Whole-process engineering consultation management design method

Abstract

The invention provides a whole-process engineering consultation management design method, which comprises the following steps of: constructing a unified data exchange standard and protocol, converting heterogeneous data output by design tools of different professions and different units into a standardized data format, realizing seamless joint and real-time sharing of data through data middleware, eliminating data barriers, and improving the data exchange efficiency. The collaborative design efficiency is improved; according to the method, digital technologies such as BIM are adopted, a digital twin model of an engineering project is constructed in a design stage, a design scheme is deepened through virtual construction simulation and optimization, factors such as a construction process, construction procedures and resource allocation are fully considered, and the construction feasibility and economic rationality of design are improved; a collaborative mechanism of a design stage and a construction stage is established, experienced construction managers and technicians are invited to participate in design review, improvement suggestions and suggestions are provided from the construction perspective for a design scheme, a designer is helped to comprehensively understand construction requirements, and the design scheme is optimized.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and particularly to a design method for the whole-process engineering consulting management. Background Art

[0002] Problem Background:

[0003] In the design method for the whole-process engineering consulting management, there are many technical problems in the collaborative work of multiple specialties and multiple units. First, different specialties and different units use different design tools and data formats, resulting in difficulties in real-time sharing and updating of data, which affects the collaborative efficiency. Second, the work progress of each specialty and each unit is inconsistent, and there are problems in the upstream and downstream connections, making it difficult to achieve seamless collaboration throughout the process. Third, although the design party intervenes earlier, it has insufficient understanding of the work content and work depth in the construction stage and is difficult to fully understand and meet the needs of the construction party. Finally, in the design stage, the resource requirements and construction period plan for the subsequent construction are not considered comprehensively, resulting in frequent design changes in the construction stage, which affects the construction progress and quality. Therefore, it is urgent to study a design method for the whole-process engineering consulting management to break through the data barriers between various specialties and units, realize real-time sharing and updating of data; overall consider the work content and work depth of each stage, coordinate the work rhythms of various specialties and units; strengthen the communication and cooperation between the design party and the construction party, fully understand and meet the needs of the construction party; and at the same time, fully consider the resource requirements and construction period plan for the subsequent construction in the design stage to minimize design changes and ensure the smooth progress of construction. Summary of the Invention

[0004] The present invention provides a design method for the whole-process engineering consulting management, mainly including:

[0005] Construct a unified data exchange standard and protocol to convert heterogeneous data output by design tools of different specialties and different units into a standardized data format, and achieve seamless docking and real-time sharing of data through a data middleware, eliminate data barriers, and improve the efficiency of collaborative design;

[0006] Build a collaborative design management system based on a cloud platform to uniformly manage and schedule the work tasks of various specialties and units, and use visual tools such as Gantt charts and milestones to monitor the work progress of each specialty in real time, discover and solve progress anomaly problems in a timely manner, and coordinate the work rhythms of various specialties and units;

[0007] Adopt digital technologies such as BIM to construct a digital twin model of the engineering project in the design stage, deepen the design scheme through virtual construction simulation and optimization, fully consider factors such as construction technology, construction processes, and resource allocation, and improve the construction feasibility and economic rationality of the design;

[0008] Establish a collaborative mechanism between the design stage and the construction stage, invite experienced construction management personnel and technical personnel to participate in the design review, put forward improvement opinions and suggestions from the construction perspective for the design scheme, help the design party fully understand the construction requirements, and optimize the design scheme;

[0009] Carry out construction schedule planning and resource requirement analysis in the design stage, adopt 4D construction schedule simulation technology, add a time dimension to the 3D model, simulate the construction process, optimize the construction schedule plan, and estimate the required resources based on the construction schedule plan to guide the optimization of the design scheme;

[0010] Establish a design change management mechanism, track and control the whole process of design changes, quantitatively evaluate the impact of design changes on construction progress, construction quality, project cost, etc. through comparative analysis of the design schemes before and after the changes, and optimize the design if necessary to minimize the adverse effects brought by design changes;

[0011] Utilize big data analysis and machine learning algorithms to mine and analyze historical engineering project data, summarize the correlation rules and risk factors between design and construction, form a risk early warning model for the design stage, conduct risk assessment on the design scheme, and take targeted measures in advance to improve the reliability and robustness of the design scheme.

[0012] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0013] The present invention discloses a BIM-based collaborative optimization method for engineering design and construction. The present invention constructs a unified data exchange standard, converts heterogeneous design data into a standard format, and realizes real-time sharing through a cloud platform. Adopt BIM technology to construct an engineering digital twin model, conduct virtual construction simulation and optimization, and improve the construction feasibility of the design. Invite construction personnel to participate in the design review and optimize the design scheme from the construction perspective. Use 4D technology to simulate the construction schedule, analyze resource requirements, and guide design optimization. Establish a design change management mechanism, quantitatively evaluate the impact of changes, and optimize if necessary. Apply big data and machine learning to analyze historical project data, establish a risk early warning model, and improve the reliability of the design scheme. The present invention breaks through the information barrier between the design and construction links, realizes the deep collaboration between design and construction, significantly improves the design quality and construction efficiency, reduces project risks and costs, and lays a foundation for the digital and intelligent management of the whole process of engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flowchart of a whole-process engineering consulting management design method of the present invention.

[0015] Figure 2 It is a schematic diagram of a whole-process engineering consulting management design method of the present invention.

[0016] Figure 3 Another schematic diagram of a whole-process engineering consulting management design method of the present invention. Specific implementation manners

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] Such as Figures 1 - 3 , a whole-process engineering consulting management design method in this embodiment may specifically include:

[0019] S101. Construct a unified data exchange standard and protocol, convert heterogeneous data output by design tools of different specialties and different units into a standardized data format, realize seamless docking and real-time sharing of data through a data middleware, eliminate data barriers, and improve the efficiency of collaborative design.

[0020] For heterogeneous data, analyze its data structure and semantics, extract key elements, map them into a unified data model, construct a standardized data format, eliminate the differences between heterogeneous data, and realize the standardized representation of data. According to the standardized data format, design a data exchange protocol, clarify the rules of data transmission, interface definition, data encoding method, etc., ensure that different systems can accurately understand and parse the data content, and realize seamless docking of data. Develop a data middleware, dock the data interfaces of each design tool, convert heterogeneous data into a standardized format through a data conversion engine, and store it in a unified database to form a consistent data view, providing a basis for data sharing. Adopt a distributed architecture, deploy the data middleware to each node, and realize real-time transmission and synchronization of data through a message queue and a data synchronization mechanism, ensure the consistency of data at each node, and provide real-time data sharing capabilities. For the problem of data barriers, through data access control and permission management, set different data access policies, desensitize sensitive data to ensure data security, and at the same time open necessary data access interfaces to promote data sharing and circulation. Integrate data quality detection and data cleaning functions in the data middleware, verify and correct the accessed heterogeneous data, identify and process missing values, outliers, etc., improve the accuracy and integrity of data, and provide high-quality data support for subsequent data analysis and applications. Establish a data sharing incentive mechanism, give rewards to units that actively participate in data sharing, commend units with excellent data quality, create a good data sharing atmosphere, promote collaborative design among various units, and improve design efficiency and quality.

