A cost audit method and system based on big data

Abstract

The present invention relates to the field of cost audit technology, and in particular to a cost audit method and system based on big data, the method comprising: obtaining audit items included in the current engineering cost audit, and integrating project file data to generate multivariate heterogeneous cost data; establishing an audit rule engine, and performing audit item compliance checks on the current project according to project types and audit specifications and standards; establishing a cost audit database based on multivariate heterogeneous cost data and historical audit records, and establishing a risk prediction model based on the cost audit database and the audit rule engine; the risk prediction model generates audit rule matching results based on historical audit records and current engineering records; analyzing the audit rule matching results and risk prediction results, generating risk warning information, and generating an audit analysis report based on the risk warning information. Through the present invention, the problems of data integration difficulties, audit lags, rule rigidity, and lack of intelligent risk prediction in traditional engineering cost audits are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of cost auditing, and in particular to a cost auditing method and system based on big data. Background Art

[0002] As the scale and complexity of construction projects continue to grow, the management and control of construction costs have become one of the key factors for the smooth implementation of projects. In traditional cost audit methods, the audit of engineering projects usually relies on manual review and post-analysis. Existing engineering cost audit methods have many shortcomings, mainly manifested in the diversity and heterogeneity of data sources, which makes it difficult to integrate information. Traditional audits rely on manual review, resulting in audit lags and the inability to discover and correct cost anomalies in real time.

[0003] In addition, audit rules are often rigid and cannot be flexibly adjusted according to changes in project types and industry standards. Existing audit methods lack intelligent risk prediction capabilities and are difficult to detect potential cost overruns or financial risks in advance. These problems are particularly prominent in large and complex projects, seriously affecting project cost management and control.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a cost audit method and system based on big data, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A cost audit method based on big data, the method comprising:

[0008] Acquire the audit items included in the current engineering cost audit, and integrate the engineering cost audit project file data according to the audit items to generate multivariate heterogeneous cost data;

[0009] Establishing an audit rule engine, which performs audit item compliance checks on the current project according to the project type and audit specification standards;

[0010] Establishing a cost audit database based on the multivariate and heterogeneous cost data and historical audit records, establishing a risk prediction model based on the cost audit database and the audit rule engine, identifying abnormal data and potential risk points to obtain risk prediction results;

[0011] The risk prediction model performs audit rule matching according to the audit entry information contained in the historical audit records and the current project records and generates an audit rule matching result;

[0012] The audit rule matching results and risk prediction results are analyzed to generate risk warning information, and an audit analysis report is generated based on the risk warning information.

[0013] Furthermore, an audit rule engine is established, including:

[0014] Establishing an audit specification set according to the type and characteristics of the engineering project, the audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications;

[0015] Setting multi-layer verification and associated audit rules, wherein the multi-layer verification and associated audit rules cross-verify the audit items through the audit specification set according to the complexity of the project, analyze the correlation between the audit items and multiple rule items, and generate a linkage audit specification group;

[0016] A rule execution feedback loop is constructed, wherein the rule execution feedback loop verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specifications according to the compliance of the audit execution process data with the audit rules.

[0017] Furthermore, the correction specification layer learns and adjusts the execution priority of the audit specifications based on historical audit data and project execution process feedback, including:

[0018] Analyze the audit execution process data and extract the audit execution process pattern in the audit rule execution, and identify the audit execution interference factors that affect the audit results based on the audit execution process pattern;

[0019] According to the audit interference factors and the feedback of the audit execution process data, the correction specification layer captures the changes in project progress, cost deviation and risk level, and estimates the project execution deviation;

[0020] The project execution deviations are integrated with the historical audit data, and a multidimensional data analysis framework is constructed based on the correction specification layer, and the multidimensional data analysis framework is used to sort the execution priority of the audit specifications.

[0021] Furthermore, a risk prediction model is established based on the cost audit database and the audit rule engine, including:

[0022] Integrate the multivariate and heterogeneous cost data in the cost audit database with the project data in the historical audit records, and generate a basic data set for risk analysis through data preprocessing and standardization;

[0023] Extracting potential risk factors from the compliance check results of the current project through the audit rule engine, wherein the potential risk factors are identified based on the audit anomaly patterns in the cost audit database;

[0024] Based on the historical audit data in the cost audit database, a risk prediction model is constructed to analyze the risk characteristics of historical projects. The risk factors of the current project are associated with similar risk characteristics in the historical data through the risk prediction model to perform correlation matching;

[0025] Using the risk prediction model and the extracted risk factors, potential risk levels are calculated at different stages of project execution to generate risk prediction results.

