Method for realizing budget dynamic prediction technology based on flexible formula and memory calculation

Abstract

The invention provides a method for realizing a budget dynamic prediction technology based on a flexible formula and memory calculation, and belongs to the technical field of enterprise data management. The method comprises the following steps: acquiring enterprise historical data, and analyzing industry characteristics and business modes of an enterprise based on the enterprise historical data; determining a budget prediction model based on the industry characteristics and the business mode, and adjusting parameters of the budget prediction model through a cross validation method; configuring a dynamic budget formula in the budget prediction model based on the business demand and the data structure, wherein the dynamic budget formula comprises a formula based on dimensions, a table sample, an Exce l function, a business preset function and an external interface; and performing time sequence analysis, regression analysis and scene analysis by using a dynamic budget formula to generate a plurality of sets of budget prediction results. By integrating enterprise historical data, flexibly configuring a budget formula and applying a memory computing technology, efficient, accurate and dynamic budget prediction adapting to different business scenes is realized, and the accuracy of budget compilation is remarkably improved.

Description

Technical Field

[0001] The present application belongs to the field of enterprise data management technology, and specifically relates to a method for implementing budget dynamic forecasting technology based on flexible formulas and memory calculations. Background Art

[0002] In modern enterprise management, budgeting is a critical step in ensuring the achievement of strategic objectives. Traditional budgeting methods rely primarily on manual input and analysis of historical data, a time-consuming and labor-intensive approach that struggles to adapt to rapidly changing market environments and complex business needs. As enterprises expand and market competition intensifies, efficient and accurate budget forecasting has become a pressing challenge.

[0003] However, existing budget systems often lack flexibility and intelligence, and cannot fully support the dynamic adjustment needs of enterprises in different business scenarios, resulting in a complicated budget formulation process and inaccurate results. Summary of the Invention

[0004] In order to solve at least one technical problem existing in the background technology, the present application provides a method for implementing budget dynamic forecasting technology based on flexible formulas and memory computing. By integrating enterprise historical data, flexibly configuring budget formulas and applying memory computing technology, efficient, accurate and dynamically adaptable budget forecasts for different business scenarios are achieved, significantly improving the accuracy of budget preparation.

[0005] The technical solutions adopted in this application are:

[0006] The first embodiment of the present application provides a method for implementing a dynamic budget prediction technology based on a flexible formula and memory calculation, including:

[0007] Obtaining historical enterprise data, and analyzing the industry characteristics and business model of the enterprise based on the historical enterprise data, wherein the historical data includes sales records, cost details, and financial statements;

[0008] Determining a budget forecasting model based on the industry characteristics and the business model, and adjusting the parameters of the budget forecasting model through a cross-validation method;

[0009] Configuring a dynamic budget formula in the budget forecasting model based on business requirements and data structure, wherein the dynamic budget formula includes a formula based on dimensions, table samples, Excel functions, business preset functions, and external interfaces;

[0010] The dynamic budget formula is used to conduct time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

[0011] According to the method for implementing dynamic budget forecasting technology based on flexible formulas and memory calculations provided by the embodiment of the first aspect of this application, first, the historical data of the enterprise is obtained, and the industry characteristics and business model of the enterprise are analyzed based on information such as sales records, cost details, and financial statements, so as to lay a data foundation for the enterprise budget modeling; secondly, according to the identified industry and business characteristics, an adaptive budget forecasting model is selected, and the model parameters are optimized and adjusted through the cross-validation method to improve the forecast accuracy and generalization ability; then, a dynamic budget formula based on dimensions, tables, Excel functions, business preset functions and external interfaces is configured in the model to achieve flexible definition and rapid adjustment of budget logic to meet complex and changing business needs; finally, the dynamic budget formula is used to carry out time series analysis, regression analysis and scenario simulation analysis to generate multiple sets of budget forecast results, supporting enterprises to make scientific decisions in different business scenarios. This method significantly improves the accuracy, response speed and system adaptability of budget forecasting by introducing a flexible formula configuration mechanism and an efficient memory computing architecture, which not only reduces the errors and costs caused by manual intervention, but also enhances the financial management capabilities and risk response capabilities of enterprises in a dynamic market environment.

[0012] According to one embodiment of the present application, the acquisition of enterprise historical data and analysis of the industry characteristics and business model of the enterprise based on the enterprise historical data, wherein the historical data includes sales records, cost details, and financial statements, specifically:

[0013] Obtain the enterprise's historical data from the enterprise's ERP system, accounting software, and internal and external databases, and perform data cleaning and preprocessing on the enterprise's historical data;

[0014] Identify market trends, seasonal changes, and macroeconomic conditions based on cleansed historical enterprise data.

[0015] According to one embodiment of the present application, the acquisition of enterprise historical data includes analyzing the industry characteristics and business model of the enterprise based on the enterprise historical data, where the historical data includes sales records, cost details, and financial statements, and also includes:

[0016] Extracting trend items, cycle items and residual items of the enterprise's historical data based on a time series decomposition method;

[0017] Use cluster analysis or classification algorithms to segment and categorize the industry in which the enterprise operates, and identify industry indicators related to the enterprise's operations;

[0018] Combined with external macroeconomic data interfaces, it dynamically obtains GDP growth rate, inflation rate and industry policy changes.

