Computational budgeting management prediction method and system based on artificial intelligence and big data

Abstract

The application provides a budget preparation management prediction method and system based on artificial intelligence and big data, which comprises the following steps: reconstructing the task scheduling control parameters of each responsibility center in the current preparation stage based on the budget task priority sequence determined based on the state, period attribute and historical execution data of the budget task in the current budget period, to obtain an adaptive task starting time sequence; screening the to-be-selected budget subjects based on the dynamic budget constraint boundary determined based on the adaptive task starting time sequence and the control mode and locking state in the preset budget control scheme, to obtain an effective budget subject range; constructing a multi-dimensional budget topology structure based on the effective budget subject range and the data classification dimension in the budget preparation data, and performing topology correction based on the data flow path of each responsibility center in the multi-dimensional budget topology structure and the budget execution actual data, to obtain a current topology structure. The application enhances the guidance of budget prediction and the effectiveness of budget control.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a computational budgeting, management, and forecasting method and system based on artificial intelligence and big data. Background Technology

[0002] As corporate organizational structures continue to expand, subordinate responsibility centers exhibit characteristics such as numerous levels, wide coverage, and significant differences in business models. This leads to an increasing variety of subject dimensions involved in budget preparation, more diverse cycle types, increasingly complex control rules, and longer data links. Consequently, companies have increasingly higher requirements for the accuracy of budget preparation and forecasting, the adaptability of task scheduling, and the flexibility of dynamic control.

[0003] Existing budget preparation and forecasting methods primarily operate through fixed budget management information systems. This involves first calculating a basic forecast value for the current budget using a pre-set algorithm, then making simple adjustments to the basic forecast value based on a fixed budget display format and pre-defined budget allocation rules, ultimately outputting the budget forecast result and budget allocation plan. However, existing methods rely entirely on fixed pre-set rules and static parameter systems, making it difficult to flexibly adapt to the differentiated business characteristics of various responsibility centers or cope with dynamically changing budget execution scenarios. This results in significant discrepancies between the budget preparation and forecast results and the company's actual operating conditions and funding needs, weakening the guiding significance of budget forecasts and the actual effectiveness of budget control, and failing to meet the company's needs for refined and dynamic budget management. Summary of the Invention

[0004] This invention provides a computational budget preparation, management, and forecasting method and system based on artificial intelligence and big data. It addresses the technical problems of existing methods that rely on fixed rules and static parameters, cannot adapt to differentiated business and dynamic execution scenarios, and have forecast results that are out of sync with actual needs. It improves the accuracy of budget preparation and forecasting, the adaptability of task scheduling, and the dynamic control capabilities, enhances the guidance of budget forecasting and the effectiveness of budget control, and meets the needs of enterprises for refined budget management.

[0005] In a first aspect, the present invention provides a computational budgeting and management forecasting method based on artificial intelligence and big data, comprising: Based on the status, cycle attributes, and historical execution data of budget tasks within the current budget cycle, the priority sequence of budget tasks is determined, and the task scheduling control parameters of each responsibility center in the current compilation stage are reconstructed to obtain an adaptive task start sequence. Based on the adaptive task startup sequence, combined with the control methods and locking states in the preset budget control scheme, the dynamic budget constraint boundaries determined by the selected budget items are screened to obtain the range of effective budget items. Based on the scope of effective budget items and the data classification dimensions in the budget preparation data, a multi-dimensional budget topology is constructed. Then, based on the data flow paths of each responsibility center in the multi-dimensional budget topology and the actual budget execution data, the topology is corrected to obtain the current topology. Based on the current topology and the display types and customization types in the preset budget display scheme, the prediction scenario mode is calculated to obtain the target budget allocation plan. Based on the target budget allocation plan and the budget release rules and cost classification, the target budget prediction result is determined.

[0006] Secondly, the present invention also provides a computational budget preparation, management, and forecasting system based on artificial intelligence and big data, applied to the computational budget preparation, management, and forecasting method based on artificial intelligence and big data as described in the first aspect; the computational budget preparation, management, and forecasting system based on artificial intelligence and big data includes: The timing reconstruction module is used to reconstruct the task scheduling control parameters of each responsibility center in the current compilation stage based on the status, cycle attributes and historical execution data of budget tasks within the current budget cycle, so as to obtain an adaptive task start timing sequence. The budget item filtering module is used to filter the budget items to be selected based on the adaptive task start sequence and the control method and locking state determined in the preset budget control scheme, so as to obtain the range of valid budget items. The topology construction and correction module is used to construct a multi-dimensional budget topology based on the scope of effective budget items and the data classification dimensions in the budget preparation data, and to correct the topology based on the data flow paths of each responsibility center in the multi-dimensional budget topology and the actual budget execution data to obtain the current topology. The budget forecast result determination module is used to calculate the forecast scenario mode based on the current topology and the display type and customized type in the preset budget display scheme, to obtain the target budget allocation plan, and to determine the target budget forecast result based on the target budget allocation plan and the budget release rules and cost classification.

[0007] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the above-described computational budgeting, management, and forecasting method based on artificial intelligence and big data.

[0008] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the above-described computational budgeting, management, and forecasting method based on artificial intelligence and big data.

[0009] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described computational budgeting, management, and forecasting method based on artificial intelligence and big data.

[0010] The present invention provides a computational budget preparation and management forecasting method based on artificial intelligence and big data. This method determines the priority sequence of budget tasks based on the budget task status, cycle attributes, and historical execution data within the current budget cycle of each responsibility center. It then reconstructs the task scheduling control parameters for the current preparation stage of each responsibility center to obtain an adaptive task startup sequence. This solves the problem of rigid task scheduling in existing methods, which cannot adapt to the differentiated business rhythms of each responsibility center. Based on this adaptive task startup sequence, and combined with the control methods and locking states in the preset budget control scheme, it determines the dynamic budget constraint boundaries and filters candidate budget items to obtain a range of effective budget items. This achieves precise selection of budget items and avoids the problems of rigid item selection and disconnect from actual business needs in existing methods. Based on the effective budget item range, and combined with the data classification dimensions of the budget preparation data, it constructs a multi-dimensional budget topology structure and... The topology is corrected based on the data flow paths and actual budget execution data of each responsibility center to obtain a current topology that fits the actual operation, solving the defects of existing methods such as messy data links and inability to accurately reflect the data correlation of each responsibility center. Based on the current topology, combined with the display types and customized type calculation and prediction scenario modes in the preset budget display scheme, a target budget allocation plan is obtained. Then, combined with budget release rules and cost classification, the target budget prediction result is determined, realizing dynamic adaptation of budget prediction and allocation. In summary, this invention effectively solves the technical problems of existing methods relying on fixed rules and static parameters, being unable to adapt to differentiated business and dynamic execution scenarios, and the disconnect between prediction results and actual needs. It improves the accuracy of budget preparation and prediction, the adaptability of task scheduling, and dynamic control capabilities, enhances the guidance of budget prediction and the effectiveness of budget control, and meets the needs of enterprises for refined budget management. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating the computational budgeting and management forecasting method based on artificial intelligence and big data provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the computational budgeting, management, and forecasting system based on artificial intelligence and big data provided in an embodiment of the present invention; Figure 3 An embodiment diagram of the electronic device provided in this invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0014] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0015] See Figure 1 , Figure 1 This is a flowchart illustrating the computational budgeting and management forecasting method based on artificial intelligence and big data provided by the present invention. In this embodiment of the invention, the executing entity of the computational budgeting and management forecasting method based on artificial intelligence and big data is the management forecasting system. Therefore, the computational budgeting and management forecasting method based on artificial intelligence and big data includes: Step 10: Based on the status, periodic attributes, and historical execution data of budget tasks within the current budget cycle, the task scheduling control parameters of each responsibility center in the current compilation stage are reconstructed to obtain an adaptive task start sequence.

[0016] Optionally, the management forecasting system acquires relevant basic data for all budget tasks within the current budget cycle. The status of a budget task refers to its current execution stage, its cycle attribute refers to specific characteristics of the current budget cycle, and historical execution data refers to the execution data of past budget tasks of the same type, responsibility center, and cycle type stored in the management forecasting system. Based on the status, cycle attribute, and historical execution data of budget tasks within the current budget cycle, a budget task priority sequence is calculated, as detailed in steps 101 to 105. Then, based on this budget task priority sequence, the task scheduling control parameters for each responsibility center in the current budgeting stage are reconstructed, ultimately yielding an adaptive task startup sequence, as detailed in steps 106 to 109. A responsibility center refers to an organizational unit within the enterprise responsible for budget preparation, execution, and control, encompassing four levels: group headquarters, second-level units, grassroots units, and departments / teams. The budget cycle refers to the time interval between budget preparation and execution, divided into annual, quarterly, monthly, and bi-weekly budget cycles.

[0017] Step 20: Based on the adaptive task startup sequence and the control method and locking status determined in the preset budget control scheme, the dynamic budget constraint boundary of the budget items to be selected is screened to obtain the range of effective budget items.

