AI-based large capital construction project management system

Abstract

The invention discloses a large capital construction project management system based on AI. The system comprises a management system background and a client communicating with the management system background. The client is used for providing a man-machine interaction access interface; wherein the background server comprises a project database, and the project database is used for storing project data; the project data comprises basic information of an engineering project, a project general situation, progress control, a responsibility subject, a project management organization, a platform user, a main contract, a downstream contract, a financial index, project creation, a project license, a total package settlement, acceptance data and a bank account. And the project management module is connected with the project database, the project management module is used for querying, creating, editing, deleting and exporting the project data, and the project data has a project number for uniqueness verification. And a systematic and intelligent project management tool is adopted, so that more systematic and comprehensive information management can be performed on capital construction projects.

Description

Technical Field

[0001] The present invention relates to the field of information management technology, and particularly to an AI-based large infrastructure project management system. Background Art

[0002] The large infrastructure project management system is an information management tool for large infrastructure construction projects, which aims to improve project management efficiency and achieve full life cycle management of projects. At present, the system design of the existing information management system still has certain defects. In actual use, it can usually only perform conventional project information management on infrastructure projects, such as contract management, etc., but cannot perform more systematic and comprehensive information management on infrastructure projects. Summary of the invention

[0003] In order to overcome the above technical defects, the present invention provides an AI-based large infrastructure project management system.

[0004] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0005] The AI-based large infrastructure project management system described in the present invention includes a management system backend and a client communicating with the management system backend; the client is used to provide an access interface for human-computer interaction;

[0006] Wherein, the backend server includes:

[0007] Project database, which is used to store project data, including basic information of engineering projects, engineering overview, progress control, responsible entities, project management organizations, platform users, main contracts, downstream contracts, financial indicators, project excellence, project licenses, general contract settlement, acceptance materials and bank accounts;

[0008] A project management module is connected to the project database, and is used to query, create, edit, delete and export the project data. The project data has a project number for uniqueness verification.

[0009] Preferably, the backend server includes:

[0010] A contract database, wherein the contract database is used to store the main contract, downstream contracts and material procurement contracts of the engineering project, wherein the material procurement contracts include contract terms information;

[0011] A contract management module connected to the contract database, the contract management module is used to query, create, edit, delete and export the material procurement contract, and the contract management module is also used to display the material procurement contract by category;

[0012] An income management module connected to the contract database, the income management module comprising an income operation unit and a notification form generation unit;

[0013] The operation unit is used to query, create, edit, delete and export the payment data associated with the material procurement contract;

[0014] The notification generating unit is used to generate a receipt notification according to the contract terms information and send it to the client; the receipt notification has a unique notification number. When the material procurement contract has a notification number, the notification generating unit cannot create a new receipt notification; the notification generating unit is also used to generate an account receipt notification according to the payment receipt information sent by the client and store it in the contract database.

[0015] Preferably, the backend server includes:

[0016] An expenditure management module connected to the contract database, the expenditure management module comprising an expenditure operation unit and a file generation unit;

[0017] The expenditure operation unit is used to query, create, edit, delete and export the approval document quotation sheet, which is used in the approval process of paying the project engineering fee;

[0018] The file generation unit is used to generate an approval document for paying project engineering funds based on a registered approval document quotation sheet. The approval document has an approval document number, which is composed of a payment category number, a four-digit year and month, a project management department number, a project number, and a two-digit sequence number of the year and month.

[0019] Preferably, the backend server includes:

[0020] A fund ledger module, which includes a collection ledger unit, a payment ledger unit, a freezing ledger unit, and a legal risk ledger unit;

[0021] The payment account unit is used to generate a project payment account according to the payment information sent by the client, and the project payment account includes project payment collection, output value, settlement, receivable and uncollected project payment and loan;

[0022] The payment ledger unit is used to generate a payment ledger according to the expenditure information sent by the client, and the payment ledger includes project name, payee name, related loan, related payment, related fund transfer, remarks, payment number, project management, whether it is borrowed, bank account type, approval number, expenditure classification, whether it is bill payment, payment date, payment amount, payment amount, principal, capital occupation fee and exemption amount;

[0023] The freezing account unit is used to generate a freezing account according to the fund freezing information sent by the client, and the freezing account includes a bank account number, a court name and a freezing amount;

[0024] The legal risk ledger unit is used to generate a legal risk ledger unit based on the litigation information sent by the client, and the legal risk ledger includes the project number, project name, engineering management department, dispute counterparty, case number, dispute handling results and case handling results.

[0025] Preferably, the backend server includes:

[0026] A cooperation unit database, which is used to store cooperation unit data, including basic information, business information, storage information, license information, award information, annual evaluation, account information, invoicing information and cooperation records of cooperation units;

[0027] A resource management module connected to the cooperation unit database, the resource management module includes a resource operation unit, a supplier recommendation unit and a supplier blacklist unit;

[0028] The resource operation unit is used to query, create, edit, delete and export the data of the cooperation unit;

[0029] The recommended supplier unit is used to classify the cooperation units according to the recommendation level to form a recommended supplier directory, and the recommendation level includes recommendation and key recommendation;

[0030] The supplier blacklist unit is used to list the cooperation unit in a blacklist directory, and the blacklist directory includes the listing time and listing reason of the cooperation unit.

