Project full-period operation management system and method based on multi-dimensional data integration

Abstract

The invention discloses a project full-period operation management system and method based on multi-dimensional data integration. The method comprises the following steps: acquiring multi-dimensional data of historical construction projects to construct and form a multi-dimensional feature matrix; screening the multi-dimensional feature matrix by using a random forest algorithm to obtain key features to generate multiple sets of initial plan templates; performing similarity matching through a BP neural network model to obtain an optimal plan template; drawing a project full-cycle operation data list map according to the optimal plan template; obtaining monthly operation data of a new construction project corresponding to the optimal plan template and generating a project operation data graph; comparing the project operation data graph with the project full-period operation data list graph, and calculating to obtain a deviation value; and based on the deviation value, calculating to obtain a deviation correction plan value of the next month. According to the invention, the problem that the decision-making efficiency of construction project operation management is affected due to many and miscellaneous operation data involved in existing construction project management is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of building construction, and in particular to a project full-cycle operation management system and method based on multidimensional data integration. Background Art

[0002] Currently, construction project management involves a lot of complex operational data, which may lead to the following problems:

[0003] 1. Data dispersion: Data such as project cash flow, output value, and labor are managed in a decentralized manner, lacking a unified view and making it difficult to achieve full-cycle, integrated analysis.

[0004] 2. Static chart defects: Traditional charts cannot dynamically update actual data and conduct actual comparisons, and are not combined with environmental factors (climate, temperature, etc.), affecting the overall risk prediction ability of the project.

[0005] 3. Insufficient visualization: The charts generated by existing tools contain redundant information, and key indicators (such as nodes and peaks) are not clearly marked, which cannot meet project requirements.

[0006] 4. Lack of standardization: The lack of unified drawing specifications and review mechanisms leads to uneven content of charts and graphs, affecting the efficiency of management decisions.

[0007] Therefore, there is an urgent need for a standardized, dynamic, and multi-dimensional integrated project operation data management system to deduce the future development and expectations of the project. Summary of the Invention

[0008] In order to overcome the defects of the existing technology, a project full-cycle operation management system and method based on multi-dimensional data integration are provided to solve the problem that the existing construction project management involves a lot of complex operation data, which affects the efficiency of construction project operation management decision-making.

[0009] To achieve the above objectives, a project full-cycle operation management system based on multi-dimensional data integration is provided, which includes the following steps:

[0010] Obtain multi-dimensional data of historical building projects to construct a multi-dimensional feature matrix;

[0011] Using a random forest algorithm to screen the multi-dimensional feature matrix to obtain key features and generate multiple sets of initial planning templates;

[0012] The optimal plan template is obtained by similarity matching through BP neural network model;

[0013] Draw an overview of the project's full-cycle operational data using the optimal plan template;

[0014] Obtaining monthly operation data of a new construction project corresponding to the optimal plan template and generating a project operation data graph;

[0015] Compare the project operation data graph with the project full cycle operation data overview graph to calculate the deviation value;

[0016] Based on the deviation value, the correction plan value for the next month is calculated.

[0017] Furthermore, the multi-dimensional data includes BIM model parameters, geological exploration data, contract duration requirements, historical project database and resource input intensity. The multi-dimensional feature matrix is:

[0018] X=[F BIM , F geo , T conteact , H DB , R input ];

[0019] Among them, F BIM is the characteristic parameter of Building Information Model (BIM);

[0020] F geo is the characteristic parameter of geological exploration;

[0021] T conteact The contract duration requirements;

[0022] H DB For historical project database features;

[0023] R input It is the resource input intensity characteristic.

[0024] Furthermore, in the step of obtaining the optimal plan template by performing similarity matching through the BP neural network model, the similarity matching calculation formula is:

[0025]

[0026] Among them, P i Provide multiple sets of initial planning templates;

[0027] φ() is the feature encoding function;

[0028] X is the multi-dimensional feature matrix of the current project.

[0029] Furthermore, when implementing the step of calculating and obtaining the deviation value, the deviation value is obtained by calculating using a deviation calculation model, and the deviation calculation model is:

[0030]

[0031] Among them, δ t is the actual deviation rate at time point t;

[0032] A t Actual operational data;

[0033] P t To plan operational data;

[0034] λ is the time decay factor;

[0035] t0 is the planned reference time point;

[0036] t is the current comparison time point.

[0037] Furthermore, the step of calculating the correction plan value for the next month includes calculating the correction plan value for the next month through an intelligent compensation model, wherein the intelligent compensation model is:

[0038]

[0039] in, is the original next period planned value;

[0040] max(0,δ t -2%) is the excess deviation;

[0041] α is the compensation coefficient.