[0026] Furthermore, the potential risk factors are identified based on the audit anomaly patterns in the cost audit database, including:

[0027] Based on the historical audit data in the cost audit database, cluster analysis is performed on abnormal data in historical projects to identify common audit abnormal patterns in different types of projects;

[0028] Extracting risk factors from the audit anomaly patterns, and classifying and rating the risk factors according to their impact on the project cost;

[0029] During the audit execution process, the compliance check results of the current project are analyzed through the audit rule engine, and the abnormal data therein is compared with the audit abnormality pattern to identify the potential risk factors similar to historical projects.

[0030] Furthermore, the risk factors are classified and graded according to their impact on the project cost, including:

[0031] Construct a risk impact analysis layer to quantify the impact of potential risk factors on the project in various audit exception patterns and generate corresponding impact coefficients;

[0032] Classifying the extracted risk factors according to the corresponding impact coefficients, wherein the classification is based on setting the threshold of the impact coefficient and different audit abnormality modes;

[0033] Adding a corresponding risk level mark to each of the classified risk factors to indicate the potential threat level of the corresponding risk factor to the project cost, and adding the risk level mark to the risk prediction model;

[0034] During the audit execution, the changes of the risk factors are monitored, the input data of the risk impact analysis layer is updated and the risk level mark is adjusted, and the risk prediction model is fed back according to the updated risk level.

[0035] Furthermore, audit rule matching is performed and audit rule matching results are generated, including:

[0036] According to the type and characteristics of the audit items, the audit items are matched by the audit rule engine, and the matching process is based on the multivariate heterogeneous cost data and historical audit records in the cost audit database;

[0037] By means of the multi-layer verification and the associated audit rules, the association between the audit items and the historical audit records is analyzed, and the audit rules are associated and matched with the risk factors in the current audit project to generate an associated matching result;

[0038] In the association matching process, rule conflict detection is performed to identify potential conflicts or overlapping parts between multiple audit entries, and conflicting entries are prioritized to generate a conflict reconciliation sequence;

[0039] An audit rule matching result is generated according to the association matching result and the conflict reconciliation sequence, and the audit rule matching result is used as input data of the risk prediction model.

[0040] Furthermore, rule conflict detection is performed, including:

[0041] Analyze the audit items in the current audit project one by one through the audit rule engine, and use the multivariate heterogeneous cost data in the cost audit database to detect potential conflicts between the audit items and the associated audit rules;

[0042] According to the identified conflict items, classify and handle the conflict types, and classify and organize the conflict items to form a conflict report;

[0043] Performing priority evaluation on the detected conflicting items through the multi-layer verification and associated audit rules to generate a conflict priority sequence, wherein the conflict priority sequence is classified according to the risk level of the conflicting items;

[0044] A conflict reconciliation sequence is generated according to the conflict priority sequence combined with the risk level. The conflict reconciliation sequence is used to reallocate the execution order of the conflicting items, and the reconciliation result is fed back to the audit rule matching process.

[0045] A cost audit system based on big data, the system comprising:

[0046] The data integration module obtains the audit items included in the current engineering cost audit, and integrates the engineering cost audit project file data according to the audit items to generate multi-dimensional heterogeneous cost data;

[0047] The rule engine module establishes an audit rule engine, which performs compliance checks on audit items of the current project based on project type and audit specifications and standards;

[0048] The risk prediction module establishes a cost audit database based on multivariate and heterogeneous cost data and historical audit records, and establishes a risk prediction model based on the cost audit database and the audit rule engine to identify abnormal data and potential risk points to obtain risk prediction results;

[0049] The rule matching module, the risk prediction model performs audit rule matching based on the audit entry information contained in the historical audit records and the current project records and generates audit rule matching results;

[0050] The risk warning module analyzes the audit rule matching results and risk prediction results, generates risk warning information, and generates an audit analysis report based on the risk warning information.

[0051] Furthermore, the rule engine module includes:

[0052] The priority adjustment unit establishes an audit specification set according to the type and characteristics of the engineering project. The audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications.

[0053] Verification specification unit, set up multi-layer verification and associated audit rules, which cross-verify audit items through audit specification sets according to the complexity of the project, analyze the correlation between audit items and multiple rule items, and generate linkage audit specification groups;

[0054] In accordance with the feedback unit, a rule execution feedback loop is constructed. The rule execution feedback loop verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specifications according to the compliance of the audit execution process data with the audit rules.