[0021] Specifically, for the processing of heterogeneous data, it is first necessary to analyze its data structure and semantics. For example, in the field of mechanical design, different CAD software may use different file formats to store 3D model data. By deeply analyzing the structures of these formats, key elements such as geometric information and material properties can be extracted. Mapping these elements into a unified data model, such as the open standard STEP format, can build a standardized data format and eliminate the differences between heterogeneous data. In terms of data exchange protocol design, general data exchange formats such as XML or JSON can be adopted to clearly define data fields, data types, data structures, etc. For example, define an XML schema for part information, including attributes such as part ID, name, material, and dimensions, to ensure that different systems can accurately parse the data content. At the same time, strict data transmission rules are formulated, such as using the HTTPS protocol to ensure transmission security. Developing data middleware is the key to realizing heterogeneous data integration. A general data conversion engine can be designed to support the data formats of mainstream CAD software. When designers upload CAD models, the middleware automatically identifies the file format, calls the corresponding parsing module to extract data, and then converts it into a standard format and stores it in a unified database. In this way, a consistent data view is formed, laying a foundation for subsequent collaborative design. Deploying data middleware using a distributed architecture can improve the scalability and reliability of the system. For example, deploy middleware nodes at the headquarters and each branch respectively, and achieve real-time data synchronization through message queues. When a certain node is offline, other nodes can still work normally, and automatically synchronize data after recovery to ensure the high availability of the system. Regarding the data barrier problem, fine-grained access control policies can be implemented. For example, product data is divided into three levels: public, internal, and confidential, and corresponding access permissions are assigned to different users. For sensitive data, such as specific material formulas, desensitization processing can be carried out, only showing the general category while hiding the detailed components. At the same time, standard API interfaces are provided to allow authorized external systems to securely access the necessary data. In terms of data quality management, data verification and cleaning functions can be integrated into the middleware. For example, perform a rationality check on part size data, and values outside the normal range will be marked as abnormal. Perform an integrity check on material properties, and missing properties will be automatically filled in or prompt the user to supplement. Through these measures, the accuracy and availability of data can be significantly improved. Establishing a data sharing incentive mechanism is an effective means to promote collaborative design. The "Best Data Contribution Award" can be set up to select excellent units according to indicators such as data volume, quality, and sharing frequency. Employees who perform outstandingly in data sharing will be given additional performance bonuses. Through these measures, the enthusiasm of all parties can be mobilized, forming a virtuous data sharing culture, and ultimately improving the overall design efficiency and product quality.

[0022] By constructing a unified data exchange standard and protocol, heterogeneous data is converted into a standardized format, and data middleware technology is adopted to achieve seamless docking and real-time sharing, thereby eliminating data barriers and improving the efficiency of collaborative design.

[0023] According to business requirements and data characteristics, formulate a unified data exchange standard and protocol, clarify data formats, field meanings, encoding methods, etc., to ensure that heterogeneous data can be converted and exchanged according to the standard. Adopt data modeling technology to analyze and abstract heterogeneous data, extract key entities, attributes, and relationships, and build a unified data model as the basis for data conversion and integration. Develop data conversion tools to convert heterogeneous data from the original format to a standardized format according to predefined mapping rules and conversion logics, eliminate data heterogeneity, and achieve unified representation of data. Build a data middleware platform to provide functions such as data transmission, routing, conversion, and filtering, support the docking of multiple data sources and target systems, and achieve seamless connection and data sharing between heterogeneous systems. Adopt a distributed architecture design. By deploying multiple data nodes and service instances, improve the system's concurrent processing ability and scalability to meet the needs of large-scale data exchange and real-time sharing. Establish a data quality monitoring and verification mechanism to check the integrity, accuracy, consistency, etc. of the accessed data, promptly discover and handle data quality problems, and ensure the reliability of the data. Based on the standardized data format and unified data interface, realize data sharing and collaboration between different systems and departments, break through data barriers, and improve the efficiency of business processes and the accuracy of decision-making.

[0024] Specifically, formulating a unified data exchange standard and protocol is the basis for realizing heterogeneous data integration. Taking the Product Lifecycle Management (PLM) system as an example, a set of standardized product data models can be defined, including product structure, part information, process parameters, etc. Clarify the meaning, data type, and value range of each field. For example, use a unified coding rule for part numbers and standardized units and representation methods for material attributes. This can ensure that data from different CAD software and ERP systems can be converted and exchanged according to a unified standard. Data modeling is a key step in building a unified data model. Taking the automotive manufacturing industry as an example, by analyzing business processes such as vehicle design, parts management, and production planning, key entities such as "vehicle model", "part", "supplier", etc. can be extracted, and their attributes and relationships can be defined. For example,

[0025] The "part" entity may contain attributes such as ID, name, material, weight, etc., and has a "supply" relationship with the "supplier" entity. In this way, a unified data model covering the entire automotive production process can be constructed, laying the foundation for subsequent data conversion and integration. Developing data conversion tools is the core of achieving heterogeneous data standardization. Taking the field of mechanical design as an example, a general CAD data converter can be developed. When designers upload a 3D model in CATIA format, the converter first parses the file structure and extracts key data such as geometric information and assembly relationships. Then, according to predefined mapping rules, this data is converted into the standard STEP format. In this process, operations such as unit conversion and coordinate system conversion may be required to ensure the consistency and accuracy of the converted data. Building a data middleware platform is the key to achieving seamless connection of heterogeneous systems. Taking the production management of manufacturing enterprises as an example, an integrated data exchange platform can be constructed. This platform connects systems such as ERP, MES, and WMS through standard interfaces to achieve real-time synchronization of order information, production plans, and inventory data. When the sales department enters a new order in ERP, the middleware automatically forwards the order information to the MES system, triggering the formulation of a production plan. At the same time, inventory data is synchronized from WMS to ERP in real time to ensure that order processing is based on the latest inventory status. This centralized data exchange mechanism greatly improves the efficiency of cross-system collaboration. Adopting a distributed architecture design can improve the performance and reliability of the system. Taking the global design collaboration of multinational enterprises as an example, multiple data nodes can be deployed in different regions. When the US design team updates the product model, the data is first synchronized to the local node and then propagated to the nodes in other regions through a data replication mechanism. This method not only reduces the latency of cross-ocean data transmission but also improves the fault tolerance of the system. Even if a node in a certain region fails, other regions can still work normally, ensuring the continuity of global design activities. Establishing a data quality monitoring and verification mechanism is an important means to ensure data reliability. Taking the quality management of pharmaceutical production as an example, multiple verifications can be set at the data access link. Integrity checks are performed on raw material batch data to ensure that key information such as production date and expiration date is not missing. Reasonableness verification is performed on production parameters such as temperature and pressure, and data outside the preset range will trigger an alarm. Through these measures, data anomalies can be detected and corrected in a timely manner to ensure the accuracy and traceability of data throughout the pharmaceutical production process. Based on standardized data formats and unified interfaces, cross-departmental and cross-system data sharing and business collaboration can be achieved. Taking the new product R & D process as an example, when the design department completes the preliminary design, the 3D model data is automatically pushed to the process department through a standard interface. Process engineers carry out process planning based on this data and feedback the results to the design department. At the same time, the procurement department can access the material information in the design data to conduct supplier inquiries in advance. This seamless data flow greatly shortens the product development cycle and improves the collaboration efficiency of each link.

