Construction project-oriented intelligent data fusion system and method

Abstract

The application provides a construction project-oriented intelligent data fusion method and system, and belongs to the technical field of data fusion. Specifically, based on monitoring processing data and associated fusion processing monitoring targets, the monitoring analysis requirement type under different data fusion modes is determined. Based on construction project data, the engineering project is divided into different combinations. Based on the fusion processing monitoring target data in the combination and the monitoring analysis requirement type under different data fusion modes, the update strategy of the fusion processing monitoring target in the combination is determined. The update processing of the fusion processing monitoring target is performed by using the update strategy. Based on the update processing result and the monitoring data under different data fusion modes, the determination of the fusion management and control mode of the combination is performed, and the balanced control of the fusion monitoring reliability and the fusion deviation is realized.

Description

Technical Field

[0001] This invention belongs to the field of data fusion technology, and in particular relates to an intelligent data fusion system and method for construction projects. Background Technology

[0002] With the deepening of refined management in the construction industry, construction companies typically operate multiple projects simultaneously, facing severe challenges in data management across four core dimensions: business, finance, taxation, and capital (collectively referred to as "business, finance, taxation, and capital"). Current management models have significant flaws, hindering companies' decision-making efficiency and risk control capabilities.

[0003] Specifically, in typical construction companies, business data (such as contracts, progress, materials, and subcontracting) is stored in the project management system; financial data (such as vouchers, ledgers, and reports) exists in accounting software; tax data (such as invoices and tax returns) is processed by the tax system; and financial data (such as receipts and payments, and cash flow records) relies on banks or independent financial systems. These systems are independently built and have inconsistent standards, resulting in isolated "islands" of business, financial, tax, and capital data. For example, a payment application for a construction project cannot automatically link to the corresponding contract progress, subcontracting invoices, tax deductions, and account balances. Approval relies on manual verification and offline transmission, which is inefficient and prone to errors.

[0004] To address the aforementioned technical issues, a unified data management platform was constructed to enable unified access and management of different engineering projects of construction companies, greatly improving the reliability of data fusion and processing. However, the following technical shortcomings exist:

[0005] Due to deviations in construction projects, the data fusion processing methods in a unified data management platform vary to some extent. In particular, with the increase in the number of connected projects, it becomes difficult for the data management platform to effectively and reliably monitor deviations during the fusion processing. Therefore, how to monitor and process the fusion process, and determine the fusion control scheme based on the monitored deviations, thereby improving the reliability and timeliness of fusion deviation control, has become an urgent technical problem to be solved.

[0006] Therefore, there is an urgent need for an intelligent data fusion system and method for construction projects. Summary of the Invention

[0007] To achieve the objectives of this invention, the following technical solution is adopted:

[0008] Specifically, this application provides an intelligent data fusion method for construction projects, which includes:

[0009] S1 determines the construction project data in the engineering project, and uses the construction project data of each engineering project in the unified management and control platform to determine the identification method of the fusion processing monitoring target in the engineering project. Based on the construction project of the fusion processing monitoring target, it determines the monitoring processing data under different data fusion methods. Based on the monitoring processing data and the associated fusion processing monitoring target, it determines the monitoring and analysis requirement type under different data fusion methods.

[0010] S2 divides the engineering project into different combinations based on the construction project data, and determines the update strategy for the fusion processing monitoring targets in the combination based on the fusion processing monitoring target data in the combination and the monitoring and analysis requirements under different data fusion methods;

[0011] S3 uses the update strategy to perform fusion processing of the monitoring target update, and determines the combined fusion control method based on the update processing result and the monitoring data under different data fusion methods.

[0012] The beneficial effects of this invention are as follows:

[0013] By using construction project data from various engineering projects within a unified management platform, a method for identifying integrated monitoring targets within engineering projects is determined. By analyzing the complexity of project types and the sheer number of projects, two different criteria for identifying "integrated monitoring targets" are dynamically selected. This improves the comprehensiveness and reliability of monitoring and processing in cases of high complexity and large scale, and applies this method to the screening of all projects. The aim is to construct a reasonably sized and highly targeted set of monitoring targets. Before integration deviations become clearly apparent or erupt on a large scale, by optimizing the screening accuracy of monitoring targets, the system pressure and management costs caused by unnecessary and broad monitoring and analysis can be significantly reduced.

[0014] Based on the updated processing results and monitoring data under different data fusion methods, a combined fusion control method is determined. By analyzing the data fusion performance across projects and similar construction projects, fusion control is implemented when there are significant deviations in the data fusion process of certain construction projects. This avoids the impact of deviations in the fused data on the accuracy of the fusion processing results. Furthermore, by combining the updated results of the fusion processing monitoring targets, it also avoids the technical problem of poor update efficiency of the fusion processing monitoring targets caused by premature fusion control processing when the reliability of the combined monitoring is poor. This achieves a balanced control over monitoring reliability and fusion deviation.

[0015] Furthermore, the construction project data in the engineering project includes the types of construction projects within the engineering project.

[0016] Furthermore, the method for determining the identification method of the fusion processing monitoring target in the aforementioned engineering project is as follows:

[0017] The type of construction project in each engineering project is determined by the construction project data in the unified management and control platform.

[0018] Based on the type of construction project in each engineering project, determine the number of engineering projects within different ranges of the number of projects of different types;

[0019] The identification method for the fusion processing monitoring targets in the engineering projects is determined based on the number of engineering projects within different types and the types of construction projects in each engineering project.

[0020] Furthermore, the data fusion method is divided according to the type of construction project. Based on the type of construction project, the number of times business data, financial data, tax data, and capital data are fused during the call process is divided into different data fusion methods.

[0021] Furthermore, the monitoring and processing data under the data fusion method includes the number of monitoring and processing processes under the data fusion method and the data fusion deviation in different monitoring and processing processes.

[0022] Furthermore, the method for determining the fusion control method of the combination is as follows:

[0023] Based on monitoring data under different data fusion methods, determine the deviation handling process of the combination under different correlation fusion methods;

[0024] Based on the update processing results, the updated data of the fusion processing monitoring target in the combination is determined;

[0025] By utilizing the updated data of the monitoring targets in the fusion processing of the combination and the deviation processing process under different correlation fusion methods, the fusion control method of the combination is determined.

[0026] Secondly, this application discloses an intelligent data fusion system for construction projects, employing the aforementioned intelligent data fusion method for construction projects, specifically including:

[0027] Fusion processing module, monitoring and processing module, fusion control module;

[0028] The fusion processing module is responsible for fusion processing of business data, financial data, tax data, and capital data from different engineering projects during the calling process;

[0029] The monitoring and processing module is responsible for determining the identification method of the fusion processing monitoring target in the project based on the construction project data in each project, and using the fusion processing monitoring target to monitor and process the deviation in the fusion processing process.

[0030] The fusion management module is responsible for dividing the engineering project into different combinations based on the construction project, and determining the fusion management method of the combination based on the update processing results of the fusion processing monitoring targets in the combination and the monitoring data under different data fusion methods.

[0031] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0032] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0033] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0034] Figure 1 This is a flowchart of an intelligent data fusion method for construction projects;

[0035] Figure 2 This is a flowchart illustrating the method for determining the identification of monitoring targets in fusion processing within an engineering project.