[0055] The technical solution of the present invention can achieve the following technical effects:

[0056] It effectively solves the problems of data integration difficulties, audit delays, rigid rules and lack of intelligent risk prediction in traditional engineering cost audits. Through big data technology and machine learning, it realizes the ability to audit, flexibly adjust rules and identify potential risks in advance.

[0057] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 This is a flowchart of the cost audit method based on big data;

[0060] Figure 2 This is a schematic diagram of the structure of the audit rule engine;

[0061] Figure 3 A schematic diagram of the structure for establishing a risk prediction model;

[0062] Figure 4 A flowchart for generating audit rule matching results;

[0063] Figure 5 The figure is a flowchart of rule conflict detection. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0066] Embodiment 1

[0067] like Figure 1 As shown, the present application provides a cost audit method based on big data, the method comprising:

[0068] S10: obtaining the audit items included in the current engineering cost audit, and integrating the engineering cost audit project file data according to the audit items to generate multivariate heterogeneous cost data;

[0069] S20: Establish an audit rule engine, which performs audit item compliance check on the current project according to the project type and audit specification standards;

[0070] S30: Establish a cost audit database based on multivariate and heterogeneous cost data and historical audit records, establish a risk prediction model based on the cost audit database and the audit rule engine, identify abnormal data and potential risk points to obtain risk prediction results;

[0071] S40: The risk prediction model performs audit rule matching based on the audit entry information contained in the historical audit records and the current project records and generates an audit rule matching result;

[0072] S50: Analyze the audit rule matching results and risk prediction results, generate risk warning information, and generate an audit analysis report based on the risk warning information.

[0073] Specifically, first obtain the audit items of the current engineering project, and analyze the relevant data of the corresponding audit items according to the obtained audit items. The data sources may include project budget, schedule management, procurement contracts, etc. For different data sources, these data are integrated into multi-dimensional heterogeneous pricing data. The multi-dimensional heterogeneous pricing data covers multi-dimensional information such as project budget, actual expenditure, schedule deviation, contract execution, etc.; establish an audit rule engine, and conduct audits based on the type, scale and characteristics of the engineering project in combination with industry standards and project-specific audit specifications. The audit rule engine checks whether the audit items in the project meet the audit specification standards item by item and conducts compliance checks. The checking process adapts to different audit requirements according to the complexity of the project to ensure the compatibility of the audit items with the specifications; in the integration Based on multi-source heterogeneous cost data and historical audit records, a cost audit database is established, and a risk prediction model is constructed in combination with the audit rule engine. By analyzing historical audit data and current project data, abnormal data or potential risk points in the project are identified, and preliminary risk prediction results are generated; the risk prediction model will match the audit rules based on historical data and the audit item information of the current project to ensure that the items in the audit process meet the actual needs of the project by matching similar project risk characteristics in historical audit records; after the audit rule matching and risk prediction are completed, risk warning information is generated, and a detailed audit analysis report is generated, which provides analysis results and recommended measures for audit anomalies and potential risks in the project, helping project managers to conduct real-time cost control and decision optimization.

[0074] The technical solution of the present invention effectively solves the problems of data integration difficulties, audit delays, rigid rules and lack of intelligent risk prediction in traditional engineering cost audits.

[0075] Further, if Figure 2As shown, establish an audit rule engine, including:

[0076] Establish an audit specification set based on the type and characteristics of the engineering project. The audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications.

[0077] Set up multi-layer verification and associated audit rules. Multi-layer verification and associated audit rules cross-verify audit items through audit specification sets according to the complexity of the project, analyze the correlation between audit items and multiple rule items, and generate linkage audit specification groups;

[0078] Construct a rule execution feedback loop, which verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specifications based on the compliance of the audit execution process data with the audit rules.

[0079] As a preferred embodiment of the above, firstly, an audit specification set is established for different types and characteristics of engineering projects. The specification set includes a basic specification layer, a project definition specification layer and a correction specification layer. The basic specification layer covers general audit rules applicable to all projects to ensure that all projects meet basic audit standards. The project definition specification layer sets specific rules for specific project requirements, such as special specifications in large projects. The correction specification layer dynamically adjusts the priority of audit rules according to feedback data and historical audit data in project execution to ensure that the rules can flexibly adapt to real-time changes in the project. At the same time, multi-layer verification and association audit rules are adopted in the audit execution process. The audit items are multi-layer verified through a cross-validation mechanism so that the project meets both basic and project-specific rules. The associated audit is associated with the current project items through historical data to generate a linkage audit specification group. The logic and correlation between different audit rules work together. A rule execution feedback loop is constructed. During the project execution process, the rules are optimized and adjusted according to the data at different stages. The data generated by the project execution will be input into the feedback loop, and the data will be verified for compliance to check whether the current audit rules match the actual situation. The audit specifications are adjusted according to the verification results. Through the feedback mechanism, key risk items are prioritized and the execution order and priority of the audit rules are adjusted.

