A project data monitoring method and system based on multi-source data fusion

Abstract

The application provides a project data monitoring method and system based on multi-source data fusion, the method comprising: obtaining project data from each project management system and performing standardization and vectorization; identifying a project type according to a project index vector; calculating a pre-evaluation score of the project according to the project index vector; constructing a quota pool and performing quota pool mapping according to the project type; calculating a post-evaluation coefficient of each department according to a post-evaluation score of a historical completed project; calculating a classified quota of each department in the quota pool according to the project type, the pre-evaluation score of the project and the post-evaluation coefficient of the department; performing project department quota matching according to the classified quota of each department; obtaining project data in real time during project execution, and re-performing quota allocation according to preset conditions; updating parameters in the calculation process according to a post-evaluation score of the project; and the system is used for implementing the method.

Description

Technical Field

[0001] This invention relates to a project data monitoring method and system, and more particularly to a project data monitoring method and system based on multi-source data fusion. Background Technology

[0002] In the process of corporate investment project management, project application data, historical investment data, revenue forecast data, actual revenue data, profit data, contract data, user scale data, project execution data, and post-evaluation data are typically stored in various systems, including project application systems, ERP systems, financial systems, CRM systems, contract management systems, project management systems, and manual spreadsheets. These different systems suffer from inconsistencies in field names, statistical periods, data formats, indicator definitions, and difficulties in tracing manual corrections. Consequently, investment project amount calculations often rely on manual aggregation, manual judgment, and offline spreadsheet calculations.

[0003] While existing enterprise investment project management systems can handle project application, approval processes, and investment ledger statistics, most remain at the level of process management and result recording. They lack the ability to automatically collect multi-source heterogeneous data, perform field mapping, indicator normalization, project type identification, and dynamic calculation of quotas. When discrepancies arise between actual and predicted revenue, project delays, abnormal project expenditures, changes in the policy environment, or the insertion of urgent projects during project execution, existing systems often require manual recalculation of project quotas, making it difficult to promptly identify the set of affected projects and complete incremental recalculation.

[0004] Furthermore, post-project evaluation results are typically stored as separate documents or manual scoring results, failing to be automatically written back into the subsequent quota calculation model. This results in the inability to convert historical project performance into model parameters. Pre-evaluation weights, departmental post-evaluation coefficients, project priority ranking rules, and quota adjustment logic cannot be continuously updated based on historical performance, causing a break in the data chain between project application, quota calculation, execution monitoring, and post-evaluation feedback.

[0005] Therefore, the existing technology has at least the following problems:

[0006] First, multi-source project data is scattered across different systems, lacking a unified data mapping and indicator generation mechanism;

[0007] Second, project data calculation relies on manual processing and cannot automatically generate project indicator vectors based on standardized data.

[0008] Third, the project types and pre-assessment indicators mainly rely on manual judgment, lacking calculable and traceable data processing procedures;

[0009] Fourth, when the deviation of project data exceeds the threshold, the system has difficulty automatically triggering a recalculation of the quota;

[0010] Fifth, the post-evaluation results cannot be automatically fed back to update the model parameters, and the quota calculation model lacks the ability to continuously optimize.

[0011] Furthermore, when project data changes, the existing system typically needs to reread all project records and repeatedly perform field transformations, indicator generation, priority calculations, and quota matching operations, making it impossible to determine the actual affected data range based on the triggering project. As the number of projects and data sources increases, the full-processing approach will increase the number of data reads, indicator calculations, and quota status updates.

[0012] Meanwhile, existing systems typically maintain project quotas, quota pool occupied quotas, and quota pool remaining quotas separately, lacking a unified status association and calculation log. When the status of multiple projects or project data changes continuously, problems such as out-of-sync project quota calculation results with quota pool status, inconsistent calculation parameter versions, and difficulty in reproducing historical calculation processes can easily occur.

[0013] Therefore, in addition to the rules for calculating project quotas, it is also necessary to solve data processing technical issues such as unified processing of multi-source heterogeneous project data, identification of the scope of affected data, local incremental calculation, synchronous update of quota pool status, and traceable access to calculation parameters. Summary of the Invention

[0014] Purpose of the invention: The technical problem to be solved by the present invention is to provide a project data monitoring method and system based on multi-source data fusion, which addresses the shortcomings of the existing technology.

[0015] To address the aforementioned technical problems, this invention discloses a project data monitoring method and system based on multi-source data fusion, the method comprising the following steps:

[0016] Step 1: Obtain project data from various project management systems and standardize and vectorize it to obtain project indicator vectors;

[0017] Step 2: Identify the project type based on the project indicator vector;

[0018] Step 3: Calculate the pre-project evaluation score based on the project indicator vector;

[0019] Step 4: Construct a credit limit pool and map the credit limit pool according to the project type;

[0020] Step 5: Calculate the post-evaluation coefficient for each department based on the post-evaluation scores of previously completed projects;

[0021] Step 6: Calculate the category quota for each department in the quota pool based on the project type, pre-project assessment score, and post-department assessment coefficient.

[0022] Step 7: Match project department quotas according to the quota categories of each department;

[0023] Step 8: Acquire project data in real time during project execution and reallocate the quota according to preset conditions;

[0024] Step 9: Update the parameters calculated in Steps 3 and 5 based on the post-project evaluation score.

[0025] The system is used to implement the above method.

[0026] Beneficial effects:

[0027] 1. By using unique project identifiers, field mapping, and time window alignment, project data from different business systems is converted into a unified project indicator vector, reducing redundant reading and conversion in subsequent calculations.

[0028] 2. Through project type identification, pre-assessment, departmental classification quota calculation, and project quota matching, a two-level quota calculation process that can be executed by computer is formed, and the departmental classification quota is constrained by the total amount of the corresponding quota pool.

[0029] 3. Determine the affected scope based on project execution deviations, and only recalculate the departmental classification quotas and project quotas within the affected scope, while retaining the original calculation results of unaffected projects, thereby reducing redundant calculations during the dynamic adjustment process.

[0030] 4. Update the department's post-evaluation coefficient and pre-evaluation dimension weights based on the post-evaluation results of completed projects, and record the data version, parameter version, and quota adjustment log to enable the quota calculation, dynamic recalculation, and parameter update process to be reviewed and traceable. Attached Figure Description

[0031] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0032] Figure 1 This is a schematic diagram of the overall process of the method of the present invention.

[0033] Figure 2 This is a system architecture diagram of the present invention.

[0034] Figure 3 This is a schematic diagram illustrating the process of standardizing and vectorizing multi-source project data.

[0035] Figure 4 This is a schematic diagram of the credit limit pool management structure.