[0026] S102. Build a collaborative design management system based on the cloud platform to uniformly manage and schedule the work tasks of each specialty and each unit. Real-time monitor the work progress of each specialty through visualization tools such as Gantt charts and milestones, promptly discover and solve progress anomaly problems, and coordinate the work rhythms of each specialty and each unit.

[0027] According to the characteristics of the cloud platform, build a Web-based collaborative design management system, input the work task information of each specialty and unit into the system to form a unified task management library. Through the analysis and processing of the work tasks in the task management library, automatically generate visualization progress monitoring tools such as Gantt charts and milestones to achieve real-time monitoring of the work progress of each specialty. Use machine learning algorithms to analyze the work progress data of each specialty, identify progress anomaly problems through anomaly detection algorithms, and automatically generate warning messages according to preset rules. Push the progress anomaly warning messages to the persons in charge of relevant specialties and units, and at the same time enter the anomaly problems into the problem management module, and automatically generate suggestions for problem solutions according to the severity and impact scope of the problems. In the problem management module, use case-based reasoning algorithms to analyze historical problem solutions, match the most similar solution to the current problem, and optimize the solution according to the characteristics of the current problem to generate the final problem solution. According to the work progress and problem-solving situations of each specialty, dynamically adjust the work tasks of each specialty and unit, reallocate tasks through task scheduling algorithms, balance the work rhythms of each specialty, and ensure that the overall project progress is not affected. The system automatically generates coordination suggestions for each specialty and unit, and promotes communication and collaboration between each specialty and unit through methods such as message pushing and task reminders, forming an efficient collaborative design management closed loop.

[0028] Specifically, the web-based collaborative design management system provides a unified platform for multi-disciplinary collaboration. The core of the system is the task management library, which centrally stores the work task information of each discipline and unit. For example, in a large-scale engineering project, disciplines such as structure, electrical, and HVAC can all enter their respective design tasks into the system. The task information includes task name, responsible person, planned start and end times, required resources, etc. The system automatically generates visual progress monitoring tools based on the task management library. Taking the Gantt chart as an example, the horizontal axis represents time, and the vertical axis lists each task. The duration and progress of the task are intuitively displayed through the length of the horizontal bar. Milestones mark key nodes, such as the determination of the design plan and the completion of construction drawings. These tools enable the project manager to have a clear understanding of the overall progress at a glance. Machine learning algorithms play an important role in progress monitoring. The system can adopt anomaly detection algorithms, such as IsolationForest or Local Outlier Factor (LOF), to compare the actual progress and planned progress of each task. When it is detected that the progress of a certain task deviates from the expectation by more than the threshold, the system will automatically generate a warning message. For example, if a sub-task of electrical design is delayed by 20% compared to the plan, the system will immediately send a warning notice to the person in charge of the electrical discipline and the project manager. The problem management module is another key component of the system. It not only records problems such as progress anomalies but also provides solution suggestions using case-based reasoning algorithms. Suppose there is a problem of design change due to delayed material supply. The system will retrieve the historical case library to find the handling methods in similar situations. Through similarity calculations, such as cosine similarity or Jaccard coefficient, the system can match the most similar cases and optimize them in combination with the current specific situation to generate practical solutions. Task scheduling is the key to ensuring the overall project progress. The system adopts dynamic task scheduling algorithms, such as priority-based scheduling or resource-constrained scheduling, to reallocate tasks according to the real-time progress. For example, when it is found that the structure design is completed ahead of schedule, the system will automatically advance the subsequent dependent tasks and adjust the task times of other disciplines accordingly to make full use of time resources. The closed-loop of collaborative design management is achieved through the coordination suggestions generated by the system and the automated communication mechanism. The system analyzes the work status and progress of each discipline and generates targeted coordination suggestions. For example, when it is found that there is a potential conflict between HVAC design and architectural design, the system will automatically arrange a coordination meeting for relevant personnel. Through message push and task reminders, it ensures that all parties can obtain information in a timely manner and take actions. This cloud-based collaborative design management system greatly improves the efficiency of multi-disciplinary collaboration by integrating functions such as task management, progress monitoring, problem-solving, and task scheduling. It not only reduces human errors and communication costs but also provides strong support for decision-making through data analysis and intelligent algorithms, ultimately achieving the optimization of the design process and the improvement of project quality.

[0029] S103. Use digital technologies such as BIM to construct a digital twin model of the engineering project during the design stage. Through virtual construction simulation and optimization, deepen the design scheme, fully consider factors such as construction techniques, construction processes, and resource allocation, and improve the construction feasibility and economic rationality of the design.

[0030] Obtain relevant data such as the design drawings, construction requirements, and site environment of the engineering project, and construct a three-dimensional digital model of the engineering project in the BIM platform to form a digital twin model. In the digital twin model, according to the requirements of construction techniques and construction processes, simulate different construction schemes. Through virtual construction technology, conduct simulation for each construction link to obtain data such as the construction progress, cost, and quality of each construction scheme. Use machine learning algorithms to analyze the simulation data of the construction schemes, classify different construction schemes through clustering algorithms, and rank the advantages and disadvantages of each category of construction schemes according to factors such as construction requirements and site environment. For the construction schemes ranked at the top, use optimization algorithms to further optimize them. By adjusting parameters such as construction techniques, construction processes, and resource allocation, obtain more economical, reasonable, and feasible construction schemes. Compare the optimized construction schemes with the original design scheme, analyze the improvements in aspects such as construction feasibility and economic rationality, and modify and improve the design scheme according to the analysis results. In the digital twin model, according to the optimized design scheme, conduct virtual construction simulation again to verify its feasibility and economy, ensure that the design scheme can meet the construction requirements, and achieve the expected goals. Output the optimized design scheme and construction scheme to form complete construction drawings and construction scheme documents, providing guidance and support for subsequent construction management and on-site construction, and ensuring the smooth implementation and efficient management of the engineering project.