[0080] Furthermore, the correction specification layer learns and adjusts the execution priority of audit specifications based on historical audit data and project execution process feedback, including:

[0081] Analyze the audit execution process data and extract the audit execution process pattern in the audit rule execution, and identify the audit execution interference factors that affect the audit results based on the audit execution process pattern;

[0082] Based on the feedback from audit interference factors and audit execution process data, the correction specification layer captures changes in project progress, cost deviation and risk level, and estimates project execution deviation;

[0083] Integrate project execution deviations with historical audit data, and build a multidimensional data analysis framework based on the correction specification layer. The multidimensional data analysis framework is used to sort the execution priority of audit specifications.

[0084] As a preferred embodiment of the above, the audit execution process data is first analyzed and the audit execution process pattern is extracted. During the actual execution of the project, relevant data such as project progress and cost are monitored, and regular audit execution process patterns are extracted from a large amount of data during the audit execution process. The audit execution process pattern reveals the execution of audit rules in different project scenarios, and helps identify key factors or interference factors that may affect the audit results during the project execution process, such as engineering progress bottlenecks and cost overruns at specific time points. Through pattern extraction, the actual execution effect of the audit rules is deeply analyzed; based on the identified audit interference factors and the data fed back during the project execution process, the process changes in the project are captured, especially the project progress deviation, cost deviation and Changes in risk levels. By monitoring and analyzing these data, we can infer the execution deviations of the current project, such as whether the actual progress lags behind the plan, whether the cost exceeds the budget, whether a risk event has occurred, etc., to ensure that the audit rules can be adjusted according to the actual situation of the project, especially when key deviations are found, they can be responded to in a timely manner; the inferred project execution deviations are integrated with historical audit data, and combined with the audit data of previous similar projects to further optimize the adjustment of audit standards. Through data integration, a multi-dimensional data analysis framework is constructed. By analyzing the comprehensive data of different dimensions of the project (such as time, cost, progress and risk), the execution priority of the audit standards is sorted, and key audit rules are implemented first according to the current status and risk level of the project.

[0085] Further, if Figure 3 As shown in the figure, a risk prediction model is established based on the cost audit database and the audit rule engine, including:

[0086] Integrate the multivariate and heterogeneous cost data in the cost audit database with the project data in the historical audit records, and generate a basic data set for risk analysis through data preprocessing and standardization;

[0087] Through the audit rule engine, the compliance check results of the current project are checked to extract potential risk factors, which are identified based on the audit anomaly patterns in the cost audit database;

[0088] Based on the historical audit data in the cost audit database, a risk prediction model is constructed to analyze the risk characteristics of historical projects. The risk factors of the current project are associated with similar risk characteristics in historical data through the risk prediction model to perform correlation matching;

[0089] Using the risk prediction model and the extracted risk factors, the potential risk level is calculated at different stages of project execution to generate risk prediction results.

[0090] As a preferred embodiment of the above, firstly, the multivariate and heterogeneous cost data of the current project are obtained from the cost audit database, including information on budgets, cost expenditures, project progress, material procurement, contract performance, etc. from different systems. Due to different data sources, these multivariate and heterogeneous cost data are integrated into a basic data set for risk analysis that can be analyzed through data preprocessing and standardization operations. The historical audit records of similar projects are retrieved from the database, including cost deviations, risk events and response measures in historical projects. By integrating current data and historical data, a complete project data view is formed. Based on the compliance check results of the current engineering project, potential risk factors that may exist in the project are extracted through the audit rule engine. The compliance check is to verify whether the current project items meet the predetermined audit specifications through the audit rule engine, such as whether the budget is within a reasonable range and whether the progress is progressing as planned. Potential risk factors are compared with the audit anomalies in the cost audit database. For example, if signs of budget overruns are detected and such overruns match abnormal patterns that have occurred in past projects, the problem will be identified as a potential risk factor; a risk prediction model is built based on historical data in the cost audit database, and different types of risk characteristics in multiple historical projects are analyzed to identify common risk patterns in different projects. The risk prediction model is used to correlate and match risk factors in the current project with similar risk characteristics in historical projects; the extracted potential risk factors and risk prediction model are used to evaluate the level of potential risks at different stages of project execution. The risk prediction model calculates the risk level of the project based on the data input of the current project. These levels may cover multiple dimensions such as cost risk, schedule risk, and resource risk. Risk warning information is generated based on the risk prediction results generated by the risk prediction model. This information not only points out the possible risk types, but also includes the potential impact range and response measures of these risks.