[0019] According to one embodiment of the present application, determining a budget forecast model based on the industry characteristics and the business model, and adjusting the parameters of the budget forecast model through a cross-validation method are specifically as follows:

[0020] The budget forecasting model includes a time series model, a regression model and a machine learning model;

[0021] The budget forecasting model is trained and validated using the K-fold cross-validation method to evaluate the model performance indicators under different parameter combinations, including mean square error, mean absolute percentage error, and coefficient of determination.

[0022] According to one embodiment of the present application, a dynamic budget formula is configured in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes a formula based on dimensions, table samples, Excel functions, business preset functions, and external interfaces, specifically:

[0023] A graphical formula configuration interface is provided in the budget preparation system to support users in defining personalized calculation logic based on different budget dimensions;

[0024] A business preset function library is set, which includes a growth rate calculation function, a year-on-year and month-on-month analysis function, a rolling forecast function, and a budget control threshold judgment function.

[0025] According to one embodiment of the present application, a dynamic budget formula is configured in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes a formula based on dimensions, table templates, Excel functions, business preset functions, and external interfaces, and further includes:

[0026] Call calculation functions or real-time data sources of external systems through API interfaces to achieve cross-system data linkage and formula collaborative calculation.

[0027] According to one embodiment of the present application, the dynamic budget formula is used to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results, specifically:

[0028] Identify trend changes and cyclical patterns in the company's historical data through time series analysis, adjust the forecast value based on seasonal factors, and generate the first-class budget forecast results;

[0029] Evaluate the impact of multiple business variables on budget targets based on regression analysis models, construct multivariate linear or nonlinear regression equations, and output second-category budget forecast results;

[0030] Setting multiple business scenario parameters to simulate the business operation status under different market environments, and performing multi-variable combination operations through dynamic budget formulas to generate third-category budget forecast results, where the business scenario parameters include optimistic scenario, neutral scenario, and pessimistic scenario;

[0031] The first category budget forecast results, the second category budget forecast results and the third category budget forecast results are compared and analyzed to determine the final budget forecast results.

[0032] A second embodiment of the present application provides a device for implementing a dynamic budget prediction technology based on a flexible formula and memory calculation, including:

[0033] A data acquisition module, adapted to acquire historical enterprise data and analyze the industry characteristics and business model of the enterprise based on the historical enterprise data, wherein the historical data includes sales records, cost details, and financial statements;

[0034] A model determination module, adapted to determine a budget forecast model based on the industry characteristics and the business model, and adjust parameters of the budget forecast model through a cross-validation method;

[0035] a configuration module adapted to configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure, wherein the dynamic budget formula includes a formula based on dimensions, table templates, Excel functions, business preset functions, and external interfaces;

[0036] The result generation module is adapted to perform time series analysis, regression analysis and scenario analysis using the dynamic budget formula to generate multiple sets of budget forecast results.

[0037] An embodiment of the third aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for implementing dynamic budget prediction technology based on flexible formulas and memory calculations as described in any embodiment of the first aspect described above is implemented.

[0038] The present application also provides a non-volatile computer storage medium having computer executable instructions stored thereon. When the computer program is executed by a processor, the method for implementing budget dynamic prediction technology based on flexible formulas and memory calculations in any embodiment of the first aspect as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0040] Figure 1A flowchart of a method for implementing dynamic budget prediction technology based on flexible formulas and memory calculations provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram of the structure of a device for implementing a dynamic budget prediction technology based on a flexible formula and memory calculation according to an embodiment of the present application;

[0042] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0043] Reference numerals:

[0044] 110. Data acquisition module; 120. Model determination module; 130. Configuration module; 140. Result generation module;

[0045] 810 , processor; 820 , communication interface; 830 , memory; 840 , communication bus. DETAILED DESCRIPTION

[0046] In order to more clearly illustrate the overall concept of the present application, a detailed description is given below in an illustrative manner in conjunction with the accompanying drawings.

[0047] The following description sets forth many specific details to facilitate a thorough understanding of the present application. However, the present application may also be implemented in other ways than those described herein, and therefore, the scope of protection of the present application is not limited by the specific embodiments disclosed below. It should be noted that the embodiments of the present application and the features of each embodiment may be combined with each other unless there is a conflict.

[0048] In this application, unless otherwise expressly specified and limited, a first feature "above" or "below" a second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments or examples.

[0049] like Figure 1 As shown, the first embodiment of the present application provides a method for implementing budget dynamic prediction technology based on flexible formulas and memory calculations, including:

[0050] Step 100: Obtain the company's historical data and analyze the company's industry characteristics and business model based on the company's historical data. The historical data includes sales records, cost details, and financial statements.

[0051] Step 200: Determine a budget forecast model based on industry characteristics and business models, and adjust the parameters of the budget forecast model through a cross-validation method.

[0052] Step 300: Configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces.