[0036] Figure 3 This is a flowchart illustrating the method for determining the types of monitoring and analysis needs under data fusion.

[0037] Figure 4 This is a framework diagram of an intelligent data fusion system for construction projects. Detailed Implementation

[0038] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0039] Example 1

[0040] Firstly, such as Figure 1As shown, this application provides an intelligent data fusion method for construction projects, specifically including:

[0041] S1 determines the construction project data in the engineering project, and uses the construction project data of each engineering project in the unified management and control platform to determine the identification method of the fusion processing monitoring target in the engineering project. Based on the construction project of the fusion processing monitoring target, it determines the monitoring processing data under different data fusion methods. Based on the monitoring processing data and the associated fusion processing monitoring target, it determines the monitoring and analysis requirement type under different data fusion methods.

[0042] Furthermore, the construction project data in the engineering project includes the types of construction projects within the engineering project.

[0043] To address potential biases arising during data fusion across numerous engineering projects, this application establishes an intelligent screening mechanism. This mechanism dynamically selects two different criteria for identifying "fusion monitoring targets" by analyzing the complexity and sheer number of project groups. This improves the comprehensiveness and reliability of monitoring in cases of high complexity and large scale, and is applied to the screening of all projects, aiming to construct a reasonably sized and highly targeted set of monitoring targets. Its fundamental purpose is to significantly reduce the system pressure and management costs associated with unnecessary and broad monitoring analysis by optimizing the screening accuracy of monitoring targets before fusion bias becomes clearly apparent or erupts on a large scale.

[0044] Specifically, such as Figure 2 As shown, the method for determining the identification method of the fusion processing monitoring target in the project is as follows:

[0045] S11 uses the construction project data of each engineering project in the unified management and control platform to determine the type of construction project in each engineering project;

[0046] "Type of construction project" refers to the professional divisions that make up the engineering project, such as "foundation engineering", "main structure", "mechanical and electrical installation", "decoration and finishing", etc. Each type represents an independent business data main line.

[0047] This is a fundamental dimension for assessing the potential data interaction complexity of a project. The more types a project includes, the more "fusion points" there are within it that require data alignment, aggregation, and verification across systems (project management, finance, tax, etc.). This doesn't necessarily mean it's active, but it provides a "stage" for it to become a highly active project. Statistical types are the first step in predicting the overall potential data interaction volume of a project group.

[0048] Example: In the management platform, Project X includes four types: "Earthwork," "Structure," "Curtain Wall," and "Interior Decoration"; Project Y only includes one type: "Road Paving." Project X objectively needs to handle a more complex data flow and is more likely to be included in the monitoring scope.

[0049] S12 determines the number of engineering projects within different type ranges based on the type of construction projects in each engineering project.

[0050] "Type Quantity Range" is a statistical grouping based on the number of types contained in the project, such as "[1-2] types", "[3-5] types", "≥6 types". The number of projects clustered within each range is counted.

[0051] This step aims to create a "structural profile" of the project portfolio, identifying which level of complexity constitutes the majority of projects. For example, if most projects fall within the [3-5] range, the entire portfolio is considered to be of "medium complexity." This profile is a crucial input for determining the appropriate activity screening criteria (tight or loose) to ensure that the subsequent set of monitoring targets matches the actual situation of the project portfolio and avoids screening failures.

[0052] Example: The platform manages 500 projects. Statistics show that there are 300 simple projects of type [1-2], 180 medium-complex projects of type [3-5], and 20 complex projects of type ≥6. This indicates that the majority of the project group consists of medium-complexity projects.

[0053] S13 determines the identification method for the fusion processing monitoring target in the engineering project based on the number of engineering projects in different type ranges and the type of construction project in each engineering project.

[0054] It is understandable that, based on the number of engineering projects within different types and quantity ranges, the method for identifying the fusion processing monitoring targets in the engineering projects specifically includes:

[0055] Case 1: Based on the types of construction projects in each engineering project, the types of construction projects are deduplicated to obtain the total number of construction project types in all engineering projects. If the total number of construction project types is greater than the preset type number threshold, then the identification method of the fusion processing monitoring target in the engineering project is determined to be the preset identification method.

[0056] The business structure is extremely complex. A lenient standard is adopted to cover a wide range of projects. When the total number of deduplicated construction project types exceeds the "preset type quantity threshold", the business structure is judged to be extremely complex.

[0057] A large total number of project types implies extremely high business diversity (e.g., more than 10 types of construction projects). Data fusion rules and scenarios vary widely, and biases can occur anywhere. In this case, using overly stringent screening standards might miss a large number of potentially active projects, leading to monitoring blind spots. Therefore, the system automatically selects a preset identification method (with a lower threshold) and applies it to all projects. The aim is to create a larger set of monitoring targets through a relatively lenient filter, ensuring broad coverage and no omissions. Even though this introduces some monitoring pressure, it is a better choice than the risk of omissions.

[0058] Scenario 2: If the total number of construction project types is not greater than the preset type number threshold, based on the number of projects within different type number intervals, determine the type number interval where the number of types is greater than the preset type number threshold, and use this interval as the target number interval. Determine whether the number of projects within the target number interval is greater than the preset project number threshold. If yes, then determine the identification method of the fusion processing monitoring target in the project as the preset identification method; otherwise, determine the identification method of the fusion processing monitoring target in the project as the second preset identification method.

[0059] If the total number of types does not exceed the threshold, then analyze whether the number of items in the "target number range" (usually referring to the range with a large number of types, such as ≥3 types) exceeds the "preset item number threshold".

[0060] This is the key logic for reducing pressure. If the number of highly complex projects is small (not exceeding the threshold), a lenient standard is used to screen all projects to identify their inefficiency. In this case, the system automatically activates a second preset identification method (with a higher threshold) applied to all projects. This "finer sieve" filters out a smaller, more active core target set. By deeply monitoring only these very few highly active projects, the overall monitoring system's load can be significantly reduced while ensuring that the most likely points of bias are captured.

[0061] It should be noted that the preset identification method is to obtain the number of data fusion processing times of the project in different construction project types. Among multiple construction project types, the project with the number of data fusion processing times greater than the preset fusion processing time threshold is taken as the fusion processing monitoring target to determine whether the data fusion result is accurate in the fusion processing process.

[0062] Preset recognition method (lenient standard, for case 1):

[0063] Standard definition: For each project, count the number of times data is fused across at least two construction project types. If the number exceeds the preset threshold for the number of fusion processing times, the project is identified as a "fusion processing monitoring target".

[0064] This standard has a relatively low threshold. When the business structure is complex, it ensures that a large number of projects with cross-type data interactions (which may cause bias) are included in the monitoring, forming a large target set. Although a large number of monitoring targets will bring some pressure, this is a necessary price to pay for preventing systemic unknown biases, and the pressure is at an acceptable "broad alert" level.

[0065] It should be noted that the second preset identification method is to obtain the number of data fusion processing times of the project in different types of construction projects, and to take the projects with more than the target number of construction projects and whose number of data fusion processing times is greater than the preset threshold of the number of fusion processing times as the fusion processing monitoring targets, and to determine whether the data fusion results in the fusion processing process are accurate.