[0091] Further, potential risk factors are identified based on audit anomaly patterns in the cost audit database, including:

[0092] Based on the historical audit data in the cost audit database, cluster analysis is performed on abnormal data in historical projects to identify common audit abnormal patterns in different types of projects;

[0093] Extract risk factors from audit anomaly patterns and classify and grade them according to the degree of impact of risk factors on project cost;

[0094] During the audit execution process, the compliance check results of the current project are analyzed through the audit rule engine, and the abnormal data is compared with the audit abnormality pattern to identify potential risk factors similar to historical projects.

[0095] As a preferred embodiment of the above, firstly, a large amount of audit data of historical projects is extracted from the cost audit database, especially the audit items marked as abnormal. The abnormal data in these historical projects are processed and classified by cluster analysis technology, and the abnormal patterns in the historical projects are classified and sorted to identify common audit abnormal patterns. Risk factors are extracted from the identified audit abnormal patterns. These risk factors include key factors that may cause significant deviations in project costs, such as abnormal procurement costs, excessive labor costs, and interruptions in the material supply chain. These risk factors are classified and graded according to their impact on the project cost. By analyzing the historical impact of each risk factor, Assign a risk level to each risk factor, such as "low risk", "medium risk", "high risk", etc. During the actual audit execution of the project, the audit items of the current project are checked for compliance through the audit rule engine. During the check, abnormal items in the project are detected, and these abnormal data are compared with the audit abnormal pattern library to identify potential risk factors in the current project. For example, the budget item of the current project shows similar characteristics to the "budget overspending" pattern in historical projects. The item is marked as a potential risk. Through pattern comparison and risk factor extraction, potential risk factors similar to historical projects are identified, and risk warnings are provided for the current project based on the impact level of the risk factors.

[0096] Furthermore, risk factors are classified and graded according to their impact on project cost, including:

[0097] Construct a risk impact analysis layer to quantify the impact of potential risk factors on the project in various audit abnormal patterns and generate corresponding impact coefficients;

[0098] The extracted risk factors are classified according to the corresponding impact coefficients, and the classification is based on setting the threshold of the impact coefficient and different audit abnormality modes;

[0099] Attach a corresponding risk level mark to each classified risk factor to indicate the potential threat level of the corresponding risk factor to the project cost, and add the risk level mark to the risk prediction model;

[0100] Monitor changes in risk factors during audit execution, update input data of the risk impact analysis layer and adjust risk level tags, and feed back risk prediction models based on the updated risk levels.

[0101] As a preferred embodiment of the above, a risk impact analysis layer is first established to analyze the specific impact of potential risk factors extracted from historical audit anomaly patterns on project costs. By performing data analysis on audit anomaly patterns of multiple historical projects, the impact of different risk factors on project costs, construction periods, resource allocation, etc. is identified. For example, some risk factors may lead to significant cost overruns, while others may lead to construction delays. Based on these impact data, the impact of each risk factor is quantified, and a corresponding impact coefficient is generated to quantify the potential threat of the risk factor to the project cost. After the impact coefficient is generated, the extracted risk factors are classified according to their impact by setting a threshold value of the impact coefficient and analyzing different types of audit anomaly patterns. Specifically, multiple impact coefficients are set. The threshold of the number is used to classify risk factors into different levels, and the risk level of each factor is determined according to the actual impact of risk factors on the project in different audit exception modes; after the risk factors are classified, a corresponding risk level mark is attached to each risk factor to reflect the potential threat degree of each risk factor to the project cost, and risk management is optimized. These risk level marks are input into the risk prediction model, so that the prediction model can combine the actual risk factors and historical data of the current project to generate more accurate risk prediction results. In this way, the focus and strategy of the audit are adjusted according to the actual data of different project stages to ensure that more attention is paid to high-risk factors; during the actual audit execution of the project, the changes in risk factors are monitored, and the input data of the risk impact analysis layer is updated according to the new data.

[0102] Further, if Figure 4 As shown, audit rule matching is performed and audit rule matching results are generated, including:

[0103] According to the audit project type and characteristics, the audit items are matched through the audit rule engine. The matching process is based on the multi-dimensional and heterogeneous cost data and historical audit records in the cost audit database.

