Intelligent project dynamic risk monitoring and evaluating method and system

Abstract

The invention discloses an intelligent project dynamic risk monitoring and evaluation method and system, and belongs to the technical field of project management, and the method comprises the steps: dynamically collecting the data of each stage and each dimension of a project, including project data, project approval, project scheduling, project progress pictures, and project bidding and tendering data, and forming a project portrait; based on risk evaluation indexes set by the system, project portraits are taken as the basis, project progress early warning, monitoring and project overall condition analysis are realized by fusing project early warning analysis and judgment and project overall analysis and judgment algorithms, and whether the project has abnormal progress or not is automatically judged and early warning is given out through a project progress early warning analysis algorithm; and related managers research and judge the overall progress of the project according to the information of each dimension of the early warning project. According to the invention, the workload and cost of manual inspection are effectively saved, and data support is provided for subsequent project file query and overall control of project risks by managers.

Description

Technical Field

[0001] The present invention relates to the technical field of project management, in particular to an intelligent project dynamic risk monitoring and evaluation method and system. Background Art

[0002] With the continuous development of information technology, the business handling of investment projects throughout the entire life cycle, from project pre-planning, project approval, project reserve, project commencement, and project completion acceptance, has become increasingly mature. At present, the application level of project management platforms is constantly improving across the country, providing basic support for investment management. The platform can benefit all types of investors and help the development and reform departments improve their investment management level. Various regions are also successively proposing new ideas, innovating management methods, exploring the practical application of emerging technologies and big data, optimizing and improving platform functions, and using information technology to solve problems in the project management process.

[0003] At present, in the process of classified management of investment projects across the country, investment projects are divided into national key projects, provincial key projects, and municipal key projects. Whether these major projects can be implemented in a timely manner, whether they are actually started, and the specific progress after the start of construction are the core points of concern for leaders. How to use intelligent analysis and judgment to provide leaders with real-time and accurate progress analysis is an urgent issue.

[0004] At present, there are still the following problems in project supervision:

[0005] First, the project monitoring and early warning capabilities are weak. In the past, projects have been managed through a monthly manual review to ensure data quality. For relatively high-quality key projects, this is barely sufficient, but when dealing with full-caliber project management, project monitoring becomes difficult and the review tasks are arduous. There is a lack of effective technical means to implement intelligent, refined, and standardized effective monitoring of full-caliber projects, and it is impossible to timely warn and discover problems and solve them.

[0006] Second, the comprehensive analysis capability is insufficient. At present, the data analysis for projects is just simple project data statistics, lacking the ability of multi-dimensional and multi-level data fusion and collision, and unable to form effective predictive analysis; in terms of data management, only production data management is carried out, lacking decision-making data management means formed by multiple factors. Summary of the invention

[0007] The technical task of the present invention is to address the above shortcomings and provide an intelligent project dynamic risk monitoring and assessment method and system, which effectively saves the workload and cost of manual inspections. The relevant inspection data automatically forms a portrait of the project, providing data support for subsequent managers to query project files and overall control of project risks.

[0008] The technical solution adopted by the present invention to solve its technical problem is:

[0009] An intelligent project dynamic risk monitoring and assessment method.

[0010] Dynamically collect data at all stages and dimensions of the project, including project data, project approval, project scheduling, project progress pictures, and project bidding data, to form a project portrait;

[0011] Based on the risk assessment indicators set by the system and project portraits, the system integrates the early warning analysis and judgment of the project and the overall analysis and judgment algorithm of the project to realize early warning, monitoring and overall situation analysis of the project progress. The early warning and monitoring of the project progress is to superimpose the image progress photos and video information of each project at different times, and judge the project progress through computer vision technology. At the same time, it integrates the progress status and approval procedures reported by the project, and automatically judges whether there is abnormal progress in the project through the project progress early warning analysis algorithm and issues an early warning;

[0012] Relevant managers will analyze the overall progress of the project based on the information of various dimensions of the early warning project, confirm whether the progress of the project is abnormal, and include abnormal projects in the progress tracking and supervision.

[0013] Based on the collected project-related data, the overall project risk is assessed through the constructed project dynamic risk monitoring and assessment model, thereby obtaining the project risk assessment result; through the set risk threshold, early warning of project risks is issued. Through this intelligent dynamic analysis, the management department can understand the risks of the project in real time, intervene in risky projects in a timely manner, reduce potential risks, and provide data support for regulatory departments to better manage projects.