[0031] Specifically, the digital twin model is a virtual representation of an engineering project. By integrating data such as design drawings, construction requirements, and on-site environment, a three-dimensional digital model is constructed on the BIM platform. Taking a bridge project as an example, details such as bridge piers, bridge decks, and steel structures can be accurately modeled to restore the real construction environment. This digital twin not only visualizes the engineering structure but also simulates the construction process. In virtual construction, different construction plans can be simulated. For example, in bridge construction, two plans, the traditional in-situ casting method and the precast and assembled method, can be simulated. Through simulation, key data such as the construction period, required human and material resources, and quality control difficulties of each plan can be obtained. These data provide a basis for subsequent analysis and decision-making. Machine learning algorithms play an important role in analyzing construction plans. Taking the clustering algorithm as an example, construction plans can be classified according to dimensions such as construction period length, cost level, and quality level. For example, bridge construction plans can be divided into fast and low-cost types, medium and balanced types, and high-quality and high-cost types. This classification helps project managers quickly understand the characteristics of each plan. Optimization algorithms are used to further improve construction plans. For example, for the precast and assembled method, by adjusting parameters such as the size of precast components, transportation routes, and hoisting sequences, the construction period can be shortened or the cost can be reduced while ensuring quality. The optimized plan may shorten the construction period by 10% and reduce the cost by 5% compared with the original plan. Comparing the optimized construction plan with the original design may reveal some room for improvement in the design. For example, through virtual construction, it is found that the construction of some structural joints is difficult, and the design team can be advised to adjust the connection method to improve construction efficiency. This feedback mechanism ensures the coordination between design and construction. Finally, final virtual construction verification is carried out in the digital twin model. For example, the construction process of the entire bridge, including foundation treatment, pier pouring, and beam installation, can be simulated. Through this comprehensive simulation, potential construction conflicts or safety hazards can be discovered, providing important references for actual construction. The optimized design and construction plans will form detailed construction drawings and plan documents. These documents not only include traditional two-dimensional drawings but may also include multimedia materials such as three-dimensional model files and construction animations. These rich materials can help the construction team better understand and execute the plan, improving construction quality and efficiency. Through this series of digital means, the design and construction processes of engineering projects are comprehensively optimized. The digital twin technology not only improves the feasibility and economy of the plan but also provides a powerful tool for project management, effectively reducing engineering risks and enhancing the overall project quality.

[0032] S104. Establish a collaborative mechanism between the design stage and the construction stage, invite experienced construction management personnel and technical personnel to participate in the design review, and put forward improvement opinions and suggestions from the construction perspective for the design plan to help the design party fully understand the construction requirements and optimize the design plan.

[0033] According to the design plan in the design phase, natural language processing technology is adopted to extract key information from the design documents, including design objectives, design focuses, technical parameters, etc., to form a structured design information table. Through knowledge graph technology, the structured design information table is associated and matched with a pre-constructed construction knowledge base to identify key nodes and risk points related to construction in the design plan. For the identified key nodes and risk points, a rule-based inference engine is used, combined with the experience rules in the construction knowledge base, to automatically generate targeted construction suggestions and optimization plans. The generated construction suggestions and optimization plans are compared and analyzed with the original design plan, a text similarity algorithm is used to calculate the similarity between the two, and the feasibility of the optimization plan is judged according to the set threshold. For the optimization plan with a similarity lower than the threshold, machine learning algorithms such as decision trees or support vector machines are used, combined with historical construction data to further optimize and adjust the plan to improve the feasibility of the plan. The optimized construction suggestions and optimization plans are fed back to the designers, and through the automatically generated visual report, the comparison effect before and after optimization is displayed to help the designers intuitively understand the value of optimization. A collaborative platform for design and construction is established, the optimized design plan is associated with the construction plan, and information sharing and real-time update between design and construction are realized to ensure that the construction process is consistent with the optimized design plan.

[0034] Specifically, according to the design plan in the design stage, natural language processing technology is used to extract key information from the design document, including design objectives, design highlights, technical parameters, etc., to form a structured design information table. For example, in the design document of a high-rise building, natural language processing technology can identify key information such as "building height of 120 meters", "seismic intensity of 8 degrees", "glass curtain wall for the exterior wall", etc., and convert it into a structured data format for subsequent processing. Through knowledge graph technology, the structured design information table is associated and matched with a pre-constructed construction knowledge base to identify key nodes and risk points related to construction in the design plan. Suppose the construction knowledge base contains a large amount of empirical data on high-rise building construction. The knowledge graph can associate "building height of 120 meters" with "high-rise building construction risks", identify key nodes in construction, such as "foundation construction", "main structure construction", etc., and point out potential risk points, such as "deep foundation pit support", "safety of working at heights", etc. For the identified key nodes and risk points, a rule-based inference engine is used, combined with the empirical rules in the construction knowledge base, to automatically generate targeted construction suggestions and optimization plans. For example, based on the risk point of "deep foundation pit support", the inference engine combines the rule in the knowledge base "segmented excavation and support measures are required for deep foundation pit construction" to generate specific construction suggestions: "Excavate in segments, with each segment not exceeding 5 meters in depth, and carry out support in a timely manner". The generated construction suggestions and optimization plans are compared and analyzed with the original design plan. A text similarity algorithm is used to calculate the similarity between the two, and the feasibility of the optimization plan is judged according to the set threshold. Suppose the similarity threshold is 0.7. If the calculation result shows that the similarity between the optimization plan and the original design plan is 0.6, it is considered that the optimization plan has a large difference from the original plan and needs further adjustment. For optimization plans with similarity lower than the threshold, machine learning algorithms, such as decision trees or support vector machines, are used, combined with historical construction data to further optimize and adjust the plan. For example, by analyzing historical data, it is found that in similar geological conditions, the plan of "excavating in segments, with each segment 3 meters in depth" is more reliable. The machine learning algorithm adjusts the optimization plan accordingly to improve its feasibility. The optimized construction suggestions and optimization plans are fed back to the designers, and through an automatically generated visual report, the comparison effect before and after optimization is displayed to help the designers intuitively understand the value of optimization. The visual report can include a comparison chart of the construction progress before and after optimization, a cost change curve, etc., enabling the designers to clearly see the specific effects of the optimization plan in shortening the construction period and reducing costs. A collaborative platform for design and construction is established to associate the optimized design plan with the construction plan, realizing information sharing and real-time update between design and construction, and ensuring that the construction process is consistent with the optimized design plan. For example, the collaborative platform can display the construction progress in real time and compare it with the expected progress of the design plan. If a deviation is found, the construction plan is adjusted in a timely manner to ensure the smooth progress of the project according to the design plan.Through the above steps, the comprehensive application of multiple technical means such as natural language processing technology, knowledge graph, inference engine, and machine learning algorithms not only improves the construction feasibility and economic rationality of the design scheme, but also realizes the efficient coordination between design and construction, ensuring the high-quality completion of engineering projects. This method of multi-technology integration effectively solves the problem of the disconnection between traditional design and construction, and improves the overall efficiency and management level of engineering projects.

[0035] S105. Carry out the analysis of construction schedule plan and resource requirements in the design stage, adopt the 4D construction schedule simulation technology, add the time dimension to the 3D model, simulate the construction process, optimize the construction schedule plan, and estimate the required resources based on the construction schedule plan to guide the optimization of the design scheme.