[0104] Through multi-layer verification and association of audit rules, the relevance of audit items and historical audit records is analyzed, and the audit rules are associated with the risk factors in the current audit project to generate association matching results;

[0105] During the association matching process, rule conflict detection is performed to identify potential conflicts or overlaps between multiple audit entries, and conflicting entries are prioritized to generate a conflict reconciliation sequence;

[0106] According to the association matching results and conflict reconciliation sequences, the audit rule matching results are generated, and the audit rule matching results are used as input data for the risk prediction model.

[0107] As a preferred embodiment of the above, firstly, according to the type and characteristics of the project, the audit items of the current project based on the multivariate and heterogeneous cost data in the cost audit database and the historical audit records are matched item by item through the audit rule engine. For example, in a complex engineering project, the specific requirements of the current project are matched according to the audit items of similar historical projects; in the process of audit item matching, the audit items of the current project are further analyzed through multi-layer verification and association audit rules. The multi-layer verification mechanism performs correlation analysis on the audit items from multiple dimensions. For example, the budget items of the current project are checked with the budget of similar projects in the historical audit records, and data comparison and verification are performed. By associating and matching the risk factors in the current audit project with the historical data, an association matching result is generated, and items with similar risk characteristics in the project are identified; in the process of audit item matching, The rule conflict detection function identifies possible conflicts or overlaps between audit items. For example, different items may involve similar budget data or schedules, and the rules between these items may be repeated or inconsistent. These conflicts are detected and relevant conflict reports are generated. The detected conflicting items are prioritized and a conflict reconciliation sequence is generated. The priority ranking is determined based on the importance and risk level of the audit items. The generated conflict reconciliation sequence reallocates the execution order of the conflicting items. Finally, based on the associated matching results and the conflict reconciliation sequence, a complete audit rule matching result is generated, including details of how the audit rules correspond to the audit items of the current project one by one, as well as the priority order after resolving the conflict. The generated audit rule matching result will be used as input data for the risk prediction model to ensure that the subsequent risk prediction process is based on the optimized audit rules.

[0108] Further, if Figure 5 As shown, rule conflict detection is performed, including:

[0109] The audit rule engine analyzes the audit items in the current audit project one by one, and uses the multivariate and heterogeneous cost data in the cost audit database to detect potential conflicts between audit items and associated audit rules;

[0110] According to the identified conflict items, classify and handle the conflict types, and classify and organize the conflict items to form a conflict report;

[0111] Through multi-layer verification and associated audit rules, the detected conflicting items are prioritized and a conflict priority sequence is generated. The conflict priority sequence is classified according to the risk level of the conflicting items.

[0112] According to the conflict priority sequence and risk level, a conflict reconciliation sequence is generated. The conflict reconciliation sequence is used to reallocate the execution order of conflicting items, and the reconciliation results are fed back to the audit rule matching process.

[0113] As a preferred embodiment of the above embodiment, firstly, each audit item of the current audit project is analyzed item by item through the audit rule engine, and the multivariate and heterogeneous cost data in the cost audit database are used to detect the possible conflicts between the audit items and the existing associated audit rules. Potential conflicts may include duplication of audit items, difference in item applicability, inconsistent execution priority, etc. For example, an item may involve two different audit rules, but when executed, the two rules may be repeated or contradictory. After identifying the potential conflicting items, the conflicting items are classified and processed, and they are classified and sorted according to the type of conflict. For example, the item conflicts are divided into categories such as rule overlap conflicts, item applicability conflicts, priority conflicts, etc. A conflict report is generated based on the classification results. The report lists in detail the specific conflict type and cause of each conflicting item, and is provided to the auditor or system for further priority evaluation and reconciliation processing; the identified conflicting items are evaluated for priority through multi-layer verification and associated audit rules, and according to the risk level of these items Classification and priority assessment are based on the impact of conflicting items on the project, especially the potential impact on project cost, progress and quality. For example, conflicting items related to budget overruns may be assessed as high priority because the item may have a serious financial impact on the entire project. After priority assessment, a conflict priority sequence is generated to list the processing priorities of all conflicting items. High-priority items in the sequence will be given priority. A conflict reconciliation sequence is generated based on the priority sequence and risk level to reallocate the execution order of conflicting items to ensure that there are no conflicts or inconsistencies in the audit rules during execution. Through the reconciliation mechanism, high-priority items are given priority, while low-risk conflicting items are merged or adjusted. For example, two rules that may cause conflict can avoid conflicts in actual execution by adjusting the execution order or changing the applicable conditions. The reconciliation results are fed back to the audit rule matching process to ensure that the updated rules and execution order can maintain consistency and coordination with other items in the audit rule matching.