[0014] Furthermore, the implementation of this method includes:

[0015] Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision;

[0016] The key image feature library of the picture uses Floyd-Warshall and edge feature algorithms to identify the features of the identification image of each project, and records the key features in the image such as color, texture, shape, contour, space, area, etc. for each project into the feature library. The distance between each node in the image is calculated by the algorithm to generate the surveying and mapping feature library of the project. This feature library is used as the basic model for each video key frame image comparison; at the same time, this feature library supports the key feature recognition when manually recording the image into the feature library after recognition;

[0017] The drone aerial photography data collection model automatically issues aerial photography tasks to projects that have entered the construction phase, and automatically collects aerial photography data based on the system's project drone aerial photography code;

[0018] Self-learning sample library, through the continuous learning of image features in aerial photography data, to achieve effective analysis of image data;

[0019] The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects.

[0020] Furthermore, it supports docking with the online approval and supervision platform for investment projects, automatically collects projects in the region, and forms a sample library of projects; it supports docking with the approval management system for engineering construction projects, obtains project approval data, and forms a project approval library; it supports docking with the major project management and scheduling system, obtains project scheduling data in stages, and forms a project scheduling library;

[0021] The project library forms a library of started projects based on the analysis of collected data, and customizes drone collection tasks for the started projects, and collects drone aerial photography image data through the project aerial photography code.

[0022] Furthermore, multiple image recognition algorithms are built in, and the grayscale histogram equalization algorithm and edge feature algorithm are used to identify the features of the identification image of each project. After identifying the key features in the image, such as buildings, machines, equipment, etc., they are recorded in the feature library for each project. This feature library is used as the basic model for comparing the drone images collected each time; at the same time, the feature library supports the recognition of key features of images when they are manually recognized and recorded in the feature library;

[0023] For the projects in the started project database, according to the collected drone images, the model forms an automatic project progress change judgment model based on the drone image feature value, feature value weight definition, image feature similarity ratio and related algorithms, automatically analyzes the project progress of each collection cycle, and obtains the results of project progress changes;

[0024] Through machine learning models, the project progress is automatically analyzed and learned, and the indicators for judging the project progress are continuously adjusted to improve the accuracy of changes in project progress.

[0025] Furthermore, after collecting the relevant data, an overall risk assessment of all projects in the entire domain is automatically conducted based on the set risk assessment model, and corresponding risk assessment results are issued to provide data support for the management department to control project risks in a timely manner.

[0026] Furthermore, the specific implementation process of this method is as follows:

[0027] 1) Risk assessment indicator item setting: the system has a built-in indicator item library to initialize and modify the indicator items involved in the risk assessment model. The modified rules are directly applied to monitoring;

[0028] 2) Initialize the project database, connect to the online approval and supervision platform for investment projects to obtain project data, and collect basic project information and project unit information data into the project database;

[0029] 3) Project scheduling information collection: connect to the major project scheduling management system to collect scheduling data for project planning, project approval, project commencement, and project completion in stages;

[0030] 4) Project approval data collection: connect to the construction project approval management system and collect and store the data of major approval items related to all projects in the project library through the unique code of the project;

[0031] 5) Project drone aerial photography data collection: After completing the above data collection work, a preliminary analysis is conducted to identify the projects that have obtained the construction permit. The results of the determination are divided into projects to generate corresponding drone aerial photography codes, which are pushed to the drone aerial photography organizations in the region through the SMS platform. After receiving the data collection task, the drone aerial photography organizations grab the order. The projects that grab the order organize drones to conduct aerial photography at a fixed time period. The aerial photography results are uploaded to the drone aerial photography data collection model by scanning the corresponding aerial photography code.

[0032] 6) Feature recognition of project-based location environment images: Use grayscale histogram equalization algorithm and edge feature algorithm to recognize the features of each project's identification image, identify key features in the image such as color, texture, shape, contour, space, area, building, equipment, material, etc., and then record them in the inspection recognition feature library for each project. This feature library is used as the comparison standard for each drone aerial photo of the project; at the same time, the feature library also supports the improvement of key feature recognition features of the image based on the original recognition through manual methods;

[0033] 7) Definition of project progress change model, which supports the definition of automatically identified image feature values ​​and feature value weights and automatically compared similarity ratios of multiple images to form a project progress change model definition;

[0034] 8) Setting of warning value for project progress changes: setting of system warning value for the final determined image similarity comparison, including the size of project progress changes; through the setting of warning value, monitoring of a large number of project drone aerial photography data, obtaining project progress analysis of each project, allowing management personnel to directly obtain the overall construction progress of the project through the changes in different periods of the corresponding project, without entering the site for viewing, and real-time dynamic monitoring can be performed, thereby improving work efficiency;

[0035] 9) Machine deep learning training. The system has a built-in automatic learning training sample library and automatic learning device. Through continuous learning of on-site image feature data of different projects and comparison of drone aerial image data of project cycle scheduling, it can determine whether the project has started, whether it is under construction in the near future, and whether it has been completed, and obtain the results of project progress analysis;

[0036] 10) Project risk assessment model: after completing the data collection and project progress data analysis of the project according to the set risk assessment indicators, conduct a comprehensive analysis of the data related to the project profile to obtain the risk assessment results of each project;

[0037] 11) Confirmation of risk assessment results. The results of project risk assessment are manually reviewed according to the risk assessment level for projects with serious risks. The review process refers to the data of project portraits and the data of project drone aerial photography for comprehensive analysis. Accurate projects are entered into the project key risk analysis database, and managers can focus on tracking and monitoring key projects.