[0031] Preferably, the backend server includes:

[0032] The invoicing module is used to generate an invoicing request based on the invoicing information sent by the client and send it to the client with the review authority. After the invoicing request passes the review, the invoicing module generates an invoicing application form based on the invoicing information;

[0033] The invoice account module is used to organize the storage address of the invoicing request, invoicing information and invoicing application form of the invoicing module into an invoice account.

[0034] Preferably, the backend server includes:

[0035] The project large screen module is used to visually display project information, engineering progress information, excellence goals, downstream contract information and on-site real-time information;

[0036] The branch BI large screen module is used to visually display business unit information, project list, branch information, project status information, cost rate information and quality improvement information;

[0037] The assessment BI large screen is used to visually display the engineering management department list, assessment rankings, overall rankings of business units, scoring details and individual rankings.

[0038] Preferably, the client is one or more of a WeChat platform applet, an enterprise WeChat information platform, a PC application, and a web client.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention provides a large infrastructure project management system based on AI, including a management system background and a client communicating with the management system background; the client is used to provide an access interface for human-computer interaction; wherein the backend server includes: a project database, the project database is used to store project data, the project data includes basic information of the engineering project, engineering overview, progress control, responsible subject, project management organization, platform users, main contract, downstream contract, financial indicators, project excellence, project license, general contract settlement, acceptance materials and bank account number. A project management module is connected to the project database, the project management module is used to query, create, edit, delete and export the project data, and the project data has a project number for uniqueness verification.

[0041] The large infrastructure project management system of the present invention aims to improve the efficiency of project management. Through the automated project management process, it reduces the time for manual input and data processing, and improves the efficiency of project management. The systematic and intelligent project management tool provides more systematic and comprehensive information management for infrastructure projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0043] Figure 1 It is a schematic diagram of the system composition of an AI-based large infrastructure project management system of the present invention. DETAILED DESCRIPTION

[0044] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0045] Embodiment 1

[0046] like Figure 1As shown, the system composition of a large infrastructure project management system based on AI described in the present invention.

[0047] In the present invention, the AI-based large infrastructure project management system includes a management system backend and a client communicating with the management system backend; the client is used to provide an access interface for human-computer interaction; wherein the backend server includes:

[0048] Project database, which is used to store project data, including basic information of engineering projects, engineering overview, progress control, responsible entities, project management organizations, platform users, main contracts, downstream contracts, financial indicators, project excellence, project licenses, general contract settlement, acceptance materials and bank accounts;

[0049] A project management module is connected to the project database, and is used to query, create, edit, delete and export the project data. The project data has a project number for uniqueness verification.

[0050] In a specific implementation, the backend server includes:

[0051] A contract database, wherein the contract database is used to store the main contract, downstream contracts and material procurement contracts of the engineering project, wherein the material procurement contracts include contract terms information;

[0052] A contract management module connected to the contract database, the contract management module is used to query, create, edit, delete and export the material procurement contract, and the contract management module is also used to display the material procurement contract by category;

[0053] An income management module connected to the contract database, the income management module comprising an income operation unit and a notification form generation unit;

[0054] The operation unit is used to query, create, edit, delete and export the payment data associated with the material procurement contract;

[0055] The notification generating unit is used to generate a receipt notification according to the contract terms information and send it to the client; the receipt notification has a unique notification number. When the material procurement contract has a notification number, the notification generating unit cannot create a new receipt notification; the notification generating unit is also used to generate an account receipt notification according to the payment receipt information sent by the client and store it in the contract database.

[0056] Through the contract and revenue management functions, not only the automation level of project management is improved, but also the control over key financial processes is strengthened, thereby improving the quality and efficiency of the entire project management work.

[0057] In a specific implementation, the backend server includes:

[0058] An expenditure management module connected to the contract database, the expenditure management module comprising an expenditure operation unit and a file generation unit;

[0059] The expenditure operation unit is used to query, create, edit, delete and export the approval document quotation sheet, which is used in the approval process of paying the project engineering fee;

[0060] The file generation unit is used to generate an approval document for paying project engineering funds based on a registered approval document quotation sheet. The approval document has an approval document number, which is composed of a payment category number, a four-digit year and month, a project management department number, a project number, and a two-digit sequence number of the year and month.

[0061] In the present invention, the expenditure management module standardizes the approval process of project engineering funds through the expenditure operation unit, improves the transparency and standardization of financial management, and improves the quality of the entire project management by effectively managing project expenditures.

[0062] In a specific implementation, the backend server includes:

[0063] A fund ledger module, which includes a collection ledger unit, a payment ledger unit, a freezing ledger unit, and a legal risk ledger unit;

[0064] The payment account unit is used to generate a project payment account according to the payment information sent by the client, and the project payment account includes project payment collection, output value, settlement, receivable and uncollected project payment and loan;

[0065] The payment ledger unit is used to generate a payment ledger according to the expenditure information sent by the client, and the payment ledger includes project name, payee name, related loan, related payment, related fund transfer, remarks, payment number, project management, whether it is borrowed, bank account type, approval number, expenditure classification, whether it is bill payment, payment date, payment amount, payment amount, principal, capital occupation fee and exemption amount;

[0066] The freezing account unit is used to generate a freezing account according to the fund freezing information sent by the client, and the freezing account includes a bank account number, a court name and a freezing amount;

[0067] The legal risk ledger unit is used to generate a legal risk ledger unit based on the litigation information sent by the client, and the legal risk ledger includes the project number, project name, engineering management department, dispute counterparty, case number, dispute handling results and case handling results.