[0042] Furthermore, it also includes obtaining historical climate data and actual climate data, and calculating the probability of climate interruption during the construction phase based on the historical climate data and actual climate data.

[0043] Furthermore, the probability of climate interruption during the construction phase is calculated using a climate interruption probability model, which is:

[0044]

[0045] Among them, ρ stop The probability of construction interruption due to climate;

[0046] R α is the actual climate parameter value;

[0047] R C is the critical threshold.

[0048] The present invention provides a project full-cycle operation management system based on multidimensional data integration, which adopts a project full-cycle operation management method based on multidimensional data integration, comprising:

[0049] An acquisition module for obtaining multi-dimensional data, monthly operation data, historical climate data, and actual climate data of historical building projects;

[0050] a modeling module comprising a modeling unit for constructing a multidimensional feature matrix based on the multidimensional data, a screening unit for screening the multidimensional feature matrix using a random forest algorithm to obtain key features to generate multiple sets of initial plan templates, and a computing unit for performing similarity matching using a BP neural network model to obtain an optimal plan template, wherein the modeling unit is connected to the acquisition module, the screening unit is connected to the modeling unit, and the computing unit is connected to the screening unit;

[0051] A first drawing module for drawing an overview of the full-cycle operation data of the project using the optimal plan template, connected to the calculation unit;

[0052] a second mapping module for collecting monthly operation data of a new construction project corresponding to the optimal plan template and generating a project operation data graph, connected to the acquisition module;

[0053] A comparison module for comparing the project operation data graph with the project full-cycle operation data overview graph and calculating a deviation value, connected to the first drawing module and the second drawing module;

[0054] A deviation correction module for calculating the deviation correction plan value for the next month based on the deviation value is connected to the comparison module.

[0055] Furthermore, it also includes a prediction module for calculating the probability of climate interruption during the construction phase based on the historical climate data and actual climate data, which is connected to the acquisition module.

[0056] The beneficial effect of the present invention is that the project full-cycle operation and management system based on multi-dimensional data integration of the present invention is calculated and drawn into charts through standardized templates and automated calculations. The project full-cycle operation and management system based on multi-dimensional data integration of the present invention conducts linkage analysis of the full-cycle data of the construction project, and supports decisions such as cash flow warning and labor deployment. The project full-cycle operation and management system based on multi-dimensional data integration of the present invention integrates gaps in environmental climate data to reduce the impact of climate factors on the construction period, and enhances the project's risk resistance. The project full-cycle operation and management system based on multi-dimensional data integration of the present invention integrates a full-cycle operation data overview of the project and a project operation data chart with multi-chart overlay and dynamic annotation design, and key information (nodes, peaks) is intuitive and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0058] Figure 1 Schematic diagram of the structure of a project full-cycle operation management system based on multi-dimensional data integration according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0060] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0061] Reference Figure 1 As shown, the present invention provides a project full-cycle operation and management system based on multi-dimensional data integration, including: an acquisition module 1, a modeling module 2, a first drawing module 3, a second drawing module 4, a comparison module 5, and a correction module 6.

[0062] Among them, the acquisition module 1 is used to obtain multi-dimensional data, monthly operation data, historical climate data and actual climate data of historical building projects.

[0063] Among them, multi-dimensional data includes more than 20 dimensional data such as project cash flow, output value (total / sub-professional), labor (construction / management personnel), and environmental data (rainfall, temperature peak).

[0064] The modeling module 2 includes a modeling unit 21, a screening unit 22, and a calculation unit 23. The modeling unit 21 is connected to the acquisition module 1. The screening unit 22 is connected to the modeling unit 21. The calculation unit 23 is connected to the screening unit 22.

[0065] The modeling unit 21 is used to construct a multi-dimensional feature matrix based on the multi-dimensional data.

[0066] The screening unit 22 uses a random forest algorithm to screen the multi-dimensional feature matrix to obtain key features and generate multiple sets of initial planning templates.

[0067] The calculation unit 23 performs similarity matching through the BP neural network model to obtain the optimal plan template.

[0068] The first drawing module 3 is connected to the calculation unit 23. The first drawing module 3 is used to draw an overview of the project's full-cycle operation data using the optimal plan template.

[0069] The second drawing module 4 is connected to the acquisition module 1. The second drawing module 4 is used to generate a project operation data graph based on the monthly operation data of the new construction project corresponding to the optimal plan template.