[0036] Figure 5 This is a diagram illustrating the event triggering mechanism for credit limit recalculation.

[0037] Figure 6 This is a schematic diagram of the post-evaluation parameter write-back mechanism. Detailed Implementation

[0038] like Figure 1 As shown, this invention provides a project data monitoring method based on multi-source data fusion, including multi-source data acquisition and standardization, project type identification, pre-project assessment, quota pool construction, departmental classification quota calculation, project quota matching, deviation-triggered recalculation, and post-assessment parameter update. Specifically, it includes the following steps:

[0039] Step 1: Obtain multi-source project data

[0040] Project data is obtained from project application systems, ERP systems, financial systems, CRM systems, contract management systems, project management systems, and post-evaluation systems.

[0041] The project data includes at least the following: project application data, including project number, project name, applying department, application amount, initial label for project type, related products, and project construction goals; financial data, including historical investment amount, projected revenue, actual revenue, projected profit, actual profit, revenue contribution value, revenue of the previous period, and projected revenue of the current period; user operation data, including user scale, target user scale, user growth rate, and user penetration rate; project execution data, including project progress and expenditure progress; contract data, including contract amount, contract term, and contract constraint period; and post-evaluation data, including acceptance results, benefit achievement data, efficiency achievement data, and effect achievement data.

[0042] A unique project identifier is generated for each project, and the relationship between project application data, financial data, user operation data, project execution data, contract data, and post-evaluation data is established based on the unique project identifier.

[0043] The above data can also be obtained through data interfaces, database synchronization, batch import of tables, or manually verified structured data tables.

[0044] Step 2: As Figure 3 As shown, standardization and vectorization are performed on multi-source project data.

[0045] Based on the field mapping table, fields from different sources but with the same meaning are uniformly converted, missing values ​​and outliers are marked or corrected, and units are unified and time windows are aligned for monetary, proportional and time data.

[0046] The field mapping table should include at least the source system name, source field name, target field name, field type, field unit, time granularity, conversion rules, and validation rules. Data from different sources is uniformly converted based on the field mapping table, and the mapping relationship between source and target fields is preserved for subsequent review and traceability.

[0047] Set outlier identification rules, including at least the following: monetary indicators less than 0, project progress greater than 100%, expenditure progress greater than 100%, revenue growth exceeding a preset limit, deviation between project expenditure progress and project execution progress exceeding a preset percentage, and deviation between actual revenue and projected revenue exceeding a preset percentage. When data meets any of these outlier rules, the data is marked as outlier and written to a verification queue or anomaly log.

[0048] After data standardization, each project is converted into a project indicator vector. Project-related data is represented as a project indicator vector. This includes basic attribute vectors, financial indicator vectors, market indicator vectors, execution status vectors, and post-evaluation indicator vectors. Project indicator vectors. It includes the project's main data and derived metrics, among which, the project The main data definitions are shown in Table 1:

[0049] Table 1 Main Data Definition Table

[0050]

[0051] Project generated based on the above data Derived indicators:

[0052] For the project The income-to-investment ratio, used to measure the expected income generated per unit of investment, is calculated using the following formula:

[0053]

[0054] For the project The profit-to-investment ratio is used to measure the expected profit generated per unit of investment, and its calculation formula is as follows:

[0055]

[0056] Indicates project The projected payback period is used to measure the expected period required for a project to recover its investment. Its calculation formula is:

[0057]

[0058] in, To prevent extremely small positive numbers with a denominator of zero;

[0059] Indicates project The user growth rate is used to measure the expected contribution of a project to user growth, and its calculation formula is as follows:

[0060]

[0061] in, To prevent extremely small positive numbers with a denominator of zero;

[0062] Indicates project The user penetration rate is calculated using the following formula:

[0063]

[0064] in, To prevent extremely small positive numbers with a denominator of zero;

[0065] Indicates project The schedule deviation rate is used to measure the degree of deviation between the project expenditure schedule and the actual construction schedule. Its calculation formula is:

[0066]

[0067] For the project The projected revenue growth rate of related products or business capabilities will be used in subsequent quota calculations.

[0068]

[0069] To generate the pre-assessment score, the system also acquires or generates the following pre-assessment auxiliary data, including: Indicates project The strategic tag matching degree is obtained by matching the project construction goals, the initial tags of the project type, the related products and the enterprise's strategic tag library; Indicates project The efficiency improvement rate is calculated from data such as process duration, man-hours, and resource utilization before and after project implementation, or estimated based on historical data of similar projects. Indicates project The technology maturity level is generated by combining technology maturity data, the acceptance results of similar historical projects, and the complexity of technology implementation. Indicates project The risk level is generated by combining compliance risk, budget deviation risk, delay risk, and contractual constraint risk. Indicates project The degree of ecosystem synergy is generated by a combination of factors, including the number of partners, the number of platforms connected, the number of API calls, the number of upstream and downstream collaborative objects, and the ability to expand business in the future. Indicates project The urgency level is generated by the remaining time bound by the contract, the required completion time, external event indicators, and the project risk level.

[0070] in, , , , , and It can be calculated directly from structured data, or it can be generated by tagging project application texts, contract information, historical project databases, and corporate strategic tag databases.

[0071] By converting project data from different sources into a unified project indicator vector at once, subsequent project type identification, pre-assessment, quota calculation, and dynamic recalculation all read data based on this project indicator vector. This eliminates the need to access the original business system separately and repeatedly perform field conversions in each calculation step, thereby reducing cross-system data reading and repeated conversion operations.

[0072] To standardize indicators with different dimensions, each indicator will be normalized. For indicators with higher numerical values ​​and higher evaluation, a positive normalization function will be used. :

[0073]

[0074] For indicators where smaller values ​​equate to higher evaluation, an inverse normalization function is used. :

[0075]

[0076] in, and Each is a set of evaluations for the same project. The maximum and minimum values ​​of the corresponding indicators. To prevent extremely small positive numbers with a denominator of zero.

[0077] Step 3: Identify Project Type

[0078] Based on project indicator vectors Identify project types and classify them into capacity-building projects, support projects, and forward-looking research projects.

[0079] First, the validity of a project is determined based on its status field. If the project status is abandoned, canceled, or invalid, the project is marked as invalid and will not proceed to the project type identification, pre-assessment sorting, and quota matching processes. If the project status is valid, the project type identification process continues.

[0080] For projects with valid status, based on the project indicator vector Generate project type characteristics. Project type characteristics include revenue contribution characteristics. User growth characteristics Capability Support Characteristics Technical pre-research characteristics .

[0081] Income contribution characteristics Based on projected revenue Profit forecast and revenue contribution value Generate data to characterize the expected contribution of a project to revenue and profit.