[0018] Optionally, the management forecasting system retrieves the adaptive task startup sequence and a preset budget control scheme, extracting the control methods and lock statuses from the budget control scheme. The budget control scheme refers to a pre-configured set of budget control rules, including core elements such as control methods, lock statuses, control periods, responsibility center scope, and budget item scope. Budget items refer to the expense, fund, and project classification entries for budget preparation and control, including daily expense items, special project items, and business item items. Control methods refer to the execution mode of budget control, categorized into five types: current control, cumulative control based on time progress, annual control, quarterly control, and monthly control. Lock status refers to the editable permission status of the budget control scheme, divided into unlocked and locked states. In the locked state, control rules cannot be modified; in the unlocked state, control rules can be adjusted. The adaptive task startup sequence is matched with the control methods and lock statuses of the budget control scheme. Using the control method's applicable period and the lock status permission requirements as core constraints, a dynamic budget constraint boundary is defined. This dynamic budget constraint boundary refers to the legal range of budget tasks, responsibility centers, and time periods allowed to participate in budget preparation and screening. Afterwards, the management forecasting system retrieves all candidate budget items within the system. These candidate budget items refer to the set of all budget preparation items preset by the enterprise. All candidate budget items are compared one by one with the dynamic budget constraint boundaries. Budget items that exceed the constraint boundaries or do not comply with the control rules are eliminated, and budget items that meet all constraint conditions are retained. Finally, the effective budget item range is obtained, which refers to the set of legal budget items that can participate in the preparation, forecasting, and control of the current budget cycle.

[0019] Step 30: Based on the scope of effective budget items and the data classification dimensions in the budget preparation data, construct a multi-dimensional budget topology structure, and then perform topology correction based on the data flow paths of each responsibility center in the multi-dimensional budget topology structure and the actual budget execution data to obtain the current topology structure.

[0020] Optionally, the management forecasting system constructs a multi-dimensional budget topology based on the effective budget item range and by retrieving data classification dimensions from the budget preparation data. This budget preparation data refers to all basic data related to the current budget preparation stored in the management forecasting system. This is illustrated in steps 301 to 305. The data classification dimensions refer to the classification statistical standards of the budget preparation data, including responsibility center dimensions, budget item dimensions, time period dimensions, cost classification dimensions, and budget attribute dimensions. Subsequently, the topology is corrected based on the data flow paths of each responsibility center in the multi-dimensional budget topology, combined with actual budget execution data, to obtain the current topology. This is illustrated in steps 306 to 309. The actual budget execution data refers to the actual budget-related data executed by each responsibility center within the current budget period, including actual revenue, costs, and expenses. This data is collected and updated in real time by the management forecasting system to ensure its authenticity and timeliness.

[0021] Step 40: Based on the current topology and the display type and customization type in the preset budget display scheme, calculate the predicted scenario mode to obtain the target budget allocation plan, and determine the target budget prediction result based on the target budget allocation plan and the budget release rules and cost classification.

[0022] Optionally, while retrieving the current topology, the management forecasting system reads the preset budget display scheme within the system and extracts two parameters from the budget display scheme: display type and customization type. The budget display scheme refers to the pre-configured set of budget data display rules, including core elements such as display type, customization type, query rules, and display method. The display type refers to the display dimension standard of budget data, including five categories: budget release number display, business execution number display, accounting execution number display, business and accounting execution number display, and budget execution detail query display. The customization type refers to the business adaptation type of the budget display, which is divided into two categories: general type and capital project type. The budget preparation management forecasting system uses the current topology as the data basis and the display type and customization type as the calculation dimensions to simulate the budget allocation logic under different business scenarios and obtain the optimal solution, which is the target budget allocation plan, as shown in steps 401 to 405.

[0023] Optionally, the management forecasting system retrieves preset budget release rules and cost classification parameters based on the target budget allocation plan. The budget release rules refer to the execution specifications for distributing budget data to lower-level responsibility centers after confirmation, including release conditions, release procedures, and release authority requirements. Cost classification refers to the cost attribute division of budget data, including four categories: regular budget, special project budget, business item budget, and daily expense budget. The budget allocation values ​​in the target budget allocation plan are matched for compliance according to the release conditions and procedures of the budget release rules, and the allocated values ​​are mapped to the four cost categories: regular budget, special project budget, business item budget, and daily expense budget. The matched budget data is then aggregated and calculated by adding the allocation values ​​according to responsibility center, budget item, and time period, forming three levels of aggregated data: group, second-level unit, and grassroots unit. Finally, the aggregated and calculated compliant budget data is combined with the budget release rules to form an executable and publishable budget forecast conclusion, ultimately yielding the target budget forecast result. This target budget forecast result includes four core components: budget preparation values, budget allocation ratio, execution control requirements, and cost classification attribution.

[0024] This invention, through its embodiments, determines the priority sequence of budget tasks based on the budget task status, cycle attributes, and historical execution data within the current budget cycle of each responsibility center. It then reconstructs the task scheduling control parameters for the current budgeting phase of each responsibility center to obtain an adaptive task startup sequence. This solves the problem of rigid task scheduling in existing methods, which cannot adapt to the differentiated business rhythms of each responsibility center. Based on this adaptive task startup sequence, and combined with the control methods and locking states in the preset budget control scheme, it determines the dynamic budget constraint boundaries, filters candidate budget items to obtain a range of effective budget items, and achieves precise selection of budget items. This avoids the problems of rigid item selection and disconnection from actual business needs in existing methods. Based on the effective budget item range, and combined with the data classification dimensions of the budget preparation data, it constructs a multi-dimensional budget topology structure, and based on the data flow paths of each responsibility center and... The actual budget execution data is used to correct the topology, resulting in a current topology that aligns with actual operations. This addresses the shortcomings of existing methods, such as cluttered data links and an inability to accurately reflect the data relationships between various responsibility centers. Based on the current topology, and combining the display types in the preset budget display scheme with customized type calculation and prediction scenario modes, a target budget allocation plan is obtained. Then, by combining budget release rules and cost classification, the target budget prediction result is determined, achieving dynamic adaptation between budget prediction and allocation. In summary, this invention effectively solves the technical problems of existing methods relying on fixed rules and static parameters, being unable to adapt to differentiated businesses and dynamic execution scenarios, and having prediction results disconnected from actual needs. It improves the accuracy of budget preparation and prediction, the adaptability of task scheduling, and dynamic control capabilities, enhancing the guidance of budget prediction and the effectiveness of budget control, thus meeting the needs of enterprises for refined budget management.

[0025] Optionally, the process of steps 101 to 105 includes: Step 101: Based on the real-time running status and task type of each budget task within the current budget cycle, feature extraction is performed to obtain the instantaneous status features of each budget task. Based on the instantaneous status features, all budget tasks are divided into multiple status category groups. The status category groups include blocked status, in progress status, and pending start status.

[0026] Optionally, the management forecasting system retrieves the real-time running status (i.e., the specific execution stage at the current moment) and task type information of all budget tasks within the current budget cycle. It then extracts real-time status features for each budget task, including four quantifiable features: task start identifier, execution progress, resource usage, and reason for blockage. This yields the real-time status features of each budget task. The task type refers to the category of tasks classified according to budget business attributes, including tasks for preparation, adjustment, reporting, review, and control. After extraction, the management forecasting system categorizes all budget tasks according to their real-time status features, assigning them to three status groups: blocked, in progress, and pending. The blocked group comprises budget tasks that cannot proceed due to insufficient resources or incomplete prerequisite tasks; the in progress group comprises budget tasks that have started and are executing normally; and the pending group comprises budget tasks that meet the start conditions but have not yet begun execution.

[0027] Step 102: Based on the periodic attributes of budget tasks in each state category group and the time axis scale of the current budget period, determine the hard deadline constraints and dependency window constraints for each budget task, and screen budget tasks with time conflict risks based on the hard deadline constraints and dependency window constraints to obtain the key budget task nodes that need to have their time windows adjusted first.

[0028] Optionally, the management forecasting system, based on the predefined state category groups, extracts periodic attribute information for each budget task within a group, and matches it with the timeline scale of the current budget period. Based on the periodic attributes and the timeline scale of the current budget period, it calculates and determines hard deadline constraints and dependency window constraints. The hard deadline constraint is the latest time node that the task must complete; the dependency window constraint is the time interval within which the preceding tasks on which the task depends must be completed. Then, the planned execution time of each budget task is compared with the above two constraints to determine if there is a risk of time conflict. Budget tasks with time conflict risks are selected to form a set of critical budget task nodes. These critical budget task nodes refer to budget tasks that have a decisive impact on the overall budgeting process progress and whose time windows must be adjusted first.

[0029] Step 103: Analyze the historical execution data of key budget task nodes and associated responsibility centers to obtain the actual time consumption deviation pattern and resource consumption fluctuation pattern of each key budget task node in the past cycle, and identify the task type labels with high-frequency delay characteristics and high-frequency resource competition characteristics based on the actual time consumption deviation pattern and resource consumption fluctuation pattern.