[0053] Step 400: Use the dynamic budget formula to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

[0054] In step 100, the system automatically collects historical enterprise data, including but not limited to sales records, cost details, and financial statements, from the enterprise's ERP system, accounting software, and internal and external databases. This raw data is then cleaned and preprocessed to remove duplicate, missing, or abnormal data and standardize the data format. Based on this, a time series decomposition method is used to extract trend terms, cyclical terms, and residual terms. Cluster analysis or classification algorithms are then used to identify the industry segment and key operating indicators of the enterprise. Furthermore, an external macroeconomic data interface is connected to dynamically obtain macroeconomic variables such as GDP growth rate, inflation rate, and policy changes, providing comprehensive data support for enterprise budget modeling.

[0055] This step, through in-depth mining and structured processing of a company's historical data, effectively identifies the company's position within the industry and the characteristics of its operating model, providing a scientific basis for subsequent budget model selection and parameter tuning. Compared to traditional methods that rely on manual experience and judgment, this step implements data-driven automated analysis, improving the intelligence of the budget forecasting system, reducing human error, and enhancing the objectivity and accuracy of forecast results.

[0056] In step 200, the system automatically matches the appropriate budget forecasting model based on the enterprise industry attributes and business characteristics analyzed in the previous stage, such as time series models (ARIMA, exponential smoothing), regression models (linear regression, ridge regression) or machine learning models (random forest, gradient boosting tree). The cleaned data is then divided into a training set and a validation set. The model is trained and validated multiple times using the K-fold cross-validation method to evaluate the model performance under different parameter combinations. The mean square error (MSE), mean absolute percentage error (MAPE) and coefficient of determination (R) are selected. 2 ) as the main evaluation index, and finally determine the optimal model parameter configuration to improve the model prediction accuracy and generalization ability.

[0057] Specifically, the manufacturing industry is characterized by significant seasonal sales fluctuations (such as increased sales during holiday promotions), sensitivity to raw material prices (particularly price fluctuations in key materials like steel and rubber, which can impact costs), and a high reliance on supply chain management. Business models include complex production processes, encompassing multiple stages from design and procurement to production and sales; diversified sales channels, encompassing direct sales, dealer networks, and online platforms; and an emphasis on inventory management and quality control.

[0058] For manufacturing companies, given their complex production processes and sensitivity to raw material prices, time series models (such as ARIMA or exponential smoothing) can be used to capture the cyclical and trend components in sales data. Regression models can also be used to assess the impact of factors like advertising investment and promotional activities on sales. Furthermore, machine learning models (such as random forests or gradient boosting trees) can handle complex nonlinear relationships and combine multiple internal and external factors for more accurate forecasts.

[0059] The retail industry is characterized by significant seasonal sales peaks, rapid response to market trends, and price sensitivity in a highly competitive environment. Business models primarily rely on multi-channel sales, both online and offline, with a focus on inventory management and supply chain optimization. Customer experience and service quality are core competitive advantages.

[0060] In the retail industry, due to distinct seasonal sales peaks and rapidly changing market demand, time series models are effective for short-term sales forecasting. To better understand market trends and the impact of promotional activities, regression analysis models can be applied to assess the impact of multiple variables on sales targets and construct multiple regression equations. Scenario analysis can also be used to simulate a company's operating conditions under different market environments.

[0061] In the service industry, industry characteristics typically include an emphasis on user growth and customer retention, with revenue primarily derived from monthly or annual recurring fees. Business models emphasize customer service and support, continuous product updates and service upgrades, and achieving stable revenue growth through a subscription model.

[0062] In the service sector, particularly SaaS companies, logistic regression or survival analysis models can be used to predict customer churn probability, while time series analysis is suitable for revenue forecasting. K-fold cross-validation allows you to optimize model parameters through multiple iterations of training and validation, ensuring good generalization and predictive accuracy.

[0063] This step improves the adaptability and stability of the budget forecasting model by introducing multiple forecasting models and a cross-validation mechanism, enabling the model to maintain high forecast accuracy across different industries and business scenarios. Compared to the traditional approach of using a single model with fixed parameters, this step significantly enhances the system's intelligent selection and self-optimization capabilities, ensuring the scientific and efficient budget forecasting process, helping companies quickly respond to market changes and make accurate decisions.

[0064] In step 300, the system provides a graphical interface for users to define personalized calculation logic based on different budget dimensions (such as department, product line, and region). It also supports the configuration of structured formulas based on budget tables, aligning formula logic with the budget table layout, improving intuitiveness and accuracy. Furthermore, the system integrates an Excel function parsing engine, compatible with the syntax of commonly used Excel formulas, facilitating the migration of existing Excel models. It also includes a built-in business pre-built function library covering functional modules such as growth rate calculation, year-on-year and month-on-month analysis, rolling forecasts, and budget control threshold determination. It also supports calling calculation functions or real-time data sources from external systems through an API interface, enabling cross-system budget linkage.

[0065] This step addresses the diverse budgeting needs of enterprises across different business scenarios by building a flexible and configurable budget formula system. This overcomes the limitations of traditional budget systems, which often feature rigid formulas and poor scalability. It allows non-technical personnel to participate in budget logic configuration, lowering the barrier to entry for system use and improving budgeting efficiency and flexibility. Furthermore, the cross-system interface design enhances the system's openness and integration capabilities, providing a solid foundation for enterprises to build integrated budget management platforms.