[0082]

[0083] in, , , Weighting coefficients for revenue contribution characteristics, and + =1, This is the positive normalization function.

[0084] User growth characteristics Based on the current user scale Target user scale User growth rate and user penetration rate Generates data to characterize a project's contribution to user growth.

[0085]

[0086] in, , , For user growth feature weight coefficients, and + =1.

[0087] Capability Support Features Based on project construction goals, initial project type labels, and revenue contribution value Efficiency improvement rate Generation is used to characterize the project's supporting role in platform construction, operation and maintenance, quality improvement, or capability foundation.

[0088]

[0089] in, The supporting attribute label scores are generated based on the initial labels of project construction goals and project type. , , For capability-supporting feature weight coefficients, and + + =1, This represents the inverse normalization function.

[0090] Technical Pre-research Characteristics Based on technology maturity Pilot verification attributes, project construction goals, and strategic tag matching degree Generates attributes that characterize a project's technological research, innovation incubation, or future capability reserves.

[0091]

[0092] in, The pilot verification label score is generated based on the pilot verification attributes or project construction goals. , , For the technical pre-research feature weight coefficients, and + + =1.

[0093] Subsequently, the items were calculated separately. Type matching degree between capacity-building projects, support projects, and forward-looking research projects:

[0094]

[0095]

[0096]

[0097] in, For capacity matching degree, To support class matching degree, For forward-looking research on matching degree, , Match weighting coefficients to capacity categories, and + =1.

[0098] when When the first type threshold is met, the project will be... Identified as a capacity-related project; when Meets the second type threshold and revenue contribution value When the revenue contribution falls below the preset threshold, the project will be... Identified as a support project; when When the third type threshold is met, the project will be... It was identified as a forward-looking research project.

[0099] When the same project meets multiple project type identification conditions simultaneously, the final project type is determined based on the type matching degree, the initial project type label, and the preset priority rules, and the project type identification result is written into the project indicator vector. .

[0100] Step 4: Generate pre-project evaluation indicators

[0101] Based on project indicator vectors Generate Project Pre-assessment score The aforementioned pre-assessment score It includes six dimensions: strategic alignment, expected benefits, efficiency improvement, technological feasibility, risk controllability, and ecological impact.

[0102] (1) Strategic fit

[0103] Strategic fit score According to the project Strategic tag matching degree generate:

[0104] = This is used to convert the degree of alignment between a project and the company's strategic direction into a standardized score. The higher the degree of alignment, the better. The higher.

[0105] (2) Expected benefit score

[0106] Expected benefit score Based on the income-to-investment ratio Profit-to-investment ratio Expected investment payback period and revenue contribution value generate:

[0107]

[0108] in, , , , For the expected benefit dimension, the weighting coefficients are as follows: .

[0109] (3)

[0110] Performance Improvement Score Based on efficiency improvement rate generate:

[0111] = This is used to measure the degree to which a project improves internal operational efficiency or resource utilization efficiency; the higher the efficiency improvement rate, the better. The higher.

[0112] (4) Technical feasibility score

[0113] Technical feasibility score Based on technology maturity generate:

[0114] = The higher the technological maturity, the higher the acceptance rate of similar historical projects, and the lower the implementation complexity, The higher, the corresponding The higher.

[0115] (5) Risk controllability score

[0116] Risk controllability score According to risk level and schedule deviation rate generate:

[0117]

[0118] in, For the project The risk level, For the project Schedule deviation rate; , For the weighting coefficients of the risk controllability dimension, and .

[0119] (6) Ecological driving force score

[0120] Ecological driving force score Based on ecological synergy generate:

[0121] = The higher the degree of ecological synergy, The higher.

[0122] (7) Basic Pre-Assessment Score

[0123] Calculate the project score based on six dimensions. Basic pre-assessment score :

[0124]

[0125] in, , , , , , These are the weight coefficients for the six dimensions, and + + + + + =1.

[0126] (8) Pre-assessment score

[0127] Calculate the project's pre-assessment score based on the scores of each dimension and their corresponding weights:

[0128]

[0129] in, This is a correction factor for strategic urgency.

[0130] The strategic urgency adjustment coefficient is determined based on strategic alignment and urgency level. When a project is classified as high-strategy and high-urgency, the adjustment coefficient is greater than 1; when a project is classified as low-strategy and low-urgency, the adjustment coefficient is less than 1.

[0131] Will Write project indicator vector It is used for subsequent quota calculation, comprehensive priority index calculation and project dynamic matching.

[0132] Step 5: Construct the credit pool data structure, where the relationships between the regular credit pool, flexible credit pool, capacity-related credit pool, support-related credit pool, forward-looking research credit pool, and emergency adjustment credit pool are as follows: Figure 4 As shown, the details are as follows:

[0133] Based on the total annual investment Create a credit limit pool data table. This table records the credit limit status and project mapping relationships for different types of credit limit pools. First, write the regular credit limit pool into the credit limit pool data table. and the flexible quota pool Two types of credit limit pool records, including the regular credit limit pool. Centralized matching for regular projects, flexible credit pool Used for emergency projects, strategic insertion projects, and mid-year dynamic adjustments.

[0134]

[0135] In the regular credit pool Next, the system further constructs a capacity-based quota pool based on the project types identified in step 3. and support quota pool Among them, the capacity-related quota pool Used to match capacity-related projects such as revenue growth, user growth, and business expansion; supporting quota pools. It is used to support projects such as platform construction, operation and maintenance, quality improvement, and capability foundation.

[0136]

[0137] In the flexible credit pool Next, the system will further construct a forward-looking pre-research quota pool. and emergency adjustment quota pool Among them, the forward-looking research quota pool Used for matching forward-looking research projects such as technology pre-research, innovation incubation, and pilot verification; emergency adjustment quota pool Used to match sudden needs, urgent tasks, or strategic insertion projects added midway through the year.

[0138]

[0139] Based on the project type identification results and the comprehensive priority index, project mapping rules are established to automatically map projects to the corresponding credit pools. When a project meets the matching conditions of multiple credit pools simultaneously, the system determines its credit pool according to the project's primary objective, type matching degree, and preset priority rules.

[0140] Each quota pool has fields such as quota pool number, quota pool type, initial quota, occupied quota, remaining quota, frozen quota, and adjustable quota. The system updates these fields in real time based on quota matching, quota freezing, project release, dynamic adjustment, and incremental recalculation results, and uses the quota pool status for subsequent quota calculation, dynamic matching, and recalculation adjustments.