[0030] Optionally, the management forecasting system statistically analyzes historical execution data based on key budget task nodes and the corresponding responsibility centers, obtaining the actual time consumption deviation pattern and resource usage fluctuation pattern of each key budget task node in past cycles. The actual time consumption deviation pattern refers to the difference between the actual and planned time consumption of the key budget task node in consecutive budget cycles, specifically including the deviation direction (actual time greater than the preset time is a positive deviation, actual time less than the preset time is a negative deviation), deviation magnitude (the proportion of deviation amount to the preset time), and deviation frequency (the proportion of cycles with deviations to the total number of cycles). By statistically analyzing the deviation data of the past three cycles, the time consumption deviation of the key budget task node is extracted. The pattern of deviation (e.g., positive deviation in three consecutive periods, with a deviation range of 10%-15%); the resource usage fluctuation pattern refers to the pattern of resource usage changes during the execution of key budget task nodes in past consecutive budget periods, specifically including the direction of fluctuation (positive fluctuation when actual resource usage is greater than the preset amount, and negative fluctuation when actual resource usage is less than the preset amount), fluctuation amplitude (the proportion of fluctuation amount to the preset resource amount), fluctuation frequency (the proportion of the number of periods with fluctuation to the total number of periods), and fluctuation peak (the largest fluctuation amplitude in past periods). By statistically analyzing the resource usage data of the past three periods, the resource usage fluctuation pattern of the key budget task node is extracted (e.g., positive fluctuation in each period, with a fluctuation amplitude between 5%-8%, and a fluctuation peak of 10%).

[0031] Optionally, the management forecasting system identifies the execution characteristics of each key budget task node based on the obtained actual time consumption deviation pattern and resource usage fluctuation pattern. If the frequency of positive deviation (delay) of a key budget task node in the past three budget cycles is not less than 60% and the average deviation amplitude is not less than 8%, it is determined that the key budget task node has a high-frequency delay characteristic. If the frequency of positive fluctuation in resource usage (insufficient resources, need to compete for resources) of a key budget task node in the past three budget cycles is not less than 70% and the average fluctuation amplitude is not less than 5%, it is determined that the key budget task node has a high-frequency resource competition characteristic. The corresponding key budget task node is assigned a task type label, which includes a high-frequency delay label, a high-frequency resource competition label, and a risk-free label.

[0032] Step 104: Analyze the task priority sequence based on the task type labels and the urgency definition rules of each status category group.

[0033] Optionally, the management forecasting system calls the system's preset urgency definition rules based on the task type label. These urgency definition rules refer to the standards pre-defined by the enterprise to define the urgency of different status categories. These rules are based on the business impact scope and task urgency of the status category. For example, the urgency of the blocked status group is the highest because the budget tasks in this group are blocked. If not handled in time, related budget tasks will not be able to proceed, affecting the overall budget preparation process. The urgency of the in-progress status group is medium. The budget tasks in this group are progressing and need to be completed as planned to avoid delays. The urgency of the pending status group is the lowest. The budget tasks in this group have not yet started, and there is sufficient time for planning and adjustments. Based on the status category of each budget task, its basic urgency level is determined (blocked > in progress > pending). Then, for critical budget task nodes, priority is adjusted according to their task type tags. If a critical budget task node exhibits high-frequency delays and / or high-frequency resource contention, its priority is increased by one level from its basic urgency level. This is because such tasks have high execution risk and require priority resource scheduling and execution to avoid delays and resource conflicts. Non-critical budget task nodes do not undergo priority adjustment and maintain their priority corresponding to their basic urgency level. After comprehensive analysis, all budget tasks are sorted from highest to lowest priority to form an ordered list, thus obtaining the budget task priority sequence. This sequence clearly defines the execution order of each budget task, prioritizing budget tasks with high urgency and high execution risk.

[0034] Step 105: Based on the budget task priority sequence, reconstruct the task scheduling control parameters of each responsibility center in the current compilation stage to obtain the adaptive task start sequence.

[0035] Optionally, the management forecasting system reconstructs the task scheduling control parameters of each responsibility center at the current compilation stage based on the budget task priority sequence to obtain an adaptive task start sequence, as shown in steps 1051 to 1054, to ensure that high-priority budget tasks can obtain resources and start execution first, avoiding the situation where low-priority tasks occupy the resources of high-priority tasks, and improving the adaptability and efficiency of task scheduling.

[0036] The embodiments of this invention realize closed-loop management of the entire process of task status classification, risk identification, priority assessment, and scheduling parameter reconstruction. It solves the technical problems of rigid task scheduling, lack of scientific basis for priority assessment, and inability to adapt to the differentiated business rhythm and risk characteristics of each responsibility center in the existing methods. It makes the task scheduling control parameters more targeted and adaptable, and the adaptive task start sequence is more in line with the actual business needs of enterprises.

[0037] Optionally, the process of steps 1051 to 1054 includes: Step 1051: Based on the execution order and the responsibility center to which each budget task belongs in the priority sequence, perform task load distribution analysis, construct a task quantity distribution map and a task time density distribution map for each responsibility center in the current compilation stage, and identify the task load based on the task quantity distribution map and the task time density distribution map to obtain the first responsibility center group with overloaded task load and the second responsibility center group with idle task load.

[0038] Optionally, the management forecasting system, based on the budget task priority sequence, first analyzes and determines the execution order of each budget task, that is, the sorting order of each budget task in the budget task priority sequence. The higher the priority of the budget task, the earlier it is executed, that is, it is started and executed first. Then, it determines the responsibility center to which each budget task belongs, that is, the responsibility center that is responsible for the preparation and execution of the budget task. It conducts comprehensive statistics and analysis on the number of budget tasks that each responsibility center needs to undertake in the current preparation stage and the distribution of task execution time. To understand the task load of each responsibility center, identify load imbalance issues, and during the task load distribution analysis, construct a task quantity distribution map for each responsibility center at the current planning stage, with the responsibility center as the horizontal axis and the task quantity as the vertical axis. This task quantity distribution map can intuitively present the total number of budget tasks that each responsibility center needs to undertake at the current planning stage, as well as the distribution of the number of tasks of different priorities. At the same time, construct a task time density distribution map for each responsibility center at the current planning stage, with the time axis of the current planning stage as the horizontal axis and the task time density as the vertical axis. This task time density distribution map can present the total time spent on budget tasks that each responsibility center needs to execute at different time nodes (such as daily, weekly, and monthly) at the current planning stage. Here, task time density refers to the preset total time spent on all budget tasks that the responsibility center needs to execute within a unit of time (such as daily or weekly), with the unit being working days / day (or working days / week). Thus, this map can clearly reflect the task busyness of each responsibility center at different time periods and identify periods of concentrated tasks and tight time.

[0039] Optionally, the management forecasting system identifies task load based on task quantity distribution maps and task time density distribution maps, identifying a first responsibility center group with overloaded tasks and a second responsibility center group with idle tasks. Overloaded tasks refer to a responsibility center's task load exceeding its preset capacity. The identification criteria are: the number of tasks in the responsibility center exceeds the company's preset upper limit for the number of tasks per responsibility center, and the task time density exceeds a preset time density threshold for three consecutive time units (e.g., three days, three weeks). This preset time density threshold is determined by the company based on the personnel configuration, resource reserves, and business capabilities of each responsibility center, and is typically the maximum daily task time that each responsibility center can handle (e.g., four working days per day). Idle tasks refer to a responsibility center's task load being far below its preset capacity. The identification criteria are: the number of tasks in the responsibility center is less than the company's preset lower limit for the number of tasks per responsibility center, and the task time density is below 50% of the preset time density threshold for three consecutive time units.

[0040] Step 1052: Based on the global scheduling order of the first responsibility center group, the second responsibility center group, and each budget task, determine the task scheduling control parameters that are configured differently for each responsibility center, and replace the default parameters of each responsibility center based on the task scheduling control parameters to obtain the target scheduling control parameters of each responsibility center in the current preparation stage; the task scheduling control parameters include the upper limit parameter of the number of concurrent tasks, the threshold parameter of the task start interval, and the task resource exclusive flag parameter.

[0041] Optionally, the management forecasting system, based on the first and second responsibility center groups, invokes the budget task priority sequence and determines the global scheduling order, i.e., the execution order, of each budget task. Based on these three types of information, it determines differentiated task scheduling control parameters for each responsibility center. These task scheduling control parameters are core parameters used by the management forecasting system to control the initiation, concurrent execution, and resource consumption of budget tasks in each responsibility center. Specifically, they include the upper limit parameter for the number of concurrent tasks, the threshold parameter for the task initiation interval, and the parameter for the task resource exclusivity flag. The upper limit parameter for the number of concurrent tasks refers to the maximum number of budget tasks that a single responsibility center can initiate and execute simultaneously within the same time period. The threshold parameter for the task initiation interval refers to the minimum time interval (in working days) between two consecutively executed budget tasks within the same responsibility center. The parameter for the task resource exclusivity flag indicates whether a budget task requires the exclusive use of some core resources (such as computing resources or dedicated personnel) of the responsibility center. Specifically, the differentiated configuration process includes: For the first responsibility center group (overloaded), configure conservative task scheduling control parameters, i.e., reduce the upper limit of task concurrency, increase the task start interval threshold, and set the task resource exclusivity flag to "yes" only for first-priority tasks, while setting it to "no" for other tasks, in order to alleviate load pressure and avoid task conflicts and resource contention; For the second responsibility center group (idle load), configure aggressive task scheduling control parameters, i.e., increase the upper limit of task concurrency, decrease the task start interval threshold, and set the task resource exclusivity flag to "yes" for third-priority tasks and above, in order to make full use of idle resources and improve task execution efficiency; For ordinary responsibility centers that do not belong to either the first or second responsibility center group, configure conventional task scheduling control parameters, i.e., configure them according to the enterprise's preset default parameter standards to balance task execution efficiency and resource utilization.