[0066] Second preset identification method (strict standard, used for case 2):

[0067] Standard definition: For each engineering project, the number of times data is fused across at least three construction project types is counted. Only when the number of times exceeds the preset fusion processing number threshold is the project recognized as a fusion processing monitoring target.

[0068] Design Significance and Pressure Considerations: This standard significantly raises the bar. When the business structure is relatively simple, only "data convergence hub" projects that are extremely active in three or more dimensions are selected. The resulting monitoring target set is small, allowing for highly concentrated monitoring resources, minimizing system analysis pressure, and maximizing efficiency while keeping risks under control.

[0069] The core value of this method lies in the fact that it is an intelligent monitoring target pre-screening system oriented towards resource constraints.

[0070] Dynamic pressure regulation: The core of this method is not to judge the level of risk, but to automatically switch between a "broadly covered alert mode" and a "precisely focused energy-saving mode" based on the characteristics of the project group. This directly addresses the core requirement of "reducing monitoring pressure when deviations are not yet in place."

[0071] By raising or lowering the threshold for "activity," the number of projects marked as "monitoring targets" can be directly controlled. This ensures that valuable computing, storage, and analytical resources are always invested in the minimum necessary set of projects that are most likely to generate data flow and thus have the best chance of observing deviations.

[0072] The selected set of "active monitoring targets" is the best target for implementing proactive data governance (such as strengthening pre-verification rules and tracking the fusion process). Proactively observing them during their high-frequency data fusion process can identify and correct deviations earlier and at a lower cost than auditing after problems have accumulated.

[0073] Furthermore, the data fusion method is divided according to the type of construction project. Based on the type of construction project, the number of times business data, financial data, tax data, and capital data are fused during the call process is divided into different data fusion methods.

[0074] Furthermore, the monitoring and processing data under the data fusion method includes the number of monitoring and processing processes under the data fusion method and the data fusion deviation in different monitoring and processing processes.

[0075] The core objective of this embodiment is to assess the current monitoring coverage and data quality performance of each specific data fusion method, thereby intelligently determining the required monitoring and analysis intensity level (i.e., the type of monitoring and analysis need). The fundamental logic lies in the fact that different fusion methods differ in monitoring scope, deviation frequency, and potential impact. By using a multi-dimensional evaluation system to classify them into different priority need types, refined management of monitoring resources can be achieved. High-intensity monitoring (Category I need) can be implemented for high-risk, low-coverage, and high-deviation fusion methods, while routine monitoring (Category II or general need) can be implemented for methods with stable performance.

[0076] Specifically, such as Figure 3 As shown, the method for determining the monitoring and analysis requirement type under the data fusion method is as follows:

[0077] S21. Using the monitoring and processing data under the data fusion method, determine the monitoring and processing process under the data fusion method;

[0078] S22 Based on the data fusion deviation in different monitoring and processing processes under the data fusion method, determine the monitoring and processing process with data fusion deviation and treat it as a deviation processing process;

[0079] "Data fusion method": refers to a specific set of data fusion rules based on the type of construction project (such as "mechanical and electrical installation" or "decoration and renovation"). For example, the "mechanical and electrical installation fusion method" defines the reconciliation and merging logic between business material requisition forms, financial cost vouchers, tax input invoices, and bank equipment payment records for this type of project.

[0080] "Monitoring and processing process": refers to the actual execution of a full-process monitoring action by the system for a certain fusion method, from data retrieval and rule application to result comparison.

[0081] "Deviation handling process": refers to a situation in which the system discovers and records a failure to successfully integrate business, financial, tax, and capital data according to rules or inconsistent results during a monitoring and processing process.

[0082] This forms the data foundation for the assessment. Each integration method corresponds to unique business logic and risk points. By statistically analyzing the number of times it was "monitored" and the number of times "deviations" occurred, we obtained two core raw indicators to measure the "level of attention" and "current state of data quality" of each method. This ensures that subsequent analysis is based on objective facts rather than subjective assumptions.

[0083] Example: Suppose that for the "Specialized Integration Method for Curtain Wall Engineering", the system initiated 150 automatic monitoring sessions (i.e., 150 monitoring and processing procedures) during the past monitoring period. Upon verification, 12 of these monitoring sessions revealed issues such as data mismatches or missing vouchers between contract quantities, settlement statements, and payment applications. These 12 sessions were marked as "deviation processing procedures".

[0084] S23 determines the monitoring and analysis requirement type under the data fusion method based on the monitoring and processing process and deviation processing process under the data fusion method, and in conjunction with the fusion processing monitoring targets associated with the data fusion method.

[0085] It should be noted that the fusion processing monitoring target associated with the data fusion method is the fusion processing monitoring target that is subjected to fusion monitoring processing under the data fusion method.

[0086] Specifically, the types of monitoring and analysis needs under the aforementioned data fusion method are determined, including:

[0087] S231 takes the fusion processing monitoring target associated with the data fusion method as the associated monitoring target, determines the monitoring ratio based on the proportion of the associated monitoring target in the engineering project with the data fusion method, and judges whether the monitoring ratio is greater than the preset monitoring ratio threshold. If yes, the monitoring analysis requirement type under the data fusion method is determined as the general requirement type. If no, proceed to step S232.

[0088] Assess the breadth of monitoring coverage (monitoring ratio):

[0089] "Monitoring ratio" refers to the percentage of projects using this fusion approach that are identified as "fusion-processed monitoring targets" (i.e., associated monitoring targets). It reflects the prevalence of this fusion approach in currently active projects.

[0090] If a particular integration method has been included in the monitoring of the vast majority of applicable projects (high monitoring rate), it indicates that its monitoring system is basically established, and it can be considered as a regular requirement (Category II or III). Conversely, if the monitoring coverage is low (low monitoring rate), it means that there are monitoring blind spots, requiring increased attention, and may lead to a higher type of requirement. This is the first "filter," prioritizing the coverage issue.

[0091] Example: Suppose the company currently has 50 projects involving "curtain wall engineering". Among them, 35 projects are identified as "related monitoring targets" due to active data interaction. Then the monitoring ratio = 35 / 50 = 70%. If the preset threshold is 80%, then 70% < 80%, which is insufficient coverage, and proceed to the next step S232 for judgment.

[0092] S232 determines whether there is a deviation processing procedure under the data fusion method. If yes, proceed to step S233. If no, determine the monitoring and analysis requirement type under the data fusion method as a general requirement type.

[0093] Assess the existence of bias: Determine whether any bias has occurred in the historical monitoring of this fusion method.

[0094] This is a fundamental risk signal. If a fusion method has never shown any deviation, even with low monitoring coverage, it may mean that its rules are simple or the risk is inherently low, and it can be temporarily categorized as a routine requirement. Once a deviation has occurred, it proves that there is a risk exposure, and a more in-depth analysis and subsequent rigorous assessment are necessary.

[0095] Example: Continuing from the previous example, it is known that the "specialized integration method for curtain wall engineering" involves 12 deviation processing steps. Therefore, it is determined that "deviation exists" and proceeds to S233.

[0096] S233 Based on the monitoring and processing data, determine the proportion of the monitoring and processing process under the data fusion method in the fusion process, and use it as the monitoring process proportion. Determine whether the monitoring process proportion is less than the preset monitoring proportion threshold. If so, determine the monitoring and analysis demand type under the data fusion method as a demand type. If not, proceed to step S234.