[0036] According to the design scheme, construct a 3D model, associate the time information of each construction stage with the 3D model to form a 4D construction schedule model. Through the 4D construction schedule model, simulate the construction process under different construction schemes, and obtain time parameters such as the start and end times and durations of each construction stage. For the simulated construction schedule data, adopt optimization algorithms such as the critical path method and linear programming to optimize and adjust the construction schedule plan, shorten the construction period, and improve construction efficiency. According to the optimized construction schedule plan, combined with factors such as construction technology and construction intensity, estimate the quantities of resources such as labor, materials, and equipment required for each construction stage. Associate the resource requirement estimation results with the 3D model, and display the resource requirement distribution of each construction stage in a visual way to identify the peaks and bottlenecks of resource requirements. For problems such as uneven resource requirement distribution and resource conflicts, adjust and optimize the construction schedule plan and resource allocation scheme to achieve the balanced utilization and reasonable allocation of resources. Feed back the optimized construction schedule plan and resource allocation scheme to the designers to guide the optimization and improvement of the design scheme and improve the construction feasibility and economic rationality of the design.

[0037] Specifically, according to the design plan, a 3D model is constructed, and the time information of each construction stage is associated with the 3D model to form a 4D construction schedule model. For example, in the design plan of a high-rise building project, it includes multiple construction stages such as the basement, main structure, and exterior curtain wall. Using BIM (Building Information Modeling) technology, a detailed 3D model is constructed, and the time information of each stage is embedded in the model. For example, the basement construction is expected to take 90 days, and the main structure construction takes 180 days, etc. In this way, the 4D model can not only display the 3D form of the building but also dynamically show the construction progress of each stage. Through the 4D construction schedule model, the construction process under different construction plans is simulated to obtain time parameters such as the start and end times and durations of each construction stage. Suppose there are two construction plans for the project: Plan One uses traditional sequential construction, and Plan Two uses segmented parallel construction. Through 4D model simulation, it is found that the total construction period of Plan One is 320 days, while for Plan Two, due to parallel operations, the total construction period is shortened to 280 days. During the simulation process, the start and end times and durations of each stage can be accurately obtained. For example, the main structure construction starts on the 91st day and ends on the 270th day in Plan One, while in Plan Two, it starts on the 61st day and ends on the 240th day. For the construction schedule data obtained from the simulation, optimization algorithms such as the critical path method and linear programming are used to optimize and adjust the construction schedule plan. For example, through the analysis of the critical path method, it is found that the exterior curtain wall construction is the critical path affecting the total construction period. So, the resource allocation is adjusted, the number of curtain wall construction teams is increased, and the curtain wall construction time is shortened from 60 days to 50 days. Linear programming is used to optimize the resource allocation to ensure the maximization of resource utilization while shortening the construction period. According to the optimized construction schedule plan, considering factors such as construction technology and construction intensity, the quantity of resources such as labor, materials, and equipment required for each construction stage is estimated. Taking the main structure construction as an example, two additional construction teams are required after optimization, 200 cubic meters of concrete, 30 tons of steel bars, and two tower crane equipment are needed per day. Through detailed estimation, the resource requirements for each stage are made clear to avoid resource shortages or waste. The resource requirement estimation results are associated with the 3D model, and the resource requirement distribution of each construction stage is displayed in a visual way. For example, using BIM software, the labor, material, and equipment requirements are marked in different colors in the model. The basement construction stage shows high demand in red, the main structure construction stage shows medium demand in yellow, and the exterior curtain wall construction stage shows low demand in green. In this way, project managers can intuitively identify the peaks and bottlenecks of resource requirements. For example, it is found that during the period from the 120th day to the 150th day, the demand for concrete surges, which may become a construction bottleneck. For problems such as uneven resource demand distribution and resource conflicts, the construction schedule plan and resource allocation plan are adjusted and optimized. Suppose during the period from the 120th day to the 150th day, the supply of concrete is insufficient. By adjusting the plan, some non-critical path construction tasks are advanced or postponed to stagger the use of resources and ensure the balanced utilization of resources. For example, some interior decoration work is advanced to start on the 100th day to reduce the concurrent demand for concrete.Feedback the optimized construction schedule plan and resource allocation plan to the designers to guide the optimization and improvement of the design plan. For example, after learning that two more teams are needed for the main structure construction, the designers optimize the design plan, adjust the dimensions and materials of some structural components to improve the construction efficiency. In this way, the design plan not only meets the design requirements but also has high construction feasibility and economic rationality. Building a 4D construction schedule model and optimizing resource allocation not only improve the construction efficiency and shorten the construction period but also enable all parties of the project to have a clearer understanding of the construction process through visualization means, reducing communication costs and construction risks. This collaborative mechanism ensures the close connection between design and construction and improves the overall management level of the project.

[0038] S106. Establish a design change management mechanism to track and control the whole process of design changes. By comparing and analyzing the design plans before and after the changes, quantitatively evaluate the impacts of design changes on construction schedule, construction quality, project cost, etc. When necessary, conduct design optimization to minimize the adverse impacts brought by design changes.

[0039] Obtain the design change application, extract key information such as the change content and reasons, and judge whether the change meets the specification requirements. If it does not meet the requirements, reject the application; if it meets the requirements, initiate the change evaluation process. According to the change content, obtain the design plans before and after the change, and use technologies such as BIM to conduct three-dimensional visual comparison and analysis of the design plans to identify the changed parts and the scope of influence. Through big data analysis of the construction schedule, quality, and cost data of historical similar projects, combined with the expert knowledge base, establish a change impact evaluation model to quantitatively analyze the impacts of design changes on the construction period, quality, and cost. If the change impact exceeds the preset threshold, trigger the design optimization process. Adopt algorithms such as parametric design and topology optimization to optimize the design plan on the premise of meeting the function and performance requirements to reduce the change impact. Compare the optimized design plan with the original plan, calculate the optimization effect, and form a design change evaluation report as the basis for change approval. If the change plan is approved, synchronize the change content to relevant documents such as construction drawings, schedule plans, material plans, and cost budgets to ensure coordinated updates of all links. Continuously track the implementation of the change during the construction process, collect on-site data through an intelligent information system, compare and analyze the actual schedule, quality, and cost differences before and after the change, dynamically evaluate the change effect, and timely warn of deviation risks to ensure the achievement of the change goal.