[0042] Optionally, after the parameters are determined, the management and prediction system replaces the default parameters of each responsibility center based on the above-mentioned differentiated task scheduling control parameters. After the replacement, the target scheduling control parameters of each responsibility center in the current planning stage are obtained. These target scheduling control parameters refer to the final scheduling control parameters that are adapted to the load status of each responsibility center and can support the subsequent task start-up timing simulation. Each responsibility center corresponds to a unique set of target scheduling control parameters to ensure the relevance and adaptability of the scheduling.

[0043] Step 1053: Based on the target scheduling control parameters of each responsibility center and the order of execution, determine the theoretical earliest start time and theoretical latest start time of each budget task in each responsibility center, and perform task start sequence deduction based on the theoretical earliest start time and theoretical latest start time to obtain a draft task start sequence.

[0044] Optionally, the management forecasting system retrieves the target scheduling control parameters of each responsibility center. First, it determines the core parameters such as the upper limit of the number of concurrent tasks, the threshold of the task start interval, and the task resource exclusivity flag for each responsibility center. At the same time, it retrieves the execution order of each budget task and then determines the start and execution order of each budget task. By combining the target scheduling control parameters of each responsibility center with the execution order of each budget task, it determines the theoretical earliest start time and theoretical latest start time of each budget task within its respective responsibility center. The calculation of both time points is based on the time axis scale of the current compilation stage and uses working days as the unit to ensure the accuracy of time calculation.

[0045] The theoretical earliest start time refers to the earliest time when the budget task can be started and executed within its respective responsibility center. The specific calculation logic is as follows: If the budget task is the highest priority and first task to be executed within its respective responsibility center, then the theoretical earliest start time is the start time of the current preparation phase, or the completion time of the task's dependent tasks (if there is a dependency relationship); if the budget task is not the first task to be executed within its respective responsibility center, then the theoretical earliest start time is the theoretical earliest start time of the previous task plus the preset time consumption of the previous task, plus the task start interval threshold parameter in the target scheduling control parameters. If the previous task has set the task resource exclusive flag parameter to yes, then the resource release time (the resource release time is fixed at 1 working day) needs to be added to ensure that resources do not conflict.

[0046] The theoretical latest start time refers to the latest time that the budgeted task can be started within its assigned responsibility center to ensure completion before the latest completion time specified by its hard deadline constraints. The calculation logic is as follows: the hard deadline of the budgeted task minus the preset time of the task, and then minus the task start interval threshold parameter in the target scheduling control parameters (if there are other tasks in the same responsibility center after this task). If the task resource exclusive flag parameter is set to yes, the resource release time (1 working day) needs to be additionally subtracted.

[0047] Optionally, after the theoretical earliest and latest theoretical start times of the management forecasting system are determined, and without violating the target scheduling control parameters or exceeding the theoretical time range, the start times of each budget task within its respective responsibility center are reasonably arranged in combination with the execution order of each budget task, forming a preliminary task start sequence plan, i.e., a task start sequence draft. This task start sequence draft refers to the preliminary start time arrangement covering all responsibility centers and all budget tasks, clarifying the planned start time and planned completion time of each budget task, as well as the execution order of tasks within each responsibility center.

[0048] Step 1054: Based on the draft task startup sequence and the hard deadline constraints and dependency window constraints of each budget task in the time dimension, a comprehensive analysis is performed to obtain an adaptive task startup sequence covering all responsibility centers and each budget task.

[0049] Optionally, the management forecasting system retrieves the draft task initiation sequence, first determining the planned start time and planned completion time of each budgeted task, as well as the task execution order within each responsibility center. Simultaneously, it retrieves the hard deadline constraints and dependency window constraints of each budgeted task in the time dimension. Based on the above information, it corrects time conflicts and constraint violations in the draft task initiation sequence, specifically including: Verify whether the planned completion time of each budgeted task in the draft task start sequence meets the hard deadline constraints. If the planned completion time of a budgeted task exceeds its hard deadline, the planned start time of the task needs to be adjusted and started in advance. At the same time, the start time of subsequent related tasks within the same responsibility center should be adjusted to ensure that the task start interval threshold parameter and the task concurrency limit parameter are not violated. If the hard deadline constraints still cannot be met after adjustment, the target scheduling control parameters of the responsibility center should be re-optimized (such as further reducing the concurrency and shortening the task start interval) until the requirements are met. Verify whether the planned start time of each budget task in the draft task start sequence meets the dependency window constraints. If the planned start time of a budget task exceeds its dependency window time range, or is not started after its dependency tasks are completed, the planned start time of the task needs to be adjusted to ensure that it starts within the dependency window time range and after the dependency tasks are completed. At the same time, the start time of related tasks within the same responsibility center should be adjusted to avoid new time conflicts. Check whether the task load of each responsibility center is balanced in the draft task start sequence. If, after adjustment, a responsibility center is overloaded or idle again, fine-tune the task start time of that responsibility center or adjust the target scheduling control parameters to ensure that the task load of each responsibility center is within a reasonable range and improve the overall task scheduling efficiency. After comprehensive analysis, the draft task startup sequence was revised and optimized. The revised sequence must meet the following requirements: the planned completion time of all budgeted tasks does not exceed the hard deadline; the planned startup time complies with dependency window constraints; it does not violate the target scheduling control parameters of each responsibility center; the task load is balanced across responsibility centers; and there are no task conflicts or resource contention. After revision and optimization, an adaptive task startup sequence covering all responsibility centers and all budgeted tasks is obtained.

[0050] The embodiments of the present invention refine the process of generating adaptive task startup sequences, and solve the technical problems of lack of targeted task scheduling, unreasonable timing arrangement, and easy occurrence of time conflicts and resource contention in existing methods. This makes the generated adaptive task startup sequences more in line with the load status and business characteristics of each responsibility center, and can effectively avoid task conflicts, alleviate load pressure and improve resource utilization.

[0051] Based on the scope of effective budget items and the data classification dimensions in the budget preparation data, a multi-dimensional budget topology is constructed. Optionally, the processes of steps 301 to 305 include: Step 301: Based on the subject attribute identifiers of each valid budget subject within the scope of valid budget subjects and the data classification dimensions in the budget preparation data, determine the multi-dimensional attribution label groups of each valid budget subject in four dimensions: organizational level, business type, time period, and cost nature.

[0052] Optionally, the management forecasting system extracts the subject attribute identifier for each valid budget subject based on the scope of valid budget subjects. This subject attribute identifier is a unique marker used to distinguish the business attributes of valid budget subjects, including subject category, scope of use, statistical scope, and control level. At the same time, it matches the data classification dimensions in the budget preparation data and generates corresponding dimension attribution labels for each valid budget subject according to four dimensions: organizational level, business type, time period, and cost nature. The four dimension labels are combined into a multi-dimensional attribution label group for the subject, thereby achieving a unique marker for the multi-dimensional attribution of each valid budget subject.

[0053] Step 302: Based on the multidimensional attribution label group, the effective budget items are mapped to the corresponding multidimensional coordinate space nodes to obtain the initial positioning point of each effective budget item in the multidimensional space.

[0054] Optionally, the management forecasting system constructs a multi-dimensional coordinate space based on the multi-dimensional attribution label groups of each valid budget item, using organizational level, business type, time period, and cost nature as four axes. This multi-dimensional coordinate space refers to a four-dimensional three-dimensional coordinate space constructed using organizational level, business type, time period, and cost nature as dimensions, used for spatial positioning of valid budget items. Each axis contains a clear scale and value range. Then, each valid budget item is mapped to a corresponding node in this multi-dimensional coordinate space, that is, the attribution information of each dimension in the multi-dimensional attribution label group is converted into the scale value of its corresponding coordinate axis. The scale values ​​of the four dimensions combine to form a four-dimensional coordinate value. The multi-dimensional coordinate space node corresponding to this coordinate value is the initial positioning point of the valid budget item in the multi-dimensional space. Each valid budget item corresponds to a unique initial positioning point, and this initial positioning point not only contains coordinate value information but also associates the item attribute identifier and multi-dimensional attribution label group information of the valid budget item.

[0055] Step 303: Based on the initial positioning points of each valid budget item in the multi-dimensional space and the account correlation relationship in the preset budget preparation rule base, extract the explicit logical association edges and implicit data dependency edges connecting different initial positioning points to obtain the association edge group. Based on the association edge group, identify the combination of account pairs with direct data flow relationship and the combination of account pairs with indirect data derivation relationship to obtain the account association network skeleton.