[0141] Step 6: Calculate the post-evaluation coefficients

[0142] The department's post-evaluation coefficient is calculated based on historical project post-evaluation data. Post-evaluation coefficient It can be represented as:

[0143]

[0144] in, For the department The average post-evaluation score of all projects in the previous period, The average post-evaluation score of all projects in the previous cycle for the company. To adjust the sensitivity coefficient, the value range is [0, 0.5]. The larger the value, the stronger the impact of historical performance on the department's quota. The smaller the amount, the more stable the department's quota; when the department has no valid historical projects. Take 1.

[0145] Post-project evaluation score It is calculated from three dimensions: benefit achievement, effectiveness achievement, and result achievement.

[0146]

[0147] in, For the project The score for achieving the desired effect For the project The performance score was achieved. For the project Achieve the desired score. , , These are the weighting coefficients. .

[0148] Calculated departmental post-evaluation coefficients Write the parameters into the model parameter table and use them as input parameters for calculating departmental classification quotas in step 7, which indirectly affects the project quota matching results in step 8.

[0149] Step 7: Calculate the quota for each department.

[0150] Based on the department's historical investment amount of similar types, the department's projected revenue growth rate, the department's post-evaluation coefficient, and the average pre-evaluation score of the department's submitted projects, the respective quotas for each department in the capacity-related quota pool, the support-related quota pool, and the forward-looking research quota pool are calculated.

[0151] in, Indicates the current department. To prevent extremely small positive numbers with a denominator of zero.

[0152] (1) Calculation of quotas for capacity-related departments

[0153] department Production capacity quota It can be represented as:

[0154]

[0155] in, For the department The initial quota for capacity-related projects. For the department The actual investment amount in capacity-related projects in the previous cycle, For the department Current period forecast revenue growth This represents the company's total capacity-related investment in the previous cycle. This represents the company's projected revenue growth rate for the current period. For the department The post-evaluation coefficient, This refers to the amount in the capacity-related quota pool. For the department The average pre-assessment score of the submitted projects. The average pre-evaluation score of all projects submitted by the company. This is the adjustment intensity coefficient for capacity category pre-assessment, with a value range of [0, 0.5].

[0156] If department If there was no actual investment in capacity-related projects in the previous cycle, then... You can use a preset base investment value or the historical average investment value of similar departments.

[0157] (2) Calculation of quota for supporting departments

[0158] department Supporting quota It can be represented as:

[0159]

[0160] in, For the department The initial quota for supporting projects, For the department The amount of investment in support projects in the previous cycle, This represents the total investment in support projects for the company in the previous cycle. To support the amount in the credit limit pool, The adjustment strength coefficient for the pre-evaluation of the support class is set, with a value range of [0, 0.5].

[0161] If department If there was no actual investment in supporting projects in the previous period, then... You can use a preset base investment value or the historical average investment value of similar departments.

[0162] (3) Calculation of quota for forward-looking research departments

[0163] department Forward-looking research quota It can be represented as:

[0164]

[0165] in, For the department The quota for forward-looking research projects For the department The strategic adjustment coefficient ranges from [0.5, 1.5]. The amount of the forward-looking research quota pool, For the department Average pre-evaluation score for forward-looking research projects It is the sum of the average pre-evaluation scores of all forward-looking research projects across all departments.

[0166] Step 8: Calculate project limits and match them.

[0167] After obtaining the departmental quotas in step 7, the specific project quotas are calculated and matched based on the project's department, project type, pre-project assessment score, project revenue forecast growth rate, urgency score, and strategic fit score.

[0168] in, Indicates the current project. Indicates project Department To prevent extremely small positive numbers with a denominator of zero.

[0169] (1) Calculate the project's overall priority index

[0170] project Comprehensive priority index It can be represented as:

[0171]

[0172] in, For the project The overall priority index, For the project Pre-assessment score, For the project Urgency rating, For the project Strategic fit score , , These are the weighting coefficients. .

[0173] (2) Calculation of quota for capacity-related projects

[0174] If the project Belongs to department For capacity-related projects, then the project Allocation factors for capacity-related projects for:

[0175]

[0176] in, For the project Projected revenue growth rate for related products or business capabilities.

[0177] project Quota for capacity-related projects It can be represented as:

[0178]

[0179] in, The department calculated in step 7 Production capacity quotas; For the department The sum of allocation factors for all candidate projects in the capacity category.

[0180] (3) Calculation of the amount of supporting projects

[0181] If the project Belongs to department For supporting projects, then the project Supporting project allocation factors for:

[0182]

[0183] project Supporting project quota It can be represented as:

[0184]

[0185] in, The department calculated in step 7 Support-type credit limits; For the department The sum of the allocation factors for all supporting candidate projects.

[0186] (4) Calculation of quota for forward-looking research projects

[0187] If the project Belongs to department Forward-looking research projects, then the project Allocation factors for forward-looking research projects for:

[0188] = *

[0189] in, For the project The strategic adjustment coefficient.

[0190] project Quota for forward-looking research projects It can be represented as:

[0191]

[0192] in, The department calculated in step 7 Forward-looking research quotas; For the department The sum of the allocation factors for all forward-looking research candidate projects.

[0193] (5) Generate project quota matching results

[0194] project Calculation limit Determined based on project type. When the project... When it is a capacity-related project, When the project When for support projects, When the project For forward-looking research projects, .

[0195] project Actual matching amount for:

[0196]

[0197] when When the priority level is higher than the preset threshold and there is a credit limit shortfall:

[0198]

[0199]

[0200] in, For the project The amount declared.

[0201] When the project When the calculated amount exceeds the declared amount, the declared amount will be used as the actual matching amount, and the difference will be written into the adjustable amount; when the project When the calculated amount is less than or equal to the declared amount, the calculated amount shall be used as the actual matching amount.

[0202] For projects that are abandoned, canceled, or invalid, no quota matching will be performed, and the corresponding quota will be written into the frozen quota or the adjustable quota.

[0203] Step 9: Trigger dynamic recalculation of credit limit, such as... Figure 5 As shown, a recalculation event is generated when the preset project data deviation exceeds a threshold, and the set of affected projects is determined, as detailed below:

[0204] During project execution, the system continuously monitors execution data such as actual revenue, projected revenue, actual profit, projected profit, project progress, and expenditure progress, and calculates project deviation indicators based on the above data.

[0205] project Income deviation rate for:

[0206]

[0207] in, For the project Actual income For the project Projected revenue.

[0208] project Profit deviation rate for:

[0209]

[0210] in, For the project The actual profit For the project The projected profit.

[0211] project The schedule deviation rate is .

[0212] when , or When any deviation indicator exceeds the corresponding preset threshold, or when the project status changes to cancel, abandon, pause or other preset status, a quota recalculation event is generated.

