is not optimal, then using the optimal search algorithm, using the control algorithm model to control again. With the help of information technology, monthly, quarterly, semi-annual and annual economic operation control and enterprise development strategy are coordinated, and a comprehensive resource coordination mechanism and operation control mechanism from strategy to operation is established, and the intelligent level of economic operation control is improved through the measurement model. Through digital technology, we can standardize the analysis of enterprise economic operation indicators, unify the economic operation regulation process, improve the overall coordination ability of operation regulation and production, supply and marketing plans, and support enterprises to make accurate decisions and respond quickly to the market.

[0133] Embodiment 2

[0134] An embodiment of the present invention provides an enterprise production and marketing management and control system 200 based on AI optimization, such as Figure 2 As shown, including:

[0135] The data analysis module 201 is used to obtain the production and sales data of the enterprise, analyze the historical production and sales data and economic operation data, and determine the current completion value and future planned quantity;

[0136] A prediction calculation module 202 is used to construct a control algorithm model to perform prediction calculation based on the current completion value and the future planned amount to obtain the annual completion value;

[0137] The control and measurement module 203 is used to perform control and measurement based on the control and management standards and the annual target value using the optimal evaluation algorithm to obtain the predicted control and measurement results;

[0138] The control evaluation result module 204 is used to control the predicted control result based on the annual target value and the annual completion value, and according to the control requirements and control objectives, using the optimal search algorithm to obtain the optimal control evaluation result.

[0139] Furthermore, the control and measurement module 203 includes:

[0140] A tax-profit analysis module, for performing tax-profit analysis using the production and sales data of the enterprise, and determining the tax-profit related data within the enterprise and the tax-profit related data of cooperative production;

[0141] The tax-profit model building module is used to calculate the annual total tax-profit target value model based on the tax-profit related data within the enterprise and the tax-profit related data of cooperative production, wherein the annual total tax-profit target value model is:

[0142]

[0143] In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

[0144] Furthermore, the control and measurement module 203 also includes:

[0145] An output value analysis module, used to use the production and sales data of the enterprise to perform output value analysis and determine the output value related data within the enterprise;

[0146] The output value model building module is used to calculate according to the output value related data within the enterprise to obtain the annual total output value target value model, wherein the annual total output value target value model is,

[0147]

[0148] In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax;

[0149] The control algorithm model construction module is used to obtain the control algorithm model based on the annual total output value target value model and the annual total tax and profit target value model.

[0150] Furthermore, the control evaluation result module 204 includes:

[0151] A key indicator determination module is used to determine the key indicator evaluation value and difference based on the control management standard and the annual target value;

[0152] The optimal evaluation algorithm construction module is used to calculate based on the key indicator evaluation value and the difference to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is:

[0153]

[0154] In the formula, S is the comprehensive evaluation value, w1 is the weight corresponding to the tax and profit evaluation value, w2 is the weight corresponding to the output value evaluation value, θ1 is the tax and profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value;

[0155] The control and measurement module is used to use the optimal evaluation algorithm to perform control and measurement to obtain predicted control results.

[0156] Furthermore, the optimal search algorithm of the control evaluation result module includes:

[0157] S41. Determine a specification set that needs to be adjusted based on the prediction and control result, wherein the specification set includes an original value;

[0158] S42, taking out two of the original values ​​from the specification set in a preset order, adjusting them according to a preset adjustment step, and calculating an adjusted evaluation value to obtain a plurality of the adjusted evaluation values;

[0159] S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification;

[0160] S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification;

[0161] S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

[0162] This embodiment is a production and marketing management and control system for enterprises based on AI optimization, including a data analysis module for obtaining enterprise production and marketing data, using historical production and marketing data and economic operation data for analysis, and determining the current completion value and future planned quantity; a prediction and calculation module for constructing a control algorithm model for prediction and calculation based on the current completion value and future planned quantity to obtain the annual completion value; a control and measurement module for control and measurement based on the control management standard and the annual target value, using the optimal evaluation algorithm to obtain the predicted control result; a control and evaluation result module for control and measurement based on the annual target value and the annual completion value, and according to the control requirements and control objectives, using the optimal search algorithm to control the predicted control result to obtain the optimal control evaluation result. Through the system to collect enterprise production and operation process data, use data modeling and AI optimization algorithm to automatically control the core economic operation indicators of tobacco enterprises, coordinate and balance the relationship between enterprise tax and profit and production and sales, and guide the production and operation activities of each quarter and month to revolve around the enterprise strategy and annual goals.

[0163] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for enterprise production and marketing management and control based on AI optimization, characterized in that: include: S1. Obtain the production and sales data of the enterprise, analyze the historical production and sales data and economic operation data, and determine the current completion value and future planned volume; S2. Based on the current completion value and the future planned amount, a control algorithm model is constructed to perform prediction calculations to obtain the annual completion value; S3. Based on the control management standards and annual target values, the optimal evaluation algorithm is used to perform control calculations to obtain predicted control results; S4. Based on the annual target value and the annual completion value, and in accordance with the control requirements and control objectives, the predicted control result is controlled using an optimal search algorithm to obtain an optimal control evaluation result.

