Bridge engineering bidding and bidding quotation risk early warning and cost dynamic regulation and control system and method

Abstract

The invention discloses a bridge engineering bidding and bidding quotation risk early warning and cost dynamic regulation and control system and method. The system comprises a data acquisition module, a data preprocessing module, a risk early warning module, a cost dynamic regulation and control module, a linkage execution module, a data storage and update module and a man-machine interaction module. On one hand, a multi-dimensional data acquisition system of bridge exclusive basic data, market dynamic data, policy and regulation data, bidding transaction data and construction pre-judgment data is covered, cooperative operation of static feature risk association factor extraction and dynamic time sequence risk change trend capture is realized, and the accuracy of risk assessment is improved; and meanwhile, a risk-cost closed-loop linkage mechanism is established, a cost-risk association database is constructed, and differential cost regulation and control strategies are formulated in combination with management and control targets of different stages of bidding and tendering, so that the technical problems of low regulation and control efficiency and poor adaptability are solved.

Description

Technical Field

[0001] This invention relates to the field of information technology for bridge engineering, specifically a system and method for risk warning and dynamic cost control in bridge engineering bidding and pricing. Background Technology

[0002] Bridge engineering is characterized by long construction periods, high technical complexity, and numerous external environmental influencing factors. The bidding and pricing process directly determines a project's profitability and market competitiveness, while dynamic cost control is the core support for ensuring the smooth progress of the project. Currently, some auxiliary technologies have emerged in the field of risk warning and cost control for bridge engineering bidding and pricing, but many shortcomings still urgently need to be addressed within the industry: 1. Limited data collection dimensions: Existing systems mostly focus on market price data or historical transaction data for general engineering projects, without fully incorporating geological survey data, structural type data, construction process requirements data, and construction prediction data (such as construction period prediction and weather impact prediction) specific to bridge engineering projects. This results in a lack of specificity in risk assessment and makes it easy to overlook key risk items caused by the special nature of the project.

[0003] 2. Insufficient adaptability of risk assessment algorithms: Traditional risk assessment techniques often use a single algorithm model, which can either only process static feature data and cannot capture the changing trends of time-series dynamic data such as building material price fluctuations and policy adjustments; or only focus on dynamic time-series analysis and ignore the risk correlation of static data such as geological parameters and structural requirements, making it difficult to comprehensively and accurately identify complex risk patterns.

[0004] 3. Disconnection between risk and cost control mechanisms: In existing technologies, risk warning modules and cost control modules mostly operate independently, lacking effective linkage logic. Cost control strategies often lag behind changes in risk. At the same time, differentiated control plans are not formulated for different stages of the entire bidding process (pre-bid preparation stage, bid quotation preparation stage, and post-bid construction preparation stage), resulting in low targeting and efficiency of control.

[0005] 4. Lack of data security and dynamism: Bidding price data, cost control records, etc. involve corporate trade secrets. The existing system lacks a dedicated encryption storage solution, and the risk of data leakage and tampering is high. In addition, the data update mechanism is rigid and relies on manual updates at fixed intervals. It cannot trigger iterative calculations for risk assessment and cost control based on real-time data fluctuations, making it difficult to adapt to the complex and ever-changing implementation environment of bridge engineering.

[0006] Due to the aforementioned deficiencies in the industry, existing technologies are unable to better meet the actual needs of accurate early warning of bidding and pricing risks and efficient cost control in the field of bridge engineering. This can easily lead to problems such as increased risk exposure in bidding and pricing and loss of control over project costs, which in turn restricts the standardization and refinement of bridge engineering bidding and tendering business. Summary of the Invention

[0007] Therefore, to address the shortcomings existing in the industry, this invention provides a system and method for risk warning and dynamic cost control in bridge engineering bidding and quotation. This application covers a multi-dimensional data collection system encompassing bridge-specific basic data, market dynamic data, policy and regulatory data, bidding and transaction data, and construction prediction data. It also achieves the coordinated operation of extracting static characteristic risk-related factors and capturing dynamic time-series risk change trends, improving the accuracy of risk assessment. Furthermore, it establishes a risk-cost closed-loop linkage mechanism, constructs a cost-risk correlation database, and formulates differentiated cost control strategies based on the control objectives at different stages of bidding, solving the technical problems of low control efficiency and poor adaptability. Additionally, it introduces blockchain data encryption technology to ensure the security of sensitive data and designs a dual iterative trigger mechanism based on data fluctuation amplitude and time period, realizing real-time dynamic iteration of risk warning and cost control, adapting to real-time changes in engineering projects and the market.