[0038] Furthermore, the project approval data collection, the main approval matters related to the project include preliminary planning data, intermediary service data, project unit information, project approval data, project approval materials, project approval results and the construction land planning permit information, construction project planning permit information, construction design document review, construction project construction permit information, and completion acceptance filing information involved in the entire life cycle of the project.

[0039] The present invention also claims protection for an intelligent project dynamic risk monitoring and assessment system, comprising:

[0040] Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision;

[0041] The key image feature library of the picture uses Floyd-Warshall and edge feature algorithms to identify the features of the identification image of each project, and records the key features in the image such as color, texture, shape, contour, space, area, etc. for each project into the feature library. The distance between each node in the image is calculated by the algorithm to generate the surveying and mapping feature library of the project. This feature library is used as the basic model for each video key frame image comparison; at the same time, this feature library supports the key feature recognition when manually recording the image into the feature library after recognition;

[0042] The drone aerial photography data collection model automatically issues aerial photography tasks to projects that have entered the construction phase, and automatically collects aerial photography data based on the system's project drone aerial photography code;

[0043] Self-learning sample library, through the continuous learning of image features in aerial photography data, to achieve effective analysis of image data;

[0044] The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects.

[0045] The system realizes the above-mentioned intelligent project dynamic risk monitoring and assessment method.

[0046] The present invention also claims protection for an intelligent project dynamic risk monitoring and assessment device, comprising at least one memory and a processor;

[0047] The at least one memory is used to store a machine-readable program;

[0048] The at least one processor is used to call the machine-readable program to implement the above method.

[0049] The present invention also claims protection for a computer-readable medium having computer instructions stored thereon, which implement the above method when executed by a processor.

[0050] Compared with the prior art, the intelligent project dynamic risk monitoring and assessment method and system of the present invention has the following beneficial effects:

[0051] 1. Supports the collection of project data from multiple sources. It can collect basic project data, approval data, and related on-site drone video and image data in accordance with the relevant standards of investment projects and engineering construction projects.

[0052] 2. Support automatic ordering of drone collection tasks through project drone codes, and collect project drone aerial data by scanning drone codes.

[0053] 3. Support multiple algorithms to obtain project image features, identify image features in multiple dimensions such as buildings, equipment, environment, materials, colors, shapes, outlines, spaces, and areas at project sites that have obtained construction permits, enrich feature comparison items, and facilitate accurate judgment of project site progress.

[0054] 4. Support the definition of the project progress change model. The model supports the definition of automatically identified image feature values ​​and feature value weights and the similarity ratio of multiple automatically compared images, forming the definition of the project progress change model.

[0055] 5. It supports automatic comparison of project progress picture data of different time periods of each project through machine deep learning training to form the results of project progress analysis, and obtain classification evaluation results such as whether the project has started, whether it is under construction in the near future, whether it has been completed, etc., to obtain the results of project progress analysis.

[0056] 6. Support the setting of project risk assessment models. According to the set risk assessment indicators, after completing the project data collection and project progress data analysis, conduct a comprehensive analysis of the data related to the project portrait to obtain the risk assessment results of each project, and formulate corresponding risk monitoring mechanisms for the dynamically monitored project risk levels, and dynamically track and supervise key risk projects.

[0057] 7. Supports automatic inspection of projects collected across the entire domain and issuance of dynamic risk assessment reports for projects. This device supports risk assessment of projects collected and stored across the entire domain. Departments can automatically obtain risky projects through risk assessment models, effectively replacing the workload of manual on-site inspections. It can also capture project risk situations in a timely manner, conduct early intervention in risky projects, and focus on tracking and monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a diagram showing the model architecture of the intelligent project dynamic risk monitoring and assessment method provided by an embodiment of the present invention;

[0059] Figure 2 It is a diagram of a project progress change model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The embodiment of the present invention provides an intelligent project dynamic risk monitoring and assessment method.

[0061] Dynamically collect data at all stages and dimensions of the project, including project data, project approval, project scheduling, project progress pictures, and project bidding data, to form a project portrait;

[0062] Based on the risk assessment indicators set by the system and project portraits, the system integrates the early warning analysis and judgment of the project and the overall analysis and judgment algorithm of the project to realize early warning, monitoring and overall situation analysis of the project progress. The early warning and monitoring of the project progress is to superimpose the image progress photos and video information of each project at different times, and judge the project progress through computer vision technology. At the same time, it integrates the progress status and approval procedures reported by the project, and automatically judges whether there is abnormal progress in the project through the project progress early warning analysis algorithm and issues an early warning;

[0063] Relevant managers will analyze the overall progress of the project based on the information of various dimensions of the early warning project, confirm whether the progress of the project is abnormal, and include abnormal projects in the progress tracking and supervision.