[0068] In the present invention, the transparency of project financial activities is enhanced through detailed ledger records. Automated ledger generation reduces the need for manual records and improves the efficiency of fund management. The frozen ledger and legal risk ledger units help project managers monitor potential financial and legal risks in real time. Providing accurate financial data support helps management make more informed financial and operational decisions.

[0069] In a specific implementation, the backend server includes:

[0070] A cooperation unit database, which is used to store cooperation unit data, including basic information, business information, storage information, license information, award information, annual evaluation, account information, invoicing information and cooperation records of cooperation units;

[0071] A resource management module connected to the cooperation unit database, the resource management module includes a resource operation unit, a supplier recommendation unit and a supplier blacklist unit;

[0072] The resource operation unit is used to query, create, edit, delete and export the data of the cooperation unit;

[0073] The recommended supplier unit is used to classify the cooperation units according to the recommendation level to form a recommended supplier directory, and the recommendation level includes recommendation and key recommendation;

[0074] The supplier blacklist unit is used to list the cooperation unit in a blacklist directory, and the blacklist directory includes the listing time and listing reason of the cooperation unit.

[0075] In the present invention, the following effects can be achieved: 1. Enhanced risk control: The blacklist mechanism helps to identify and manage potential cooperation risks. 2. Optimized supplier selection: The recommended supplier catalog helps the project team quickly identify and select high-quality suppliers.

[0076] In a specific implementation, the backend server includes:

[0077] The invoicing module is used to generate an invoicing request based on the invoicing information sent by the client and send it to the client with the review authority. After the invoicing request passes the review, the invoicing module generates an invoicing application form based on the invoicing information;

[0078] The invoice account module is used to organize the storage address of the invoicing request, invoicing information and invoicing application form of the invoicing module into an invoice account.

[0079] In a specific implementation, the backend server includes:

[0080] The project large screen module is used to visually display project information, engineering progress information, excellence goals, downstream contract information and on-site real-time information;

[0081] The branch BI large screen module is used to visually display business unit information, project list, branch information, project status information, cost rate information and quality improvement information;

[0082] The assessment BI large screen is used to visually display the engineering management department list, assessment rankings, overall rankings of business units, scoring details and individual rankings.

[0083] Preferably, the client is one or more of a WeChat platform applet, an enterprise WeChat information platform, a PC application, and a web client.

[0084] Embodiment 2

[0085] The second embodiment of the present invention provides an AI-based large infrastructure project management system, which includes an AI safety construction subsystem for infrastructure construction sites; the application scenario of the AI ​​safety construction subsystem is mainly in infrastructure construction sites. Usually, there are a large number of construction equipment, electrical equipment and construction personnel at infrastructure construction sites. During the construction process, it is necessary to supervise the construction behavior and issue timely alarms when any illegal behavior is found.

[0086] In a specific implementation, the AI ​​safety construction subsystem includes a sensor and an edge processing terminal, wherein the sensor is connected to the edge processing terminal, and the edge processing terminal is connected to the management system backend. The sensor is used to collect video data from the construction site, and the sensor transmits the data to the edge computing terminal for real-time processing. Finally, the edge computing terminal transmits the processing results and the corresponding video data back to the management system backend. Visual display is performed on the client through the management system backend.

[0087] In the specific implementation, the behavior of workers at the infrastructure site needs to be automatically identified and supervised by effective monitoring means. The AI ​​safety construction subsystem of the present invention uses optical sensors to collect and perceive workers. Specifically, the optical sensor is a camera mounted on a drone. The drone can be a multi-rotor drone.

[0088] In a preferred implementation, the optical sensor uses a camera mounted on a drone, a plurality of fixed cameras deployed at an infrastructure construction site, and a mobile control ball.

[0089] Specifically, the illegal behaviors of workers are identified through video analysis technology, and intelligent analysis of videos (such as whether safety helmets are worn during construction) is realized based on deep learning methods. The identification and early warning of illegal construction behaviors are completed by combining data transmission and storage technology and behavior recognition models.

[0090] In a specific implementation, the video analysis method includes target detection, target classification, target tracking, behavior recognition, etc. The present invention adopts a video analysis method based on deep learning, starting from the characteristics of the data, comprehensively considering static image data and dynamic video data, and introducing a temporal and spatial convolutional network architecture, thereby realizing video analysis of dynamic and static data of illegal behaviors.

[0091] In the AI ​​safety construction subsystem, the coordinated work of sensors, edge computing terminals, and management system background is crucial to achieving the real-time and reliability of the system. Through a specific transmission processing link, the three are explained in terms of real-time and reliability:

[0092] 1. Real-time:

[0093] 1. Real-time data collection by sensors: Sensors are responsible for real-time monitoring of the environment and activities at the construction site, and can instantly capture any abnormalities or critical events.