[0213] The scope of impact is determined based on the department to which the triggering project belongs, the project type, the quota pool to which it belongs, related products, and project dependencies.

[0214] When a recalculation event affects the departmental classification quota calculation parameters, steps 7 and 8 are re-executed for the affected departments; when a recalculation event only affects the project status, project priority, or project allocation factor within a department, only step 8 is re-executed. Projects not within the affected scope retain their original quota calculation results.

[0215] Recalculated Project credit limit It can be represented as:

[0216]

[0217] in, This is a recalculated limit; For the project Recalculated project allocation factors; This is the sum of the project allocation factors after recalculation of all candidate projects within the affected area.

[0218] Step 10: Update the model parameters based on the post-evaluation results, such as... Figure 6 As shown, the post-project evaluation results are used to update the department's post-evaluation coefficient and the weights of the pre-evaluation dimensions, and are then used for the next cycle's quota calculation, as detailed below:

[0219] Upon project completion, the system collects actual project execution data, including actual revenue, actual profit, acceptance results, delays, schedule deviation rate, and efficiency improvement data. Based on this data, the system calculates a post-project evaluation score. The department's post-evaluation coefficient and pre-evaluation dimension weights are updated based on the project's post-evaluation score.

[0220] Updated departmental post-evaluation coefficients It can be represented as:

[0221]

[0222] in, The sensitivity coefficient is adjusted based on the post-evaluation coefficient; For the department Average post-evaluation score of projects completed in the previous cycle; This represents the average post-evaluation score of projects completed by the company in the previous cycle.

[0223] No. Temporary updated weights for each pre-assessment dimension for:

[0224]

[0225] in, For the first Weights of each pre-evaluation dimension before update; The learning rate is [0,1], preferably in the range of [0.01,0.20]. When the value is smaller, the weight changes are stable but the response is slower, while when the value is larger, it is more sensitive to recent projects but fluctuates more. For the first The weight adjustment amount for each pre-evaluation dimension.

[0226] Weight adjustment It can be represented as:

[0227]

[0228] in, For the project In the Dimensional scores on each of the pre-assessment dimensions; For the project The sum of the scores for each pre-assessment dimension.

[0229] To ensure that the sum of the weights of each pre-evaluation dimension is 1, the temporary update weights are normalized:

[0230]

[0231] in, For the first The updated weights of each pre-evaluation dimension; The sum of the temporary updated weights for each of the pre-evaluation dimensions.

[0232] Updated departmental post-evaluation coefficients and pre-evaluation dimension weights Write the parameters to the model parameter table. Simultaneously generate parameter update records, which include the value before update, the value after update, the relevant project, the update time, and the reason for the update.

[0233] When calculating departmental quotas, project quotas, comprehensive priority ranking, and dynamic matching in the next cycle, the quota calculation model calls the updated departmental post-evaluation coefficients and pre-evaluation dimension weights, thus forming a closed-loop update mechanism of "project execution - post-evaluation calculation - parameter write-back - next cycle call".

[0234] This invention also provides a project data monitoring system based on multi-source data fusion, such as... Figure 2 As shown, it includes the following modules.

[0235] 1. Multi-source data acquisition module: Used to acquire project-related data from project application system, ERP system, financial system, CRM system, contract management system, project management system and post-evaluation system.

[0236] 2. Data Standardization Module: Used to perform field mapping, format conversion, missing value handling, outlier identification, unit standardization, and time window alignment on project-related data, and generate project indicator vectors.

[0237] 3. Project Identification Module: Used to identify project types based on project indicator vectors.

[0238] 4. Pre-assessment indicator generation module: This module generates pre-assessment scores for six dimensions based on project indicator vectors: strategic fit, expected benefits, performance improvement, technical feasibility, risk controllability, and ecological impact.

[0239] 5. Quota Pool Management Module: Used to build and maintain regular quota pools, flexible quota pools, capacity-related quota pools, support quota pools, forward-looking research quota pools, and emergency adjustment quota pools.

[0240] 6. Quota Calculation and Matching Module: Used to calculate departmental category quotas, project quotas, basic matching quotas, and final matching quotas.

[0241] 7. Event Trigger Module: Used to monitor the deviations between actual revenue, predicted revenue, actual profit, predicted profit, project progress, and expenditure progress, and generate a recalculation event when the deviation exceeds a preset threshold.

[0242] 8. Incremental Recalculation Module: Used to monitor project execution deviations, determine the affected scope, and perform incremental recalculation of departmental classification quotas and / or project quotas within the affected scope;

[0243] 9. Post-evaluation feedback module: Used to generate post-evaluation scores based on the actual execution data after project completion, and update the department's post-evaluation coefficients and pre-evaluation dimension weights.

[0244] 10. Log Traceability Module: Used to record the entire process of data collection, indicator generation, quota calculation, quota matching, quota recalculation and parameter update, forming a traceable record of project quota processing.

[0245] Example

[0246] The following example, using the matching of annual investment project quotas of a digital lifestyle company, illustrates the method of this invention. The data in this embodiment is only used to illustrate the processing flow of this invention and does not constitute a limitation on the scope of protection of this invention.

[0247] 1. Multi-source data acquisition

[0248] The system extracts annual investment project data from project application data, financial data, project execution data, and post-evaluation data, using a subset of projects from the project ledger as representative projects. Specifically, the "serial number" in the project ledger is converted into a unique project identifier (Project_ID), and the "approved amount (excluding tax)" is used as the project application amount. The process status is uniformly converted into a project status field, which is used for subsequent quota matching, freezing or release judgment.

[0249] 2. Data standardization processing

[0250] The system performs field mapping, unit unification, and time window alignment on data from different sources. For monetary fields, the system uniformly converts them to ten thousand yuan; for status fields, the system converts them into calculable execution status fields according to preset rules; for records with empty fields, " / ", "waste", or "cancelled", the system handles missing values, marks invalid status, or marks canceled status, respectively.

[0251] In one exemplary data structure, the system can generate the following basic data table for the project, as shown in Table 2:

[0252] Table 2 Project Basic Data Table

[0253]

[0254] After the above processing, the system generates a project indicator vector for each project. The project indicator vector It includes a basic attribute vector, a financial indicator vector, a market indicator vector, an execution status vector, and a post-evaluation indicator vector. The basic attribute vector includes the department, project name, and initial project type labels; the financial indicator vector includes the declared amount, historical investment amount, projected revenue, and projected profit; the market indicator vector includes user scale, target user scale, and user penetration rate; the execution status vector includes project progress, expenditure progress, and project status; and the post-evaluation indicator vector is used to record the benefits, effectiveness, and achievement of the project after completion.