[0064] Based on the collected project-related data, the overall project risk is assessed through the constructed project dynamic risk monitoring and assessment model, thereby obtaining the project risk assessment result; through the set risk threshold, early warning of project risks is issued. Through this intelligent dynamic analysis, the management department can understand the risks of the project in real time, intervene in risky projects in a timely manner, reduce potential risks, and provide data support for regulatory departments to better manage projects.

[0065] It supports docking with the online approval and supervision platform for investment projects, automatically collects projects in the local area, and forms a sample library of projects.

[0066] Supports docking with the engineering construction project approval management system to obtain project approval data and form a project approval database.

[0067] Supports docking with major project management and scheduling systems, obtains project scheduling data by stage, and forms a project scheduling library.

[0068] The project library is supported to form a project library based on the collected data analysis, and drone collection tasks are customized for the started projects, and drone aerial photography image data is collected through the project aerial photography code.

[0069] It has multiple built-in image recognition algorithms, and uses the grayscale histogram equalization algorithm and edge feature algorithm to identify the features of the identification image of each project. After identifying the key features in the image, such as buildings, machines, equipment, etc., it records them in the feature library for each project. This feature library is used as the basic model for comparing the drone images collected each time. At the same time, the feature library supports the recognition of key features of images when they are manually recognized and recorded into the feature library.

[0070] For the projects in the started project database, according to the collected drone images, the model forms an automatic project progress change judgment model based on the drone image feature values, feature value weight definitions, image feature similarity ratios and related algorithms, automatically analyzes the project progress of each collection cycle, and obtains the results of project progress changes.

[0071] Through machine learning models, the project progress is automatically analyzed and learned, and the indicators for judging the project progress are continuously adjusted to improve the accuracy of changes in project progress.

[0072] After collecting the relevant data, it automatically conducts an overall risk assessment of all projects in the entire domain based on the set risk assessment model, and issues corresponding risk assessment results, providing data support for the management department to control project risks in a timely manner.

[0073] The implementation of this method includes:

[0074] Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision;

[0075] The key image feature library of the picture uses Floyd-Warshall and edge feature algorithms to identify the features of the identification image of each project, and then records the key features in the image such as color, texture, shape, contour, space, area, etc. for each project into the feature library. The distance between each node in the picture is calculated through the algorithm to generate the surveying and mapping feature library of the project. This feature library is used as the basic model for each video key frame image comparison; at the same time, this feature library supports the key feature recognition when manually recording the image in the feature library after recognition.

[0076] The drone aerial photography data collection model allows the system to automatically issue aerial photography tasks to projects that have entered the construction phase, and automatically collect aerial photography data based on the system's project drone aerial photography code.

[0077] The self-learning sample library continuously learns the image features in the aerial photography data to achieve effective analysis of the image data.

[0078] The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects.

[0079] The specific implementation process is as follows:

[0080] 1. Risk assessment indicator item setting: The system has a built-in indicator item library to initialize and modify the indicator items involved in the risk assessment model. The modified rules can be directly applied to monitoring.

[0081] 2. Initialize the project database, connect to the online approval and supervision platform for investment projects to obtain project data, and collect project basic information and project unit information data into the project database.

[0082] 3. Collect project scheduling information and connect to the major project scheduling management system to collect scheduling data for project planning, project approval, project commencement, project completion and other stages in stages.

[0083] 4. Project approval data collection, connecting to the engineering construction project approval management system, through the unique code of the project, the relevant preliminary planning data, intermediary service data, project unit information, project approval data, project approval materials, project approval results and the construction land planning permit information, construction project planning permit information, construction design document review, construction project construction permit information, completion acceptance filing information and other major approval matters involved in the entire life cycle of the project are collected and stored in the database.

[0084] 5. Project drone aerial photography data collection: After completing the above data collection work, a preliminary analysis will be conducted to identify the projects that have completed the construction permit application. The corresponding drone aerial photography codes will be generated for each project, and pushed to the drone aerial photography organizations in the region through the SMS platform. After receiving the data collection task, the drone aerial photography organizations will grab the order. The projects that have grabbed the order will organize drones to conduct aerial photography at fixed time periods. The aerial photography results will scan the corresponding aerial photography codes to upload the corresponding image data and video data to the drone aerial photography data collection model.

[0085] 6. Feature recognition of project-based location environment images: Use grayscale histogram equalization algorithm and edge feature algorithm to recognize the features of each project's identification image, identify key features in the image such as color, texture, shape, outline, space, area, building, equipment, material, etc., and then record them in the inspection recognition feature library for each project. This feature library is used as the comparison standard for each drone aerial photo of the project; at the same time, the feature library also supports the improvement of key feature recognition features of the image based on the original recognition through manual methods;

[0086] 7. Definition of the project progress change model. The model supports the definition of automatically identified image feature values ​​and feature value weights and the similarity ratio of multiple automatically compared images to form the definition of the project progress change model.