[0094] 2. Instant processing of edge computing: After receiving sensor data, the edge computing terminal can process it immediately, such as real-time analysis of video streams to quickly identify security issues or violations.

[0095] 3. Fast data transmission: The edge computing terminal quickly transmits the processing results back to the background server, reducing the transmission time of video data in the network.

[0096] 4. Instant response from the background: After the background system receives the data processing results from the edge computing terminal, it can instantly update the status, trigger an alarm or notify relevant personnel.

[0097] 5. Real-time display on the client: The client displays the data and analysis results of the backend server in real time, allowing users to understand the current status of the construction site at any time.

[0098] 2. Reliability:

[0099] 1. The system design takes network fluctuations and delays into consideration, and uses edge computing terminals to process data in real time, reducing the time it takes to transmit a large amount of video data back to the backend server, network instability, delays, etc.

[0100] 2. Improve system reliability through system redundancy design, such as multiple sensors monitoring the same area or multiple edge computing terminals processing the same data.

[0101] Through the above measures, the AI ​​safe construction subsystem can ensure real-time performance while also ensuring the reliability and stability of the system, providing users with a safety management tool that is both fast-responding and highly reliable.

[0102] Embodiment 3

[0103] This embodiment three also provides a preferred implementation of an AI-based large infrastructure project management system. Specifically, the large infrastructure project management system includes an intelligent assisted invoicing and signature subsystem, a cost strategy intelligent recommendation subsystem, and a project investment strategy intelligent optimization subsystem.

[0104] The intelligent assisted invoicing and signature subsystem includes:

[0105] Data processing module: Use regular expressions or natural language processing technology to parse the invoicing information entered by the user and extract key information such as invoice number, amount, date, product details, etc. Check whether the input format meets the invoice requirements of the country or region and provide real-time data verification feedback.

[0106] Intelligent invoicing engine: Automatically fill in invoice content based on invoice template and input information. Introduce rule engine to support tax regulations in different regions, such as VAT rate, tax preferential policies, etc., which are automatically applied to invoice generation.

[0107] Electronic signature module: supports the application of digital certificates to generate legally binding electronic signatures. The seal generation function can generate a unique signature based on the seal image and signature style preset by the user. It supports multiple signature algorithms, such as RSA, DSA, etc., to ensure the security and tamper-proof capability of the signature.

[0108] Security verification module: Integrate SSL / TLS encryption to ensure the security of invoicing data transmission. Verify the integrity of the invoice file through hash algorithms (such as SHA-256) to ensure that the invoice has not been tampered with. Support compliance testing to ensure that the generated invoice complies with the requirements of laws and regulations, and provide compliance reports.

[0109] In the present invention, the intelligent assisted invoicing and signature subsystem is used to implement the following algorithm steps:

[0110] S100: Obtain the invoicing information input by the user through the interface, including the invoice number, amount, invoicing date, product details, etc. Perform basic format verification on the invoicing information to ensure that the input data is legal.

[0111] In the present invention, the user enters the billing information through the system interface, including but not limited to the following:

[0112] Invoice number: a number that uniquely identifies the invoice;

[0113] Amount: the transaction amount involved in the invoice;

[0114] Invoicing date: the date on which the invoice is generated;

[0115] Product details: product name, quantity, unit price and other information.

[0116] Format verification: The system first performs format verification on the invoicing information entered by the user to ensure that the data is legal. For example, the invoice number must comply with the format specified by the national tax regulations, the amount should be a positive number, and the invoicing date cannot be earlier than the transaction date. If the data is not in compliance (such as missing key fields or incorrect format), the system will prompt the user to make corrections.

[0117] S200: The data processing module calls the parsing engine and uses regular expressions or NLP technology to extract key information from the input information. At the same time, it detects whether the selected invoice template is suitable for the invoicing information to avoid format errors.

[0118] Specifically, the system calls the data processing module to parse the information entered by the user. Regular expressions are used to detect and extract key information in the invoice, such as amount, product details, tax rate, etc. In cases where free text needs to be processed (such as product details or notes), the system uses natural language processing (NLP) technology to identify and classify information. The system detects whether the selected invoice template is applicable to the current invoicing information. For example, if the template selected by the user is not suitable for the amount range or product type entered, the system will prompt the user to change to a suitable template to avoid format errors.

[0119] S300: The intelligent invoicing engine generates draft invoices based on the extracted key information, automatically applies corresponding tax rates, discounts and other parameters, and generates draft invoices that meet national tax requirements.

[0120] In the specific implementation, invoice draft generation: the system uses the intelligent invoicing engine to automatically generate invoice drafts based on the key information extracted by S200, including automatic arrangement of product details, amount summary, etc.

[0121] In specific implementation, tax rates and discounts are applied: the system automatically applies the corresponding tax rates (such as VAT) and possible discounts to ensure that the invoice meets the requirements of the State Taxation Bureau. The tax rate is automatically matched according to the industry or product type provided by the user.

[0122] In specific implementation, compliance check: After the invoice draft is generated, the system will conduct a basic compliance check to ensure that the invoice content complies with local or national tax regulations. This includes checking the invoice number, tax rate, amount and other fields.