[0114] Embodiment 2

[0115] Based on the same inventive concept as a method in the aforementioned embodiment, the present invention further provides a cost audit system based on big data, the system comprising:

[0116] The data integration module obtains the audit items included in the current engineering cost audit, and integrates the engineering cost audit project file data according to the audit items to generate multi-dimensional heterogeneous cost data;

[0117] The rule engine module establishes an audit rule engine, which performs compliance checks on audit items of the current project based on project type and audit specifications and standards;

[0118] The risk prediction module establishes a cost audit database based on multivariate and heterogeneous cost data and historical audit records, and establishes a risk prediction model based on the cost audit database and the audit rule engine to identify abnormal data and potential risk points to obtain risk prediction results;

[0119] The rule matching module, the risk prediction model performs audit rule matching based on the audit entry information contained in the historical audit records and the current project records and generates audit rule matching results;

[0120] The risk warning module analyzes the audit rule matching results and risk prediction results, generates risk warning information, and generates an audit analysis report based on the risk warning information.

[0121] The above-mentioned adjustment system in the present invention can be effectively implemented, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0122] Specifically, the rule engine module includes:

[0123] The priority adjustment unit establishes an audit specification set according to the type and characteristics of the engineering project. The audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications.

[0124] Verification specification unit, set up multi-layer verification and associated audit rules, which cross-verify audit items through audit specification sets according to the complexity of the project, analyze the correlation between audit items and multiple rule items, and generate linkage audit specification groups;

[0125] In accordance with the feedback unit, a rule execution feedback loop is constructed. The rule execution feedback loop verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specifications according to the compliance of the audit execution process data with the audit rules.

[0126] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0127] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A cost audit method based on big data, characterized in that: The method comprises: Acquire the audit items included in the current engineering cost audit, and integrate the engineering cost audit project file data according to the audit items to generate multivariate heterogeneous cost data; Establishing an audit rule engine, which performs audit item compliance checks on the current project according to the project type and audit specification standards; Establishing a cost audit database based on the multivariate and heterogeneous cost data and historical audit records, establishing a risk prediction model based on the cost audit database and the audit rule engine, identifying abnormal data and potential risk points to obtain risk prediction results; The risk prediction model performs audit rule matching according to the audit entry information contained in the historical audit records and the current project records and generates an audit rule matching result; Analyze the audit rule matching results and risk prediction results, generate risk warning information, and generate an audit analysis report based on the risk warning information; Establish an audit rules engine, including: Establishing an audit specification set according to the type and characteristics of the engineering project, the audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications; Setting multi-layer verification and associated audit rules, wherein the multi-layer verification and associated audit rules cross-verify the audit items through the audit specification set according to the complexity of the project, analyze the correlation between the audit items and multiple rule items, and generate a linkage audit specification group; Constructing a rule execution feedback loop, wherein the rule execution feedback loop verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specification according to the compliance of the audit execution process data with the audit rules; The correction specification layer learns and adjusts the execution priority of audit specifications based on historical audit data and project execution process feedback, including: Analyze the audit execution process data and extract the audit execution process pattern in the audit rule execution, and identify the audit execution interference factors that affect the audit results based on the audit execution process pattern; According to the audit interference factors and the feedback of the audit execution process data, the correction specification layer captures the changes in project progress, cost deviation and risk level, and estimates the project execution deviation; The project execution deviations are integrated with the historical audit data, and a multidimensional data analysis framework is constructed based on the correction specification layer, and the multidimensional data analysis framework is used to sort the execution priority of the audit specifications.

2. The cost audit method based on big data according to claim 1 is characterized in that: Establishing a risk prediction model based on the cost audit database and the audit rule engine includes: Integrate the multivariate and heterogeneous cost data in the cost audit database with the project data in the historical audit records, and generate a basic data set for risk analysis through data preprocessing and standardization; Extracting potential risk factors from the compliance check results of the current project through the audit rule engine, wherein the potential risk factors are identified based on the audit anomaly patterns in the cost audit database; Based on the historical audit data in the cost audit database, a risk prediction model is constructed to analyze the risk characteristics of historical projects. The risk factors of the current project are associated with similar risk characteristics in the historical data through the risk prediction model to perform correlation matching; Using the risk prediction model and the extracted risk factors, potential risk levels are calculated at different stages of project execution to generate risk prediction results.