[0066] In step 400, the system performs time series analysis, regression analysis, and scenario simulation analysis based on the configured dynamic budget formula. Time series analysis identifies trends and cyclical patterns in historical data and adjusts forecasts based on seasonal factors. Regression analysis assesses the impact of multiple business variables on budget targets and constructs a multivariate regression model. Scenario analysis simulates business operations under different market conditions by setting optimistic, neutral, and pessimistic business scenarios, generating corresponding budget forecasts. Finally, the system compares and analyzes these three types of results and selects the optimal forecast based on the forecast error index, which serves as the basis for formal budget compilation.

[0067] This step integrates multiple mainstream forecasting and analysis methods to achieve diverse and comparable budget forecast results, helping companies comprehensively assess their financial performance under different operating scenarios and enhancing the scientific and forward-looking nature of budget decisions. Compared to traditional single linear forecasting methods, this step not only enhances the flexibility and accuracy of budget forecasts but also improves the system's ability to cope with uncertainty, helping companies develop more flexible and risk-controlled financial strategies.

[0068] According to the method for implementing dynamic budget forecasting technology based on flexible formulas and memory calculations provided by the embodiment of the first aspect of this application, first, the historical data of the enterprise is obtained, and the industry characteristics and business model of the enterprise are analyzed based on information such as sales records, cost details, and financial statements, so as to lay a data foundation for the enterprise budget modeling; secondly, according to the identified industry and business characteristics, an adaptive budget forecasting model is selected, and the model parameters are optimized and adjusted through the cross-validation method to improve the forecast accuracy and generalization ability; then, a dynamic budget formula based on dimensions, tables, Excel functions, business preset functions and external interfaces is configured in the model to achieve flexible definition and rapid adjustment of budget logic to meet complex and changing business needs; finally, the dynamic budget formula is used to carry out time series analysis, regression analysis and scenario simulation analysis to generate multiple sets of budget forecast results, supporting enterprises to make scientific decisions in different business scenarios. This method significantly improves the accuracy, response speed and system adaptability of budget forecasting by introducing a flexible formula configuration mechanism and an efficient memory computing architecture, which not only reduces the errors and costs caused by manual intervention, but also enhances the financial management capabilities and risk response capabilities of enterprises in a dynamic market environment.

[0069] In some embodiments of the present application, historical data of an enterprise is obtained, and the industry characteristics and business model of the enterprise are analyzed based on the historical data. The historical data includes sales records, cost details, and financial statements, specifically:

[0070] Obtain enterprise historical data from the enterprise ERP system, accounting software, and internal and external databases, and perform data cleaning and pre-processing on the enterprise historical data;

[0071] Identify market trends, seasonal changes, and macroeconomic conditions based on cleansed historical enterprise data.

[0072] In this step, historical data is first collected from the company's internal ERP system, accounting software, and any external databases. This data includes, but is not limited to, key business information such as sales records, cost details, and financial statements. The acquired data often contains a significant amount of noise and inconsistencies, necessitating a series of data cleaning and preprocessing operations. This includes, but is not limited to, removing duplicate records, filling in missing values, correcting erroneous data, and standardizing the data format for subsequent analysis. This series of data preparation ensures that subsequent analysis is based on an accurate and consistent dataset.

[0073] Based on the cleaned and preprocessed data, further analysis is performed to identify market trends, seasonal changes, and the specific impact of the macroeconomic environment on the company. This part of the analysis usually involves time series decomposition techniques to distinguish long-term trends, seasonal fluctuations, and random variation components in the data. For example, statistical models or machine learning algorithms can be used to identify seasonal patterns in sales records and the reasons behind them, such as the impact of holiday promotions; at the same time, combined with external macroeconomic indicators (such as GDP growth rate, inflation rate, etc.), the specific impact of macroeconomic conditions on corporate operations can be evaluated. This kind of analysis not only helps companies better understand their market position and industry characteristics, but also provides a solid foundation for building more accurate budget forecasting models. In this way, companies can formulate budget plans more accurately and adjust their strategic direction accordingly to adapt to the ever-changing market environment.

[0074] In some embodiments of the present application, historical enterprise data is obtained and analyzed based on the historical enterprise data to determine the industry characteristics and business model of the enterprise. The historical data includes sales records, cost details, financial statements, and also includes:

[0075] Extract trend items, cycle items and residual items of enterprise historical data based on time series decomposition method;

[0076] Use cluster analysis or classification algorithms to segment and categorize the industry in which the enterprise operates, and identify industry indicators related to the enterprise's operations;

[0077] Combined with external macroeconomic data interfaces, it dynamically obtains GDP growth rate, inflation rate and industry policy changes.

[0078] On the basis of obtaining the historical data of the enterprise and analyzing the industry characteristics and business model of the enterprise based on it, this application further introduces a deeper data analysis method. First, the time series decomposition method is used to process the cleaned historical data of the enterprise to extract trend items, cycle items and residual items. The trend item reflects the long-term development direction of the enterprise, the cycle item reveals the regular fluctuation characteristics in the data (such as seasonal sales peaks), and the residual item represents the random disturbance factor. Through this process, the development trajectory and operating laws of the enterprise at different stages can be more clearly identified.