[0123] S400: Obtain a signature request, which is submitted after the user confirms that the content of the draft invoice is correct. The system calls the signature module.

[0124] It is understandable that after the invoice draft is generated, the user will receive a draft preview from the system and can submit a signature request only after confirming that it is correct. After confirming the invoice draft, the user issues a signature request. The system will automatically call the signature module to prepare for the electronic signature and seal operation.

[0125] S500: The electronic signature module uses the user's digital certificate to generate an electronic signature and seal. The signature is applied to the designated location of the invoice to ensure the tamper-proof nature of the content.

[0126] In the present invention, digital certificate verification: the system performs identity verification through the user's digital certificate (issued by an authoritative third-party organization) to ensure the legality and compliance of the signature operation.

[0127] Generate electronic signature and seal: Based on the user's digital certificate, the system generates an electronic signature and generates an electronic seal in the specified location. These two steps ensure the uniqueness and non-tamperability of the invoice.

[0128] Signature and seal application: The system applies the generated signature and seal to the designated location of the invoice to ensure that it meets the signature and seal requirements and is visually neat and correct.

[0129] S600: The security verification module performs integrity and security checks on invoices, verifies invoice files using a hash algorithm, and generates a security report.

[0130] In the specific implementation, in order to ensure that the content of the invoice is not tampered with after signing, the system uses a hash algorithm to check the integrity of the invoice. By calculating the hash value of the invoice and comparing it with the original hash value, it is verified whether the invoice has been modified after signing. The system generates a security verification report that records the signing process, the results of the hash verification, and the integrity and security status of the invoice.

[0131] S700: The invoice is generated and the user can choose to download the electronic invoice or send it to a designated email address or third-party platform through the system.

[0132] In one example, the user can choose to download the signed electronic invoice, and the system will provide the invoice to the user in PDF format to ensure the legality of its format and signature. In addition to downloading, the system can automatically send the invoice to a designated mailbox according to the user's instructions, or send it to a third-party platform (such as a tax system, customer management system, etc.) through an integrated API. After the invoice is generated, the system will automatically archive it for the user to query and track it later.

[0133] This intelligent assisted invoicing and signature subsystem greatly improves invoicing efficiency and ensures the accuracy and compliance of invoices through automated invoicing, signature and security verification processes, while ensuring the tamper-proofness and legal effectiveness of invoices through electronic signatures.

[0134] Specifically, the cost strategy intelligent recommendation subsystem includes:

[0135] Data collection module: collects project cost-related data, such as material costs, labor costs, etc., from internal data sources such as project management systems and financial systems. Integrates external data sources (such as market price index, supplier prices, etc.) to provide comprehensive data support. Supports data cleaning, deduplication and formatting to ensure data accuracy and consistency.

[0136] Cost analysis engine: Use statistical methods (such as linear regression, time series analysis) and machine learning algorithms (such as random forest, XGBoost) to analyze historical cost data. Analyze cost trends, periodicity and anomalies, and predict cost changes for future projects. Introduce optimization algorithms (such as heuristic algorithms, genetic algorithms) to help analyze the optimal cost control solution.

[0137] Strategy generation module: Based on the results of the analysis engine, it automatically generates a variety of cost control strategies (such as reducing procurement costs, reducing project hours, etc.). It provides visualization tools to show the cost impact of different strategies to help users make choices.

[0138] User feedback module: collects user feedback on the effectiveness of strategy execution and forms historical evaluation records. Use feedback data to optimize algorithms and improve the accuracy of future strategy recommendations.

[0139] Report generation module: Automatically generate cost analysis reports, including trend forecasts, strategy recommendations, and risk assessments. Supports customized reports, and users can choose the indicators and display methods they are interested in.

[0140] In the present invention, the cost strategy intelligent recommendation subsystem is used to implement the following algorithm steps:

[0141] S100: The data collection module collects project cost data from multiple internal and external data sources. The data is cleaned, deduplicated and normalized to ensure consistency.

[0142] In the present invention, internal data sources: project cost-related data such as material costs, labor costs, equipment usage costs, operating costs, etc. are obtained from the enterprise's project management system, financial management system, human resource management system and other internal systems. External data sources: including market prices, industry standards, supplier quotations, economic environment data, etc., which are of great reference significance for cost assessment.

[0143] Specifically, data processing includes:

[0144] (1) Data cleaning: Remove erroneous, incomplete or redundant data to ensure data accuracy. For example, exclude duplicate purchase records or invalid price information.

[0145] (2) Deduplication: Ensure that each data item has only one source to avoid duplicate records caused by multi-source data.

[0146] (3) Normalization: Normalize data from different data sources and dimensions to make them suitable for model analysis. For example, standardize the monetary unit into one currency and the time unit into days or months.

[0147] S200: The cost analysis engine analyzes cost data based on statistical models and machine learning models to identify historical trends, cost fluctuation cycles and abnormal data points. At the same time, it predicts future cost trends.

[0148] In the specific implementation, use statistical methods such as linear regression and moving average to analyze historical cost data and identify the historical trend of project costs. Analyze the cost fluctuation cycle of the project to find out which factors affect the high and low fluctuations of costs, such as seasonal changes or changes in external market conditions.