[0040] Specifically, in the process of design change management, it is first necessary to obtain a design change application. For example, in a certain engineering project, due to changes in geological conditions, the foundation design needs to be adjusted. The designer submits a change application, which details the reasons, content, and expected effects of the change. The change content involves the reinforcement of the foundation structure and the replacement of materials. After extracting the key change information, it is necessary to determine whether it meets the specification requirements. Suppose the change application mentions using a new type of high-strength steel to replace the original material. It is necessary to consult relevant specifications to confirm whether the new material meets the requirements of safety and durability. If it does not meet the requirements, the application will be rejected and the reasons will be explained. If it meets the requirements, the change evaluation process will be initiated. Next, use BIM technology to conduct a three-dimensional visual comparison and analysis of the design schemes before and after the change. Through the BIM model, the changes in the foundation structure can be visually seen, and the changed parts and their influence ranges can be identified. For example, after the change, the foundation depth increases, which may affect the layout of surrounding underground pipelines. To quantitatively analyze the change impact, use big data to analyze the construction progress, quality, and cost data of historical similar projects. Suppose it is found through analysis that similar changes have increased the construction period by an average of 10% and the cost by 15% in historical projects. Combining with the expert knowledge base, establish a change impact evaluation model to predict the specific impact of this change on the construction period and cost. If the change impact exceeds the preset threshold, such as the construction period extension exceeding 5% of the total project construction period, the design optimization process will be triggered. Use parametric design and topology optimization algorithms to optimize the design scheme on the premise of meeting the functional performance requirements. For example, adjust the foundation structure parameters through parametric design to reduce the material usage, and use the topology optimization algorithm to optimize the structure layout to reduce the construction difficulty. Compare the optimized design scheme with the original scheme and calculate the optimization effect. Suppose the construction period is shortened by 5% and the cost is reduced by 8% after optimization. Form a design change evaluation report as the basis for change approval. If the change scheme is approved, synchronize the change content to relevant documents such as construction drawings, progress plans, material plans, and cost budgets. For example, update the foundation structure design in the construction drawings, adjust the key nodes in the progress plan, and recompile the material plan and cost budget. During the construction process, continuously track the implementation of the change. Collect on-site data through an intelligent information system and compare and analyze the actual progress, quality, and cost differences before and after the change. For example, use Internet of Things sensors to monitor the foundation construction progress in real time, compare the construction efficiency and quality indicators before and after the change through a data analysis platform, and discover deviations and give early warnings in a timely manner to ensure the achievement of the change goal. Through the above steps, ensure the scientificity and systematicness of design change management, improve the efficiency and effectiveness of project management, and reduce the risks and costs brought by changes.

[0041] By obtaining the design schemes before and after the design change, using comparative analysis techniques, judging the impact degree of the design change on aspects such as construction progress, construction quality, and project cost, and determining whether design optimization is required according to the quantitative evaluation results, an optimized design scheme is obtained.

[0042] Obtain the design drawing before the change, extract the drawing information through optical character recognition technology, apply the pre-established three-dimensional model library to generate the three-dimensional model before the change, automatically calculate the bill of quantities before the change based on the three-dimensional model before the change, synchronously capture and interpret the construction log file, determine the qualified rate before the change, the rework rate before the change, the material cost before the change, the labor cost before the change, and the machinery cost before the change, and then obtain the total cost before the change by applying the regression analysis algorithm. Obtain the design drawing after the change, extract the drawing information through optical character recognition technology, apply the pre-established three-dimensional model library to generate the three-dimensional model after the change, automatically calculate the bill of quantities after the change based on the three-dimensional model after the change, and at the same time retrieve data of similar projects from the central database, and obtain the qualified rate after the change, the rework rate after the change, the material cost after the change, the labor cost after the change, and the machinery cost after the change through the random forest algorithm, and calculate the total cost after the change. Compare the bill of quantities before the change and the bill of quantities after the change to obtain the bill of quantities difference data, and compare the total cost data before the change and the total cost data after the change to obtain the total cost difference data. Obtain the graphic change data based on the three-dimensional models before and after the change. If there are cost increase items in the bill of quantities difference data and the sum of the cost increase items is greater than the preset cost threshold, then activate the early warning information push module to give construction early warning prompts. At the same time, if the total cost after the change is higher than the total cost before the change and the difference is greater than the preset price difference threshold, it is pushed to the manual analysis process. Compare the quality change data based on the graphic change data, combined with the qualified rate before the change and the qualified rate after the change. According to the result of the quality change data comparison, if the qualified rate after the change is lower than the qualified rate before the change, then calculate the decline rate of the qualified rate of all components with design changes in the design scheme after the change, and the decline rate of the qualified rate = (qualified rate before the change - qualified rate after the change) / qualified rate before the change * 100%. Generate a list of modification suggestions based on the pre-established expert database. Judge whether there are design changes with a dimension modification exceeding 20% in the graphic change data. If the proportion of dimension modification exceeds the preset dimension ratio threshold, then automatically traverse multiple modification schemes through computer-aided design software, and use the comparative analysis technology to compare different design change schemes. Select a scheme with the lowest cost and the dimension change closest to the expert modification opinion from multiple modification schemes. Determine the design change optimization suggestion through comprehensive cost calculation, combined with the qualified rate after the change, the rework rate after the change, and the expert modification opinion, and use the pre-trained support vector machine algorithm to judge whether the design scheme reaches the deliverable state. If it is deliverable, then output the results. If it cannot be delivered, then return to the modification suggestion list interface.

[0043] Specifically, obtain the design drawing before the change, extract the drawing information through optical character recognition (OCR) technology, and apply the pre-established three-dimensional model library to generate the three-dimensional model before the change. For example, in a certain construction project, the design drawing contains structural information such as walls, beams, and columns. The OCR technology identifies parameters such as the dimensions and materials of each component, and automatically generates a three-dimensional model in combination with the model library to visually display the building structure. According to the three-dimensional model before the change, automatically calculate the bill of quantities before the change, synchronously capture and interpret the construction log file, and determine the qualified rate before the change, the rework rate before the change, the material cost before the change, the labor cost before the change, and the machinery cost before the change. Assume that the qualified rate before the change is 95%, the rework rate is 3%, the material cost is 5 million yuan, the labor cost is 3 million yuan, and the machinery cost is 2 million yuan. Using the regression analysis algorithm, the total cost before the change is approximately 10 million yuan. Obtain the design drawing after the change, extract the information through OCR technology in the same way, and generate the three-dimensional model after the change. Assume that the wall thickness increases and the beam-column structure is adjusted after the change, and the model library is automatically updated to generate a new three-dimensional model. According to the three-dimensional model after the change, automatically calculate the bill of quantities after the change. At the same time, retrieve data of similar projects from the central database. Through the random forest algorithm, the qualified rate after the change is 90%, the rework rate is 5%, the material cost is 5.5 million yuan, the labor cost is 3.2 million yuan, and the machinery cost is 2.1 million yuan. Calculate that the total cost after the change is approximately 10.8 million yuan. Compare the bills of quantities before and after the change, and find that the usage of wall materials increases by 50 cubic meters, and the usage of beam-column materials decreases by 30 cubic meters, obtaining the difference data of the bill of quantities. Compare the total cost of 10 million yuan before the change and the total cost of 10.8 million yuan after the change, obtaining the difference data of the total cost of 800,000 yuan. According to the three-dimensional models before and after the change, obtain the graphic change data, such as the wall thickness increases from 300 mm to 350 mm. If there are cost increase items in the difference data of the bill of quantities, and the sum of the cost increase items is greater than the preset cost threshold of 500,000 yuan, then activate the warning information push module to give a construction warning prompt to remind the project management team to pay attention to cost control. At the same time, if the total cost after the change is higher than the total cost before the change, and the difference is greater than the preset price difference threshold of 500,000 yuan, then push it to the manual analysis link, and the experts will further evaluate the rationality of the change. According to the graphic change data, combined with the qualified rate of 95% before the change and the qualified rate of 90% after the change, compare the quality change data. If the qualified rate after the change is lower than the qualified rate before the change, then calculate the decline rate of the qualified rate. The decline rate of the qualified rate = (95% - 90%) / 95% × 100% = 5.26%. According to the pre-established expert database, generate a list of modification suggestions, such as suggesting to optimize the wall construction process to improve the qualified rate. Judge whether there are design changes with a dimension modification exceeding 20% in the graphic change data. Assume that the wall thickness increases by 16.7% (from 300 mm to 350 mm), exceeding the preset dimension ratio threshold of 20%. Then, automatically traverse multiple modification schemes through computer-aided design software, and use the comparative analysis technology to compare different design change schemes.Select a solution with the lowest cost and the smallest dimensional change closest to the expert's modification opinion from multiple modification options. For example, use a new lightweight material to replace the traditional material, which not only meets the design requirements but also reduces costs. Through comprehensive cost calculation, combined with a 90% qualified rate after the change, a 5% rework rate after the change, and the expert's modification opinion, determine the optimization suggestions for design changes. Use the pre-trained support vector machine algorithm to judge whether the design solution reaches the deliverable state. If it is deliverable, output the results, such as generating a detailed change report and an optimization plan; if it cannot be delivered, return to the modification opinion list interface to re-evaluate and optimize the design solution. This can not only effectively control costs and improve construction quality but also enhance the efficiency and accuracy of design change management through intelligent means, ensuring the smooth progress of the project as planned. Through 3D models and data analysis, the project management team can intuitively understand the impact of changes, adjust the construction strategy in a timely manner, and avoid construction delays and cost overruns caused by design changes.