[0056] Optionally, the management forecasting system retrieves a preset budget preparation rule base. This preset budget preparation rule base refers to a set of rules pre-formulated by the enterprise to standardize the budget preparation process, account accounting, and data association. The rule base includes core content such as account reconciliation relationships, accounting standards, and data flow rules. Among them, account reconciliation relationships refer to the inherent data association relationships between different valid budget accounts. Specifically, they are divided into explicit reconciliation relationships and implicit reconciliation relationships. Explicit reconciliation relationships refer to explicit data association relationships that are directly reflected in the budget preparation rules and can be directly identified. Implicit reconciliation relationships refer to indirect data association relationships that are not directly reflected in the budget preparation rules but can be deduced through business logic and historical data. Based on the initial positioning points of each valid budget item, and combined with the item correlation relationships in the preset budget preparation rule base, all item pairs are traversed to extract explicit logical association edges and implicit data dependency edges connecting different initial positioning points, forming association edge groups. Explicit logical association edges refer to the connection edges corresponding to explicit correlation relationships, used to connect two initial positioning points with a direct data flow relationship, with the direction of the edge pointing from the data source item to the data receiving item (e.g., raw material cost budget pointing to production cost budget). Implicit data dependency edges refer to the connection edges corresponding to implicit correlation relationships, used to connect two initial positioning points with an indirect data derivation relationship, with the direction of the edge pointing from the cause item in the causal relationship to the result item (e.g., sales revenue budget pointing to raw material cost budget).

[0057] Optionally, the management forecasting system, based on this set of related edges, identifies subject pairs with direct data flow relationships and subject pairs with indirect data derivation relationships: subject pairs with direct data flow relationships refer to combinations consisting of two valid budget subjects connected by explicit logical related edges; subject pairs with indirect data derivation relationships refer to combinations consisting of two valid budget subjects connected by implicit data dependency edges. These two types of combinations are non-overlapping and complete, fully covering the relationships between all valid budget subjects. After identification, the initial positioning points of all valid budget subjects are used as nodes, and all related edges are used as edges connecting the nodes, forming the subject relationship network skeleton.

[0058] Step 304: Based on the subject association network skeleton and the responsibility center affiliation relationship of each effective budget subject, determine the data flow direction indication set across different responsibility centers and the data interaction node set across departments, and splice the subject association network skeleton based on the data flow direction indication set and the data interaction node set to obtain the directed data flow path chain.

[0059] Optionally, the management forecasting system retrieves the responsibility center affiliation relationship of each valid budget item based on the nodes, associated edges, and relationships between items in the subject association network skeleton. This responsibility center affiliation relationship refers to the correspondence between each valid budget item and its respective responsibility center. Then, based on the subject association network skeleton and the responsibility center affiliation relationship of each valid budget item, the system analyzes the responsibility center affiliation of each pair of valid budget items connected by an associated edge: if the two valid budget items belong to the same responsibility center, the data flow corresponding to that associated edge is an intra-responsibility center data flow and does not require separate labeling; if the two valid budget items belong to different responsibility centers, the data flow corresponding to that associated edge is a cross-responsibility center data flow. For cross-responsibility center data flows, it is necessary to determine the data flow direction indicator set and the cross-departmental data interaction node set. The data flow direction indicator set refers to the collection containing all cross-responsibility center data flow direction information. Each data flow direction indicator clearly marks the source responsibility center, target responsibility center, data source subject, data target subject, and data flow direction of the data flow. The data interaction node set refers to the collection containing all cross-responsibility center data interaction nodes. Data interaction nodes refer to the nodes responsible for data transmission, verification, and aggregation in cross-responsibility center data flows, corresponding to the initial positioning point of specific valid budget subjects. Each cross-responsibility center data flow corresponds to two data interaction nodes.

[0060] Optionally, the management forecasting system, based on a defined set of data flow direction indicators and data interaction nodes, assembles the subject association network skeleton. This involves adding cross-responsibility center data flow direction indicators and data interaction nodes to the subject association network skeleton, clearly marking the data flow direction and interaction nodes of each cross-responsibility center association edge, while retaining the association edges and nodes within each responsibility center. This assembles the scattered subject association fragments, divided by responsibility center, into a complete network structure containing cross-responsibility center data flow, resulting in a directed data flow path chain. Each path chain consists of a series of nodes (initial positioning points of effective budget subjects) and directed edges (association edges), clearly presenting the complete path of data flowing from a subject in one responsibility center to a subject in another responsibility center, as well as the data flow path within each responsibility center.

[0061] Step 305: Based on the initial positioning points of each valid budget item in the multidimensional space as topology nodes and the directed data flow path chain as topology connection edges, construct a multidimensional budget topology structure containing node attribute information and edge direction information.

[0062] Optionally, the management forecasting system, based on the initial positioning points of each effective budget item in the multi-dimensional space, uses each initial positioning point as a topological node in the multi-dimensional budget topology structure according to its coordinates, associated item attribute identifiers, multi-dimensional affiliation label groups, and responsibility center affiliation relationships. Each topological node is the core unit in the multi-dimensional budget topology structure used to carry effective budget item information, and each topological node contains complete node attribute information. Simultaneously, the system retrieves directed data flow path chains. Based on the data flow paths and directions within and across all responsibility centers, each path in the directed data flow path chain is used as a topological connection edge in the multi-dimensional budget topology structure. Each topological connection edge is a connecting unit used to connect two topological nodes and reflect the data flow relationship between the nodes, and each topological connection edge contains complete edge direction information. Based on this, all topology nodes and topology connecting edges are integrated and constructed to form a multidimensional budget topology structure containing node attribute information and edge direction information. This multidimensional budget topology structure is a four-dimensional three-dimensional network structure that clearly presents the direct and indirect data association relationships (topology connecting edges) between each effective budget item (topology node), clarifies the direction of data flow, association type, and responsibility center flow status, and fully presents the core attributes and multidimensional affiliation of each effective budget item through node attribute information, realizing the visualization and structured presentation of effective budget item data association.

[0063] The embodiments of the present invention solve the technical problems of messy data links and inability to accurately reflect the data associations of each responsibility center in the existing methods. Moreover, the constructed multi-dimensional budget topology can clearly and structurally present the associations and data flow paths between each effective budget item, and clarify the data flow logic within and across responsibility centers.

[0064] Optionally, the process of steps 306 to 309 includes: Step 306: Logically align the data input ports and data output ports of each responsibility center based on the multi-dimensional budget topology to obtain a prototype of the statically constructed budget topology.

[0065] Optionally, the management forecasting system is based on a multi-dimensional budget topology. It traverses all responsibility centers and performs logical alignment operations on the data input and output ports of each responsibility center within the topology. This involves matching lower-level output ports with higher-level input ports and peer interaction ports according to the top-down organizational hierarchy and bottom-up data flow rules, eliminating port misalignment, missing, and duplicate issues. This results in a prototype budget topology that is structurally complete, has smooth paths, and only possesses static relationships. Here, a data input port refers to the logical access point in the multi-dimensional budget topology where a responsibility center receives budget data from higher-level / other departments; a data output port refers to the logical output point in the multi-dimensional budget topology where a responsibility center sends budget data to lower-level / other departments.

[0066] Step 307: Based on the prototype of the budget topology and the real-time data status of each valid budget item, perform status marking to obtain the candidate group of correction nodes and the candidate group of connection edges to be corrected in the topology.

[0067] Optionally, the management forecasting system retrieves the budget topology prototype and simultaneously obtains the real-time data status of each valid budget item in the current period. It then compares the real-time data status with the static nodes and connecting edges in the topology prototype item by item: topology nodes with inconsistent, missing, or expired data status are marked as objects to be corrected and integrated into a candidate group for correction; connecting edges with data flow directions or dependencies inconsistent with actual business are marked as objects to be corrected and integrated into a candidate group for connecting edges to be corrected. Here, real-time data status refers to the actual business status of budget preparation, reporting, execution, and adjustment corresponding to the valid budget items within the current period.

[0068] Step 308: Based on the candidate group of corrected nodes, the candidate group of connection edges to be corrected, and the actual execution data, analyze the data to obtain the group of outdated node instances and the group of failed connection edges.

[0069] Optionally, the management forecasting system retrieves actual budget execution data and uses it as the verification standard. It then verifies the budget item attributes and planned execution status of each candidate node against the actual business status in the actual budget execution data (e.g., whether actual expenditures have occurred, whether the node has been deactivated, or whether it has been adjusted to another item). Topology nodes that are indeed deactivated, have no actual data, or are no longer used are included in the outdated node instance group. Simultaneously, for the candidate group of connection edges to be corrected, the system verifies the data flow direction and associated item relationships corresponding to each candidate connection edge against the actual cross-item and cross-responsibility center data flow relationships in the actual budget execution data. Connection edges that are completely disconnected from actual business and no longer have data transmission are classified into the failed connection edge instance group.

[0070] Step 309: Based on the outdated node instance group and the failed connection edge instance group, perform node replacement and edge redirection on the budget topology prototype to obtain the current topology.

[0071] Optionally, the management forecasting system performs two correction operations based on the budget topology prototype. The first operation is node replacement, which involves matching all failed nodes in the outdated node instance group with the corresponding valid budget items in the current budget execution data, generating new topology nodes that conform to the actual business, and replacing the outdated nodes with the new nodes to ensure that the replaced nodes are consistent with the real business. The second operation is edge redirection, which involves performing an edge cleanup operation on all failed edges in the failed connection edge instance group, deleting such connection edges that do not have actual data flow support from the topology. At the same time, for the remaining valid connection edges (edges in the original candidate group of connection edges to be corrected but not included in the failed connection edge instance group), the data flow direction and correlation are reconfirmed according to the actual budget execution data, and the edge redirection is completed. After the two operations are completed, a current topology structure that is accurate in structure, valid in nodes, realistic in edge relationships, and closely matches the actual business situation is formed.