[0087] 8. Setting of warning values ​​for project progress changes. Setting of system warning values ​​for the final determined image similarity comparison, including the magnitude of project progress changes. Through the setting of the warning values, a large amount of drone aerial photography data of projects is monitored to obtain project progress analysis of each project. Management personnel can directly obtain the overall construction progress of the project through the changes in different periods of the corresponding projects. Real-time dynamic monitoring is possible without entering the site for viewing, thereby improving work efficiency.

[0088] 9. Machine deep learning training. The system has a built-in automatic learning training sample library and automatic learning device. Through continuous learning of on-site picture feature data of different projects and comparison of drone aerial picture data of project cycle scheduling, it can determine whether the project has started, whether it is under construction in the near future, and whether it has been completed, and obtain the results of the project progress analysis.

[0089] 10. The project risk assessment model, in accordance with the set risk assessment indicators, after completing the project data collection and project progress data analysis, conducts a comprehensive analysis of the data related to the project portrait to obtain the risk assessment results for each project.

[0090] 11. Confirmation of risk assessment results. The results of project risk assessment are manually reviewed according to the risk assessment level for projects with serious risks. The review process can refer to the project portrait data and the project drone aerial photography data for comprehensive analysis. Accurate projects will be entered into the project key risk analysis database, and managers can focus on tracking and monitoring key projects.

[0091] The model described in this method supports the collection of project data and project drone data for projects within the entire domain, automatically generates project portraits, and continuously improves and optimizes the project progress judgment model and project risk assessment model during the collection and calculation process. By monitoring the risks of projects within the entire domain, the workload and cost of manual inspections are effectively saved. At the same time, the relevant inspection data automatically forms a project portrait, providing data support for subsequent managers to query project files and overall control of project risks.

[0092] After the application of this project dynamic risk assessment method, it is necessary to connect with the local online investment project approval management platform, engineering construction project approval management system, and major project scheduling management system to collect project, project approval data, and project scheduling data. For projects that have entered the start-up stage, the project progress is automatically judged through the project progress change model through the pictures of the project site taken by drones, and the project portrait is formed based on the collected data. This method will evaluate the overall risk of the project according to the set risk assessment model and form a project risk assessment result. Management personnel can review and store projects based on the analyzed risk assessment results to form a key project monitoring library. According to the projects in the library and the results of risk monitoring, the problems in the project implementation process are tracked and handled in a targeted manner, and project risks are promptly resolved to ensure the timely completion of the project with quality and quantity, and improve the project conversion rate.

[0093] The present invention also claims protection for an intelligent project dynamic risk monitoring and evaluation system, which implements the intelligent project dynamic risk monitoring and evaluation method described in the above embodiment, and the system comprises:

[0094] Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision;

[0095] The key image feature library of the picture uses Floyd-Warshall and edge feature algorithms to identify the features of the identification image of each project, and records the key features in the image such as color, texture, shape, contour, space, area, etc. for each project into the feature library. The distance between each node in the image is calculated by the algorithm to generate the surveying and mapping feature library of the project. This feature library is used as the basic model for each video key frame image comparison; at the same time, this feature library supports the key feature recognition when manually recording the image into the feature library after recognition;

[0096] The drone aerial photography data collection model automatically issues aerial photography tasks to projects that have entered the construction phase, and automatically collects aerial photography data based on the system's project drone aerial photography code;

[0097] Self-learning sample library, through the continuous learning of image features in aerial photography data, to achieve effective analysis of image data;

[0098] The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects.

[0099] The specific implementation process is as follows:

[0100] 1. Risk assessment indicator item setting: The system has a built-in indicator item library to initialize and modify the indicator items involved in the risk assessment model. The modified rules can be directly applied to monitoring.

[0101] 2. Initialize the project database, connect to the online approval and supervision platform for investment projects to obtain project data, and collect project basic information and project unit information data into the project database.

[0102] 3. Collect project scheduling information and connect to the major project scheduling management system to collect scheduling data for project planning, project approval, project commencement, project completion and other stages in stages.

[0103] 4. Project approval data collection, connecting to the engineering construction project approval management system, through the unique code of the project, the relevant preliminary planning data, intermediary service data, project unit information, project approval data, project approval materials, project approval results and the construction land planning permit information, construction project planning permit information, construction design document review, construction project construction permit information, completion acceptance filing information and other major approval matters involved in the entire life cycle of the project are collected and stored in the database.