3. The cost audit method based on big data according to claim 2 is characterized in that: The potential risk factors are identified based on the audit anomaly patterns in the cost audit database, including: Based on the historical audit data in the cost audit database, cluster analysis is performed on abnormal data in historical projects to identify common audit abnormal patterns in different types of projects; Extracting risk factors from the audit anomaly patterns, and classifying and rating the risk factors according to their impact on the project cost; During the audit execution process, the compliance check results of the current project are analyzed through the audit rule engine, and the abnormal data therein is compared with the audit abnormality pattern to identify the potential risk factors similar to historical projects.

4. The cost audit method based on big data according to claim 3 is characterized in that: The risk factors are classified and graded according to their impact on the project cost, including: Construct a risk impact analysis layer to quantify the impact of potential risk factors on the project in various audit exception patterns and generate corresponding impact coefficients; Classifying the extracted risk factors according to the corresponding impact coefficients, the classification being based on setting the thresholds of the impact coefficients and the different audit anomaly modes; Adding a corresponding risk level mark to each of the classified risk factors to indicate the potential threat level of the corresponding risk factor to the project cost, and adding the risk level mark to the risk prediction model; During the audit execution, the changes of the risk factors are monitored, the input data of the risk impact analysis layer is updated and the risk level mark is adjusted, and the risk prediction model is fed back according to the updated risk level.

5. The cost audit method based on big data according to claim 1 is characterized in that: Perform audit rule matching and generate audit rule matching results, including: According to the type and characteristics of the audit items, the audit items are matched by the audit rule engine, and the matching process is based on the multivariate heterogeneous cost data and historical audit records in the cost audit database; By means of the multi-layer verification and the associated audit rules, the association between the audit items and the historical audit records is analyzed, and the audit rules are associated and matched with the risk factors in the current audit items to generate an associated matching result; In the association matching process, rule conflict detection is performed to identify potential conflicts or overlapping parts between multiple audit entries, and conflicting entries are prioritized to generate a conflict reconciliation sequence; An audit rule matching result is generated according to the association matching result and the conflict reconciliation sequence, and the audit rule matching result is used as input data of the risk prediction model.

6. The cost audit method based on big data according to claim 5 is characterized in that: Perform rule conflict detection, including: Analyze the audit items in the current audit project one by one through the audit rule engine, and use the multivariate heterogeneous cost data in the cost audit database to detect potential conflicts between the audit items and the associated audit rules; According to the identified conflict items, classify and handle the conflict types, and classify and organize the conflict items to form a conflict report; Performing priority evaluation on the detected conflicting items through the multi-layer verification and associated audit rules to generate a conflict priority sequence, wherein the conflict priority sequence is classified according to the risk level of the conflicting items; A conflict reconciliation sequence is generated according to the conflict priority sequence combined with the risk level. The conflict reconciliation sequence is used to reallocate the execution order of the conflicting items, and the reconciliation result is fed back to the audit rule matching process.

7. The cost audit system based on big data is characterized by: Using the cost audit method based on big data as claimed in claim 1, the system includes: The data integration module obtains the audit items included in the current engineering cost audit, and integrates the engineering cost audit project file data according to the audit items to generate multi-dimensional heterogeneous cost data; The rule engine module establishes an audit rule engine, which performs compliance checks on audit items of the current project based on project type and audit specifications and standards; The risk prediction module establishes a cost audit database based on multivariate and heterogeneous cost data and historical audit records, and establishes a risk prediction model based on the cost audit database and the audit rule engine to identify abnormal data and potential risk points to obtain risk prediction results; The rule matching module, the risk prediction model performs audit rule matching based on the audit entry information contained in the historical audit records and the current project records and generates audit rule matching results; The risk warning module analyzes the audit rule matching results and risk prediction results, generates risk warning information, and generates an audit analysis report based on the risk warning information; The rule engine module comprises: The priority adjustment unit establishes an audit specification set according to the type and characteristics of the engineering project. The audit specification set includes a basic specification layer, a project definition specification layer, and a correction specification layer. The correction specification layer learns based on historical audit data and feedback from the project execution process and adjusts the execution priority of the audit specifications. Verification specification unit, set up multi-layer verification and associated audit rules, which cross-verify audit items through audit specification sets according to the complexity of the project, analyze the correlation between audit items and multiple rule items, and generate linkage audit specification groups; In accordance with the feedback unit, a rule execution feedback loop is constructed. The rule execution feedback loop verifies the audit execution process data based on multi-layer verification and associated audit rules, and adjusts the audit specifications according to the compliance of the audit execution process data with the audit rules.

Images