2. The enterprise production and marketing management and control method based on AI optimization according to claim 1 is characterized in that: The S2 includes: S21. Using the production and sales data of the enterprise, conduct tax and profit analysis to determine the tax and profit related data within the enterprise and the tax and profit related data of cooperative production; S22. Calculate the annual total tax and profit target value model based on the enterprise tax and profit related data and the cooperative production tax and profit related data, wherein the annual total tax and profit target value model is: In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

3. The enterprise production and marketing management and control method based on AI optimization according to claim 2 is characterized in that: The S2 further includes: S23, using the production and sales data of the enterprise, performing output value analysis to determine the output value-related data within the enterprise; S24. Calculate the annual total output value target value model based on the enterprise output value related data to obtain the annual total output value target value model, wherein the annual total output value target value model is: In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax; S25. Based on the annual total output value target model and the annual total tax and profit target model, obtain the control algorithm model.

4. The enterprise production and marketing management and control method based on AI optimization according to claim 1 is characterized in that: The S3 includes: S31. Determine the key indicator evaluation value and difference based on the control management standard and the annual target value; S32, based on the key indicator evaluation value and the difference, calculate to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is: In the formula, S is the comprehensive evaluation value, w1 is the weight corresponding to the tax and profit evaluation value, w2 is the weight corresponding to the output value evaluation value, θ1 is the tax and profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value; S33. Utilize the optimal evaluation algorithm to perform control calculations and obtain predicted control results.

5. The enterprise production and marketing management and control method based on AI optimization according to claim 1 is characterized in that: The optimal search algorithm of S4 includes: S41. Determine a specification set that needs to be adjusted based on the prediction and control result, wherein the specification set includes an original value; S42, taking out two of the original values ​​from the specification set in a preset order, adjusting them according to a preset adjustment step, and calculating an adjusted evaluation value to obtain a plurality of the adjusted evaluation values; S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification; S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification; S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

6. An enterprise production and marketing management and control system based on AI optimization, characterized in that: include: The data analysis module is used to obtain the production and sales data of the enterprise, analyze the historical production and sales data and economic operation data, and determine the current completion value and future planned volume; A prediction and calculation module is used to construct a control algorithm model to perform prediction and calculation based on the current completion value and the future planned amount to obtain the annual completion value; The control and measurement module is used to perform control and measurement based on the control management standards and the annual target values ​​using the optimal evaluation algorithm to obtain the predicted control results; The control evaluation result module is used to control the predicted control result based on the annual target value and the annual completion value, and according to the control requirements and control objectives, using the optimal search algorithm to obtain the optimal control evaluation result.

7. The enterprise production and marketing management and control system based on AI optimization according to claim 6 is characterized in that: The control and measurement module includes: A tax-profit analysis module, for performing tax-profit analysis using the production and sales data of the enterprise, and determining the tax-profit related data within the enterprise and the tax-profit related data of cooperative production; The tax-profit model building module is used to calculate the annual total tax-profit target value model based on the tax-profit related data within the enterprise and the tax-profit related data of cooperative production, wherein the annual total tax-profit target value model is: In the formula, y is the annual total tax and profit target value, A is the completed tax and profit, B is the limited tax and profit, N is the tax and profit adjustment specification allowed within the enterprise, and x is j is the amount that can be transferred within the enterprise, β j Single box tax system within the enterprise, K is the cooperative production specification, x' j The amount of cooperative production transfer, β' j For cooperative production of single box tax and profit.

8. The enterprise production and marketing management and control system based on AI optimization according to claim 7 is characterized in that: The control and measurement module also includes: An output value analysis module, used to use the production and sales data of the enterprise to perform output value analysis and determine the output value related data within the enterprise; The output value model building module is used to calculate according to the output value related data within the enterprise to obtain the annual total output value target value model, wherein the annual total output value target value model is, In the formula, p is the annual total output target, C is the completed output, D is the limited total output, R is the output specification allowed to be adjusted, and z j is the total remaining output, g j The transfer unit price is excluding tax; The control algorithm model construction module is used to obtain the control algorithm model based on the annual total output value target value model and the annual total tax and profit target value model.

9. The enterprise production and marketing management and control system based on AI optimization according to claim 6 is characterized in that: The control evaluation result module includes: A key indicator determination module, used to determine the key indicator evaluation value and difference based on the control management standard and the annual target value; The optimal evaluation algorithm construction module is used to calculate based on the key indicator evaluation value and the difference to obtain the optimal evaluation algorithm, wherein the optimal evaluation algorithm is: In the formula, S is the comprehensive evaluation value, w1 is the weight corresponding to the tax and profit evaluation value, w2 is the weight corresponding to the output value evaluation value, θ1 is the tax and profit evaluation value, θ2 is the output value evaluation value, θ3 is the output evaluation value, and α is the deviation value; The control and measurement module is used to use the optimal evaluation algorithm to perform control and measurement to obtain predicted control results.

10. The enterprise production and marketing management and control system based on AI optimization according to claim 6, characterized in that: The optimal search algorithm of the control evaluation result module includes: S41. Determine a specification set that needs to be adjusted based on the prediction and control result, wherein the specification set includes an original value; S42, taking out two of the original values ​​from the specification set in a preset order, adjusting them according to a preset adjustment step, and calculating an adjusted evaluation value to obtain a plurality of the adjusted evaluation values; S43, comparing a plurality of the adjustment evaluation values ​​to determine the maximum adjustment evaluation value, and then determining the current optimal adjustment specification; S44, after adding the adjustment step again, return to S42 to perform calculation, obtain a new adjustment evaluation value and compare it with the current optimal adjustment specification, until the new adjustment evaluation value is less than the current optimal adjustment specification; S45. After a round of search is completed, the adjustment step size is reduced by half and rounded up, and then the process returns to S42 for a new round of calculation, and the process ends when the reduced adjustment step size is less than a preset control value or 0.

Images