[0070] In this embodiment, the first drawing module 3 and the second drawing module 4 respectively use Origin software to perform drawing.

[0071] Specifically, 1. Data preparation and import.

[0072] Based on the optimal plan template, obtain multi-dimensional data including the overall progress plan, output value of each profession, number of laborers, etc.

[0073] Import multi-dimensional data into the worksheet of Origin2024 and classify it according to preset rules (such as cash flow, output value, and labor).

[0074] 2. Chart generation and annotation

[0075] Execute the software script to generate the basic chart and adjust it according to the following rules:

[0076] Cash flow bar chart (light blue) superimposed on line chart, with labels centered;

[0077] The per capita output value bar chart (green) and line chart are displayed on two axes;

[0078] The professional output value / number of people are displayed in a stacked bar chart, with red, blue and yellow as the colors.

[0079] Mark key nodes (differentiated by color), climate data (divided by horizontal axis), legend (lower right corner) and project overview (upper left corner).

[0080] 3. Output and dynamic management.

[0081] The second drawing module updates the project operation data graph every month based on the monthly operation data of the new construction project corresponding to the optimal plan template.

[0082] The comparison module 5 is connected to the first drawing module 3 and the second drawing module 4. The comparison module 5 is used to compare the monthly updated project operation data map with the project full cycle operation data overview map and calculate the deviation value.

[0083] The deviation correction module 6 is connected to the comparison module 5. The deviation correction module 6 is used to calculate the deviation correction plan value for the next month based on the deviation value.

[0084] Monthly actual data is compared with planned data in real time, with deviations noted and alerts triggered to support project management and correction. Project managers organize special meetings to analyze the causes of deviations and develop correction plans based on the correction plan values, which are then marked on the chart.

[0085] When encountering major changes in the construction period, regenerate the charts and submit them for review to ensure that the data is synchronized with the plan.

[0086] In this embodiment, the project full-cycle operation and management system based on multi-dimensional data integration of the present invention further includes a prediction module 7. The prediction module 7 is connected to the acquisition module 1. The prediction module 7 is used to calculate the probability of climate disruption during the construction phase based on historical climate data and actual climate data.

[0087] The prediction module integrates the climate data of the past five years (rainfall, snowfall, temperature peak / valley values) and marks it on the project's full-cycle operation data overview to assist in construction cycle planning and emergency plan formulation.

[0088] In this embodiment, the project lifecycle operations management system based on multidimensional data integration of the present invention further includes a display module. The display module is connected to the first drawing module, the second drawing module, the error correction module, and the forecasting module. The display module is used to display an overview of the project lifecycle operations data, multiple charts of the project operations data, risk forecast results, and other information.

[0089] The project full-cycle operation management system based on multi-dimensional data integration of the present invention shortens the compilation time by more than 50% through standardized templates, automatic calculation and drawing of charts.

[0090] The project full-cycle operation management system based on multi-dimensional data integration of the present invention conducts linkage analysis of full-cycle data of construction projects and supports decisions such as cash flow warning and labor deployment.

[0091] The project full-cycle operation management system based on multi-dimensional data integration of the present invention integrates gaps in environmental climate data to reduce the impact of climate factors on the construction period and improve the project's risk resistance.

[0092] The project full-cycle operation management system based on multi-dimensional data integration of the present invention integrates a project full-cycle operation data overview diagram and a project operation data diagram, and multiple charts are superimposed and dynamically labeled, so that key information (nodes, peaks) is intuitive and efficient.

[0093] The present invention provides a project full-cycle operation management method based on multidimensional data integration, comprising the following steps:

[0094] S1. Obtain multi-dimensional data of historical building projects to construct a multi-dimensional feature matrix.

[0095] Specifically, multi-dimensional data includes BIM model parameters, geological exploration data, contract duration requirements, historical project database and resource input intensity.

[0096] The multi-dimensional feature matrix is:

[0097] X=[F BIM , F geo , T conteact , H DB , R input ];

[0098] Among them, E BIM It is a characteristic parameter of the Building Information Model (BIM), which includes structured data such as the project's three-dimensional geometric information, construction process, and project type.

[0099] F geo These are characteristic parameters of geological exploration, including geomechanical data such as foundation bearing capacity, groundwater level, and rock and soil types.

[0100] T conteact The contract duration requirement is the contract duration constraint, which refers to the key node times and total duration requirements agreed in the contract.

[0101] H DB is the historical project database feature. In this embodiment, H DB Covers experience data such as construction period, resource consumption, risk events, etc. of more than 1,000 historical projects.