[0149] Use machine learning algorithms (such as random forests, support vector machines, etc.) to analyze hidden factors affecting project costs and identify abnormal data points (such as abnormally high or low cost records). Forecast future costs based on historical data, such as predicting the cost of material procurement for the next quarter. By combining statistical analysis and machine learning prediction results, a comprehensive assessment of the overall cost of the project is formed, including identifying potential high-risk cost factors and controllable cost parts.

[0150] S300: The strategy generation module generates multiple cost control strategies based on the cost analysis results and applies optimization algorithms. The cost saving potential, implementation difficulty and risk factors of each strategy are calculated in detail.

[0151] In the present invention, heuristic algorithms, genetic algorithms or linear programming algorithms are used to optimize the cost data. According to the characteristics of different projects, a variety of cost control strategies are generated, such as optimizing the procurement process, adjusting the allocation of human resources, etc. For each generated strategy, the system will evaluate its cost saving potential. For example, by selecting different suppliers or changing the construction time, the possible cost saving space is calculated. The implementation difficulty of each strategy is analyzed in detail, including the impact on the existing workflow and supply chain. The potential risks of the strategy are evaluated, and the challenges that may be encountered during the implementation process are considered, such as market price fluctuations or increased labor costs.

[0152] S400: Users evaluate the effectiveness of strategy execution through the feedback module. The system collects user feedback information as a basis for future algorithm adjustments.

[0153] In the present invention, the system collects user feedback information after executing the cost control strategy through the feedback module, including the actual implementation effect of the strategy and unexpected problems that occurred. The system evaluates the effectiveness of each strategy in actual operation based on the execution effect of user feedback, and compares the difference between the predicted value generated by the system and the actual result. User feedback will serve as an important basis for adjusting the algorithm. The system analyzes the successful experience and failed lessons of strategy execution in the feedback to provide higher accuracy for future strategy recommendations.

[0154] S500: The system optimizes the strategy generation algorithm based on user feedback to improve the accuracy of future recommendations.

[0155] In specific implementation, the system automatically optimizes the strategy generation algorithm based on user feedback. For example, if a strategy performs poorly in multiple projects, the system will automatically reduce the weight of recommending the strategy or find a new alternative strategy. The system uses the supervised learning model in machine learning and user feedback data to further adjust the prediction and optimization model to improve the accuracy and reliability of future recommendations.

[0156] S600: The report generation module generates a detailed report of the cost control strategy, including cost forecasts, strategy implementation plans and their impact on the project.

[0157] In one example, the system automatically generates a detailed cost control strategy report, which includes historical cost data analysis, future cost forecasts, savings potential of recommended strategies, implementation difficulty and risk assessment. Report format: The report is presented in a graphical form, such as cost fluctuation trend chart, savings potential distribution chart, risk matrix chart, etc., to help decision makers understand the report content more intuitively. The report supports customization, and decision makers can choose the key indicators and report formats they are concerned about.

[0158] S700: The system provides the final generated report to the decision maker, who selects and executes the cost control strategy based on the report.

[0159] In the present invention, the system provides the generated report to the decision maker, who selects the appropriate cost control strategy based on the content of the report and arranges for its implementation. The decision maker can weigh different strategies based on the savings potential and risk factors provided in the report. The system continuously monitors the implementation of the selected strategy and tracks the actual changes in project costs in real time. If there is a discrepancy with the forecast, the system will automatically issue a reminder and recommend adjusting the strategy. As the project progresses and the market environment changes, the system can dynamically adjust the recommended strategy to ensure the continuous optimization and adaptability of the strategy, helping the project achieve the best cost control effect.

[0160] This cost strategy intelligent recommendation subsystem helps enterprises manage costs more efficiently through powerful data analysis and optimization algorithms, and provides continuous strategy optimization functions to ensure that project costs are effectively controlled and managed.

[0161] In the present invention, the project investment strategy intelligent optimization system includes:

[0162] Market data integration module: Integrate market data from multiple data sources, including financial markets, industry trends, competitor dynamics, policies and regulations, etc. Update data in real time to ensure the accuracy and timeliness of analysis results. Provide data cleaning, denoising and standardization functions to ensure compatibility between various data sources.

[0163] Investment analysis engine: Use time series analysis, ARIMA model, neural network and other technologies to analyze market trends and predict market changes. Combine macroeconomic indicators, industry development trends and company financial status to evaluate potential investment opportunities. Support multi-asset portfolio analysis to find the best investment portfolio.

[0164] Risk Assessment Module: Use VaR (Value at Risk), Monte Carlo simulation and other risk assessment methods to analyze the risk level of potential investments. Provide risk warnings based on market fluctuations, policy changes and other risk factors to support users in risk avoidance.

[0165] Strategy Optimization Module: Use heuristic algorithms (such as genetic algorithms and simulated annealing algorithms) to optimize investment strategies and find the optimal risk-return ratio. Provide strategy visualization tools to display the return rate, risk distribution and risk-return ratio of different investment portfolios.