[0255] 3. Project type identification

[0256] The system is based on project indicator vectors The system categorizes projects. Specifically, it calculates the type matching degree between a project and capacity-related, support-related, and forward-looking research-related projects based on initial project type tags, project name, applying department, revenue contribution characteristics, user growth characteristics, and project construction goals. The final project type is then determined based on the type matching degree and preset priority rules.

[0257] Among them, P019 "AI and Ultra-HD Capabilities" and P022 "Digital Operation Platform" are related to business capability upgrades, user development, and revenue growth, and are identified as capacity-related projects; P009 "Dynamic Defense System" is mainly used for security assurance and platform support, and is identified as a support-related project; P030 "Robotics Laboratory" and P042 "AI Platform Multimodal" involve laboratory capability building, AI capability building, or cutting-edge technology verification, and are identified as forward-looking research projects.

[0258] For records with a project status of "abandoned", such as P059 "Beautiful Home Lab", the system marks them as invalid projects and will not enter the normal quota matching process. Their corresponding quota will enter the freezing or release process.

[0259] 4. Generation of Pre-Evaluation Indicators

[0260] The system is based on project indicator vectors Pre-project assessment score generation Among them, strategic alignment is generated based on the matching results between the project construction goals and the company's strategic labels; expected benefits are generated based on projected revenue, projected profit, revenue contribution value, and declared amount; efficiency improvement is generated based on the expected improvement in process efficiency, resource utilization, or reduction in man-hours after project implementation; technical feasibility is generated based on technology maturity and the implementation results of similar projects in the past; risk controllability is generated based on schedule deviation rate, project delay risk, and compliance risk; and ecosystem driving force is generated based on the degree to which the project drives platform capabilities, partners, or subsequent business expansion.

[0261] In one exemplary calculation result, the system can generate the following pre-evaluation results, as shown in Table 3:

[0262] Table 3 Pre-assessment Results

[0263]

[0264] The system calculates a basic pre-assessment score based on scores across six dimensions and preset weights, and then generates a final pre-assessment score by incorporating a strategic urgency correction coefficient. For projects with a status of abandoned, canceled, or invalid, the system will no longer proceed to the normal pre-assessment sorting and quota matching process.

[0265] 5. Credit pool construction

[0266] The system is based on the total annual investment. Establish a credit limit pool management structure. First, the system divides the total annual investment amount into a regular credit limit pool. and the flexible quota pool Subsequently, a capacity-based quota pool was constructed under the regular quota pool. and support quota pool Construct a forward-looking pre-research quota pool under the flexible quota pool. and emergency adjustment quota pool .

[0267] In this embodiment, the credit pool can be divided according to the total annual investment of 58 million yuan. Among them, the regular credit pool... It accounts for 80%, or 46.4 million yuan; the flexible quota pool This accounts for 20%, or 11.6 million yuan. The system maps each project to its corresponding quota pool based on the project type identification results.

[0268] 6. Initial credit limit calculation

[0269] The system calculates the quota for each department based on the project type, departmental post-evaluation coefficient, and pre-evaluation score, respectively by calling the quota calculation models for capacity, support, and forward-looking research categories.

[0270] Among them, P009 belongs to the support category, with an application amount of 6.6 million yuan; P019 and P022 belong to the capacity category, with application amounts of 2.3474 million yuan and 4.7094 million yuan respectively; P030 and P042 belong to the forward-looking research category, with application amounts of 5.3459 million yuan and 1.58 million yuan respectively. The above projects are matched from the corresponding category quotas of their respective departments; P059 is not included in the normal quota calculation because its status is abandoned.

[0271] 7. Comprehensive priority calculation and quota matching

[0272] The system is based on the pre-project assessment score. The urgency level indicator and strategic alignment score are used to calculate the comprehensive priority index. The system sorts projects based on their comprehensive priority index. It prioritizes matching quotas from the corresponding quota pool of the department to which the project belongs; if a gap still exists and the project's comprehensive priority index is higher than a preset threshold, the system matches quotas from the contingency quota pool.

[0273] In this embodiment, the system can generate the following credit limit matching results, as shown in Table 4:

[0274] Table 4. Credit Limit Matching Results

[0275]

[0276] The system generates a quota matching result table, which records the project number, quota pool to which each project belongs, matched quota, matching time, triggering rules, and matching basis for each project.

[0277] 8. Deviation monitoring and incremental recalculation

[0278] During project execution, the system continuously monitors deviations between projected revenue, projected profit, project progress, and expenditure progress. When the deviations in revenue forecast, profit forecast, progress, or expenditure exceed preset thresholds, the system generates a credit limit recalculation event. For existing project status fields in the project ledger, the system can use statuses such as "abandoned," "cancelled," or "invalid" as event identifiers to trigger credit limit freezing or release.

[0279] Taking the P059 Beautiful Home Laboratory Construction Project as an example, the project application amount was 500,000 yuan, the project status was abandoned, the system generated a quota release or freeze event, removed the project from the normal quota matching process, and transferred the corresponding quota to the adjustable quota or frozen quota processing process.

[0280] After a credit limit recalculation event is generated, the system determines the set of affected projects based on the triggering project number, department, credit limit pool, project type, associated products, and project priority. The system only performs incremental recalculation on the credit limit, remaining credit limit in the credit pool, and project priority order for projects within the affected project set, and does not perform a full recalculation on all projects. After the recalculation is completed, the system generates the credit limit before adjustment, the credit limit after adjustment, the reason for adjustment, and an adjustment log for subsequent review and auditing.

[0281] 9. Post-evaluation feedback and parameter updates

[0282] Upon project completion, the system retrieves actual execution results, acceptance results, schedule deviation rate, delay status, and efficiency improvement results from the financial system, project management system, and post-evaluation system, and generates a project post-evaluation score. .

[0283] The system is based on the post-project evaluation score. Updated departmental evaluation coefficients And based on the pre-project assessment score Post-evaluation score Deviation between them, pre-correction evaluation dimension weights The updated departmental post-evaluation coefficients and pre-evaluation dimension weights will be used in the initial quota calculation, priority sorting, and dynamic matching of the next cycle.

[0284] In this embodiment, when the execution data of a project exceeds a preset deviation threshold, the system does not reload and calculate all project data. Instead, it reads the corresponding affected project records based on the triggering project number, department, project type, and quota pool, and recalculates the allocation factor, project quota, and remaining quota pool quota for the affected projects. Projects not included in the affected scope retain their original calculation results. Therefore, the number of data records processed in the dynamic recalculation process depends on the number of affected projects, rather than always on the total number of projects submitted annually.