[0104] 5. Project drone aerial photography data collection: After completing the above data collection work, a preliminary analysis will be conducted to identify the projects that have completed the construction permit application. The corresponding drone aerial photography codes will be generated for each project, and pushed to the drone aerial photography organizations in the region through the SMS platform. After receiving the data collection task, the drone aerial photography organizations will grab the order. The projects that have grabbed the order will organize drones to conduct aerial photography at fixed time periods. The aerial photography results will scan the corresponding aerial photography codes to upload the corresponding image data and video data to the drone aerial photography data collection model.

[0105] 6. Feature recognition of project-based location environment images: Use grayscale histogram equalization algorithm and edge feature algorithm to recognize the features of each project's identification image, identify key features in the image such as color, texture, shape, outline, space, area, building, equipment, material, etc., and then record them in the inspection recognition feature library for each project. This feature library is used as the comparison standard for each drone aerial photo of the project; at the same time, the feature library also supports the improvement of key feature recognition features of the image based on the original recognition through manual methods;

[0106] 7. Definition of the project progress change model. The model supports the definition of automatically identified image feature values ​​and feature value weights and the similarity ratio of multiple automatically compared images to form the definition of the project progress change model.

[0107] 8. Setting of warning values ​​for project progress changes. Setting of system warning values ​​for the final determined image similarity comparison, including the magnitude of project progress changes. Through the setting of the warning values, a large amount of drone aerial photography data of projects is monitored to obtain project progress analysis of each project. Management personnel can directly obtain the overall construction progress of the project through the changes in different periods of the corresponding projects. Real-time dynamic monitoring is possible without entering the site for viewing, thereby improving work efficiency.

[0108] 9. Machine deep learning training. The system has a built-in automatic learning training sample library and automatic learning device. Through continuous learning of on-site picture feature data of different projects and comparison of drone aerial picture data of project cycle scheduling, it can determine whether the project has started, whether it is under construction in the near future, and whether it has been completed, and obtain the results of the project progress analysis.

[0109] 10. The project risk assessment model, in accordance with the set risk assessment indicators, after completing the project data collection and project progress data analysis, conducts a comprehensive analysis of the data related to the project portrait to obtain the risk assessment results for each project.

[0110] 11. Confirmation of risk assessment results. The results of project risk assessment are manually reviewed according to the risk assessment level for projects with serious risks. The review process can refer to the project portrait data and the project drone aerial photography data for comprehensive analysis. Accurate projects will be entered into the project key risk analysis database, and managers can focus on tracking and monitoring key projects.

[0111] This system configures risk monitoring indicator items through the risk monitoring indicator setting module, and the configured indicator items are subsequently used in the risk assessment of the project.

[0112] This system collects and stores samples of the overall project library by collecting project data from the online approval and supervision platform for investment projects. The data collection standards are implemented in accordance with the national standards for data collection for investment projects and engineering construction projects.

[0113] This system collects data from the engineering construction project approval management system to store data on the project database including relevant preliminary planning data, intermediary service data, project information, project approval data, project approval materials, project approval results, and major approval matters involved in the entire project life cycle, including construction land planning permit information, construction project planning permit information, construction design document review, construction project construction permit information, and completion acceptance filing information.

[0114] This system collects data on the actual construction progress and project milestone achievement of each stage of the major project scheduling management system according to the time stages of project planning, project preliminary, project commencement, project completion, and project commissioning.

[0115] After completing the above data collection work, this system will conduct a preliminary analysis to determine the projects that have completed the construction permit application, and generate corresponding drone aerial photography codes for each project. The codes are pushed to the drone aerial photography organizations in the region through the SMS platform. After receiving the data collection tasks, the drone aerial photography organizations will grab the orders. The projects in the middle will organize drones to conduct aerial photography at fixed time periods. The aerial photography results are uploaded to this model by scanning the corresponding aerial photography codes.

[0116] This system relies on the project basic database and project analysis database data constructed by the project theme database to build a comprehensive three-dimensional portrait for each project, including a project image progress portrait and a full-dimensional portrait of the project.

[0117] This system is based on project profiling and integrates project early warning analysis and judgment with project overall analysis and judgment algorithms to realize project progress judgment, abnormal progress warning, monitoring and overall project situation analysis.

[0118] This system realizes the superposition of image progress photos and video information of each project at different times. It relies on computer vision technology and image feature recognition algorithms to identify key feature objects (such as buildings, equipment, and materials) in each image. According to the changes of key feature objects at different times, it integrates the project progress calculation algorithm to comprehensively judge the actual progress of the project, such as whether the project is under construction, whether it is under construction in the near future, whether it is completed, etc.

[0119] Relying on the results of project progress judgment, this system conducts an integrated analysis of the project progress status and project approval procedures when the project progress is reported. Based on the project early warning analysis and judgment algorithm, it analyzes the deviation of each indicator in the project risk judgment index. According to various risk monitoring rules, it makes a grade judgment on possible abnormal progress. When the abnormal progress warning level is reached, the system automatically triggers the warning to realize the active judgment of abnormal project progress warning.