[0079] Based on this, the system uses cluster analysis or classification algorithms to segment and categorize companies' industries, identifying industry indicators highly relevant to their business models. For example, the retail industry is segmented into categories such as fast-moving consumer goods and durable goods, and corresponding key indicators such as inventory turnover and customer repurchase rates are matched. Simultaneously, it accesses external macroeconomic data interfaces to dynamically capture macroeconomic variables such as GDP growth rates, inflation rates, and changes in industry policies, to assess the impact of the macroeconomic environment on corporate budgetary objectives. This information serves as a key input parameter for the budget forecasting model, ensuring the accuracy and adaptability of subsequent forecasts.

[0080] This step significantly enhances the depth of historical data mining and industry positioning accuracy by incorporating advanced technologies such as time series decomposition, cluster analysis, and external macroeconomic data interfaces. Compared to traditional approaches that rely on empirical judgment, this method achieves data-driven automated industry identification and feature extraction, effectively avoiding human bias and enhancing the intelligence of the budget forecasting system.

[0081] Furthermore, by incorporating external macroeconomic factors, budget forecasts become more forward-looking and dynamically adaptable, enabling timely responses to policy changes and economic fluctuations. This multi-dimensional, multi-layered data analysis mechanism not only enhances the scientific nature and flexibility of budget compilation but also provides strong support for businesses to make accurate financial decisions in complex market environments, demonstrating promising practical application value and widespread adoption.

[0082] In some embodiments of the present application, a budget forecasting model is determined based on industry characteristics and business models, and the parameters of the budget forecasting model are adjusted through a cross-validation method, specifically:

[0083] Budget forecasting models include time series models, regression models, and machine learning models;

[0084] The budget forecasting model is trained and validated using the K-fold cross-validation method to evaluate the model performance indicators under different parameter combinations. The performance indicators include mean square error, mean absolute percentage error, and coefficient of determination.

[0085] Based on the company's industry characteristics and business model identification, the system automatically matches the appropriate budget forecasting model type, including multiple mainstream forecasting algorithms such as time series models, regression models, and machine learning models. Time series models (such as ARIMA and exponential smoothing) are suitable for data with obvious trends or cyclical characteristics; regression models (such as linear regression and ridge regression) are used to analyze the quantitative relationship between multiple variables; and machine learning models (such as random forests and gradient boosting trees (GBDT)) can process high-dimensional nonlinear data and meet the budget forecasting needs of complex business scenarios.

[0086] Furthermore, the system uses the K-fold cross-validation method to train and optimize the parameters of the selected model. Specifically, the historical data is divided into K subsets, one of which is selected as the validation set, and the remaining K-1 subsets are selected as the training set. After repeating K times, the average performance indicators are calculated, including the mean square error (MSE), mean absolute percentage error (MAPE), and determination coefficient (R 2 By comparing the performance under different parameter combinations, the optimal parameter configuration is screened out to improve the prediction accuracy and generalization ability of the model.

[0087] This step significantly enhances the intelligence and adaptability of the budget forecasting system by introducing multiple budget forecasting models and a cross-validation mechanism. Compared to the traditional approach of using a single model with fixed parameters, this method automatically selects the optimal forecasting model based on the company's industry and data characteristics, and dynamically optimizes its parameters through a data-driven approach, ensuring high forecast accuracy across diverse business scenarios.

[0088] Furthermore, the application of K-fold cross-validation enhances the stability and reliability of model evaluation, effectively avoiding overfitting or underfitting issues caused by improper data partitioning, and improving the robustness of the model in practical applications. This multi-model adaptive selection and automatic parameter optimization mechanism not only improves the efficiency and quality of budget preparation, but also provides enterprises with more scientifically based financial decision-making support, with promising promotional value and commercial application prospects.

[0089] In some embodiments of the present application, a dynamic budget formula is configured in the budget forecasting model based on business needs and data structure. The dynamic budget formula includes a formula based on dimensions, table templates, Excel functions, business preset functions, and external interfaces, specifically:

[0090] A graphical formula configuration interface is provided in the budget preparation system to support users in defining personalized calculation logic based on different budget dimensions;

[0091] Set up a business preset function library, which includes a growth rate calculation function, a year-on-year and month-on-month analysis function, a rolling forecast function, and a budget control threshold judgment function.

[0092] After the budget forecasting model is built, the system further provides a flexible formula configuration mechanism to meet the budget preparation needs of different business scenarios. Specifically, a graphical formula configuration interface is integrated into the budget preparation system. Users can define personalized calculation logic based on different budget dimensions (such as department, product line, region, time period, etc.) through interactive methods such as dragging and dropping, and selecting from drop-down menus. For example, financial personnel can set up revenue forecast formulas based on sales regions, or configure corresponding expense control rules based on different cost centers, thereby achieving multi-dimensional and refined budget management.

[0093] In addition, the system includes a comprehensive library of pre-built business functions, covering commonly used budget calculation functions, including but not limited to growth rate calculation functions, year-on-year and month-on-month analysis functions, rolling forecast functions, and budget control threshold judgment functions. These functions can be directly embedded into budget formulas, enabling rapid modeling of complex business logic. For example, by calling the year-on-year growth function, the system can automatically calculate the growth rate of a product line in the current month relative to the same period the previous year. By setting the budget control threshold judgment function, the system can provide real-time warnings of overspending risks during budget execution, enhancing budget control capabilities.