[0102] R input It is the resource input intensity characteristic, including the input per unit time of manpower, machinery and materials.

[0103] S2. Use the random forest algorithm to screen the multi-dimensional feature matrix to obtain key features and generate multiple sets of initial planning templates.

[0104] In this embodiment, a multi-dimensional feature matrix is ​​screened by a random forest algorithm to obtain key features to generate three sets of initial planning templates.

[0105] S3. Obtain the optimal plan template through similarity matching using the BP neural network model.

[0106] Specifically, in the step of obtaining the optimal plan template by performing similarity matching through the BP neural network model, the similarity matching calculation formula is:

[0107]

[0108] Among them, P i These are multiple sets of initial plan templates, referring to three sets of initial plans (P1, P2, and P3) generated by the random forest algorithm.

[0109] φ() is a feature encoding function implemented by BP neural network, which maps the input data into a low-dimensional feature space to calculate the similarity.

[0110] X is the multi-dimensional feature matrix of the current project.

[0111] 90% CI: represents the reliable range of the planned base recommendation.

[0112] S4. Use the optimal plan template to draw an overview of the project's full-cycle operational data.

[0113] S5. Obtain monthly operation data of the new construction project corresponding to the optimal plan template and generate a project operation data graph.

[0114] S6. Compare the project operation data chart with the project full-cycle operation data overview chart to calculate the deviation value.

[0115] Specifically, when performing the step of calculating and obtaining the deviation value, the deviation value is obtained by calculating using a deviation calculation model, and the deviation calculation model is:

[0116]

[0117] Among them, δ t It is the actual deviation rate at time point t, reflecting the percentage difference between the actual value and the planned value.

[0118] A t It is the actual operation data, the completion amount or progress value at time point t.

[0119] P t is the planned operation data, the expected target value at time point t.

[0120] λ is the time decay factor, ranging from 0.1 to 0.3, which is used to reduce the impact of historical deviations on the current assessment.

[0121] t0 is the planning benchmark time point, usually the project launch date or the planning model start date.

[0122] t is the current comparison time point.

[0123] The time decay factor is calculated as:

[0124] λ=0.1+0.02×(t-t0) (0.1+λ≤0.3).

[0125] S7. Based on the deviation value, calculate the correction plan value for the next month.

[0126] Specifically, the step of calculating the correction plan value for the next month includes calculating the correction plan value for the next month through an intelligent compensation model. The intelligent compensation model is:

[0127]

[0128] in, It is the original planned value for the next period, which is the initial setting value without considering the impact of deviation.

[0129] max(0,δ t-2%) is the excess deviation, and only the deviation exceeding the 2% threshold is compensated.

[0130] α is the compensation coefficient, which is dynamically adjusted according to the degree of deviation (0.3 to 0.5).

[0131] The selection rule of compensation coefficient α is:

[0132]

[0133] When the deviation δ≥ compensation difference, the next period plan value is automatically corrected.

[0134] S8. Obtain historical climate data and actual climate data, and calculate the probability of climate disruption during the construction phase based on the historical climate data and the actual climate data.

[0135] Specifically, the probability of climate interruption during the construction phase is calculated using the climate interruption probability model, which is:

[0136]

[0137] Among them, ρ stop The probability of construction interruption due to climate is capped at 90%.

[0138] R α are actual climate parameter values ​​(such as daily rainfall and temperature extremes);

[0139] R c It is a critical threshold, for example, the rainfall threshold in the earthwork stage is 30 mm / day.

[0140] In addition, the probability of climate risk occurrence is calculated through the Monte Carlo risk simulation algorithm.

[0141] Monte Carlo risk simulation algorithm (N = 10,000 times):

[0142]

[0143] Among them, π risk is the probability of occurrence of climate risk, and the mean value is calculated through 10,000 simulations.

[0144] The extreme climate value (such as the number of consecutive high temperature days) generated by the k-th simulation.

[0145] is the 95% quantile threshold of historical climate data.

[0146] I(): indicator function, which takes 1 when the simulation value exceeds the threshold, otherwise it takes 0.

[0147] The probability of climate risk occurrence and the probability of climate interruption during the construction phase constitute a multi-level early warning coordination mechanism. t ≥10%, and π risk When the risk is ≥0.5, a company-level emergency response will be initiated.

[0148] The present invention provides a project full-cycle operation data management system. Through multi-dimensional data integration, dynamic visualization charts and environmental prediction modules, it solves the problems of data dispersion, update lag and insufficient risk prediction in the existing technology, and significantly improves project management efficiency and decision-making scientificity.