[0072] The current topology obtained by this invention completely breaks away from the limitations of static rules, and truly reflects the real-time, accurate and dynamic data flow relationships and business connections between various responsibility centers and budget items. It thoroughly solves the problems of messy links and disconnection from actual operations in traditional budget models from the data structure level.

[0073] Optionally, the processes of steps 401 to 405 include: Step 401: Match the business attribute identifiers of each node in the current topology with the display types in the budget display scheme to obtain a standardized scenario deduction rule group. Based on the standardized scenario deduction rule group and the customized types in the budget display scheme, implement the standardized scenario deduction rule group to obtain an executable scenario mode instruction sequence.

[0074] Optionally, the management forecasting system retrieves the current topology and preset budget display scheme, extracts the business attribute identifiers contained in each node of the current topology. These business attribute identifiers are unique identifiers assigned to each node in the current topology to identify the core business characteristics of that node. The identifiers clearly define the core information of the corresponding node, such as the business type, responsible entity, and time range. At the same time, the system extracts the display type from the budget display scheme and accurately matches the business dimensions contained in the node's business attribute identifiers with the dimensions of the display type, ensuring that each node can be matched with the corresponding display type. These are then integrated into a standardized scenario simulation rule set. This standardized scenario simulation rule set refers to a set of general rules built based on the display type to standardize scenario simulation. This rule set clarifies the core logic, data source, and simulation process of scenario simulation and is applicable to all scenario simulation scenarios that meet the display type requirements.

[0075] Optionally, the management forecasting system, based on a standardized scenario simulation rule set, combines it with customized types from a preset budget display scheme. This involves implementing the standardized scenario simulation rule set by combining the general rules within the set with the personalized requirements of the customized types, transforming them into specific, directly executable rules tailored to the company's actual needs. This clarifies the specific execution standards, parameter settings, and judgment conditions for each rule, ultimately forming an executable scenario mode instruction sequence. This executable scenario mode instruction sequence refers to an ordered set of specific, operable scenario simulation instructions. Each instruction corresponds to a combination of standardized rules and customized requirements, thus clarifying the specific operations that the management forecasting system must perform during scenario simulation, including data retrieval, simulation calculation, and status judgment.

[0076] Step 402: Based on the executable scenario mode instruction sequence and the data flow path of each responsibility center in the current topology, scenario transmission is performed to obtain the scenario variable transmission node group and scenario state transmission edge group arranged along the data flow direction. Based on the scenario variable transmission node group and scenario state transmission edge group, a transmission chain is constructed to obtain the topology path scenario transmission chain.

[0077] Optionally, the management and prediction system retrieves the execution scenario mode instruction sequence and the current topology. Based on the data flow paths of each responsibility center clearly defined in the current topology—that is, the direction, link, and correlation of data transmission between each responsibility center and node—and in conjunction with the specific instructions in the executable scenario mode instruction sequence, scenario transmission is performed. That is, according to the requirements of the executable scenario mode instruction sequence, along the data flow path of the current topology, scenario simulation instructions are progressively transmitted to each relevant node, prompting each node to perform corresponding scenario simulation operations, achieving complete scenario transmission from the starting node to the terminal node. During scenario transmission, nodes participating in scenario simulation along the data flow direction are identified one by one. These nodes are then grouped into a scenario variable transmission node group. This group refers to the set of all nodes participating in scenario simulation and undertaking scenario variable transmission and processing along the data flow direction of the current topology. Each node has a clearly defined scenario transmission responsibility: receiving scenario instructions and data from the previous node, performing the corresponding simulation operation, and then transmitting the result to the next node.

[0078] Simultaneously, the management and prediction system identifies the connection relationships between the various scenario variable transmission nodes, that is, the connecting edges in the current topology that connect the various scenario variable transmission nodes. These connecting edges are summarized into a scenario state transmission edge group. This scenario state transmission edge group refers to the set of all connecting edges that connect each node in the scenario variable transmission node group and are used to transmit scenario state data and inference instructions. Each connecting edge clearly corresponds to the scenario transmission relationship between two nodes, including the data transmission direction, transmission content, and transmission sequence, to ensure that scenario states and inference instructions can be transmitted accurately and orderly.

[0079] Optionally, after the scenario variable transmission node group and the scenario state transmission edge group are determined, the management prediction system constructs a transmission chain based on the two. That is, a chain structure is constructed with the scenario variable transmission node group as the core node and the scenario state transmission edge group as the connecting link, according to the data flow direction of the current topology, to realize the complete transmission of scenario inference instructions and scenario state data. This transmission chain clarifies the sequence of scenario inference and the relationship between nodes, ensuring that the scenario transmission process is orderly and efficient, and avoiding problems such as transmission interruption and timing disorder.

[0080] Step 403: Based on the executable scenario mode instruction sequence of each scenario variable transmission node in the topological path scenario transmission chain and the initial budget state value of the associated nodes in the current topological structure, perform multi-branch scenario state deduction to obtain multiple hypothesis state value groups, and based on the multiple hypothesis state value groups, perform state value transmission and logical replacement along the topological path scenario transmission chain to obtain a multi-branch scenario state evolution map.

[0081] Optionally, the management forecasting system retrieves the topology path scenario transmission chain, the executable scenario mode instruction sequence, and the initial budget state values ​​of associated nodes in the current topology. These initial budget state values ​​refer to the initial budget values ​​of each associated node (all nodes in the scenario variable transmission node group) before the scenario simulation begins. Based on the retrieved topology path scenario transmission chain, for each scenario variable transmission node, the system calls specific instructions from the executable scenario mode instruction sequence corresponding to that node. Combined with the initial budget state values ​​of the associated nodes, a multi-branch scenario state simulation is performed. That is, for the budget state of each node, based on different business scenarios that may occur during enterprise operations (such as market demand growth, cost increases, business adjustments, etc.), multiple hypothetical scenarios are set, and the budget state value of the node under each hypothetical scenario is simulated separately, forming multiple branch scenario states. This ensures the comprehensiveness of the scenario simulation and covers various business scenarios that the enterprise may face. Through multi-branch scenario state simulation, the budget state values ​​of each scenario variable transmission node under different hypothetical scenarios are obtained. The different hypothetical state values ​​of all nodes are then summarized to form a multi-hypothetical state value group.

[0082] Optionally, the management and forecasting system, based on multiple hypothetical state value groups, performs state value transfer and logical replacement operations along the transmission direction of the topological path scenario transmission chain. State value transfer refers to passing the multiple hypothetical state values ​​of each node along the connection link of the scenario state transmission edge group to the next associated node. Logical replacement refers to the next associated node, upon receiving the hypothetical state value from the previous node, updating and replacing its own hypothetical state value in conjunction with its own executable scenario mode instructions, ensuring that the hypothetical state value of each node reflects the transmission impact of the previous node and conforms to data association logic. After the state value transfer and logical replacement are completed, the multiple hypothetical state values ​​and transmission relationships of all nodes are integrated to construct a multi-branch scenario state evolution graph. This multi-branch scenario state evolution graph, based on the topological path scenario transmission chain, is a visual graph showing the evolution process of budget state values ​​of each scenario variable transmission node under different hypothetical scenarios and the transmission relationships between nodes. This graph clearly presents the complete evolution path of the budget state value from the starting node to the terminal node under each hypothetical scenario, as well as the correlation relationships between the state values ​​of each node.

[0083] Step 404: Based on the distribution of state values ​​of each branch in the multi-branch scenario state evolution graph and the preset budget balance constraint rule base, conflict and contradiction detection is performed to obtain the scenario branch group with resource over-allocation conflict and the state node group with data logic contradiction. Then, based on the scenario branch group and the state node group, conflict pruning and state value backtracking adjustment are performed to obtain a consistent scenario state evolution graph.

[0084] Optionally, the management forecasting system retrieves the multi-branch scenario state evolution graph and a pre-set budget balance constraint rule library. This pre-set budget balance constraint rule library refers to a set of rules pre-defined by the enterprise to standardize budget allocation, ensure logical consistency of budget data, and avoid resource allocation conflicts. This rule library covers two core categories: resource allocation constraint rules and data logic constraint rules. All rules comply with the enterprise's operational management requirements and budget preparation standards. Specifically, resource allocation constraint rules are used to constrain the allocation of budget resources for each responsibility center and budget item; data logic constraint rules are used to ensure logical consistency and consistency of budget data. Subsequently, based on the budget balance constraint rule library, the budget status value of each scenario branch and each node in the multi-branch scenario state evolution graph is checked one by one. This checks whether the resource allocation of each responsibility center and budget item in each scenario branch complies with the resource allocation constraint rules, and whether there are issues such as total over-allocation, proportional imbalance, or priority inversion. It also checks whether the budget status value of each node in each scenario branch complies with the data logic constraint rules, and whether there are logical contradictions such as related proportion deviations, numerical anomalies, or temporal sequence errors. After the detection is completed, the scenario branches with resource over-allocation conflicts are summarized to form a scenario branch group with resource over-allocation conflicts. At the same time, the nodes with data logic contradictions are summarized to form a state node group with data logic contradictions. Among them, the scenario branch group with resource over-allocation conflicts refers to the set of scenario branches in the multi-branch scenario state evolution graph that have problems such as budget resource allocation exceeding the preset amount, imbalance in proportion, or inversion of priority; the state node group with data logic contradictions refers to the set of nodes in the multi-branch scenario state evolution graph that have logical contradictions such as correlation ratio deviation, numerical anomalies, and temporal disorder in budget state values.