[0097] "Monitoring process ratio" refers to the proportion of the number of monitoring and processing processes initiated by this fusion method to the total number of fusion processes that should have occurred for all associated monitoring targets within the corresponding time period. It reflects the frequency and depth of monitoring (whether it is monitoring every time or sampling monitoring).

[0098] Some integration methods may cover the project, but the monitoring is sampled or infrequent (the monitoring process accounts for a low proportion). The fact that deviations can still be detected under low-frequency monitoring often means that the problem occurs frequently or is easily triggered, and the risks are more hidden and serious. Therefore, it is necessary to increase the intensity of monitoring (leading to a certain type of requirement).

[0099] Example: According to statistics, the 35 related monitoring targets mentioned above should have generated approximately 1000 data fusion events related to "curtain wall engineering" within the period. However, the system only initiated monitoring for 150 of these events. Therefore, the monitoring rate is 150 / 1000 = 15%. If the preset threshold is 20%, then 15% < 20%, indicating insufficient monitoring depth and potential risks. The initial assessment is that this is a Class I demand.

[0100] S234 Determine whether the proportion of the deviation processing process in the monitoring processing process is greater than the preset deviation proportion threshold. If yes, determine the monitoring and analysis demand type under the data fusion method as a demand type. If no, proceed to step S235.

[0101] "Deviation ratio" refers to the proportion of deviation handling processes to the total number of monitoring and handling processes.

[0102] This directly measures the frequency with which problems are exposed under implemented monitoring. A high deviation rate indicates that the fusion method is extremely unstable, the data quality is questionable, and high-intensity, high-frequency analysis is necessary (a type of requirement). This is a very direct risk signal.

[0103] Example: To demonstrate different branches, assume that the monitoring rate of another "simplified integration method for earthwork engineering" is 25% (above the threshold), then proceed to this step. It was monitored 80 times, with deviations detected in 20 of them. Deviation rate = 20 / 80 = 25%. If the preset deviation rate threshold is 15%, then 25% > 15%, indicating a high frequency of problem exposure, and its requirement type is determined to be a Class I requirement.

[0104] S235 determines a monitoring deviation factor based on the monitoring ratio, the monitoring process ratio, and the proportion of the deviation handling process in the monitoring process, and determines the monitoring analysis requirement type under the data fusion method based on the monitoring deviation factor.

[0105] Indeed, the lower the monitoring ratio and the monitoring process ratio, the higher the proportion of the deviation processing process in the monitoring processing process, and the larger the monitoring deviation factor.

[0106] It is understood that if the monitoring deviation factor is greater than the preset deviation factor threshold, the monitoring and analysis demand type under the data fusion method is determined to be a type I demand type, and in other cases it belongs to a type II demand type. Type I demand type is greater than type II demand type, and type II demand type is greater than general demand type.

[0107] The "monitoring deviation factor" is a composite indicator that integrates the monitoring ratio, the monitoring process ratio, and the deviation ratio. According to the rules, the lower the monitoring ratio and the monitoring process ratio, and the higher the deviation ratio, the larger the factor value.

[0108] Monitoring Deviation Factor (MDF) = ((1 - CR) * w1 + (1 - MDR) * w2) * (ER * w3);

[0109] Formula explanation:

[0110] (1 - CR): An inverse indicator of insufficient monitoring coverage. The lower the CR, the larger this value.

[0111] (1 - MDR): The inverse indicator of insufficient monitoring depth. The lower the MDR, the larger this value.

[0112] ER: Deviation rate index. The higher the ER, the larger this value.

[0113] w1, w2, w3: These are the weighting coefficients for each item (w1 + w2 + w3 = 1), used to adjust the importance of different dimensions. For example, if more attention is paid to coverage blind spots, w1 can be increased; if more attention is paid to problem frequency, w3 can be increased.

[0114] This formula multiplies the "degree of monitoring weakness" (the sum of the first two terms) by the "severity of the problem" (the third term), rather than simply adding them together. This means that in cases of weak monitoring, even a low absolute deviation rate can amplify the overall risk; conversely, with adequate monitoring, a high deviation rate will be assessed more cautiously. This aligns with the core principles of risk management.

[0115] Threshold determination: If MDF > preset deviation factor threshold (δ, for example, 0.01), it is determined to be a type of demand; otherwise, it is determined to be a type of demand.

[0116] For the "intermediate zone" situations that the previous steps could not directly determine (i.e., coverage is not too low, depth is acceptable, and deviation exists but the proportion is not high), a more refined quantitative tool is needed for comprehensive evaluation. This factor penalizes both "inadequate monitoring" (low coverage, low frequency) and "substandard quality" (high deviation), and can more sensitively identify fusion methods that require upgraded monitoring.

[0117] Example: Assume that the monitoring ratio of the "conventional building material procurement integration method" is 75%, the monitoring process ratio is 22%, and the deviation ratio is 10%. After calculating its monitoring deviation factor, if it does not exceed the preset "deviation factor threshold", it is determined as a type II demand type. If the calculated factor of another method exceeds the threshold, it is determined as a type I demand type.

[0118] S2 Based on the construction project data, divide the engineering project into different combinations, and determine the update strategy of the fusion processing monitoring target in the combination according to the fusion processing monitoring target data in the combination and the monitoring analysis demand types under different data fusion methods;

[0119] The core goal of this embodiment is: for a specific "combination" (such as a collection of projects in a specific area, specific business line, or specific time period), formulate a dynamic update strategy for the "fusion processing monitoring target" within it. This strategy dynamically selects two update precisions according to the overall monitoring coverage of the combination, project scale, and risk level of the associated data fusion method: comprehensive and reliable update or focused and general update. Its fundamental logic is that when the overall monitoring of the combination is weak or risks are concentrated, it is necessary to expand the update scope to ensure comprehensive coverage; when the combination monitoring is relatively complete and risks are controllable, focus on the highest-risk links for precise update, so as to optimize the update computing resources while ensuring data quality.

[0120] Specifically, the method for determining the update strategy of the fusion processing monitoring target in the combination is as follows:

[0121] S31 Determine the fusion processing monitoring target in the combination based on the fusion processing monitoring target data in the combination;

[0122] [[ID= "15"]]"Combination" refers to a group of engineering projects divided according to management requirements (such as by project type). The "fusion processing monitoring target ratio" refers to the percentage of the number of active projects that have been identified as key monitoring targets in the combination to the total number of projects in the combination.

[0123] This is the first and most macroscopic checkpoint for evaluating the urgency of the update. The monitoring target ratio directly and comprehensively reflects the coverage breadth of the existing monitoring system for this combination. A too low ratio means that there are huge loopholes in the monitoring network, and the data fusion activities of a large number of projects are in a blind test state, with a very high risk of systematic deviation. Therefore, once it is below the threshold, the most comprehensive update strategy must be unconditionally activated to expand the monitoring coverage as quickly as possible, which is the bottom-line thinking for ensuring data quality and security.

[0124] Example: The "East China residential project combination" currently has 80 projects, of which only 30 are listed as monitoring targets, and the monitoring ratio is 37.5%. The preset threshold is 60%. Since 37.5% < 60%, the system determines that the monitoring coverage of this combination is seriously insufficient.