[0120] Through this system, relevant management personnel can view various types of project information based on early warning projects, combined with project portraits and project progress overview related modules, and manually check the actual situation of the project to conduct a comprehensive assessment of the overall progress of the project, confirm whether the progress of the project is abnormal, and include abnormal projects in progress tracking and supervision.

[0121] Ultimately, this system will automatically assess the risky projects in the system according to the set risk assessment model, and provide the assessment results to the relevant departments. The relevant departments can use the assessment results to track and supervise the risky projects in a timely manner, provide corresponding guidance services for the projects in a timely manner, and provide corresponding auxiliary services for the improvement of the business environment and the smooth implementation of the projects.

[0122] The embodiment of the present invention also provides an intelligent project dynamic risk monitoring and evaluation device, comprising at least one memory and a processor;

[0123] The at least one memory is used to store a machine-readable program;

[0124] The at least one processor is used to call the machine-readable program to implement the intelligent project dynamic risk monitoring and assessment method described in the above embodiment.

[0125] The embodiment of the present invention further provides a computer-readable medium, on which computer instructions are stored, and when the computer instructions are executed by a processor, the processor executes the intelligent project dynamic risk monitoring and assessment method described in the above embodiment. Specifically, a system or device equipped with a storage medium can be provided, on which software program codes implementing the functions of any of the above embodiments are stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0126] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0127] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0128] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0129] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0130] The present invention is shown and described in detail above through the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the protection scope of the present invention.

Claims

1. An intelligent project dynamic risk monitoring and assessment method, characterized in that: Dynamically collect data at all stages and dimensions of the project, including project data, project approval, project scheduling, project progress pictures, and project bidding data, to form a project portrait; Based on the risk assessment indicators set by the system and project portraits, the system integrates the early warning analysis and judgment of the project and the overall analysis and judgment algorithm of the project to realize early warning, monitoring and overall situation analysis of the project progress. The early warning and monitoring of the project progress is to superimpose the image progress photos and video information of each project at different times, and judge the project progress through computer vision technology. At the same time, it integrates the progress status and approval procedures reported by the project, and automatically judges whether there is abnormal progress in the project through the project progress early warning analysis algorithm and issues an early warning; Relevant managers will analyze the overall progress of the project based on the information of various dimensions of the early warning project, confirm whether the progress of the project is abnormal, and include abnormal projects in the progress tracking and supervision.

2. According to claim 1, an intelligent project dynamic risk monitoring and assessment method is characterized in that: The implementation of this method includes: Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision; Image key image feature library, using Floyd-Warshall and edge feature algorithms to identify the features of each project's identification image, after identifying the key features in the image, record them in the feature library for each project, calculate the distance between each node in the image through the algorithm, generate the project's mapping feature library, and use this feature library as the basic model for each video key frame image comparison; at the same time, this feature library supports the recognition of key features when manually recording images into the feature library after recognition; The drone aerial photography data collection model automatically issues aerial photography tasks to projects that have entered the construction phase, and automatically collects aerial photography data based on the system's project drone aerial photography code; Self-learning sample library, through the continuous learning of image features in aerial photography data, to achieve effective analysis of image data; The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects.

3. According to an intelligent project dynamic risk monitoring and assessment method according to claim 1 or 2, it is characterized in that: It supports docking with the online approval and supervision platform for investment projects, automatically collects projects in the region, and forms a sample library of projects; it supports docking with the approval management system for engineering construction projects, obtains project approval data, and forms a project approval library; it supports docking with the major project management and scheduling system, obtains project scheduling data in stages, and forms a project scheduling library; The project library forms a library of started projects based on the analysis of collected data, and customizes drone collection tasks for the started projects, and collects drone aerial photography image data through the project aerial photography code.

4. According to claim 3, an intelligent project dynamic risk monitoring and assessment method is characterized in that: The grayscale histogram equalization algorithm and edge feature algorithm are used to identify the features of the identification image of each project. After identifying the key features in the image, they are recorded in the feature library for each project. This feature library is used as the basic model for comparing the drone images collected each time. At the same time, the feature library supports the recognition of key features of images when they are manually recognized and recorded in the feature library. For the projects in the started project database, according to the collected drone images, the model forms an automatic project progress change judgment model based on the drone image feature value, feature value weight definition, image feature similarity ratio and related algorithms, automatically analyzes the project progress of each collection cycle, and obtains the results of project progress changes; Through machine learning models, the project progress is automatically analyzed and learned, and the indicators for judging the project progress are continuously adjusted to improve the accuracy of changes in project progress.

5. According to an intelligent project dynamic risk monitoring and assessment method according to claim 1 or 2, it is characterized in that: After collecting the relevant data, an overall risk assessment is automatically conducted on all projects in the entire domain, and corresponding risk assessment results are issued.

