CN-122023041-A - Budget intelligent monitoring management method and system

CN122023041ACN 122023041 ACN122023041 ACN 122023041ACN-122023041-A

Abstract

The application relates to the technical field of budget management, and discloses an intelligent budget monitoring and management method and system, which not only display alarm information when budget deviation alarm is monitored, and the deviation amount is quantitatively analyzed through a contribution degree decomposition model, the main factor path is locked, multi-level automatic drill-down is performed, and the root data unit is accurately positioned. And meanwhile, the analysis result is intelligently associated with unstructured information, related business events are searched by utilizing key entities and time ranges, and financial numbers are combined with real business causes to form contextual insights. And finally, integrating all information and inputting the information into a budget deviation intelligent monitoring engine based on a large language model to generate a diagnosis report with clear logic and strong readability, thereby improving the budget management and strategic execution capacity.

Inventors

WANG JIANYING
ZHANG LEI
BAO YUZHE

Assignees

中国计量大学现代科技学院

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. The budget intelligent monitoring and managing method is characterized by comprising the following steps: the acquired actual expenditure data are aggregated according to a preset time window, and summarized according to granularity in budget plan data to obtain summarized actual expenditure data; Calculating the deviation amount and the deviation rate of each budget unit based on the summarized actual expenditure data, and carrying out budget deviation early warning triggering judgment based on the deviation amount and the deviation rate of each budget unit; responding to budget deviation alarm, extracting a target analysis node from the budget deviation alarm and performing primary contribution degree decomposition on the deviation amount of the target analysis node to obtain a primary decomposition report; extracting the item with the highest contribution from the first-level decomposition report as a main factor path, and carrying out multi-level automatic drill-down along the main factor path to obtain a drill-down path result; retrieving relevant documents from unstructured documents based on key entities in the drill-down path results and the time range in which the alert occurred; Extracting keywords from the related documents, and carrying out information association on the extracted keywords and the drill-down path result to obtain contextual insights; Inputting the budget deviation alarm, the drill-down path result and the contextual insights into a large language model-based budget deviation intelligent monitoring engine to obtain a budget deviation diagnosis report.
2. The budget intelligent monitoring management method according to claim 1, wherein said budget plan data comprises budget subjects, departments, projects, time periods and plan amounts.
3. The budget intelligent monitoring and managing method according to claim 2, wherein calculating a deviation amount and a deviation rate of each budget unit based on the summarized actual expenditure data, and performing budget deviation early warning triggering judgment based on the deviation amount and the deviation rate of each budget unit comprises: and generating a budget deviation alarm in response to the absolute value of the deviation amount of the budget unit being greater than or equal to a first preset threshold and the absolute value of the deviation rate being greater than or equal to a second preset threshold, wherein the budget deviation alarm comprises information of the target analysis node, the deviation amount and the deviation rate.
4. The budget intelligent monitoring and managing method according to claim 3, wherein calculating a deviation amount and a deviation rate of each budget unit based on the summarized actual expenditure data, and performing budget deviation early warning triggering judgment based on the deviation amount and the deviation rate of each budget unit comprises: The deviation amount and deviation rate of each budget unit are calculated by the following formula: Wherein, the method comprises the steps of, For dynamically planning the amount, In order to be a practical expenditure, In order to find the sum function, Representing the total period plan amount of money, In order for the number of days to elapse in the cycle, As a total number of days in a cycle, Indicating the amount of deviation and, Representing the deviation rate.
5. The budget intelligent monitoring management method according to claim 1, wherein in response to a budget deviation alert, extracting a target analysis node from the budget deviation alert and performing a first degree contribution decomposition on a deviation amount of the target analysis node to obtain a first degree decomposition report, comprising: S100, determining the direct subordinate dimension of the target analysis node; s101, traversing each direct subordinate dimension and calculating the deviation amount of each direct subordinate dimension to obtain a set of direct subordinate dimension deviation amounts; S102, calculating the contribution percentage of each direct subordinate dimension deviation amount in the set of direct subordinate dimension deviation amounts relative to the deviation amount of the target analysis node.
6. The budget intelligent monitoring and management method according to claim 5, wherein extracting a most contributing item from said primary decomposition report as a main cause path, and performing multi-level automatic drill-down along said main cause path to obtain a drill-down path result, comprises: S200, extracting the direct subordinate dimension with the largest contribution percentage from the first-level decomposition report as the item with the highest contribution degree; And S201, taking the item with the highest contribution degree as a new target analysis node, and repeatedly executing the processing logic of the steps S100 to S102 until a preset condition is met to obtain the drill-down path result.
7. The budget intelligent monitoring and managing method according to claim 6, wherein said preset condition is that the contribution degree of data or a certain level drilled down to the finest granularity is distributed uniformly.
8. The budget intelligent monitoring and management method according to claim 5, wherein extracting a most contributing item from said primary decomposition report as a main cause path, and performing multi-level automatic drill-down along said main cause path to obtain a drill-down path result, comprises: extracting a plurality of principal paths from the primary decomposition report; Thread callback registration is carried out on each main factor path; Performing hierarchical cyclic polling on the thread callbacks of all main cause paths to obtain a plurality of hierarchical cyclic polling results; and carrying out the asynchronous emptying and queue tail addition of the micro-tasks with the same subordinate dimension on the multiple levels of circulating polling results to obtain the drill-down path result.
9. The budget intelligent monitoring management method according to claim 1, wherein inputting said budget deviation alarm, said drill-down path result and said contextual insights into a large language model based budget deviation intelligent monitoring engine for obtaining a budget deviation diagnostic report, comprises: Constructing a budget deviation diagnostic report prompt based on the budget deviation alert, the drill-down path result, and the contextual insights; inputting the budget deviation diagnosis report prompt into the large language model-based budget deviation intelligent monitoring engine to obtain the budget deviation diagnosis report.
10. An intelligent budget monitoring and management system, comprising: the expense aggregation and summarization module is used for aggregating the acquired actual expense data according to a preset time window and summarizing according to granularity in the budget plan data to obtain summarized actual expense data; The budget deviation analysis module is used for calculating the deviation amount and the deviation rate of each budget unit based on the summarized actual expenditure data and carrying out budget deviation early warning triggering judgment based on the deviation amount and the deviation rate of each budget unit; The deviation attribution analysis module is used for responding to the budget deviation alarm, extracting a target analysis node from the budget deviation alarm and carrying out primary contribution degree decomposition on the deviation amount of the target analysis node so as to obtain a primary decomposition report; The main factor path analysis module is used for extracting the item with the highest contribution from the first-level decomposition report as a main factor path, and carrying out multi-level automatic drill-down along the main factor path to obtain a drill-down path result; An event-related document retrieval module for retrieving related documents from unstructured documents based on the key entities in the drill-down path result and the time range in which the alarm occurred; The context insight association module is used for extracting keywords from the related documents and carrying out information association on the extracted keywords and the drill-down path result so as to obtain context insights; And the intelligent budget diagnosis report generation module is used for inputting the budget deviation alarm, the drill-down path result and the contextual insight into an intelligent budget deviation monitoring engine based on a large language model to obtain a budget deviation diagnosis report.