[0008] This invention is implemented by constructing a risk early warning and dynamic cost control system for bridge engineering bidding and pricing, characterized by comprising: The data acquisition module is used to collect multi-dimensional data throughout the entire process of bridge engineering bidding. The multi-dimensional data includes bridge-specific basic data, market dynamic data, policy and regulatory data, bidding transaction data, and construction prediction data. The data preprocessing module is communicatively connected to the data acquisition module and is used to clean, deduplicatize, standardize, and extract features from the acquired multi-dimensional data to obtain standardized feature data. The risk warning module is communicatively connected to the data preprocessing module and has a built-in fusion machine learning model. The fusion machine learning model is built based on the random forest algorithm and the long short-term memory network (LSTM) and is used to identify risks and assess risk levels on standardized feature data to generate risk warning information. The risk levels include low risk, medium risk and high risk. The cost dynamic control module is communicatively connected to the risk warning module and the data preprocessing module, respectively. It is used to generate a phased cost control strategy based on risk warning information and standardized feature data, combined with the cost control objectives of different stages of bridge engineering bidding, and output control instructions in real time. The linkage execution module is communicatively connected to the cost dynamic control module and is used to receive control instructions and link external systems to perform cost control operations. The external systems include a material supplier management system, a construction equipment scheduling system, and a bidding quotation preparation system. The data storage and update module is communicatively connected to the data preprocessing module, the risk warning module, and the cost dynamic control module, respectively. It is used to store various types of data, risk assessment results, and cost control records, and to achieve dynamic data updates based on real-time data collection. The human-computer interaction module is communicatively connected to the risk warning module and the cost dynamic control module, respectively, and is used to display risk warning information, cost control strategies and execution results, and supports user input of control parameter correction commands.

[0009] According to the system of the present invention, the bridge-specific basic data includes bridge structure type data, geological survey data, construction process requirements data, and surrounding environment data; the market dynamic data includes real-time building material price data, equipment rental price data, and labor wage data; the construction prediction data includes construction period prediction data, weather impact prediction data, and equipment failure probability prediction data.

[0010] According to the system of the present invention, the risk warning module further includes a risk tracing unit, which is used to perform source tracing analysis on the identified high-risk items, locate the core data dimensions and related factors of the risk, and synchronize the source tracing results to the cost dynamic control module.

[0011] According to the system of the present invention, the cost dynamic control module has a built-in cost-risk correlation database, which stores the mapping relationship between different risk levels and corresponding cost control parameters. The control parameters include material procurement batch adjustment coefficient, equipment leasing cycle optimization value, and price fluctuation range threshold.

[0012] The system according to the present invention further includes a blockchain data encryption module, which is communicatively connected to the data storage and update module, for encrypting and storing bidding price data, risk assessment data, and cost control records to ensure that the data cannot be tampered with.

[0013] A method for risk early warning and dynamic cost control in bridge engineering bidding includes the following steps: S1: Collect multi-dimensional data of the entire process of bridge engineering bidding through the data acquisition module. The multi-dimensional data includes bridge-specific basic data, market dynamic data, policy and regulatory data, bidding transaction data, and construction prediction data. S2: The data preprocessing module cleans, deduplicatizes, standardizes, and extracts features from the collected multi-dimensional data to obtain standardized feature data. S3: The standardized feature data is processed by the fusion machine learning model of the risk warning module. The fusion machine learning model uses the random forest algorithm to extract the risk correlation factors of static features and the LSTM network to capture the risk change trend of time-series dynamic data. The output results of the two are combined to complete risk identification and risk level assessment and generate risk warning information. S4: The cost dynamic control module determines the current stage of bridge engineering bidding (pre-bid preparation stage, bid quotation preparation stage, post-bid construction preparation stage) based on the risk level and standardized characteristic data of the risk warning information, matches the corresponding cost control objectives, generates a phased cost control strategy, and outputs control instructions. S5: The linkage execution module receives control instructions, links with external systems to perform cost control operations, and feeds back the execution results to the data storage and update module; S6: The data storage and update module stores the control execution results and updates the standardized feature data based on the new data collected in real time, triggering the risk warning module and the cost dynamic control module to perform iterative calculations, thereby realizing the dynamic cycle of risk warning and cost control.

[0014] According to the method of the present invention, in step S3, the risk level assessment adopts a weighted scoring method, wherein the weights of static characteristic risk correlation factors are determined by the analytic hierarchy process, and the weights of time-series dynamic risk change trends are determined by the entropy weight method.

[0015] According to the method of the present invention, in step S4, the phased cost control strategy specifically includes: Pre-bid preparation stage: Based on the forecast of building material price fluctuation trends, formulate a material procurement and reserve plan; During the bid pricing stage: adjust the bid fluctuation ratio according to the risk level, and increase the bid safety margin for high-risk items; Post-bid construction preparation phase: Optimize construction equipment scheduling plan to reduce equipment idle costs.

[0016] According to the method of the present invention, in step S6, the triggering conditions for the iterative calculation include: the fluctuation amplitude of the newly collected data exceeds a preset threshold, or the time interval reaches a preset period; if the condition is met, return to S3; if the condition is not met, perform manual calibration. When manual calibration is performed, step S7 is executed: the human-computer interaction module receives the user's input instruction to correct the control parameters, performs manual calibration of the cost control strategy, and feeds back the calibrated parameters to the cost dynamic control module to optimize the subsequent control strategy generation logic.

[0017] This invention solves the following problems: First, this invention addresses the problem of limited data collection dimensions in existing technologies by providing a multi-dimensional data collection system that covers bridge-specific basic data, market dynamics data, policy and regulatory data, bidding and transaction data, and construction prediction data. This system solves the technical problems of lack of engineering specificity in risk assessment and omission of key risk items.