[0094] This step significantly improves the flexibility and usability of the budget system by introducing a graphical formula configuration interface and a rich library of pre-built business functions. Compared to traditional budget systems, where formulas are rigid and difficult to modify, this method enables dynamic configuration of budget logic, making it easier for non-technical personnel to participate in the budget modeling process, lowering the barrier to entry for system users and improving budget compilation efficiency.

[0095] Furthermore, the built-in business pre-built function library not only covers the core calculation logic of financial management but also offers excellent scalability, supporting enterprises to conduct secondary development or function expansion based on their own business characteristics. This flexible and open formula system provides a solid foundation for enterprises to build a unified, standardized, and configurable budget management platform, helping to improve the accuracy and responsiveness of budget forecasts and enhance enterprises' financial adaptability in complex market environments.

[0096] In some embodiments of the present application, a dynamic budget formula is configured in the budget forecasting model based on business needs and data structure. The dynamic budget formula includes a formula based on dimensions, table templates, Excel functions, business preset functions, and external interfaces, and also includes:

[0097] Call calculation functions or real-time data sources of external systems through API interfaces to achieve cross-system data linkage and formula collaborative calculation.

[0098] When configuring dynamic budget formulas within budget forecasting models, the system not only supports flexible definition based on dimensions, table templates, Excel functions, and pre-defined business functions, but also provides open interfaces to external systems. Specifically, through the integration of standard APIs, the system can call calculation functions or real-time data sources from external systems, such as inventory data from ERP systems, customer order information from CRM systems, and industry benchmarks from external market data platforms.

[0099] This cross-system linkage mechanism allows budget formulas to dynamically access external data during execution and incorporate it into the budget calculation logic. For example, when forecasting sales expenses for a specific region, the system can automatically obtain the latest logistics cost index for that region through an API; or when performing cash flow forecasts, it can use real-time interest rate data from the bank system as an input parameter. Furthermore, the system supports pushing the output of internal budget models back to other business systems, enabling collaborative budget calculation and data sharing across multiple systems.

[0100] This step significantly enhances the budget forecasting system's scalability and integration capabilities by introducing an API interface to enable data interoperability with external systems. Compared to traditional closed budget systems, this approach breaks down data silos, enabling the budget preparation process to integrate real-time information from multiple systems, improving the comprehensiveness and accuracy of forecast results.

[0101] At the same time, this design supports enterprises in building a unified data management platform, achieving high-level coordination between budget logic and core business processes. This not only enhances the intelligence of budget systems but also provides strong support for enterprises in advancing digital transformation and achieving integrated business and financial management. This technical approach is highly versatile and scalable, applicable to budget management needs across a wide range of industry scenarios, and possesses significant industrial application value.

[0102] In some embodiments of the present application, a dynamic budget formula is used to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results, specifically:

[0103] Identify trend changes and cyclical patterns in the company's historical data through time series analysis, adjust the forecast value based on seasonal factors, and generate the first-class budget forecast results;

[0104] Evaluate the impact of multiple business variables on budget targets based on regression analysis models, construct multivariate linear or nonlinear regression equations, and output second-category budget forecast results;

[0105] Set multiple business scenario parameters to simulate the business operation status under different market environments, perform multi-variable combination operations through dynamic budget formulas, and generate third-category budget forecast results. The business scenario parameters include optimistic scenario, neutral scenario, and pessimistic scenario.

[0106] Compare and analyze the budget forecast results of the first category, the second category and the third category to determine the final budget forecast results.

[0107] After configuring the dynamic budget formula, the system conducts multi-dimensional budget forecast analysis based on this formula system, generating multiple sets of budget forecast results to meet decision-making needs in different business scenarios. First, the system models the company's historical data using time series analysis to identify trends and cyclical patterns. It then dynamically adjusts the forecast values ​​based on seasonal factors to generate the first type of budget forecast results. For example, in the retail industry, this can identify cyclical sales peaks caused by holidays and promotional seasons, and use this information to optimize the revenue forecast model.

[0108] Next, the system constructs a regression analysis model to assess the impact of multiple business variables (such as market investment, customer base, and product price) on budget targets (such as sales revenue and cost expenditure). Based on the correlations between variables, a multivariate linear or nonlinear regression equation is established, and a second-category budget forecast is output. This process not only considers the impact of individual factors but also integrates the interactions between variables, enhancing the scientific nature and accuracy of the forecast model.

[0109] Furthermore, the system sets parameters for various business scenarios, including optimistic, neutral, and pessimistic scenarios, to simulate a company's operations under different market conditions. For example, the optimistic scenario assumes strong market demand and stable raw material prices, while the pessimistic scenario considers factors such as economic downturn and tightening policies. By combining these variables through dynamic budgeting formulas, the system can generate a third-category budget forecast, providing comprehensive risk assessment and response strategy support for companies.

[0110] Finally, the system compares and analyzes the above three types of budget forecast results, combines the forecast error indicators (such as mean square error MSE, mean absolute percentage error MAPE) and the company's current strategic orientation, and selects the optimal forecast plan as the final budget forecast result, providing management with a scientific and flexible decision-making basis.