[0166] Decision support system: Provides users with strategy evaluation tools, supports multiple scenario simulations, and allows users to select the optimal strategy based on different market conditions. Provides a visual interface to display market data analysis results and optimized investment portfolio recommendations.

[0167] In the present invention, the project investment strategy intelligent optimization system is used to implement the following algorithm steps:

[0168] S100: The market data integration module collects real-time market and industry data from multiple sources such as financial markets, industry data, and policies and regulations, and performs data cleaning, denoising, and formatting.

[0169] In specific implementations, the data source is the system collecting real-time market data from multiple data sources, including financial market prices, industry development trends, company financial reports, economic indicators, policies and regulations, etc. Data sources may include APIs (financial data providers), industry reports, and economic data released by the government.

[0170] In a specific implementation, the data cleaning includes removing noise, duplicates or missing values ​​in the data to ensure the accuracy and consistency of the data, such as removing abnormal price fluctuations or invalid financial data.

[0171] In one specific implementation, the denoising process smoothes short-term fluctuations in the data by applying a filter (such as a Kalman filter) or a moving average technique.

[0172] In a specific implementation, the formatting: standardizes raw data from different data sources into a unified format to facilitate subsequent model processing.

[0173] S200: The investment analysis engine uses time series models (such as ARIMA) and machine learning algorithms to analyze market trends and investment opportunities, and makes predictions based on economic indicators, market fluctuations, and industry developments.

[0174] In this invention, time series analysis uses the ARIMA model to perform time series analysis on financial market prices and predict market trends and fluctuations. The model uses historical data to predict future price trends and provide quantitative support. Combining cyclical and seasonal factors, further adjustments are made to industry development and market trends.

[0175] Machine learning algorithms use historical market data and economic indicators to explore potential investment opportunities through algorithms such as random forests and support vector machines. Based on the input economic data (such as GDP growth rate, interest rate changes, etc.), market fluctuations are predicted to provide support for investment opportunity evaluation. Comprehensive forecasts combine market trend forecasts and industry development dynamics to generate a list of investment opportunities, covering potential investment opportunities in different industries and asset types.

[0176] S300: The risk assessment module uses VaR and Monte Carlo simulation methods to assess the risk level of potential investment portfolios. Risk modeling is performed by considering factors such as market fluctuations, economic cycles, and external policy changes.

[0177] It should be noted that VaR (Value at Risk): uses VaR models (such as historical simulation method, Monte Carlo method) to evaluate the maximum potential loss of an investment portfolio within a certain confidence interval. This method is used to quantify the risks brought by market fluctuations.

[0178] In the present invention, Monte Carlo simulation: Monte Carlo simulation is performed based on historical data, and the performance of different investment portfolios under market conditions is simulated through a large number of random samples, so as to estimate the potential loss distribution and return range.

[0179] In the present invention, risk modeling: combines market fluctuations, economic cycles, and external policy changes to establish a comprehensive risk model, dynamically evaluate various risks (such as market risk, credit risk, policy risk, etc.), and provide a basis for investment decisions.

[0180] S400: The strategy optimization module uses genetic algorithms or other optimization algorithms to iteratively optimize the investment portfolio and find the investment strategy with the best risk-return ratio.

[0181] In specific implementation, genetic algorithm: Use genetic algorithm to iteratively optimize investment strategies. Through population selection, crossover, mutation and other operations, the investment portfolio is gradually optimized to achieve a balance between return and risk. The fitness function is based on the risk-return ratio of the investment portfolio to ensure that the selected investment portfolio has a high rate of return and low risk. Simulated annealing algorithm: Use simulated annealing algorithm to further optimize the investment portfolio, avoid local optimal solutions, and find the global optimal investment strategy. Optimization iteration: The optimization algorithm is iterated for multiple rounds to gradually improve the overall performance of the investment portfolio, and dynamically adjust the strategy according to market changes.

[0182] S500: The decision support system provides visualization tools based on the optimization results to help users evaluate different investment strategies, simulate different market environments, and calculate the potential returns and risks of different investment strategies.

[0183] Specifically, the system provides graphical visualization tools to display the performance of investment portfolios, such as risk distribution, return rate curve, portfolio asset weights, etc. Users can simulate strategies through different hypothetical scenarios, and the system generates different portfolio performance forecasts based on changes in the market environment. Through intuitive charts (such as risk-return rate charts and asset distribution pie charts), the potential returns and risks of different investment strategies are displayed to help users better understand the impact of decisions.

[0184] S600: The user selects a suitable strategy through the decision support system, and the system evaluates and adjusts the strategy based on the user's selection.

[0185] In the present invention, users can compare multiple investment strategies through the decision support system and select the best strategy that meets their risk preferences and return goals. Users can dynamically adjust strategy parameters (such as risk tolerance, return expectations, etc.) through the system according to different market environments and personal preferences, and the system automatically re-evaluates and optimizes strategies.

[0186] S700: Determine the final investment strategy and execute the investment decision. The strategy is continuously monitored and optimized.

[0187] Preferably, the finalized investment strategy will be implemented by the execution module. The system automatically monitors the performance of the investment portfolio and provides real-time feedback on the impact of market changes. The system continuously optimizes the investment portfolio based on actual market performance. When the market fluctuates greatly or policies change, the system can automatically adjust the investment strategy to ensure maximum returns and minimize risks.