[0149] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention herein is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the inventive concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features having similar functions disclosed in this application.

Claims

1. A project full-cycle operation management method based on multi-dimensional data integration, characterized by: The following steps are involved: Obtain multi-dimensional data of historical building projects to construct a multi-dimensional feature matrix; Using a random forest algorithm to screen the multi-dimensional feature matrix to obtain key features and generate multiple sets of initial planning templates; The optimal plan template is obtained by similarity matching through BP neural network model; Draw an overview of the project's full-cycle operational data using the optimal plan template; Obtaining monthly operation data of a new construction project corresponding to the optimal plan template and generating a project operation data graph; Compare the project operation data graph with the project full cycle operation data overview graph to calculate the deviation value; Based on the deviation value, the correction plan value for the next month is calculated.

2. The project full-cycle operation management method based on multidimensional data integration according to claim 1 is characterized in that: The multi-dimensional data includes BIM model parameters, geological exploration data, contract duration requirements, historical project database and resource input intensity. The multi-dimensional feature matrix is: X=[F BIM ,F geo ,T conteact ,H DB ,R input ]; Among them, F BIM is the characteristic parameter of Building Information Model (BIM); F geo is the characteristic parameter of geological exploration; T conteact The contract duration requirements; H DB For historical project database features; R input It is the resource input intensity characteristic.

3. The project full-cycle operation management method based on multidimensional data integration according to claim 1 is characterized in that: In the step of obtaining the optimal plan template by performing similarity matching through the BP neural network model, the similarity matching calculation formula is: Among them, P i Provide multiple sets of initial planning templates; φ() is the feature encoding function; X is the multi-dimensional feature matrix of the current project.

4. The project full-cycle operation management method based on multidimensional data integration according to claim 1 is characterized in that: When performing the step of calculating and obtaining the deviation value, the deviation value is obtained by calculating using a deviation calculation model, and the deviation calculation model is: Among them, δ t is the actual deviation rate at time point t; A t Actual operational data; P t To plan operational data; λ is the time decay factor; t0 is the planned reference time point; t is the current comparison time point.

5. The project full-cycle operation management method based on multidimensional data integration according to claim 4 is characterized in that: The step of calculating and obtaining the correction plan value for the next month includes calculating and obtaining the correction plan value for the next month through an intelligent compensation model, wherein the intelligent compensation model is: in, is the original next period planned value; max(0,δ t -2%) is the excess deviation; α is the compensation coefficient.

6. The project full-cycle operation management method based on multidimensional data integration according to claim 1 is characterized in that: The method also includes obtaining historical climate data and actual climate data, and calculating the probability of climate interruption during the construction phase based on the historical climate data and the actual climate data.

7. The project full-cycle operation management method based on multidimensional data integration according to claim 6 is characterized in that: The probability of climate interruption during the construction phase is calculated using a climate interruption probability model, which is: Among them, ρ stop The probability of construction interruption due to climate; R α is the actual climate parameter value; R C is the critical threshold.

8. A project full-cycle operation management system based on multidimensional data integration, which adopts the project full-cycle operation management method based on multidimensional data integration as claimed in any one of claims 1 to 9, characterized in that: include: An acquisition module for obtaining multi-dimensional data, monthly operation data, historical climate data, and actual climate data of historical building projects; a modeling module comprising a modeling unit for constructing a multidimensional feature matrix based on the multidimensional data, a screening unit for screening the multidimensional feature matrix using a random forest algorithm to obtain key features to generate multiple sets of initial plan templates, and a computing unit for performing similarity matching using a BP neural network model to obtain an optimal plan template, wherein the modeling unit is connected to the acquisition module, the screening unit is connected to the modeling unit, and the computing unit is connected to the screening unit; A first drawing module for drawing an overview of the full-cycle operation data of the project using the optimal plan template, connected to the calculation unit; a second mapping module for collecting monthly operation data of a new construction project corresponding to the optimal plan template and generating a project operation data graph, connected to the acquisition module; A comparison module for comparing the project operation data graph with the project full-cycle operation data overview graph and calculating a deviation value, connected to the first drawing module and the second drawing module; A correction module for calculating the correction plan value for the next month based on the deviation value is connected to the comparison module.

9. The project full-cycle operation management system based on multi-dimensional data integration according to claim 8 is characterized in that: It also includes a prediction module for calculating the probability of climate interruption during the construction phase based on the historical climate data and actual climate data, and is connected to the acquisition module.

Images