[0125] It is understandable that the above steps include the following:

[0126] Scenario 1: If the proportion of the fusion processing monitoring targets in the combination is less than the preset monitoring target proportion threshold, then the update strategy of the fusion processing monitoring targets in the combination is determined to be a reliable update strategy.

[0127] The "Reliable Update Strategy" is a comprehensive screening strategy that examines the data activity of all items within the portfolio under any associated fusion method, and includes them in the monitoring target list as long as they are active in any method.

[0128] In situations where coverage is severely inadequate, the primary objective of decision-making is to "fill the gaps and eliminate blind spots," rapidly establishing a basic monitoring network. Using lenient entry criteria (active in any mode) for a full-network scan ensures that no potentially active project is missed. While this may include some less active projects, this is a necessary trade-off between efficiency and secure coverage, aiming to quickly eliminate systemic monitoring risks.

[0129] Example: Continuing from the previous example, due to insufficient coverage, the system re-screened all 80 residential projects in the East China region. If a project's data interaction count exceeds the threshold under any of the "civil engineering," "installation," or "decoration" integration methods, it will be updated as a monitoring target.

[0130] Case 2: If the proportion of the fusion processing monitoring targets in the combination is not less than the preset monitoring target proportion threshold, obtain the number of engineering projects in the combination, and determine whether the number of engineering projects in the combination is greater than the preset engineering project number threshold. If yes, determine that the update strategy of the fusion processing monitoring targets in the combination is a reliable update strategy. If no, proceed to step S32.

[0131] The "project quantity threshold" is a critical value used to determine whether the combined scale has reached the "massive" level.

[0132] Even if the monitoring ratio meets the target, if the portfolio itself is extremely large, the absolute number of unmonitored projects can still be considerable. A large base means management complexity increases exponentially, and the few uncovered projects may accumulate into significant risks due to economies of scale. Therefore, when the total number of portfolio projects exceeds the size threshold, a comprehensive update strategy is also necessary to address the complexity brought about by large-scale management and ensure that the monitoring system can match the business volume.

[0133] Example: The "East China Residential Project Portfolio" currently contains 80 projects, of which only 30 are listed as monitoring targets, representing a monitoring rate of 37.5%. Assume the preset threshold is 30%. The portfolio now comprises 80 projects, exceeding the threshold of 60 projects. Therefore, the system still determines that a reliable update strategy needs to be initiated.

[0134] S32 uses the construction projects of the engineering projects in the combination to determine the monitoring and analysis requirement type under the association and fusion method of the combination;

[0135] "Association and fusion method" refers to the set of various data fusion rules involved in all projects within the combination, such as "civil engineering", "installation", or "decoration". "Type of demand" is the highest risk level assessed in the previous embodiment, representing a fusion method with high deviation rate or low monitoring depth.

[0136] When the breadth of portfolio coverage and project scale do not trigger a full update, decision-making needs to delve into the specific business risks. Different integration methods carry different business risks. By screening for high-risk methods for a particular type of demand, it can be determined whether the portfolio is exposed to specific, known high-risk business processes. This is a crucial step in linking macro-level portfolio management with micro-level data quality risks.

[0137] Example: For a "smart building project portfolio", the system queries the risk rating of its associated "civil engineering", "installation" or "decoration" methods.

[0138] The above steps include the following:

[0139] S321 Based on the monitoring and analysis demand type under the association fusion method in the combination, determine whether there is an association fusion method for a certain demand type under the association fusion method. If yes, proceed to step S322. If no, determine that the update strategy for the fusion processing monitoring target in the combination is no update processing required.

[0140] "There is a type of demand associated with a fusion method" is a Boolean condition used to trigger a risk-oriented update process.

[0141] If all association methods are low-risk (Category II and III requirements), it indicates that the overall data quality in the business area where the combination operates is stable, the monitoring pressure is low, and there is no need to initiate an update process to consume resources. Conversely, if even one high-risk method exists, it proves that the combination has a clear hidden danger in a certain business dimension, and it must proceed to the next step of evaluation to determine the strength of the update strategy.

[0142] Example: Upon investigation, among the association methods of "Installation", "Installation" was rated as a Class 1 requirement due to its recent high deviation rate. Therefore, it is determined to "exist" and proceeds to S322.

[0143] S322 Obtain the number of association fusion methods for the first type of demand, and determine whether the number of association fusion methods for the first type of demand is greater than the preset value of the number of methods. If yes, determine that the update strategy of the fusion processing monitoring target in the combination is a reliable update strategy. If not, proceed to step S33.

[0144] The number of ways in which a type of demand is associated and integrated is a quantitative indicator of risk concentration. The preset value of the number of ways is the critical point for judging whether the risk has evolved from an "isolated case" to a "common phenomenon".

[0145] A single high-risk scenario may be an isolated problem with limited impact; however, multiple high-risk scenarios indicate data fusion issues across multiple business lines or process nodes, pointing to potential systemic management flaws or rule conflicts. When the number of high-risk scenarios exceeds a preset value, it means the risk has become widespread, necessitating a comprehensive update strategy to address this pervasive risk.

[0146] Example: Continuing from the previous example, if only one method of "installation" in the combination is a type of requirement, and it does not exceed the preset value (e.g., 2), the system considers the risk to be relatively concentrated, does not immediately trigger a full update, and proceeds to S33 for a more refined quantitative assessment.

[0147] S33 determines the update strategy for the fusion processing monitoring targets in the combination based on the fusion processing monitoring targets in the combination and the monitoring and analysis requirements under the associated fusion method.

[0148] Specifically, the association and fusion method of the combination is determined according to the construction project of the engineering project in the combination, and the data fusion method corresponding to the construction project is used as the association and fusion method.

[0149] It should be noted that, based on the monitoring and analysis demand types under the correlation and fusion method, the demand weight coefficients under the correlation and fusion method are determined. If the sum of the demand weight coefficients under different correlation and fusion methods is greater than the preset weight coefficient threshold, then the update strategy of the fusion processing monitoring target in the combination is determined to be a reliable update strategy. If the sum of the demand weight coefficients under different correlation and fusion methods is not greater than the preset weight coefficient threshold, then the update strategy of the fusion processing monitoring target in the combination is determined to be a general update strategy.

[0150] Specifically, the reliable update strategy is to update projects whose data fusion processing times under any association fusion method exceed a preset fusion processing time threshold as fusion processing monitoring targets. The general update strategy is to update projects whose data fusion processing times under any association fusion method of a certain demand type exceed a preset fusion processing time threshold as fusion processing monitoring targets.

[0151] The "Demand Weighting Coefficient" is a numerical value assigned to the risk level of each integration method. Typically, the first type of demand has the highest weight (e.g., 1.5), followed by the second type (e.g., 1.0), and the third type has the lowest weight (e.g., 0.5). The "Total Demand Weighting Coefficient Sum" is the cumulative value of the weights of all related methods within the combination, representing the overall risk load faced by the combination.

[0152] This is a precise quantitative tool for dealing with "gray areas." For combinations with a few high-risk options but not exceeding a certain number, simple decision-making may be inaccurate. By using weighted summation, not only high-risk options are considered, but also the cumulative effect of a large number of medium- and low-risk options are incorporated. This allows for the identification of combinations that, while not having multiple prominent high-risk points, still have a heavy overall risk burden, thus enabling more scientific and balanced decision-making.