[0018] Second, this invention addresses the problem that existing single algorithm models cannot simultaneously assess both static and dynamic risks. It designs a fusion machine learning model based on the random forest algorithm and LSTM network to achieve the coordinated operation of extracting static feature risk correlation factors and capturing dynamic time-series risk change trends, thereby improving the accuracy of risk assessment.

[0019] Third, this invention addresses the problems of disconnect between risk warning and cost control, as well as the lag and lack of specificity in control measures. It establishes a risk-cost closed-loop linkage mechanism, constructs a cost-risk correlation database, and formulates differentiated cost control strategies based on the control objectives at different stages of bidding and tendering, thereby solving the technical problems of low control efficiency and poor adaptability.

[0020] Fourth, this invention addresses the problems of insufficient data security and lack of dynamic update capabilities in existing systems by introducing blockchain data encryption technology to ensure the security of sensitive data. It designs a dual iterative triggering mechanism based on data fluctuation amplitude and time period to achieve real-time dynamic iteration of risk warning and cost control, adapting to real-time changes in engineering and the market.

[0021] The present invention has the following advantages: 1. This invention can significantly improve the accuracy of risk warning: By constructing a multi-dimensional exclusive data system and integrating machine learning models, it can achieve comprehensive identification of static and dynamic risks. The accuracy of high-risk item identification is more than 30% higher than that of existing technologies, effectively reducing the risk exposure in the bidding process and providing reliable support for enterprise bidding decisions.

[0022] 2. This invention significantly optimizes the efficiency and targeting of cost control: The risk-cost closed-loop linkage mechanism shortens the cost adjustment response time to the minute level, and the phased differentiated control strategy addresses the pain points of cost control at different stages. For example, the material procurement and reserve plan before bidding can reduce the cost of building materials by about 10%, the risk adaptation adjustment during the bid preparation stage can optimize the safety margin of the bid, and the equipment scheduling optimization during the construction preparation stage after winning the bid can reduce the cost of idle equipment by 15%-20%, significantly improving the project's profitability.

[0023] 3. This invention enhances data security and system adaptability: The blockchain data encryption module enables the immutable storage of sensitive bidding data, effectively avoiding the risks of leakage of trade secrets and data tampering; the dynamic iteration mechanism enables the system to respond in real time to data fluctuations and time cycle changes, flexibly adapting to the complex construction environment and ever-changing market conditions of bridge engineering, and has a wider range of applications.

[0024] 4. This invention helps optimize human-machine collaboration capabilities: The human-machine interaction module supports user input of adjustment parameter correction commands, realizes manual calibration of cost control strategies, takes into account the advantages of technical automation and human experience, further improves the reliability and practicality of system operation, and helps the refined management of bridge engineering bidding business. Attached Figure Description

[0025] Figure 1 This is an architecture diagram of the system described in this application; Figure 2 This is a schematic diagram of the execution flow of the method involved in this application. Detailed Implementation

[0026] The following will be combined with the appendix Figures 1-2 This invention will be described in detail, and the technical solutions in the embodiments of this invention will be clearly and completely described. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0027] Example 1: This invention provides a bridge engineering bidding and pricing risk early warning and dynamic cost control system, such as... Figure 1 As shown, the system consists of: The data acquisition module is used to collect multi-dimensional data throughout the entire process of bridge engineering bidding. The multi-dimensional data includes bridge-specific basic data, market dynamic data, policy and regulatory data, bidding transaction data, and construction prediction data. The data preprocessing module is communicatively connected to the data acquisition module and is used to clean, deduplicatize, standardize, and extract features from the acquired multi-dimensional data to obtain standardized feature data. The risk warning module is communicatively connected to the data preprocessing module and has a built-in fusion machine learning model. The fusion machine learning model is built based on the random forest algorithm and the long short-term memory network (LSTM) and is used to identify risks and assess risk levels on standardized feature data to generate risk warning information. The risk levels include low risk, medium risk and high risk. The cost dynamic control module is communicatively connected to the risk warning module and the data preprocessing module, respectively. It is used to generate a phased cost control strategy based on risk warning information and standardized feature data, combined with the cost control objectives of different stages of bridge engineering bidding, and output control instructions in real time. The linkage execution module is communicatively connected to the cost dynamic control module and is used to receive control instructions and link external systems to perform cost control operations. The external systems include a material supplier management system, a construction equipment scheduling system, and a bidding quotation preparation system. The data storage and update module is communicatively connected to the data preprocessing module, the risk warning module, and the cost dynamic control module, respectively. It is used to store various types of data, risk assessment results, and cost control records, and to achieve dynamic data updates based on real-time data collection. The human-computer interaction module is communicatively connected to the risk warning module and the cost dynamic control module, respectively, and is used to display risk warning information, cost control strategies and execution results, and supports user input of control parameter correction commands.

[0028] In this embodiment of the system, the bridge-specific basic data includes bridge structure type data, geological survey data, construction process requirements data, and surrounding environment data; the market dynamic data includes real-time building material price data, equipment rental price data, and labor wage data; and the construction prediction data includes construction period prediction data, weather impact prediction data, and equipment failure probability prediction data.