[0285] Through the above processing, this embodiment realizes a closed-loop data processing process from multi-source data collection, project indicator generation, quota calculation, dynamic matching, deviation-triggered recalculation to post-evaluation parameter write-back.

[0286] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a project data monitoring method and system based on multi-source data fusion, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0287] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MCU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0288] This invention provides a method and system for project data monitoring based on multi-source data fusion. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A project data monitoring method based on multi-source data fusion, characterized in that, Includes the following steps: Step 1: Obtain project data from various project management systems and standardize and vectorize it to obtain project indicator vectors; Step 2: Identify the project type based on the project indicator vector; Step 3: Calculate the pre-project evaluation score based on the project indicator vector; Step 4: Construct a credit limit pool and map the credit limit pool according to the project type; Step 5: Calculate the post-evaluation coefficient for each department based on the post-evaluation scores of previously completed projects; Step 6: Calculate the category quota for each department in the quota pool based on the project type, pre-project assessment score, and post-department assessment coefficient. Step 7: Match project department quotas according to the quota categories of each department; Step 8: Acquire project data in real time during project execution and reallocate the quota according to preset conditions; Step 9: Update the parameters calculated in Steps 3 and 5 based on the post-project evaluation score.

2. The project data monitoring method based on multi-source data fusion according to claim 1, characterized in that, The project indicator vector mentioned in step 1 includes: The vectors consist of basic attribute vectors, financial indicator vectors, market indicator vectors, execution status vectors, and post-evaluation indicator vectors; among which, the project indicator vector... Specifically, it includes: project The declared amount Historical investment amount Forecasted income Actual income Forecast profits Actual profit Revenue contribution value Revenue from the previous period for related products or business capabilities User scale Target user scale Project progress expenditure progress Contract amount Acceptance results ; project Income-to-investment ratio Profit-to-investment ratio Expected investment payback period User growth rate User penetration rate Schedule Deviation Rate Revenue growth forecast for related products or business capabilities ; project Strategic tag matching degree Efficiency improvement rate Technology maturity Risk level Ecological synergy urgency .

3. The project data monitoring method based on multi-source data fusion according to claim 2, characterized in that, Step 2, which involves identifying project types based on project indicator vectors, includes: Step 2-1: Determine the validity of the project based on its status. If the project status is abandoned, canceled, or invalid, mark the project as invalid; otherwise, continue with project type identification. Step 2-2, based on the project indicator vector Generate project type characteristics; among which, project type characteristics include revenue contribution characteristics. User growth characteristics Capability Support Characteristics Technical pre-research characteristics The specific calculation method is as follows: Income contribution characteristics Based on projected revenue Profit forecast and revenue contribution value The generation process is as follows: ； in, , , Weighting coefficients for revenue contribution characteristics, and + =1, It is a positive normalization function; User growth characteristics Based on the current user scale Target user scale User growth rate and user penetration rate The generation process is as follows: ； in, , , For user growth feature weight coefficients, and + =1; Capability Support Features Based on project construction goals, initial project type labels, and revenue contribution value Efficiency improvement rate The generation process is as follows: ； in, The supporting attribute label scores are generated based on the initial labels of project construction goals and project type. , , For capability-supporting feature weight coefficients, and + + =1, This represents the inverse normalization function; Technical Pre-research Characteristics Based on technology maturity Pilot verification attributes, project construction goals, and strategic tag matching degree The generation process is as follows: ； in, The pilot verification label score is generated based on the pilot verification attributes or project construction goals. , , For the technical pre-research feature weight coefficients, and + + =1; Steps 2 and 3: Calculate the items respectively. The type matching degree between the project and the projects is as follows: (Specific details are provided below.) ； ； ； in, For capacity matching degree, To support class matching degree, For forward-looking research on matching degree, , Match weighting coefficients to capacity categories, and + =1; when When the first type threshold is met, the project will be... Identified as a capacity-related project; when Meets the second type threshold and revenue contribution value When the revenue contribution falls below the preset threshold, the project will be... Identified as a support project; when When the third type threshold is met, the project will be... Identified as a forward-looking research project; Steps 2-4: When the same project meets multiple project type identification conditions simultaneously, the final project type is determined based on the type matching degree, the initial project type label, and the preset priority rules, and the project type identification result is written into the project indicator vector. .

4. The project data monitoring method based on multi-source data fusion according to claim 3, characterized in that, Step 3, which involves calculating the pre-project evaluation score based on the project indicator vector, includes: Step 3-1, according to the project Strategic tag matching degree Calculate the strategic fit score The details are as follows: ； Step 3-2, based on the income-to-investment ratio Profit-to-investment ratio Expected investment payback period and revenue contribution value Calculate the expected benefit score The details are as follows: ； in, , , , For the expected benefit dimension, the weighting coefficients are as follows: ; Step 3-3, based on the efficiency improvement rate Improved computational performance score , means as follows: ； Steps 3-4, based on technology maturity Calculation technology feasibility score The details are as follows: ； Steps 3-5, based on risk level and schedule deviation rate Calculate the risk controllability score The details are as follows: ； in, , For the weighting coefficients of the risk controllability dimension, and ; Steps 3-6, based on ecological synergy Calculate the ecological driving force score The details are as follows: ； Steps 3-7, calculate the project Basic pre-assessment score The details are as follows: ； in, , , , , , These are the weight coefficients for the six dimensions, and + + + + + =1; Steps 3-8: Calculate the pre-assessment score , means as follows: ； in, Adjustment coefficient for strategic urgency; Steps 3-9, will Write project indicator vector .

5. A project data monitoring method based on multi-source data fusion according to claim 4, characterized in that, The construction of the credit pool described in step 4 includes: Step 4-1, based on the total annual investment. Establish a credit limit pool data table to record the credit limit status and project mapping relationship of different types of credit limit pools; Step 4-2: Write the regular credit limit pool into the credit limit pool data table. and the flexible quota pool Two types of credit limit pool records; Step 4-3, in the regular credit limit pool Next, based on the project types identified in step 3, construct a capacity-based quota pool. and support quota pool ; Step 4-4, in the flexible credit pool Next, construct a forward-looking research and development quota pool. and emergency adjustment quota pool ; Steps 4-5: Based on the project type identification results, map the projects to the corresponding credit pools; Each credit pool has fields for credit pool number, credit pool type, initial credit limit, occupied credit limit, remaining credit limit, frozen credit limit, and adjustable credit limit.