[0153] Example: A combination is associated with 5 methods, with weights of 1.5 (Category 1), 1.0 (Category 2), 1.0 (Category 2), 0.5 (Category 3), and 0.5 (Category 3). The total weight is 4.5. Assume the preset weight threshold is 3.0. Since 4.5 > 3.0, it indicates a high overall risk load.

[0154] Reliable Update Strategy (Comprehensive Mode): This strategy is adopted when the total weights exceed a threshold. The execution logic is as follows: scan each item within the portfolio; if its data fusion processing count exceeds the limit under any association fusion method, it is updated as a monitoring target. The significance lies in conducting a comprehensive, blanket investigation of high-risk load portfolios to ensure no omissions.

[0155] General update strategy (focused mode): This strategy is adopted when the total weight is less than or equal to a threshold. The execution logic is as follows: only scan the data activity of each item within the combination under a certain type of demand fusion method, and only update to the target if the limit is exceeded under this high-risk method. The significance is that limited update computing resources are precisely allocated to the confirmed highest-risk areas, achieving highly efficient "precision strikes" and significantly reducing system overhead.

[0156] This embodiment constructs a "three-level progressive, quantitatively driven" intelligent update decision-making system. Its core value lies in achieving a "efficiency-security" balance in the maintenance of monitoring targets. Through three layers of filtering—"coverage / scale → risk existence → risk quantification load"—the system can automatically determine when a comprehensive survey (reliable update) is needed and when a focused review (general update) can be implemented. This avoids the update process itself becoming a bottleneck for system performance. The micro-risk ratings (Category I, II, and III) of the fusion method directly provide feedback and guide the macro-level monitoring target update strategy, forming a rapid closed loop from "risk identification" to "monitoring adjustment," thereby improving the agility and proactivity of the data governance system.

[0157] S3 uses the update strategy to update the monitoring targets of the fusion processing, and determines the update control method for the monitoring targets of the fusion processing in the combination based on the update processing results and the monitoring data under different data fusion methods.

[0158] This embodiment focuses on a combination of large-scale standardized engineering projects. By comprehensively analyzing the data fusion deviation performance of similar construction projects across these projects and the dynamic update efficiency of the monitoring target system, it identifies a fusion management strategy that can effectively control the spread of data fusion deviation while ensuring the self-optimization capability of the monitoring system. Its core logic lies in identifying and balancing two key dimensions: data fusion risk (reflected by the degree of deviation) and monitoring reliability (reflected by monitoring coverage and update activity). This method aims to avoid inhibiting the evolutionary capability of the monitoring system due to premature or excessive business control, while ensuring necessary intervention in high-risk business processes, thereby achieving a dynamic balance between maintaining data quality in the long term and ensuring business continuity in the short term.

[0159] Furthermore, the method for determining the fusion control method of the combination is as follows:

[0160] S41 determines the deviation handling process of the combination under different correlation fusion methods based on the monitoring data under different data fusion methods;

[0161] "Association and fusion method" here refers to the type of construction project (T-Foundation, T-Steel, T-Mechanical, T-Decoration). "Deviation handling process" refers to the abnormal records that occur in the data fusion of a specific project instance under a certain type.

[0162] This is the initial scan of the overall risk profile of the portfolio. Its core significance lies in quickly determining whether the problem is global or localized. By examining whether deviations occur in all business types, the severity and potential root causes of the risk can be preliminarily assessed. If deviations are present in all types, it may indicate a systemic flaw in the underlying data source or general rules, requiring immediate attention at the highest level. If deviations are present only in some types, the risk is relatively isolated, allowing for more granular analysis. This step provides the basis for qualitative risk assessment in subsequent decisions.

[0163] Specifically, the above steps include:

[0164] Case 1: If the combination has a deviation processing process under different association fusion methods, then the fusion control method of the combination is determined to be that no further fusion processing is performed on the monitoring targets except for the fusion processing.

[0165] The statement that "deviation handling processes exist under different association and fusion methods" is an "absolutely universal" condition. It requires that for each construction project type defined within the combination, at least one project instance experiences fusion deviation during the monitoring period.

[0166] This situation sets a red line for the highest level of risk warning. When all business types within the portfolio "fail," it strongly suggests that the risk is not isolated but may stem from common, systemic failures, such as: abnormal basic data interfaces, global business rule logic errors, or widespread problems with financial / tax system integration. In this case, the overall credibility of the data fusion environment is severely challenged. Therefore, the strictest "circuit breaker" mechanism must be activated immediately (i.e., "no more fusion processing is performed except for the fusion processing monitoring targets"), allowing only those projects already under close monitoring to continue operating, so as to concentrate resources on systemic troubleshooting, while freezing most business operations to prevent the widespread spread of erroneous data. This is an "emergency braking" measure to control global risks.

[0167] Case 2: If the combination has a deviation processing process under different association fusion methods, obtain the association fusion method with deviation processing process, and determine whether the proportion of the association fusion method with deviation processing process under the association fusion method is less than the preset association method proportion threshold. If yes, determine that the fusion control method of the combination does not require control processing. If not, proceed to step S42.

[0168] The "proportion of correlation fusion methods with deviation handling processes among all correlation fusion methods" is a supplement and quantitative refinement of Case 1. It calculates the percentage of construction project types with deviations out of the total number of construction project types within the combination. This is a "relatively universal" indicator.

[0169] The establishment of Scenario 2 embodies the "gray-scale decision-making" concept in risk management. Even if deviations occur in all types (satisfying Scenario 1), their severity may vary. By calculating the proportion of deviation types and comparing it with a preset threshold (λ), it can be determined whether the severity of this prevalence is within an acceptable "tolerable range." If the proportion is greater than the threshold, it indicates that the prevalence of deviations has exceeded tolerance, the risk signal is significant, and further in-depth assessment (S42) is necessary. If the proportion is less than the threshold, it means that although there are many deviation types, they have not yet reached the "unacceptable" threshold, and can be temporarily judged as "not requiring control," giving the system time to self-correct or observe. This step adds a buffer zone based on a quantitative threshold between "full circuit breaker" and "complete laissez-faire," making decision-making more refined and flexible.

[0170] S42 determines the updated data of the fusion processing monitoring target in the combination based on the update processing result;

[0171] "Updated processing results" and "updated data" are comprehensive inputs for evaluating the current effectiveness of the monitoring network, reflecting the system's static defense capability (coverage) and dynamic evolution capability (updating activity).

[0172] This step is the first key decision point in realizing the "balanced control" concept. It recognizes that implementing business controls solely based on deviation risk can have side effects: if the monitoring system itself is unreliable (low coverage) or inactive (not updated), strict controls will further reduce data flow, making it even harder for the monitoring system to discover new active items to update its target list, leading to a vicious cycle of "inefficient monitoring target updates." Therefore, the core logic of S42 is: prioritize assessing the reliability of the monitoring system itself. Only when the monitoring system is both "weak" (low coverage) and (inactive updates) should comprehensive business controls be considered; if the monitoring system demonstrates strong self-updating capabilities (high update activity), even if current coverage is insufficient, it should be given an "observation period" and "learning opportunity," temporarily suspending strict controls to protect its evolutionary potential and avoid damaging long-term monitoring effectiveness.