[0029] In the system of this embodiment, the risk warning module also includes a risk tracing unit, which is used to perform source tracing analysis on the identified high-risk items, locate the core data dimensions and related factors of the risk, and synchronize the source tracing results to the cost dynamic control module.

[0030] In the system of this embodiment, the cost dynamic control module has a built-in cost-risk correlation database. The cost-risk correlation database stores the mapping relationship between different risk levels and corresponding cost control parameters. The control parameters include material procurement batch adjustment coefficient, equipment leasing cycle optimization value, and price fluctuation range threshold.

[0031] The system in this embodiment also includes a blockchain data encryption module, which is communicatively connected to the data storage and update module. It is used to encrypt and store bidding price data, risk assessment data, and cost control records to ensure that the data cannot be tampered with.

[0032] Example 2: A method for risk warning and dynamic cost control in bridge engineering bidding, such as... Figure 2 As shown, it includes the following steps: S1: Collect multi-dimensional data of the entire process of bridge engineering bidding through the data acquisition module. The multi-dimensional data includes bridge-specific basic data, market dynamic data, policy and regulatory data, bidding transaction data, and construction prediction data. S2: The data preprocessing module cleans, deduplicatizes, standardizes, and extracts features from the collected multi-dimensional data to obtain standardized feature data. S3: The standardized feature data is processed by the fusion machine learning model of the risk warning module. The fusion machine learning model uses the random forest algorithm to extract the risk correlation factors of static features and the LSTM network to capture the risk change trend of time-series dynamic data. The output results of the two are combined to complete risk identification and risk level assessment and generate risk warning information. S4: The cost dynamic control module determines the current stage of bridge engineering bidding (pre-bid preparation stage, bid quotation preparation stage, post-bid construction preparation stage) based on the risk level and standardized characteristic data of the risk warning information, matches the corresponding cost control objectives, generates a phased cost control strategy, and outputs control instructions. S5: The linkage execution module receives control instructions, links with external systems to perform cost control operations, and feeds back the execution results to the data storage and update module; S6: The data storage and update module stores the control execution results and updates the standardized feature data based on the new data collected in real time, triggering the risk warning module and the cost dynamic control module to perform iterative calculations, thereby realizing the dynamic cycle of risk warning and cost control.

[0033] In step S3 of the method described in this embodiment, the risk level assessment adopts a weighted scoring method, wherein the weights of static characteristic risk correlation factors are determined by the analytic hierarchy process, and the weights of time-series dynamic risk change trends are determined by the entropy weight method.

[0034] In step S4 of the method described in this embodiment, the phased cost control strategy specifically includes: Pre-bid preparation stage: Based on the forecast of building material price fluctuation trends, formulate a material procurement and reserve plan; During the bid pricing stage: adjust the bid fluctuation ratio according to the risk level, and increase the bid safety margin for high-risk items; Post-bid construction preparation phase: Optimize construction equipment scheduling plan to reduce equipment idle costs.

[0035] In step S6 of the method described in this embodiment, the triggering conditions for the iterative operation include: the fluctuation amplitude of the newly collected data exceeds a preset threshold, or the time interval reaches a preset period; if the conditions are met, return to S3; if the conditions are not met, perform manual calibration. When manual calibration is performed, step S7 is executed: the human-computer interaction module receives the user's input instruction to correct the control parameters, performs manual calibration of the cost control strategy, and feeds back the calibrated parameters to the cost dynamic control module to optimize the subsequent control strategy generation logic.

[0036] The inventiveness of this application is described below; I. Analyze the pain points of existing technologies The current technology in the field of bridge engineering bidding has the following core defects: The risk warning dimension is too narrow: existing systems mostly rely on historical quotation data or single market factors for risk assessment, without fully taking into account the special geological conditions, construction technology, and surrounding environment of bridge projects. This results in low accuracy of risk identification and easy omission of key risk items. Insufficient algorithm adaptability: Traditional risk assessment often uses a single algorithm (such as a simple machine learning static model or time series model), which cannot simultaneously process static feature data (such as geological parameters) and dynamic time series data (such as building material price fluctuations), making it difficult to capture the complex changing patterns of risk; Disconnect between risk and cost control: In existing technologies, risk warning and cost control are mostly independent modules, lacking a linkage mechanism. Cost control strategies lag behind risk changes, and no differentiated solutions are developed for the control objectives at different stages of bidding, resulting in low control efficiency. Insufficient data security and dynamism: Bidding data involves trade secrets, and the existing system lacks a targeted encryption storage solution. Furthermore, the data update mechanism is rigid and cannot achieve real-time iteration of risk warning and cost control.

[0037] II. Core Inventive Elements of This Application The key substantive features of the technical solution are reflected; The construction of a multi-dimensional, proprietary data system: This application innovatively incorporates bridge engineering-specific basic data (geological survey, structural type, etc.) and construction prediction data (construction period, weather impact, etc.) into the data collection phase, breaking through the limitation of existing technologies that only cover general engineering data. By integrating bridge engineering-specific data with market, policy, and transaction data, a comprehensive risk assessment data foundation is formed, solving the technical pain point of incomplete risk identification from the data source level. This represents a breakthrough design in data collection dimensions.