[0188] For other structures of the AI-based large infrastructure project management system described in this embodiment, refer to the prior art.

[0189] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A large infrastructure project management system based on AI, characterized by: Including management system backend and client communicating with management system backend; The client is used to provide an access interface for human-computer interaction; Wherein, the backend server includes: Project database, which is used to store project data, including basic information of engineering projects, engineering overview, progress control, responsible entities, project management organizations, platform users, main contracts, downstream contracts, financial indicators, project excellence, project licenses, general contract settlement, acceptance materials and bank accounts; A project management module is connected to the project database, and is used to query, create, edit, delete and export the project data. The project data has a project number for uniqueness verification.

2. According to claim 1, a large infrastructure project management system based on AI is characterized in that: The backend server comprises: A contract database, wherein the contract database is used to store the main contract, downstream contracts and material procurement contracts of the engineering project, wherein the material procurement contracts include contract terms information; A contract management module connected to the contract database, the contract management module is used to query, create, edit, delete and export the material procurement contract, and the contract management module is also used to display the material procurement contract by category; An income management module connected to the contract database, the income management module comprising an income operation unit and a notification form generation unit; The operation unit is used to query, create, edit, delete and export the payment data associated with the material procurement contract; The notification generating unit is used to generate a receipt notification according to the contract terms information and send it to the client; the receipt notification has a unique notification number. When the material procurement contract has a notification number, the notification generating unit cannot create a new receipt notification; the notification generating unit is also used to generate an account receipt notification according to the payment receipt information sent by the client and store it in the contract database.

3. According to claim 2, a large infrastructure project management system based on AI is characterized in that: The backend server comprises: An expenditure management module connected to the contract database, the expenditure management module comprising an expenditure operation unit and a file generation unit; The expenditure operation unit is used to query, create, edit, delete and export the approval document quotation sheet, which is used in the approval process of paying the project engineering fee; The file generation unit is used to generate an approval document for paying project engineering funds based on a registered approval document quotation sheet. The approval document has an approval document number, which is composed of a payment category number, a four-digit year and month, a project management department number, a project number, and a two-digit sequence number of the year and month.

4. According to claim 3, a large infrastructure project management system based on AI is characterized in that: The backend server comprises: A fund ledger module, which includes a collection ledger unit, a payment ledger unit, a freezing ledger unit, and a legal risk ledger unit; The payment account unit is used to generate a project payment account according to the payment information sent by the client, and the project payment account includes project payment collection, output value, settlement, receivable and uncollected project payment and loan; The payment ledger unit is used to generate a payment ledger according to the expenditure information sent by the client, and the payment ledger includes project name, payee name, related loan, related payment, related fund transfer, remarks, payment number, project management, whether it is borrowed, bank account type, approval number, expenditure classification, whether it is bill payment, payment date, payment amount, payment amount, principal, capital occupation fee and exemption amount; The freezing account unit is used to generate a freezing account according to the fund freezing information sent by the client, and the freezing account includes a bank account number, a court name and a freezing amount; The legal risk ledger unit is used to generate a legal risk ledger unit based on the litigation information sent by the client, and the legal risk ledger includes the project number, project name, engineering management department, dispute counterparty, case number, dispute handling results and case handling results.

5. According to claim 4, a large infrastructure project management system based on AI is characterized in that: The backend server comprises: A cooperation unit database, which is used to store cooperation unit data, including basic information, business information, storage information, license information, award information, annual evaluation, account information, invoicing information and cooperation records of cooperation units; A resource management module connected to the cooperation unit database, the resource management module includes a resource operation unit, a supplier recommendation unit and a supplier blacklist unit; The resource operation unit is used to query, create, edit, delete and export the data of the cooperation unit; The recommended supplier unit is used to classify the cooperation units according to the recommendation level to form a recommended supplier directory, and the recommendation level includes recommendation and key recommendation; The supplier blacklist unit is used to list the cooperation unit in a blacklist directory, and the blacklist directory includes the listing time and listing reason of the cooperation unit.

6. According to claim 5, a large infrastructure project management system based on AI is characterized in that: The backend server comprises: The invoicing module is used to generate an invoicing request based on the invoicing information sent by the client and send it to the client with the review authority. After the invoicing request passes the review, the invoicing module generates an invoicing application form based on the invoicing information; The invoice account module is used to organize the storage address of the invoicing request, invoicing information and invoicing application form of the invoicing module into an invoice account.

7. The AI-based large infrastructure project management system according to claim 6 is characterized in that: The backend server comprises: The project large screen module is used to visually display project information, engineering progress information, excellence goals, downstream contract information and on-site real-time information; The branch BI large screen module is used to visually display business unit information, project list, branch information, project status information, cost rate information and quality improvement information; The assessment BI large screen is used to visually display the engineering management department list, assessment rankings, overall rankings of business units, scoring details and individual rankings.

8. A large infrastructure project management system based on AI according to any one of claims 1 to 7, characterized in that: The client is one or more of a WeChat platform applet, an enterprise WeChat information platform, a PC application, and a web client.

Images