[0173] Specifically, the above steps may also include the following:

[0174] S421 uses the updated data of the fusion processing monitoring targets in the combination to determine the number of fusion processing monitoring targets in the combination, and determines whether the proportion of fusion processing monitoring targets in the combination is less than a preset monitoring target proportion threshold. If yes, proceed to step S422; otherwise, determine that the fusion control method of the combination is no longer to perform fusion processing except for fusion processing monitoring targets.

[0175] The "proportion of monitored targets" is a core indicator for measuring the static coverage integrity of the monitoring system.

[0176] This is the cornerstone of risk management. Low coverage means there are many "monitoring blind spots," where errors in data fusion activities will not be detected in a timely manner, resulting in a large overall risk exposure. When the coverage falls below the safety threshold, in order to ensure the efficiency of updating the monitoring targets for fusion processing and the reliability of monitoring fusion deviations, it is necessary to proceed to the next step for further judgment and identification.

[0177] S422 determines whether the number of updates to the monitored target of the fusion processing is greater than the preset update number threshold. If yes, proceed to step S43. If no, determine that the fusion control method of the combination is no control processing required.

[0178] "Update quantity" is a key indicator for quantifying the dynamic activity, sensitivity, and "learning ability" of the monitoring system.

[0179] This step is a core design element to avoid compromising the long-term evolutionary capability of the monitoring system due to premature control. When the monitoring system is actively and massively updating its target list (update count > ν), it indicates that its identification algorithm is effective and is striving to respond to business changes and capture new risk points or activity points. At this time, if comprehensive business control is implemented due to temporary insufficient static coverage, the data flow of these newly identified targets will be cut off. The direct consequence is that the system cannot verify the actual activity of these new targets, the update process becomes meaningless, and the optimization process of the monitoring network stagnates, resulting in the technical problem of "poor update efficiency." Therefore, determining it as "not requiring control" essentially provides a protected "learning window" for the monitoring system's self-improvement and adaptive optimization, prioritizing the long-term health and effectiveness of the monitoring system and improving the efficiency of updating monitored targets through integrated processing.

[0180] S43 uses the updated data of the monitoring target in the fusion processing of the combination and the deviation processing process under different association fusion methods to determine the fusion control method of the combination.

[0181] Furthermore, in the above steps, the updated fusion processing monitoring targets based on the associated fusion method are obtained, and it is determined whether the proportion of the updated fusion processing monitoring targets based on the associated fusion method is greater than a preset update proportion threshold. If so, the combined fusion control method is determined to be no fusion control processing under the associated fusion method. If not, fusion control processing is performed under the associated fusion method whenever the number of fusion monitoring targets with deviation processing under the associated fusion method does not meet the requirements, i.e., no more fusion processing is performed.

[0182] This step refines the decision-making granularity down to each construction project type with deviations. It performs an independent two-dimensional assessment for each type: monitoring reliability (the proportion of projects under this type that are monitored) and risk control effectiveness (the proportion of projects under this type with known deviations that are monitored).

[0183] This represents the final implementation and refined application of the "balanced control" concept. Within the combination, the impact of different construction projects on the updates of the integrated monitoring projects varies. For construction project types with high contribution rates, no control measures are needed given the existing poor monitoring reliability. However, for other construction project types, if the integration deviation is severe, control measures should be implemented as soon as possible, thus achieving balanced control over both monitoring reliability and integration deviation.

[0184] In one possible specific embodiment, the following is included:

[0185] Group name: "National Data Center Prefabricated Module Construction Project Group".

[0186] Composition and Scale: The project comprises 28 standardized data center module projects (numbered M-001 to M-028). Each project includes six types of prefabricated construction projects: Base module assembly (Type: C-Base), Power module installation (Type: C-Power), Cooling module installation (Type: C-Cooling), Network module installation (Type: C-Network), Security module installation (Type: C-Security), and Fire protection module installation (Type: C-Fire).

[0187] 3. Execution of coherent steps and decision deduction:

[0188] Step S41: Initial screening for the generality of deviations:

[0189] Example execution: Among the six types of construction projects, four types, C-Base, C-Power, C-Cooling, and C-Security, showed deviations, while two types, C-Network and C-Fire, showed no deviations.

[0190] Case 1: The condition "deviation exists in all types" is not met.

[0191] Scenario 2 Judgment: Deviation type proportion = 4 / 6 ≈ 66.7%. Judgment: 66.7% < λ(70%)? Yes.

[0192] Decision: The prevalence of the deviation is within the tolerable range, and it is directly determined that "no control or intervention is required". Process ends.

[0193] (To demonstrate the subsequent logic, a deduction branch is created: λ is reduced to 60%)

[0194] Derivation branch: Assuming λ=60%, then 66.7% > 60%, the deviation generally exceeds the threshold. Proceed to S42.

[0195] Step S42: Overall effectiveness assessment of the monitoring system;

[0196] S421 (Static Coverage Assessment): The proportion of all monitored targets = 21 integrated monitoring targets / 28 engineering projects = 75%. Judgment: 75% is greater than μ(65%), and the coverage rate meets the standard.

[0197] Decision: Based on the rules, the decision is directly made that "no further fusion processing will be performed on any monitoring targets except those that have undergone fusion processing." Process ends.

[0198] (Create another branch to enter S43: Increase μ to 80%)

[0199] Derivation Branch 2: Given λ=60%, let μ=80%. Then, in S421, we determine: 75% < 80% → Yes. Proceed to S422.

[0200] S422 (Update Activity Assessment): Total number of updates for the entire portfolio = 36. Judgment: 36 > ν(20), the overall monitoring system is actively updating, proceed to step S43;

[0201] Step S43: Assess update activity and risk by category, and determine a precise control plan:

[0202] This step involves an independent assessment for each type of construction project where deviations exist:

[0203] The "monitoring update activity" for this type refers to the proportion of items updated as monitoring targets for fusion processing under this type (the proportion of monitoring targets for fusion processing updated based on this association fusion method). If this proportion is high (>π), it indicates that this type is fully monitored due to active data fusion, making a significant contribution and having a high impact on the overall update activity; therefore, no control measures are required.

[0204] If updates are inactive, assess the necessity of risk management for this type: check if the "number of fusion monitoring targets with deviation handling processes" meets the requirement (i.e., deviation coverage ratio ≥ ρ). If it does not meet the requirement, it indicates a high risk and insufficient monitoring, requiring strict control; if it meets the requirement, implement general control measures (such as strengthening audits).

[0205] S43 Example Execution (Based on Derivation Branch 3):

[0206] Data Supplement (Proportion of Monitoring Targets Based on Update-Based Fusion Processing for Each Type):

[0207] C-Base type: In this update, 8 items were added as monitoring targets, accounting for approximately 28.6% of the total number of items in this type.

[0208] C-Power type: Two new monitoring targets have been added, with a ratio of 2 / 28 ≈ 7%.

[0209] C-Cooling type: 5 new monitoring targets added, proportion = 5 / 28 ≈ 17.9%.

[0210] C-Security type: 4 new monitoring targets added, proportion = 4 / 28 ≈ 14%.