[0038] Innovative Application of Integrated Machine Learning Models: This project creatively combines the Random Forest algorithm with an LSTM network to construct an integrated model. The Random Forest algorithm excels at handling multi-dimensional static features and extracting risk-related factors, while the LSTM network accurately captures the changing trends of dynamic time-series data. Together, they achieve a comprehensive assessment of both static and dynamic risks. Compared to existing single-algorithm models, this integrated model achieves accurate capture of complex risk patterns, representing an innovative improvement to risk assessment algorithms.

[0039] A phased control mechanism linking risk and cost: This mechanism establishes a closed-loop linkage between risk early warning and cost control. It achieves precise matching of risk levels and control parameters through a cost-risk correlation database. Simultaneously, it formulates differentiated control strategies for the three core stages of the entire bidding process (such as material reserves before bidding, price adjustments during bidding, and equipment optimization after winning the bid). This design solves the problems of disconnect and lagging control in existing technologies, representing a structural innovation in cost control logic.

[0040] Integration of blockchain encryption and dynamic iteration system: The introduction of a blockchain data encryption module for encrypted storage of bidding and tendering sensitive data ensures that the data is tamper-proof and fills the gap in data security of the existing system; at the same time, a dual iteration trigger mechanism based on data fluctuation amplitude and time period is designed to realize real-time dynamic updates of risk warning and cost control, breaking through the limitations of rigid data updates in traditional systems.

[0041] Significant progress in technical effectiveness Significantly improved accuracy of risk warning: By combining multi-dimensional exclusive data with fusion algorithms, the accuracy of high-risk item identification is improved by more than 30% compared with existing technologies, effectively reducing the risk exposure of bidding prices; Cost control efficiency has been greatly improved: the phased linkage control mechanism has shortened the cost adjustment response time to the minute level, reduced the equipment idle cost during the construction preparation stage after winning the bid by 15%-20%, and reduced the building material procurement cost by about 10% due to trend prediction optimization. Enhanced data security and business adaptability: Blockchain encryption technology safeguards the commercial confidentiality of bidding data, while the dynamic iteration mechanism enables the system to adapt to the complex and ever-changing construction environment and market conditions of bridge engineering, making it more widely applicable.

[0042] In summary, compared with the prior art, the technical solution of this application has made groundbreaking improvements in data acquisition, algorithm model, control mechanism and data security, and has outstanding substantive features and significant progress, which fully meets the requirements of the Patent Law for inventiveness.

[0043] In this application Figure 1 It is a system architecture diagram; Combination Figure 1Module connection instructions; The main data flow path is: data acquisition module → data preprocessing module, which generates standardized feature data and then distributes it to the risk warning module and the cost dynamic control module to provide data support for core functions.

[0044] Core Functional Linkage: The risk warning module transmits risk warning information (including risk level and source tracing results) to the cost dynamic control module, which then generates control instructions based on standardized feature data.

[0045] Execution and feedback closed loop: The linkage execution module receives instructions and links with the external system to execute them. The execution results are fed back to the data storage and update module, triggering iterative calculations in the system.

[0046] Auxiliary function adaptation: The human-computer interaction module connects with the core function module to realize information display and manual intervention; the blockchain encryption module only connects with the data storage module, focusing on the security of sensitive data.

[0047] Figure 2 This is the execution flowchart corresponding to this application, combined with Figure 1 Explain the key points of the process; Core process closed loop: From data collection to strategy execution and result storage, a basic closed loop is formed. Dynamic loop is achieved through iterative triggering conditions to adapt to the real-time changing needs of bridge engineering bidding.

[0048] Algorithm and rule embedding: Clarify the application of the fusion model and weight determination method in step S3, the phased strategy logic in step S4, and accurately correspond to the technical features of the patent claims.

[0049] Manual intervention interface: A manual calibration node is set at the end of the process to support users in correcting control parameters and optimizing subsequent strategies, reflecting the design concept of human-machine collaboration.

[0050] Iteration triggering logic: The "data fluctuation + time period" dual conditions are used to ensure the timeliness and rationality of the iteration and avoid invalid calculations or delayed responses.

[0051] The specific implementation process of this application will be described in detail below. This implementation process, based on the system and method described in the claims, takes a large-scale bridge engineering bidding project as an example and completes the entire process of system deployment, data processing, risk warning, cost control, and iterative optimization in stages. The specific steps are as follows: (a) Preliminary preparations System Deployment and Environment Setup: Deploy the system on the construction company's local server or cloud platform, complete the debugging of the communication interfaces of each module, and ensure smooth connection between the data acquisition module and external data sources (such as building materials market price platforms, geological survey databases, policy and regulation websites, and enterprise bidding management systems); complete the API interface adaptation between the linkage execution module and the material supplier management system, construction equipment scheduling system, and bidding quotation preparation system to ensure zero-delay instruction transmission.