[0044] S107. Use big data analysis and machine learning algorithms to mine and analyze historical engineering project data, summarize the correlation rules and risk factors between design and construction, form a risk early warning model in the design stage, conduct risk assessment on the design solution, and take targeted measures in advance to improve the reliability and robustness of the design solution.

[0045] Collect multi-dimensional data of historical engineering projects, including engineering data, project type, construction unit, design unit, design parameters, number of changes, risk category, material type, and service life. Establish a structured historical engineering database through data association, and determine a unique number for each project data. According to the risk categories contained in the completed engineering projects in the database, determine the completed engineering risk results corresponding to each unique number in the actual engineering project, and count the number of different risk categories according to the risk results, sort the numbers from large to small, take out the top ten risk categories as specific risk categories, and determine the specific risk category set. According to each unique number in the database, construct the design parameter vector of each design scheme. The elements of the design parameter vector are composed of the characteristic values ​​of risk category, project type, construction unit, design unit, design parameter, number of changes, and material type. Use the K-means clustering algorithm to divide the design parameter vector into different specific risk categories to determine the risk category of each design scheme. The risk level prediction model is obtained by training the structured historical engineering database through the LightGBM machine learning algorithm. The design scheme classified in the previous step, including the design parameters, project type, construction unit, design unit, number of changes, and material type feature values, is input into the risk level prediction model to determine the risk level prediction value. The XGBoost machine learning algorithm is used to train the design scheme robustness prediction model to determine the design parameters of each design scheme in the structured historical engineering database, including length, width, and height. Each design parameter is multiplied by the perturbation coefficient to obtain each perturbation design parameter. If each perturbation design parameter is input into the risk level prediction model, the more perturbation parameter groups below the preset threshold in the output result, the lower the design robustness of this unique numbered scheme. The risk category of the design scheme, the risk level prediction value, and the judgment result of the robustness are comprehensively analyzed to obtain the risk category of the design scheme. By comparing the current risk category with the specific risk category set, it is determined whether it belongs to the specific risk category. If it belongs to the specific risk category and the robustness is low, the unique number is included in the key focus list. By focusing on each unique number in the list, historical corresponding project information can be obtained from the database, including the completion drawings and change notices of the corresponding projects. The design parameters with the most changes in similar projects can be obtained as key parameters, and adjustment suggestions can be made in advance.

[0046] Specifically, the establishment of a historical engineering database is the foundation of risk management. For example, a large bridge project may contain the following data: project number B001, project type is suspension bridge, construction unit is Company A, design unit is Institute B, main span length is 1000 meters, tower height is 200 meters, number of changes is 5 times, risk category is structural safety, material type is high-strength steel, design service life is 100 years, etc. Through the association of these multi-dimensional data, the project characteristics can be comprehensively reflected. The determination of risk categories is crucial for preventing risks in similar projects. Suppose in the database, the structural safety risk appears 100 times, the construction quality risk appears 80 times, the material defect risk appears 60 times, etc. Selecting the top ten as the set of specific risk categories helps to focus on high-incidence risks. The risk classification of design schemes uses the K-means clustering algorithm. Taking the bridge project as an example, parameters such as main span length, tower height, and steel strength can be used as vector elements, and the design schemes can be divided into different risk categories through clustering, such as structural safety category, construction quality category, etc. The establishment of a risk level prediction model uses the LightGBM algorithm. Inputting the feature vector of a certain bridge design scheme, such as main span 1200 meters, tower height 220 meters, high-strength steel, 3 changes, etc., the model can output the predicted risk level value, such as medium risk. The prediction of the robustness of design schemes uses the XGBoost algorithm. For example, increasing the main span length of the bridge by 5% and decreasing the tower height by 3% are used as perturbation parameters. If the risk level remains stable under multiple groups of perturbation parameters, it indicates that the design scheme has a high level of robustness. After comprehensive analysis, if a certain bridge design scheme belongs to the structural safety risk category (one of the specific risk categories), the predicted risk level is relatively high, and the robustness is relatively low, then it will be included in the key attention list. By analyzing the historical data of such projects, it may be found that the ratio of main span length to tower height is a key parameter, and based on this, optimization suggestions can be put forward, such as adjusting the tower height to improve structural stability. This risk management method based on historical data and machine learning can effectively identify potential risks and improve the reliability and safety of design schemes. By mining and analyzing a large amount of historical data, hidden risk patterns and key influencing factors can be discovered, providing strong support for the risk prevention and control of new projects. At the same time, the application of machine learning models makes risk assessment more objective and accurate, can quickly process complex multi-dimensional data, and improve the efficiency and accuracy of risk management.

[0047] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these changes and modifications.