Description

Budget intelligent monitoring management method and system Technical Field The application relates to the technical field of budget management, in particular to an intelligent budget monitoring and management method and system. Background Financial data and business data within organizations are rapidly expanding, and data sources are increasingly diversified, covering a plurality of heterogeneous systems such as Enterprise Resource Planning (ERP) systems, purchasing platforms, reimbursement systems, project management tools, and the like. This massive, multi-source data environment makes traditional, manual or simple spreadsheet-dependent budget monitoring approaches a serious challenge. The traditional budget management is low in efficiency, time-consuming and labor-consuming, and is extremely prone to error in the data aggregation and verification process, so that a monitoring result is lagged, and high requirements of modern management on instantaneity and accuracy cannot be met. More importantly, when a deviation exists between the budget and the actual expenditure, the root cause of the deviation is quickly and accurately located from the vast data, and the underlying business cause is understood, which is an extremely complex and difficult task for management personnel. Therefore, a set of management schemes capable of automatically and intelligently carrying out budget monitoring and analysis are constructed, so that decision efficiency and management depth are improved, and urgent demands of various organizations are met. To address the challenges described above, some budget monitoring schemes based on business intelligence tools or data analysis platforms have emerged in the prior art. The schemes can realize automatic collection and aggregation of the multisource actual expenditure data, compare the multisource actual expenditure data with a budget plan, show deviation conditions through a visual instrument panel or report, and even set simple threshold rules to trigger alarms. However, these existing solutions have significant drawbacks in terms of depth analysis and intelligent diagnostics. When an alarm is triggered, the analysis process is often split and manual. The analyst needs to drill down manually layer by layer in the system based on personal experience to try to locate the root cause of the problem, the process is low in efficiency, the analysis path is not clear, key information is easy to miss, and the comprehensiveness and accuracy of analysis are difficult to ensure. Therefore, in order to overcome the defects of the prior art, the application provides an intelligent budget monitoring management scheme. Disclosure of Invention The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an intelligent budget monitoring management method and system, which aim to overcome the defects of low budget monitoring analysis efficiency, data and service disjoint, insufficient analysis depth and non-visual result presentation in the prior art. According to one aspect of the application, an intelligent budget monitoring management method is provided, and the intelligent budget monitoring management method comprises the steps of aggregating acquired actual expenditure data according to a preset time window, summarizing according to granularity in budget plan data to obtain summarized actual expenditure data, calculating deviation amount and deviation rate of each budget unit based on the summarized actual expenditure data, carrying out budget deviation early warning trigger judgment based on the deviation amount and the deviation rate of each budget unit, responding to budget deviation warning, extracting a target analysis node from the budget deviation warning, carrying out primary contribution degree decomposition on the deviation amount of the target analysis node to obtain a primary decomposition report, extracting a most contributed item from the primary decomposition report as a primary factor path, carrying out multi-level automatic drill-down along the primary factor path to obtain a drill-down path result, searching related documents from unstructured documents based on key entities and the time range of occurrence of alarms in the drill-down path result, extracting key words from the related documents, carrying out information correlation judgment on the extracted key words and the drill-down path result to obtain context, and inputting the key words and the extracted key words into an intelligent budget monitoring language diagnosis engine based on the budget deviation monitoring model. In the budget intelligent monitoring and management method, the budget plan data comprises budget subjects, departments, projects, time periods and plan amounts. In the budget intelligent monitoring management method, the deviation amount and the deviation rate of each budget unit are calculated based on the summarized actual expen