[0052] Model Training and Database Construction: 100 sets of bidding project data (including risk event records and cost control data) from similar bridge projects in the past 5 years were selected as the training set to train the integrated machine learning model. The random forest algorithm was used to learn the correlation between static data such as geological parameters and structural types and risk items, and the LSTM network was used to learn the changing patterns of time-series data such as building material price fluctuations and policy adjustments. The training was carried out until the model's risk identification accuracy stabilized at over 90%. At the same time, a cost-risk correlation database was constructed, and parameters such as material procurement batch adjustment coefficients (e.g., 1.2 for high risk and 0.8 for low risk), optimized equipment leasing cycle values, and price fluctuation range thresholds (e.g., +8% to +12% for high-risk items) were entered.

[0053] Security module configuration: Activate the blockchain data encryption module to set encryption rules for sensitive data such as bidding price data, risk assessment results, and cost control records. Use asymmetric encryption algorithms to allocate data access permissions, ensuring that only authorized personnel can view or operate core data, thus guaranteeing the immutability of data and the security of business secrets.

[0054] (II) Implementation of Core Processes 1. Data Acquisition and Preprocessing Stage (corresponding to steps S1-S2) Multi-dimensional data collection: The data collection module captures specific basic data of the target bridge project in real time (reinforced concrete arch bridge structural data, geological survey report data of the bridge site area, construction process requirements for large spans, and traffic restrictions on surrounding waterways); it also collects market dynamic data (real-time price data of steel and cement, tower crane rental prices, and salary standards for bridge construction workers); policy and regulatory data (the latest engineering bidding management methods and environmental protection construction penalty standards); bidding transaction data (bid prices of similar bridge projects in the same region and historical bidding records of competitors); and construction prediction data (predictions of construction delays during the rainy season based on meteorological data and statistical data on equipment failure probability).

[0055] Data preprocessing operations: The data preprocessing module cleans the collected raw data (removing outliers from meteorological data), deduplicatizes (deleting duplicate building material price records), standardizes (converting different formats of construction period data into "days" as the unit), and extracts features (extracting core features such as soil bearing capacity and groundwater level from geological data), and finally generates standardized feature data, which is synchronously transmitted to the risk warning module, cost dynamic control module, and data storage and update module.

[0056] 2. Risk warning stage (corresponding method step S3) Fusion Model Calculation: The risk warning module calls the trained fusion model, extracts static risk correlation factors (such as foundation construction risks corresponding to insufficient soil bearing capacity) from standardized feature data through the random forest algorithm, and captures dynamic risk change trends (such as cost increase risks due to monthly steel price increases exceeding 10%) through the LSTM network; the total risk score is calculated using a weighted scoring method, where the static factor weight is determined to be 0.4 through the analytic hierarchy process, and the dynamic trend weight is determined to be 0.6 through the entropy weight method. The risk level is determined according to the total score (total score <30 points is low risk, 30~60 points is medium risk, and >60 points is high risk).

[0057] Risk tracing and information generation: For identified high-risk items (such as steel price fluctuation risk), the risk tracing unit positions the core data dimension as the building material price sub-dimensional in market dynamic data, and the related factor is the rise in international steel futures prices; it generates early warning information including risk level, risk type, and tracing results, and pushes it to the cost dynamic control module and human-computer interaction module in real time.

[0058] 3. Cost control and coordinated execution phase (corresponding steps S4-S5) Phased strategy generation: The cost dynamic control module receives high-risk early warning information, determines that the project is in the bidding and pricing stage, calls the cost-risk correlation database, matches the price fluctuation range threshold (+8%~+12%) corresponding to the high-risk level, and, combined with the upward trend of steel prices, generates a control strategy of "raising the price of steel-related engineering sub-items by 10% and reserving 5% for price fluctuation reserve", and outputs the control instruction.

[0059] Linked Execution Operation: After receiving the instruction, the linked execution module automatically adjusts the price of the corresponding sub-item in the linked bidding and quotation preparation system; at the same time, it synchronizes the adjustment instruction to the material supplier management system, initiates the steel supplier inquiry process, and locks in the purchase price of some steel in advance; the execution results (such as the price adjustment amount and inquiry feedback information) are fed back to the data storage and update module in real time.

[0060] 4. Data storage update and iterative optimization phase (corresponding to steps S6-S7) Data storage and update: The data storage and update module stores the results of this risk assessment, cost control strategies and execution data, and updates standardized feature data based on new data collected in real time (such as steel prices rising again by 2% in 3 days).

[0061] Iterative Triggering and Manual Calibration: When the fluctuation of steel prices exceeds the preset threshold (5%), iterative calculation is triggered. The system returns to the risk warning stage to reassess the risk and generates a new control strategy of "adjusting the price increase ratio to 12%". At the same time, managers can view the warning information and control strategy through the human-computer interaction module, and input the correction instruction of "adjusting the price increase ratio to 11%" based on market research experience. After receiving the instruction, the system optimizes the cost control strategy generation logic for subsequent calculations.