6. An intelligent project dynamic risk monitoring and assessment method according to claim 1 or 2, characterized in that: The specific implementation process of this method is as follows: 1) Risk assessment indicator item setting: the system has a built-in indicator item library to initialize and modify the indicator items involved in the risk assessment model. The modified rules are directly applied to monitoring; 2) Initialize the project database, connect to the online approval and supervision platform for investment projects to obtain project data, and collect basic project information and project unit information data into the project database; 3) Project scheduling information collection: connect to the major project scheduling management system to collect scheduling data for project planning, project approval, project commencement, and project completion in stages; 4) Project approval data collection: connect to the construction project approval management system and collect and store the data of major approval items related to all projects in the project library through the unique code of the project; 5) Project drone aerial photography data collection: After completing the above data collection work, a preliminary analysis is conducted to identify the projects that have obtained the construction permit. The results of the determination are divided into projects to generate corresponding drone aerial photography codes, which are pushed to the drone aerial photography organizations in the region through the SMS platform. After receiving the data collection task, the drone aerial photography organizations grab the order. The projects that grab the order organize drones to conduct aerial photography at a fixed time period. The aerial photography results are uploaded to the drone aerial photography data collection model by scanning the corresponding aerial photography code. 6) Feature recognition of project basic location environment images: Use grayscale histogram equalization algorithm and edge feature algorithm to recognize the features of each project's identification image, and record the key features in the image into the inspection recognition feature library for each project. This feature library is used as the comparison standard for each project's drone aerial photography images. At the same time, the feature library also supports the improvement of the key feature recognition features of the image based on the original recognition through manual methods; 7) Definition of project progress change model, which supports the definition of automatically identified image feature values ​​and feature value weights and automatically compared similarity ratios of multiple images to form a project progress change model definition; 8) Setting of warning values ​​for project progress changes: setting system warning values ​​for the final determined image similarity comparison, including the magnitude of project progress changes; through the setting of the warning values, monitoring of a large number of project drone aerial photography data, and obtaining project progress analysis of each project; 9) Machine deep learning training. The system has a built-in automatic learning training sample library and automatic learning device. Through continuous learning of on-site image feature data of different projects and comparison of drone aerial image data of project cycle scheduling, it can determine whether the project has started, whether it is under construction in the near future, and whether it has been completed, and obtain the results of project progress analysis; 10) Project risk assessment model: after completing the data collection and project progress data analysis of the project according to the set risk assessment indicators, conduct a comprehensive analysis of the data related to the project profile to obtain the risk assessment results of each project; 11) Confirmation of risk assessment results. The results of project risk assessment are manually reviewed according to the risk assessment level for projects with serious risks. The review process refers to the data of project portraits and drone aerial photography data for comprehensive analysis. Accurate projects will be included in the project key risk analysis library.

7. According to claim 1, an intelligent project dynamic risk monitoring and assessment method is characterized in that: The project approval data collection includes the main approval matters related to the project, including preliminary planning data, intermediary service data, project unit information, project approval data, project approval materials, project approval results and the construction land planning permit information, construction project planning permit information, construction design document review, construction project construction permit information, and completion acceptance filing information involved in the entire life cycle of the project.

8. An intelligent project dynamic risk monitoring and assessment system, characterized in that: include: Risk monitoring indicator data, each indicator item corresponds to an assessment score. The sample data of the project is analyzed through the risk monitoring indicator data, and each project is scored for risk. The scoring results and risk coefficients are fed back to the regulatory department, which reviews the data. Projects that have passed the review are entered into the risk database for key tracking and supervision; Image key image feature library, using Floyd-Warshall and edge feature algorithms to identify the features of each project's identification image, after identifying the key features in the image, record them in the feature library for each project, calculate the distance between each node in the image through the algorithm, generate the project's mapping feature library, and use this feature library as the basic model for each video key frame image comparison; at the same time, this feature library supports the recognition of key features when manually recording images into the feature library after recognition; The drone aerial photography data collection model automatically issues aerial photography tasks to projects that have entered the construction phase, and automatically collects aerial photography data based on the system's project drone aerial photography code; Self-learning sample library, through the continuous learning of image features in aerial photography data, to achieve effective analysis of image data; The risk assessment model assigns risk scores to projects according to risk assessment indicators. Projects with risks are automatically identified, providing data support for risk tracking and supervision of subsequent projects. The system implements the intelligent project dynamic risk monitoring and assessment method as described in any one of claims 1 to 7.

9. An intelligent project dynamic risk monitoring and assessment device, characterized in that: comprising at least one memory and a processor; The at least one memory is used to store a machine-readable program; The at least one processor is used to call the machine-readable program to implement the method described in any one of claims 1 to 7.

10. A computer readable medium, characterized in that The computer readable medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images