[0111] This step integrates multiple forecasting methods, including time series analysis, regression analysis, and multi-scenario simulation, to achieve a diversified output of budget forecast results and an intelligent optimization mechanism. Compared to traditional approaches that rely on a single model or empirical judgment, this method significantly improves the accuracy and adaptability of budget forecasts, meeting the financial planning needs of enterprises at different operating stages and market environments.

[0112] Furthermore, by introducing a multi-scenario simulation mechanism, the system enhances risk warning and response capabilities, helping companies develop pre-emptive plans in an increasingly uncertain market and improving the foresight and robustness of financial management. Ultimately, by comparing and optimizing multi-model results, it assists companies in selecting the most appropriate budget plan, improving overall budgeting efficiency and decision-making quality, demonstrating promising application prospects and potential for widespread adoption.

[0113] like Figure 2 As shown, the second embodiment of the present application provides a device for implementing a budget dynamic prediction technology based on a flexible formula and memory calculation, including:

[0114] The data acquisition module 110 is adapted to acquire the enterprise's historical data and analyze the enterprise's industry characteristics and business model based on the enterprise's historical data. The historical data includes sales records, cost details, and financial statements.

[0115] The model determination module 120 is adapted to determine a budget forecast model based on industry characteristics and business models, and adjust parameters of the budget forecast model through a cross-validation method;

[0116] Configuration module 130, adapted to configure dynamic budget formulas in the budget forecasting model based on business requirements and data structures, the dynamic budget formulas including formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces;

[0117] The result generation module 140 is adapted to perform time series analysis, regression analysis, and scenario analysis using the dynamic budget formula to generate multiple sets of budget forecast results.

[0118] The device for implementing dynamic budget prediction technology based on flexible formulas and memory calculations provided in the second aspect embodiment of this application can implement the method for implementing dynamic budget prediction technology based on flexible formulas and memory calculations in any embodiment of the first aspect above, and thus can achieve any technical effect of the method for implementing dynamic budget prediction technology based on flexible formulas and memory calculations above, which will not be repeated here.

[0119] An embodiment of the third aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for implementing budget dynamic prediction technology based on flexible formulas and memory calculations in any embodiment of the first aspect above is implemented.

[0120] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the method for implementing a dynamic budget prediction technology based on a flexible formula and memory calculation in any embodiment of the first aspect above, the method comprising:

[0121] Step 100: Obtain the company's historical data and analyze the company's industry characteristics and business model based on the company's historical data. The historical data includes sales records, cost details, and financial statements.

[0122] Step 200: Determine a budget forecast model based on industry characteristics and business models, and adjust the parameters of the budget forecast model through a cross-validation method.

[0123] Step 300: Configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces.

[0124] Step 400: Use the dynamic budget formula to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

[0125] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0126] In another aspect, the present invention further provides a computer program product, comprising a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for implementing a dynamic budget prediction technology based on a flexible formula and memory calculation provided by the above methods, the method comprising:

[0127] Step 100: Obtain the company's historical data and analyze the company's industry characteristics and business model based on the company's historical data. The historical data includes sales records, cost details, and financial statements.

[0128] Step 200: Determine a budget forecast model based on industry characteristics and business models, and adjust the parameters of the budget forecast model through a cross-validation method.

[0129] Step 300: Configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces.

[0130] Step 400: Use the dynamic budget formula to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

[0131] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for implementing the budget dynamic prediction technology based on flexible formulas and memory calculations provided by the above methods is implemented. The method includes:

[0132] Step 100: Obtain the company's historical data and analyze the company's industry characteristics and business model based on the company's historical data. The historical data includes sales records, cost details, and financial statements.

[0133] Step 200: Determine a budget forecast model based on industry characteristics and business models, and adjust the parameters of the budget forecast model through a cross-validation method.

[0134] Step 300: Configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces.

[0135] Step 400: Use the dynamic budget formula to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

[0136] Finally, the present invention also provides a non-volatile computer storage medium having computer executable instructions stored thereon. When the computer program is executed by a processor, the method for implementing the budget dynamic prediction technology based on flexible formulas and memory calculation provided by the above methods is implemented. The method includes:

[0137] Step 100: Obtain the company's historical data and analyze the company's industry characteristics and business model based on the company's historical data. The historical data includes sales records, cost details, and financial statements.

[0138] Step 200: Determine a budget forecast model based on industry characteristics and business models, and adjust the parameters of the budget forecast model through a cross-validation method.

[0139] Step 300: Configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure. The dynamic budget formula includes formulas based on dimensions, table templates, Excel functions, business preset functions, and external interfaces.

[0140] Step 400: Use the dynamic budget formula to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results. Anything not described in this application can be achieved by adopting or drawing on existing technologies.

[0141] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0142] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included in the protection scope of the present application.