6. A project data monitoring method based on multi-source data fusion according to claim 5, characterized in that, Step 5, which involves calculating the project post-evaluation score and the post-evaluation coefficients for each department based on historical project data, includes: Step 5-1: Calculate the departmental post-evaluation coefficient based on historical project post-evaluation data, as follows: ； in, For the department The post-evaluation coefficient, For the department The average post-evaluation score of all projects in the previous period, The average post-evaluation score of all projects in the previous cycle for the company. To adjust the sensitivity coefficient; Step 5-2: Calculate the post-project evaluation score. The details are as follows: ； in, For the project The score for achieving the desired effect For the project The performance score was achieved. For the project Achieve the desired score. , , These are the weighting coefficients. .

7. A project data monitoring method based on multi-source data fusion according to claim 6, characterized in that, Step 6, which involves calculating the categorized credit limit for each department in the credit pool based on project type, pre-project assessment score, and post-department assessment coefficient, includes: Step 6-1, when the department If there are candidate projects related to production capacity, then the department Production capacity quota The calculation method is as follows: ； in, For the department The initial quota for capacity-related projects. For the department The actual investment amount in capacity-related projects in the previous cycle, For the department Current period forecast revenue growth This represents the company's total capacity-related investment in the previous cycle. This represents the company's projected revenue growth rate for the current period. For the department The post-evaluation coefficient, This refers to the amount in the capacity-related quota pool. For the department The average pre-assessment score of the submitted projects. The average pre-evaluation score of all projects submitted by the company. This is the adjustment intensity coefficient for capacity-related pre-assessment; Step 6-2, when the department If there are supporting candidate projects, then the department Supporting quota The calculation method is as follows: ； in, For the department The initial quota for support projects, For the department The amount of investment in support projects in the previous cycle, This represents the total investment in support projects for the company in the previous cycle. To support the amount in the credit limit pool, Adjust the strength coefficient for the pre-assessment of the support category; Step 6-3, when the department If there are candidate projects for forward-looking research, then the department Forward-looking research quota The calculation method is as follows: ； in, For the department The quota for forward-looking research projects For the department Strategic adjustment coefficient, The amount of the forward-looking research quota pool, For the department Average pre-evaluation score for forward-looking research projects It is the sum of the average pre-evaluation scores of all forward-looking research projects across all departments.

8. A project data monitoring method based on multi-source data fusion according to claim 7, characterized in that, Step 7, which involves matching project department quotas based on departmental classification quotas, includes: Step 7-1, Calculate the project Comprehensive priority index The details are as follows: ； in, For the project The overall priority index, For the project Pre-assessment score, For the project Urgency rating, For the project Strategic fit score , , Weighting coefficients; Step 7-2, if the project Belongs to department For capacity-related projects, the project calculation is as follows: Allocation factors for capacity-related projects The details are as follows: ； in, For the project Forecasted revenue growth rate for related products or business capabilities; Then the project Quota for capacity-related projects Represented as: ； in, For the department Production capacity quotas; For the department The sum of allocation factors for all candidate projects in the capacity category; Step 7-3, if the project Belongs to department For supporting projects, then calculate the project. Supporting project allocation factors Specifically: ； Then the project Supporting project quota Represented as: ； in, For the department Support-type credit limits; For the department The sum of the allocation factors for all supporting candidate projects; Step 7-4, if the project Belongs to department For forward-looking research projects, then the calculation project Allocation factors for forward-looking research projects Specifically: ； in, For the project The strategic adjustment coefficient; Then the project Quota for forward-looking research projects Represented as: ； in, For the department Forward-looking research quotas; For the department The sum of allocation factors for all forward-looking pre-research candidate projects; Step 7-5 involves matching project quotas, as detailed below: project Calculation limit Determined based on project type, when the project When it is a capacity-related project, When the project When for support projects, When the project For forward-looking research projects, ; Then the project Actual matching amount for: ； when When the priority level is higher than the preset threshold and there is a credit limit shortfall: ； ； in, For the project The amount declared.

9. A project data monitoring method based on multi-source data fusion according to claim 8, characterized in that, Step 8, which involves reallocating the credit limit according to preset conditions, includes: Step 8-1: Calculate the project deviation index based on the project data acquired in real time during project execution, as detailed below: project Income deviation rate for: ； in, For the project Actual income For the project Projected revenue; project Profit deviation rate for: ； in, For the project The actual profit For the project Forecasted profits; project The schedule deviation rate is ; Step 8-2, when the income deviation rate Profit Deviation Rate or schedule deviation rate If any of the following exceeds the corresponding preset threshold, the quota will be reassigned. That is, the affected scope will be determined based on the department to which the project triggering the reassignment belongs, the project type, the quota pool to which it belongs, the related products, and the project dependencies. When the event triggering the reassignment affects the department classification quota parameters, steps 6 and 7 will only be re-executed for the affected departments. When the event triggering the reassignment only affects the project status, priority, or project allocation factor within a department, step 7 will only be re-executed for the affected projects.

10. A project data monitoring system based on multi-source data fusion, characterized in that, To implement the method of any one of claims 1-9, comprising: The system includes a multi-source data acquisition module, a data standardization module, a project identification module, a pre-assessment indicator generation module, a credit pool management module, a credit calculation module, a dynamic matching module, an event triggering module, an incremental recalculation module, a post-assessment feedback module, and a log traceability module; among these, The multi-source data acquisition module is used to acquire project application data, financial data, revenue forecast data, contract data, project execution data, and post-evaluation data. The data standardization module is used to perform field mapping, format conversion, missing value handling, outlier identification, unit unification, and time window alignment on the data, and generate project indicator vectors. The project identification module is used to identify project types based on project indicator vectors; The pre-assessment indicator generation module is used to generate pre-assessment scores for projects. The quota pool management module is used to build and maintain regular quota pools, flexible quota pools, capacity-related quota pools, support quota pools, forward-looking research quota pools, and emergency adjustment quota pools. The quota calculation module is used to calculate the departmental category quota in the corresponding quota pool based on the actual investment amount of similar projects in the previous period, the projected revenue growth rate, the department's post-evaluation coefficient, and the average pre-evaluation score of the projects applied for by the department. The dynamic matching module is used to calculate the project quota based on the project's overall priority index and the corresponding departmental classification quota. The event triggering module is used to generate a credit limit recalculation event when the deviation exceeds a preset threshold; The incremental recalculation module is used to perform incremental recalculation of the credit limit for the affected project set based on the credit limit recalculation event; The post-evaluation feedback module is used to calculate the post-evaluation score of the completed project based on the actual execution results, and to update the department's post-evaluation coefficient and the weight of the pre-evaluation dimensions. The log traceability module is used to record the process of quota calculation, dynamic matching, incremental recalculation, and parameter update.

Images