[0085] Optionally, the management forecasting system performs conflict pruning and state value backtracking adjustment based on scenario branch groups with resource over-allocation conflicts and state node groups with data logic contradictions. Conflict pruning refers to removing a scenario branch with resource over-allocation conflicts from the multi-branch scenario state evolution graph if the conflict cannot be resolved by adjusting the node state value (e.g., the branch as a whole exceeds the enterprise's total budget and there is no room for adjustment), while retaining branches without resource allocation conflicts or whose conflicts can be resolved through adjustment. If the conflict can be resolved, the branch is retained and adjustments are made at the conflict location. State value backtracking adjustment refers to backtracking along the topological path scenario transmission chain to find the root node of the contradiction, adjusting the assumed state value of the root node, and then retransmitting it to subsequent nodes to correct the state values ​​of subsequent nodes until the budget state values ​​of all nodes meet the requirements of the budget balance constraint rule base, thus eliminating data logic contradictions. After conflict pruning and state value backtracking adjustment are completed, a consistent scenario state evolution map is obtained. This consistent scenario state evolution map refers to the scenario state evolution map after conflict detection, pruning and adjustment, in which all scenario branches have no resource allocation conflicts, all node budget state values ​​meet data logic constraints, and the correlation and transmission sequence between nodes are consistent.

[0086] Step 405: Based on the state values ​​of each scenario branch in the consistent scenario state evolution graph and the optimization strategy of the preset budget display scheme, the scenario branches are screened to obtain the target scenario branch and the corresponding full node budget allocation value set. Based on the target scenario branch and the full node budget allocation value set, the data is summarized and formatted according to the organizational hierarchy of the current topology to obtain the target budget allocation plan.

[0087] Optionally, the management forecasting system retrieves a consistent scenario state evolution graph and a preferred strategy for the preset budget display scheme. The consistent scenario state evolution graph includes all conflict-free and logically consistent scenario branches and the budget state values ​​of each node. The preferred strategy for the preset budget display scheme refers to the judgment criteria and screening rules pre-defined by the enterprise to select the optimal scenario branch from multiple conflict-free scenario branches and generate a target budget allocation plan. These criteria include budget feasibility indicators, budget rationality indicators, risk controllability indicators, and target suitability indicators. The budget feasibility indicator refers to whether the budget allocation scheme corresponding to the scenario branch is suitable for the target budget allocation scheme. The budget rationality indicator refers to whether the budget status value and correlation ratio of each node in the scenario branch are consistent with the company's historical execution data and the current market environment, and whether they meet the budget balance constraint requirements; the risk controllability indicator refers to whether the operating risks corresponding to the scenario branch (such as cost overruns and revenue shortfalls) are within the company's preset risk tolerance range, and whether the probability of risk occurrence is low; the target suitability indicator refers to whether the budget allocation plan corresponding to the scenario branch is suitable for the company's current operating goals (such as revenue growth and cost control), and whether it can support the company's business development plan.

[0088] Furthermore, the optimization strategy clearly defines the priority of indicators: target suitability indicators have the highest priority, followed by risk controllability indicators, budget feasibility indicators, and finally budget rationality indicators. During the screening process, scenario branches that meet the target suitability indicators are first selected, then those that meet the risk controllability indicators are selected, and so on, until the scenario branch with the highest overall score is selected as the target scenario branch. If multiple branches have the same overall score, the branch that aligns with the core business and has a higher priority is prioritized, considering the company's current business priorities. Specifically, for each scenario branch in the consistency scenario state evolution graph, the scores of the four core indicators are calculated one by one (each indicator has a maximum score of 100 points, scored based on branch performance), and the overall score is calculated (overall score = target suitability score × 40% + risk controllability score × 30% + budget feasibility score × 20% + budget rationality score × 10%). Based on the overall scores, the scenario branch with the highest overall score is selected as the target scenario branch.

[0089] Optionally, the management forecasting system extracts the budget status values ​​of all scenario variable transmission nodes under this branch, summarizes them to form a full set of node budget allocation values, and formats them according to the organizational hierarchy of the current topology. The process is as follows: First, according to the responsibility center level, the node values ​​belonging to the same responsibility center in the full set of node budget allocation values ​​are summarized to obtain the total budget amount and budget amount of each business type for each responsibility center; then, according to the budget item level, the node values ​​belonging to the same budget item under the same responsibility center are summarized to obtain the budget amount of each budget item for each responsibility center; finally, according to the time unit level, the node values ​​belonging to the same time unit under the same budget item are summarized to obtain the budget amount of each budget item for each responsibility center in different time units. At the same time, according to the preset budget display scheme, the format of the budget allocation values ​​is standardized (such as numerical precision and unit), and necessary explanatory information (such as the basis for budget allocation and execution requirements) is supplemented to finally form the target budget allocation plan. This target budget allocation plan is the final budget allocation scheme that is tailored to the company's actual operating conditions, complies with budget balance constraints, and can support the company's operating objectives, after scenario simulation, conflict detection, branch screening, and summary formatting.

[0090] This invention addresses the shortcomings of existing budget forecasting methods, such as rigid scenario extrapolation, failure to consider individual enterprise needs, disconnect between budget allocation and actual business operations, resource conflicts, and logical contradictions. It improves the accuracy and dynamic control capabilities of budget preparation and forecasting, ensuring that the generated target budget allocation plan accurately matches the business rhythm and data relationships of each responsibility center, aligning with the enterprise's refined budget management needs. It provides reliable guidance for enterprise budget execution and control, ultimately resolving the technical problems of existing methods relying on fixed rules and static parameters, being unable to adapt to differentiated business and dynamic execution scenarios, and having forecast results disconnected from actual needs. This enhances the guidance of budget forecasting and the effectiveness of budget control, meeting the core needs of refined enterprise budget management.

[0091] Furthermore, the computational budget preparation, management, and forecasting system based on artificial intelligence and big data provided by this invention will be described below. The computational budget preparation, management, and forecasting system based on artificial intelligence and big data described below can be referred to in correspondence with the computational budget preparation, management, and forecasting method based on artificial intelligence and big data described above.

[0092] Optional, refer to Figure 2 , Figure 2 This is a schematic diagram of the structure of the computational budgeting, management, and forecasting system based on artificial intelligence and big data provided by the present invention. The computational budgeting, management, and forecasting system based on artificial intelligence and big data includes: The timing reconstruction module 210 is used to reconstruct the task scheduling control parameters of each responsibility center in the current compilation stage based on the status, cycle attributes and historical execution data of the budget tasks within the current budget cycle to obtain an adaptive task start timing sequence. The budget item screening module 220 is used to screen the budget items to be selected based on the adaptive task start sequence and the control mode and locking state determined in the preset budget control scheme, so as to obtain the effective budget item range. The topology construction and correction module 230 is used to construct a multi-dimensional budget topology based on the effective budget item range and the data classification dimensions in the budget preparation data, and to correct the topology based on the data flow path of each responsibility center in the multi-dimensional budget topology and the actual budget execution data to obtain the current topology. The budget forecast result determination module 240 is used to calculate the forecast scenario mode based on the current topology and the display type and customization type in the preset budget display scheme, to obtain the target budget allocation plan, and to determine the target budget forecast result based on the target budget allocation plan and the budget release rules and cost classification.

[0093] The embodiments of the present invention effectively solve the technical problems of existing methods relying on fixed rules and static parameters, being unable to adapt to differentiated business and dynamic execution scenarios, and having prediction results that are out of sync with actual needs. They improve the accuracy of budget preparation and forecasting, the adaptability of task scheduling and dynamic control capabilities, enhance the guidance of budget forecasting and the effectiveness of budget control, and meet the needs of enterprises for refined budget management.

[0094] Please see Figure 3 , Figure 3 An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it implements steps 10 to 40.

[0095] Please see Figure 4 , Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it implements steps 10 to 40.

[0096] On the other hand, the present invention also provides a computer program product, which includes 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 is able to execute the computational budgeting, management and forecasting method based on artificial intelligence and big data provided by the above methods, which includes steps 10 to 40.

[0097] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

Claims

1. A computational budgeting and management forecasting method based on artificial intelligence and big data, characterized in that, include: Based on the status, cycle attributes, and historical execution data of budget tasks within the current budget cycle, the priority sequence of budget tasks is determined, and the task scheduling control parameters of each responsibility center in the current compilation stage are reconstructed to obtain an adaptive task start sequence. Based on the adaptive task startup sequence, combined with the control methods and locking states in the preset budget control scheme, the dynamic budget constraint boundaries determined by the selected budget items are screened to obtain the range of effective budget items. Based on the scope of effective budget items and the data classification dimensions in the budget preparation data, a multi-dimensional budget topology is constructed. Then, based on the data flow paths of each responsibility center in the multi-dimensional budget topology and the actual budget execution data, the topology is corrected to obtain the current topology. Based on the current topology and the display types and customization types in the preset budget display scheme, the prediction scenario mode is calculated to obtain the target budget allocation plan. Based on the target budget allocation plan and the budget release rules and cost classification, the target budget prediction result is determined.

2. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 1, characterized in that, The budget task priority sequence, determined based on the status, cycle attributes, and historical execution data of budget tasks within the current budget cycle, reconstructs the task scheduling control parameters of each responsibility center in the current compilation phase to obtain an adaptive task startup sequence, including: Feature extraction is performed based on the real-time running status and task type of each budget task within the current budget cycle to obtain the instantaneous status features of each budget task. Based on the instantaneous status features, all budget tasks are divided into multiple status category groups. The status category groups include blocked status, in progress status, and pending start status. Based on the periodic attributes of budget tasks in each state category group and the time axis scale of the current budget period, the hard deadline constraints and dependency window constraints of each budget task are determined. Based on the hard deadline constraints and dependency window constraints, budget tasks with time conflict risks are screened to obtain key budget task nodes that need to have their time windows adjusted first. Based on the historical execution data of the key budget task nodes and associated responsibility centers, the actual time consumption deviation pattern and resource consumption fluctuation pattern of each key budget task node in the past cycle are obtained. Based on the actual time consumption deviation pattern and resource consumption fluctuation pattern, task type labels with high frequency delay characteristics and high frequency resource contention characteristics are obtained. Based on the task type labels and the urgency definition rules of each status category group, an analysis is performed to obtain the budget task priority sequence; Based on the budget task priority sequence, the task scheduling control parameters of each responsibility center in the current compilation stage are reconstructed to obtain the adaptive task startup sequence.

3. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 2, characterized in that, The process of reconstructing the task scheduling control parameters of each responsibility center at the current compilation stage based on the budget task priority sequence to obtain the adaptive task startup sequence includes: Based on the execution order of each budget task priority sequence and the corresponding responsibility center, task load distribution analysis is performed to construct a task quantity distribution map and a task time density distribution map for each responsibility center in the current planning stage. Based on the task quantity distribution map and the task time density distribution map, task load is identified to obtain the first responsibility center group with overloaded task load and the second responsibility center group with idle task load. Based on the global scheduling order of the first responsibility center group, the second responsibility center group, and each budget task, the task scheduling control parameters for each responsibility center are determined to be configured differently. The default parameters of each responsibility center are then replaced based on the task scheduling control parameters to obtain the target scheduling control parameters for each responsibility center in the current planning stage. Based on the target scheduling control parameters of each responsibility center and the order of execution, the theoretical earliest start time and theoretical latest start time of each budget task in each responsibility center are determined. Based on the theoretical earliest start time and theoretical latest start time, the task start sequence is deduced to obtain a draft task start sequence. Based on the aforementioned task startup sequence draft and the hard deadline constraints and dependency window constraints of each budget task in the time dimension, an adaptive task startup sequence covering all responsibility centers and each budget task is obtained.

4. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 3, characterized in that, Task scheduling control parameters include the upper limit of concurrent task quantity, the threshold of task start interval, and the task resource exclusivity flag.

5. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 1, characterized in that, The multi-dimensional budget topology structure, constructed based on the scope of effective budget items and the data classification dimensions in the budget preparation data, includes: Based on the subject attribute identifiers of each effective budget subject within the scope of effective budget subjects and the data classification dimensions in the budget preparation data, we determine the multi-dimensional attribution label groups of each effective budget subject in four dimensions: organizational level, business type, time period, and cost nature. Based on the multidimensional attribution label group, the effective budget items are mapped to the corresponding multidimensional coordinate space nodes to obtain the initial positioning point of each effective budget item in the multidimensional space. Based on the initial positioning points of each effective budget item in the multidimensional space and the account correlation relationship in the preset budget preparation rule base, the explicit logical association edges and implicit data dependency edges connecting different initial positioning points are extracted to obtain the association edge group. Based on the association edge group, the combination of account pairs with direct data flow relationship and the combination of account pairs with indirect data deduction relationship are identified to obtain the account association network skeleton. Based on the subject association network skeleton and the responsibility center affiliation relationship of each effective budget subject, determine the data flow direction indication set across different responsibility centers and the data interaction node set across departments, and splice the subject association network skeleton based on the data flow direction indication set and the data interaction node set to obtain a directed data flow path chain; Based on the initial positioning points of each effective budget item in the multidimensional space as topological nodes and the directed data flow path chain as topological connecting edges, the multidimensional budget topology structure containing node attribute information and edge direction information is constructed.

6. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 5, characterized in that, The data flow paths of each responsibility center in the multi-dimensional budget topology are combined with actual budget execution data to perform topology correction, resulting in the current topology, including: Based on the multidimensional budget topology, the data input ports and data output ports of each responsibility center are logically aligned to obtain a prototype of the statically constructed budget topology. Based on the budget topology prototype and the real-time data status of each valid budget item, the status is marked to obtain the candidate group of correction nodes and the candidate group of connection edges to be corrected in the topology. Based on the analysis of the candidate group of corrected nodes, the candidate group of connection edges to be corrected, and the actual execution data, the group of outdated node instances and the group of failed connection edges are obtained. Based on the group of outdated node instances and the group of invalid connection edges, the budget topology prototype is subjected to node replacement and edge redirection to obtain the current topology.

7. The computational budgeting and management forecasting method based on artificial intelligence and big data according to claim 1, characterized in that, The calculation of the predicted scenario mode based on the current topology and the display type and customization type in the preset budget display scheme, to obtain the target budget allocation plan, includes: Based on the business attribute identifiers of each node in the current topology and the display types in the budget display scheme, a standardized scenario deduction rule group is obtained. Based on the standardized scenario deduction rule group and the customized types in the budget display scheme, the standardized scenario deduction rule group is implemented to obtain an executable scenario mode instruction sequence. Based on the executable scenario mode instruction sequence and the data flow path of each responsibility center in the current topology, scenario transmission is performed to obtain a scenario variable transmission node group and a scenario state transmission edge group arranged along the data flow direction. Based on the scenario variable transmission node group and the scenario state transmission edge group, a transmission chain is constructed to obtain the topology path scenario transmission chain. Based on the executable scenario mode instruction sequence of each scenario variable transmission node in the topological path scenario transmission chain and the initial budget state value of the associated nodes in the current topological structure, multi-branch scenario state deduction is performed to obtain multiple hypothesis state value groups. Based on the multiple hypothesis state value groups, state value transmission and logical replacement are performed along the topological path scenario transmission chain to obtain a multi-branch scenario state evolution map. Based on the distribution of state values ​​of each branch in the multi-branch scenario state evolution graph and the preset budget balance constraint rule base, conflict and contradiction detection is performed to obtain scenario branch groups with resource over-allocation conflicts and state node groups with data logic contradictions. Conflict pruning and state value backtracking adjustment are then performed based on the scenario branch groups and the state node groups to obtain a consistent scenario state evolution graph. Based on the state values ​​of each scenario branch in the consistent scenario state evolution graph and the preferred strategy of the preset budget display scheme, scenario branches are screened to obtain the target scenario branch and the corresponding full node budget allocation value set. Based on the target scenario branch and the full node budget allocation value set, the data is summarized and formatted according to the organizational hierarchy of the current topology to obtain the target budget allocation plan.

8. A computational budgeting, management, and forecasting system based on artificial intelligence and big data, characterized in that: The method for computational budgeting, management, and forecasting based on artificial intelligence and big data, as described in any one of claims 1 to 7, is applied; the computational budgeting, management, and forecasting system based on artificial intelligence and big data comprises: The timing reconstruction module is used to reconstruct the task scheduling control parameters of each responsibility center in the current compilation stage based on the status, cycle attributes and historical execution data of budget tasks within the current budget cycle, so as to obtain an adaptive task start timing sequence. The budget item filtering module is used to filter the budget items to be selected based on the adaptive task start sequence and the control method and locking state determined in the preset budget control scheme, so as to obtain the range of valid budget items. The topology construction and correction module is used to construct a multi-dimensional budget topology based on the scope of effective budget items and the data classification dimensions in the budget preparation data, and to correct the topology based on the data flow paths of each responsibility center in the multi-dimensional budget topology and the actual budget execution data to obtain the current topology. The budget forecast result determination module is used to calculate the forecast scenario mode based on the current topology and the display type and customized type in the preset budget display scheme, to obtain the target budget allocation plan, and to determine the target budget forecast result based on the target budget allocation plan and the budget release rules and cost classification.

9. An electronic device, characterized in that, include: Memory, used to store computer software programs; A processor is configured to read and execute the computer software program, wherein when the processor executes the computer software program, it implements the computational budgeting, management, and forecasting method based on artificial intelligence and big data as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, The storage medium stores a computer software program, which, when executed by a processor, implements the computational budgeting, management, and forecasting method based on artificial intelligence and big data as described in any one of claims 1 to 7.

Images