Claims

1. A method for implementing budget dynamic forecasting technology based on flexible formulas and memory calculations, characterized in that: include: Obtaining historical enterprise data, and analyzing the industry characteristics and business model of the enterprise based on the historical enterprise data, wherein the historical data includes sales records, cost details, and financial statements; Determining a budget forecasting model based on the industry characteristics and the business model, and adjusting the parameters of the budget forecasting model through a cross-validation method; Configuring a dynamic budget formula in the budget forecasting model based on business requirements and data structure, wherein the dynamic budget formula includes a formula based on dimensions, table samples, Excel functions, business preset functions, and external interfaces; The dynamic budget formula is used to conduct time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results.

2. The method for implementing budget dynamic forecasting technology based on flexible formula and memory calculation according to claim 1, characterized in that: The acquisition of enterprise historical data, and analysis of the industry characteristics and business model of the enterprise based on the enterprise historical data, wherein the historical data includes sales records, cost details, and financial statements, specifically: Obtain the enterprise's historical data from the enterprise's ERP system, accounting software, and internal and external databases, and perform data cleaning and preprocessing on the enterprise's historical data; Identify market trends, seasonal changes, and macroeconomic conditions based on cleansed historical enterprise data.

3. The method for implementing budget dynamic forecasting technology based on flexible formula and memory calculation according to claim 2, characterized in that: The acquisition of enterprise historical data and analysis of the industry characteristics and business model of the enterprise based on the enterprise historical data, wherein the historical data includes sales records, cost details, financial statements, and also includes: Extracting trend items, cycle items and residual items of the enterprise historical data based on a time series decomposition method; Use cluster analysis or classification algorithms to segment and categorize the industry in which the enterprise operates, and identify industry indicators related to the enterprise's operations; Combined with external macroeconomic data interfaces, it dynamically obtains GDP growth rate, inflation rate and industry policy changes.

4. The method for implementing budget dynamic forecasting technology based on flexible formula and memory calculation according to claim 1, characterized in that: The budget forecasting model is determined based on the industry characteristics and the business model, and the parameters of the budget forecasting model are adjusted through a cross-validation method, specifically: The budget forecasting model includes a time series model, a regression model and a machine learning model; The budget forecasting model is trained and validated using the K-fold cross-validation method to evaluate the model performance indicators under different parameter combinations, including mean square error, mean absolute percentage error, and coefficient of determination.

5. The method for implementing dynamic budget forecasting technology based on flexible formulas and memory calculation according to claim 1, characterized in that: The dynamic budget formula is configured in the budget forecast model based on business requirements and data structure. The dynamic budget formula includes a formula based on dimensions, table samples, Excel functions, business preset functions, and external interfaces, specifically: A graphical formula configuration interface is provided in the budget preparation system to support users in defining personalized calculation logic based on different budget dimensions; A business preset function library is set, which includes a growth rate calculation function, a year-on-year and month-on-month analysis function, a rolling forecast function, and a budget control threshold judgment function.

6. The method for implementing dynamic budget forecasting technology based on flexible formulas and memory calculation according to claim 5, characterized in that: The dynamic budget formula is configured in the budget forecast model based on business requirements and data structure. The dynamic budget formula includes a formula based on dimensions, table samples, Excel functions, business preset functions, and external interfaces, and further includes: Call calculation functions or real-time data sources of external systems through API interfaces to achieve cross-system data linkage and formula collaborative calculation.

7. The method for implementing dynamic budget forecasting technology based on flexible formulas and memory calculation according to claim 1, characterized in that: The dynamic budget formula is used to perform time series analysis, regression analysis, and scenario analysis to generate multiple sets of budget forecast results, specifically: Identify trend changes and cyclical patterns in the company's historical data through time series analysis, adjust the forecast value based on seasonal factors, and generate the first-class budget forecast results; Evaluate the impact of multiple business variables on budget targets based on regression analysis models, construct multivariate linear or nonlinear regression equations, and output second-category budget forecast results; Setting multiple business scenario parameters to simulate the business operation status under different market environments, and performing multi-variable combination operations through dynamic budget formulas to generate third-category budget forecast results, where the business scenario parameters include optimistic scenario, neutral scenario, and pessimistic scenario; The first category budget forecast results, the second category budget forecast results and the third category budget forecast results are compared and analyzed to determine the final budget forecast results.

8. A device for implementing budget dynamic prediction technology based on flexible formulas and memory calculations, characterized in that: include: A data acquisition module, adapted to acquire historical enterprise data and analyze the industry characteristics and business model of the enterprise based on the historical enterprise data, wherein the historical data includes sales records, cost details, and financial statements; a model determination module adapted to determine a budget forecast model based on the industry characteristics and the business model, and to adjust parameters of the budget forecast model through a cross-validation method; a configuration module adapted to configure a dynamic budget formula in the budget forecasting model based on business requirements and data structure, wherein the dynamic budget formula includes a formula based on dimensions, table templates, Excel functions, business preset functions, and external interfaces; The result generation module is adapted to perform time series analysis, regression analysis and scenario analysis using the dynamic budget formula to generate multiple sets of budget forecast results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for implementing dynamic budget prediction technology based on flexible formulas and memory calculations as described in any one of claims 1 to 7 is implemented.

10. A non-volatile computer storage medium having computer executable instructions stored thereon, characterized in that: When the computer program is executed by a processor, the method for implementing a dynamic budget prediction technology based on flexible formulas and memory calculations as claimed in any one of claims 1 to 7 is implemented.

Images