[0211] Decisions are made for each type of deviation:

[0212] C-Base type:

[0213] Update activity is assessed: if the update ratio (28.6%) is greater than π (10%), the update activity is high, and according to the rules, no fusion control processing will be performed under this association fusion method. Similarly, no fusion control processing will be performed for C-Cooling and C-Security types.

[0214] C-Power type:

[0215] Assessing update activity: The update ratio (7%) is less than π (10%). This type of project exhibits fusion deviation in 9 projects, with 9 / 28 = 32%, which is greater than ρ (20%). This satisfies the condition that "the number of fusion monitoring targets in the deviation handling process does not meet the requirements." Therefore, fusion control processing will be implemented under this associated fusion method, meaning fusion processing will no longer be performed. The fusion function for all unmonitored projects under this type will be suspended.

[0216] This revised embodiment accurately identifies "healthy and active" business types by distinguishing between "updates generated by active data": by judging the proportion (π) of updates to monitored targets under a certain type, the system can identify those types that naturally attract monitoring attention (active updates) due to frequent data fusion activities. For these types, the system trusts their existing monitoring density and response speed, and does not impose additional controls, thereby avoiding interference with their efficient business flow. For types with inactive updates, the system further investigates their risk situation. If the risk situation is high, strict business suspension must be implemented to prevent the risk from spreading; if the risk situation is low, no control measures are required.

[0217] Example 2

[0218] Secondly, such as Figure 4 As shown, this application discloses an intelligent data fusion system for construction projects, employing the aforementioned intelligent data fusion method for construction projects, specifically including:

[0219] Fusion processing module, monitoring and processing module, fusion control module;

[0220] The fusion processing module is responsible for fusion processing of business data, financial data, tax data, and capital data from different engineering projects during the calling process;

[0221] The monitoring and processing module is responsible for determining the identification method of the fusion processing monitoring target in the project based on the construction project data in each project, and using the fusion processing monitoring target to monitor and process the deviation in the fusion processing process.

[0222] The fusion management module is responsible for dividing the engineering project into different combinations based on the construction project, and determining the fusion management method of the combination based on the update processing results of the fusion processing monitoring targets in the combination and the monitoring data under different data fusion methods.

[0223] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0224] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0225] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A smart data fusion method for construction projects, characterized in that, Specifically, it includes: The construction project data in the engineering project is determined. In the unified management and control platform, based on the number of engineering projects in different types and the types of construction projects in each engineering project, the identification method of the fusion processing monitoring target in the engineering project is determined. Based on the construction project of the fusion processing monitoring target, the monitoring processing data under different data fusion methods is determined. Based on the monitoring processing data and the associated fusion processing monitoring target, the monitoring and analysis requirement type under different data fusion methods is determined. The method for identifying the fusion processing monitoring targets in the project includes a preset identification method and a second preset identification method. Specifically, the method for identifying the fusion processing monitoring targets in the project is as follows: Based on the types of construction projects in each engineering project, the types of construction projects are deduplicated to obtain the total number of construction project types in all engineering projects. If the total number of construction project types is greater than a preset type number threshold, then the identification method of the fusion processing monitoring target in the engineering project is determined to be the preset identification method. If the total number of construction project types is not greater than the preset type number threshold, based on the number of projects in different type number intervals, determine the type number interval where the number of types is greater than the preset type number threshold, and take it as the target number interval. Determine whether the number of projects in the target number interval is greater than the preset project number threshold. If yes, then determine the identification method of the fusion processing monitoring target in the project as the preset identification method. If no, then determine the identification method of the fusion processing monitoring target in the project as the second preset identification method. The preset identification method is to obtain the number of data fusion processing times of the project in different construction project types, and to take the project with the number of data fusion processing times greater than the preset fusion processing time threshold as the fusion processing monitoring target in multiple construction project types, and to determine whether the data fusion result is accurate in the fusion processing process. The second preset identification method is to obtain the number of data fusion processing times of the project in different construction project types, and to take the projects with more than the target number of construction projects whose number of data fusion processing times is greater than the preset fusion processing time threshold as fusion processing monitoring targets, and to determine whether the data fusion results in the fusion processing process are accurate. Based on the construction project data, the project is divided into different combinations. The update strategy for the monitoring targets in the combination is determined by the fusion processing monitoring target data in the combination and the monitoring and analysis requirements under different data fusion methods. The update strategy is used to update the monitoring targets through fusion processing. Based on the update processing results and the monitoring data under different data fusion methods, the combined fusion control method is determined.

2. The intelligent data fusion method for construction projects as described in claim 1, characterized in that, The construction project data in the engineering project includes the types of construction projects within the engineering project.

3. The intelligent data fusion method for construction projects as described in claim 1, characterized in that, The data fusion method is divided according to the type of construction project. Based on the type of construction project, the number of times business data, financial data, tax data, and capital data are fused during the call process is divided into different data fusion methods.

4. The intelligent data fusion method for construction projects as described in claim 1, characterized in that, The monitoring and processing data under the data fusion method includes the number of monitoring and processing processes under the data fusion method and the data fusion deviation in different monitoring and processing processes.

5. The intelligent data fusion method for construction projects as described in claim 1, characterized in that, The method for determining the monitoring and analysis requirement type under the aforementioned data fusion approach is as follows: Based on the monitoring and processing data under the aforementioned data fusion method, determine the monitoring and processing process under the aforementioned data fusion method; Based on the data fusion deviation in different monitoring and processing processes under the aforementioned data fusion method, the monitoring and processing processes with data fusion deviation are identified and treated as deviation processing processes. Based on the monitoring and processing procedures and deviation handling procedures under the data fusion method, and in conjunction with the fusion processing monitoring targets associated with the data fusion method, the monitoring and analysis requirement type under the data fusion method is determined.

6. The intelligent data fusion method for construction projects as described in claim 5, characterized in that, The fusion processing monitoring target associated with the data fusion method is the fusion processing monitoring target that undergoes fusion monitoring processing under the data fusion method.

7. The intelligent data fusion method for construction projects as described in claim 1, characterized in that, The method for determining the fusion control method of the combination is as follows: Based on monitoring data under different data fusion methods, determine the deviation handling process of the combination under different correlation fusion methods; Based on the update processing results, the updated data of the fusion processing monitoring target in the combination is determined; By utilizing the updated data of the monitoring targets in the fusion processing of the combination and the deviation processing process under different correlation fusion methods, the fusion control method of the combination is determined.

8. An intelligent data fusion system for construction projects, employing the intelligent data fusion method for construction projects as described in any one of claims 1-7, characterized in that, Specifically, it includes: Fusion processing module, monitoring and processing module, fusion control module; The fusion processing module is responsible for fusion processing of business data, financial data, tax data, and capital data from different engineering projects during the calling process; The monitoring and processing module is responsible for determining the identification method of the fusion processing monitoring target in the project based on the number of projects in different types and the type of construction project in each project, and using the fusion processing monitoring target to monitor and process the deviation in the fusion processing process. The fusion management module is responsible for dividing the engineering project into different combinations based on the construction project, and determining the fusion management method of the combination based on the update processing results of the fusion processing monitoring targets in the combination and the monitoring data under different data fusion methods.

Images