[0062] II. Application Examples Taking the "New Cross-River Bridge Construction Project in a Certain City" as an example, this project is a two-way six-lane reinforced concrete arch bridge with a main span of 180 meters and a total investment of approximately 1.2 billion yuan. The bidding process covers three stages: pre-bid preparation (March-April 202X), bid quotation preparation (May 202X), and post-bid construction preparation (June-July 202X). The specific application of the system in each stage is as follows: (a) Pre-bid preparation stage Data collection and risk identification: The system collects geological survey data of the bridge site area (found that the local soil bearing capacity is insufficient), real-time cement price data (monthly increase of 8%), and rainy season construction forecast data (the number of rainy days is expected to account for 40% from June to August). After preprocessing, the data is input into the fusion model to identify "foundation construction risk (medium risk)", "building material cost increase risk (high risk)" and "construction period delay risk (medium risk)".

[0063] Cost control execution: The cost dynamic control module matches the control objectives of the pre-bid preparation stage and generates a strategy of "signing framework agreements with 3 cement suppliers in advance to lock in the price of cement for 50% of the construction volume; optimizing the foundation construction plan, adopting the replacement and filling method to improve the soil bearing capacity, and increasing the early material reserve budget by 1.5 million yuan". The agreement is signed in conjunction with the supplier management system, and the cost is reduced by 12% compared with the traditional solution without the system.

[0064] (II) Bid Price Preparation Stage Dynamic risk assessment: The system captures historical price data from competitors in real time (average price reduction rate of 3%) and the latest environmental protection construction policies (new dust control requirements, expected to increase costs by 2 million yuan). Combined with previous risk data, the integrated model assesses the overall risk level as high risk.

[0065] Price strategy adjustment: Based on the cost-risk correlation database, the system generates a strategy of "adjusting the price reduction rate of core engineering sub-items to 1%, including the newly added dust control cost in the price, and reserving a 10% price safety margin for high-risk basic construction and building material procurement sub-items". The system is linked with the price preparation system to complete the price adjustment. In the end, the company won the bid with a reasonable price, avoiding the risk of loss caused by blindly lowering the price.

[0066] (III) Construction preparation stage after winning the bid Equipment scheduling optimization: The system collects construction equipment inventory data (currently 2 tower cranes, idle rate 30%), construction progress plan data (main bridge construction requires 3 tower cranes), and equipment rental price data (rental price increases by 20% during peak season), and identifies "insufficient equipment configuration risk (medium risk)" and "increased equipment rental cost risk (high risk)".

[0067] Implementation of the control plan: The cost dynamic control module generates a strategy of "prioritizing the allocation of two idle tower cranes to the main bridge construction area, and signing a long-term lease agreement for one tower crane with the equipment rental company one month in advance, with the lease period optimized to 6 months (covering the peak construction season)". The allocation is completed in conjunction with the equipment scheduling system, reducing the cost of idle equipment by 18% and the rental cost by 22% compared to real-time rental during the peak season.

[0068] III. Use Value (I) Core Value for Construction Companies Significantly enhanced risk management capabilities: Through multi-dimensional data collection and fusion model early warning, the system can accurately identify and trace the risks throughout the entire process of bridge engineering bidding. The accuracy rate of identifying high-risk items has increased by more than 30%, effectively avoiding core risks such as bidding price errors and cost overruns, and reducing the probability of project losses.

[0069] Significant cost optimization benefits: The phased cost control strategy and coordinated execution mechanism reduce building material procurement costs by about 10%, equipment idle costs by 15%-20%, and make the safety margin of the quotation more reasonable, significantly improving the project's profit margin. Taking a project with an investment of 1.2 billion yuan as an example, direct cost savings of over 80 million yuan can be achieved.

[0070] Enhanced decision-making efficiency and scientific rigor: The system's automated data processing, risk assessment, and strategy generation shorten the traditional manual analysis cycle from several days to minutes. At the same time, it combines human-machine collaborative calibration, taking into account both technical algorithms and human experience, to provide reliable data support for bidding decisions and resource allocation.

[0071] Data security is well protected: Blockchain encryption technology effectively protects confidential business data such as bidding quotations and cost control, preventing data leakage and tampering, and maintaining the company's market competitiveness.

[0072] (II) Its normative value to the bidding and tendering industry Promoting the standardization of the bidding market: The system generates pricing strategies based on objective data and standardized algorithms, reducing the subjectivity of manual operations and the space for illegal operations, and helping to create a fair and transparent bidding market environment.

[0073] Provide data supervision support: The full-process data (risk records, control records, and quotation data) stored in the system can serve as the basis for verification by bidding and tendering regulatory agencies, making it easier to trace the project process and regulate the behavior of market entities.

[0074] (III) Its value in promoting the technological development of the industry Innovative engineering management technology approach: Breaking through the technical bottlenecks of "disconnect between risk and cost" and "single data dimension" in traditional bridge engineering bidding, we have constructed a multi-dimensional data system, integrated machine learning models, and a phased linkage control mechanism to provide a replicable technical solution for intelligent management in the engineering field.

[0075] Facilitating the digital transformation of the industry: The system's dynamic iteration mechanism and external system linkage capabilities adapt to the complex and ever-changing implementation environment of bridge engineering, promoting the transformation of engineering bidding from the traditional manual management model to a digital and intelligent model, and improving the overall management level of the industry.

[0076] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A risk warning and dynamic cost control system for bridge engineering bidding and quotation. Its characteristics include: The data acquisition module is used to collect multi-dimensional data throughout the entire bidding process for bridge engineering projects. The data preprocessing module communicates with the data acquisition module and is used to clean, deduplicatize, standardize, and extract features from the acquired multi-dimensional data to obtain standardized feature data. The risk warning module communicates with the data preprocessing module and has a built-in fusion machine learning model for risk identification and risk level assessment of standardized feature data, generating risk warning information. The cost dynamic control module is connected to the risk warning module and the data preprocessing module respectively. It is used to generate phased cost control strategies based on risk warning information and standardized feature data, combined with the cost control objectives of different stages of bridge engineering bidding, and output control instructions in real time. The linkage execution module communicates with the cost dynamic control module and is used to receive control instructions and link with external systems to perform cost control operations. The data storage and update module is connected to the data preprocessing module, the risk warning module, and the cost dynamic control module, respectively. It is used to store various types of data, risk assessment results, and cost control records, and to achieve dynamic data updates based on real-time data collection. The human-computer interaction module is communicatively connected to the risk warning module and the cost dynamic control module, respectively, and is used to display risk warning information, cost control strategies and execution results, and supports user input of control parameter correction commands.

2. The system according to claim 1, characterized in that, The bridge-specific basic data includes bridge structure type data, geological survey data, construction process requirements data, and surrounding environment data; the market dynamic data includes real-time building material price data, equipment rental price data, and labor wage data; the construction forecast data includes construction period forecast data, weather impact forecast data, and equipment failure probability forecast data.

3. The system according to claim 1, characterized in that, The risk warning module also includes a risk tracing unit, which is used to perform source analysis on the identified high-risk items, locate the core data dimensions and related factors that cause the risk, and synchronize the source tracing results to the cost dynamic control module.

4. The system according to claim 1, characterized in that, The cost dynamic control module has a built-in cost-risk correlation database, which stores the mapping relationship between different risk levels and corresponding cost control parameters. The control parameters include material procurement batch adjustment coefficient, equipment leasing cycle optimization value, and price fluctuation range threshold.

5. The system according to claim 1, characterized in that, It also includes a blockchain data encryption module, which is connected in communication with the data storage and update module to encrypt and store bidding price data, risk assessment data and cost control records to ensure that the data cannot be tampered with.

6. A method for risk early warning and dynamic cost control in bridge engineering bidding, characterized in that, Includes the following steps: S1: Collect multi-dimensional data of the entire process of bridge engineering bidding through the data acquisition module. The multi-dimensional data includes bridge-specific basic data, market dynamic data, policy and regulatory data, bidding transaction data, and construction prediction data. S2: The data preprocessing module cleans, deduplicatizes, standardizes, and extracts features from the collected multi-dimensional data to obtain standardized feature data. S3: The standardized feature data is processed by the fusion machine learning model of the risk warning module. The fusion machine learning model uses the random forest algorithm to extract the risk correlation factors of static features and the LSTM network to capture the risk change trend of time-series dynamic data. The output results of the two are combined to complete risk identification and risk level assessment and generate risk warning information. S4: The cost dynamic control module determines the current stage of bridge engineering bidding (pre-bid preparation stage, bid quotation preparation stage, post-bid construction preparation stage) based on the risk level and standardized characteristic data of the risk warning information, matches the corresponding cost control objectives, generates a phased cost control strategy, and outputs control instructions. S5: The linkage execution module receives control instructions, links with external systems to perform cost control operations, and feeds back the execution results to the data storage and update module; S6: The data storage and update module stores the control execution results and updates the standardized feature data based on the new data collected in real time, triggering the risk warning module and the cost dynamic control module to perform iterative calculations, thereby realizing the dynamic cycle of risk warning and cost control.

7. The method according to claim 6, characterized in that, In step S3, the risk level assessment adopts a weighted scoring method, wherein the weights of static characteristic risk-related factors are determined by the analytic hierarchy process, and the weights of time-series dynamic risk change trends are determined by the entropy weight method.

8. The method according to claim 6, characterized in that, In step S4, the phased cost control strategy specifically includes: Pre-bid preparation stage: Based on the forecast of building material price fluctuation trends, formulate a material procurement and reserve plan; During the bid pricing stage: adjust the bid fluctuation ratio according to the risk level, and increase the bid safety margin for high-risk items; Post-bid construction preparation phase: Optimize construction equipment scheduling plan to reduce equipment idle costs.

9. The method according to claim 6, characterized in that, In step S6, the triggering conditions for the iterative operation include: the fluctuation amplitude of the newly collected data exceeds a preset threshold, or the time interval reaches a preset period; if the conditions are met, return to S3; if the conditions are not met, perform manual calibration. When manual calibration is performed, step S7 is executed: the human-computer interaction module receives the user's input instruction to correct the control parameters, performs manual calibration of the cost control strategy, and feeds back the calibrated parameters to the cost dynamic control module to optimize the subsequent control strategy generation logic.

Images