Claims

1. A full-process engineering consulting management design method, characterized in that: The method comprises: Build unified data exchange standards and protocols, convert heterogeneous data output by design tools of different disciplines and units into standardized data formats, realize seamless connection and real-time sharing of data through data middleware, eliminate data barriers, and improve collaborative design efficiency; Build a collaborative design management system based on the cloud platform to uniformly manage and schedule the work tasks of various disciplines and units, monitor the work progress of various disciplines in real time through visual tools such as Gantt charts and milestones, promptly discover and solve abnormal progress problems, and coordinate the work rhythm of various disciplines and units; Adopt digital technologies such as BIM to build a digital twin model of the project during the design phase. Through virtual construction simulation and optimization, deepen the design plan, fully consider factors such as construction technology, construction process, resource allocation, etc., and improve the construction feasibility and economic rationality of the design. Establish a coordination mechanism between the design and construction stages, invite experienced construction management personnel and technical personnel to participate in the design review, and put forward improvement opinions and suggestions for the design plan from the construction perspective, so as to help the design party fully understand the construction requirements and optimize the design plan; During the design phase, we conduct construction schedule and resource demand analysis, use 4D construction schedule simulation technology, add the time dimension to the 3D model, simulate the construction process, optimize the construction schedule, and estimate the required resources based on the construction schedule to guide the optimization of the design plan; Establish a design change management mechanism to track and control the entire process of design changes. By comparing and analyzing the design plans before and after the changes, quantitatively evaluate the impact of design changes on construction progress, construction quality, project cost, etc., optimize the design when necessary, and minimize the adverse effects of design changes; By using big data analysis and machine learning algorithms, we can mine and analyze historical engineering project data, summarize the correlation patterns and risk factors between design and construction, form a risk warning model for the design stage, conduct risk assessments on design solutions, take targeted measures in advance, and improve the reliability and robustness of design solutions.

2. The method according to claim 1, characterized in that The construction of unified data exchange standards and protocols converts heterogeneous data output by design tools of different disciplines and units into standardized data formats, realizes seamless connection and real-time sharing of data through data middleware, eliminates data barriers, and improves collaborative design efficiency, including: Analyze the data structure and semantics of heterogeneous data, extract key elements, map them to a unified data model, build a standardized data format, eliminate the differences between heterogeneous data, and achieve standardized representation of data; Design data exchange protocols based on standardized data formats, clarify data transmission rules, interface definitions, and data encoding methods, etc., to ensure that different systems can accurately understand and parse data content and achieve seamless data connection; Develop data middleware to connect to the data interfaces of various design tools. Through the data conversion engine, convert heterogeneous data into a standardized format and store it in a unified database to form a consistent data view, providing a basis for data sharing. Adopting a distributed architecture, data middleware is deployed to each node. Through message queues and data synchronization mechanisms, real-time data transmission and synchronization are achieved to ensure data consistency in each node and provide real-time data sharing capabilities. In order to solve the data barrier problem, different data access strategies are set through data access control and authority management to desensitize sensitive data and ensure data security. At the same time, necessary data access interfaces are opened to promote data sharing and circulation. In the data middleware, data quality detection and data cleaning functions are integrated to verify and correct the heterogeneous data, identify and process missing values ​​and abnormal values, etc., to improve the accuracy and completeness of the data and provide high-quality data support for subsequent data analysis and application; Establish a data sharing incentive mechanism, reward units that actively participate in data sharing, commend units with excellent data quality, create a good data sharing atmosphere, promote collaborative design among units, and improve design efficiency and quality; It also includes: by building unified data exchange standards and protocols, converting heterogeneous data into standardized formats, and using data middleware technology to achieve seamless docking and real-time sharing, thereby eliminating data barriers and improving collaborative design efficiency.

3. The method according to claim 2, characterized in that The above mentioned process is to build a unified data exchange standard and protocol, convert heterogeneous data into a standardized format, and use data middleware technology to achieve seamless connection and real-time sharing, thereby eliminating data barriers and improving collaborative design efficiency, including: According to business needs and data characteristics, formulate unified data exchange standards and protocols, clarify data format, field meaning, encoding method, etc., to ensure that heterogeneous data can be converted and exchanged according to standards; Use data modeling technology to analyze and abstract heterogeneous data, extract key entities, attributes and relationships, and build a unified data model as the basis for data conversion and integration; Develop data conversion tools to convert heterogeneous data from original formats to standardized formats based on predefined mapping rules and conversion logic, eliminate data heterogeneity, and achieve unified data representation; Build a data middleware platform to provide data transmission, routing, conversion, filtering and other functions, support the connection of multiple data sources and target systems, and realize seamless connection and data sharing between heterogeneous systems; Adopting a distributed architecture design, by deploying multiple data nodes and service instances, the system's concurrent processing capability and scalability are improved to meet the needs of large-scale data exchange and real-time sharing; Establish a data quality monitoring and verification mechanism to check the integrity, accuracy, consistency, etc. of the accessed data, promptly identify and handle data quality issues, and ensure data reliability; Based on standardized data formats and unified data interfaces, data sharing and collaboration between different systems and departments can be achieved, data barriers can be broken down, and the efficiency of business processes and the accuracy of decision-making can be improved.

4. The method according to claim 1, characterized in that: The collaborative design management system based on the cloud platform is used to uniformly manage and schedule the work tasks of various disciplines and units, monitor the work progress of various disciplines in real time through visual tools such as Gantt charts and milestones, promptly discover and solve abnormal progress problems, and coordinate the work rhythm of various disciplines and units, including: According to the characteristics of the cloud platform, a Web-based collaborative design management system is built to input the work task information of various disciplines and units into the system to form a unified task management library; By analyzing and processing the work tasks in the task management library, visual progress monitoring tools such as Gantt charts and milestones are automatically generated to achieve real-time monitoring of the progress of each professional work; Use machine learning algorithms to analyze the work progress data of each discipline, identify abnormal progress issues through anomaly detection algorithms, and automatically generate warning information based on preset rules; The abnormal progress warning information is pushed to the heads of relevant disciplines and units, and the abnormal problems are entered into the problem management module, and the problem solution suggestions are automatically generated according to the severity and impact scope of the problem; In the problem management module, the case-based reasoning algorithm is used to analyze historical problem solutions, match the solution most similar to the current problem, and optimize the solution according to the characteristics of the current problem to generate the final problem solution; Dynamically adjust the work tasks of each discipline and unit according to the work progress and problem solving status of each discipline, reallocate tasks through task scheduling algorithms, balance the work rhythm of each discipline, and ensure that the overall progress of the project is not affected; The system automatically generates coordination suggestions for various disciplines and units, and promotes communication and collaboration among disciplines and units through message push and task reminders, forming an efficient collaborative design management closed loop.

5. The method according to claim 1, characterized in that The digital twin model of the project is constructed in the design stage by using BIM and other digital technologies. Through virtual construction simulation and optimization, the design scheme is deepened, and factors such as construction technology, construction process, and resource allocation are fully considered to improve the construction feasibility and economic rationality of the design, including: Obtain relevant data such as the design drawings, construction requirements, and site environment of the project, and build a three-dimensional digital model of the project on the BIM platform to form a digital twin model; In the digital twin model, different construction plans are simulated according to the requirements of construction technology and construction procedures. Through virtual construction technology, each construction link is simulated to obtain data such as construction progress, cost and quality of each construction plan; The simulation data of the construction schemes are analyzed using machine learning algorithms, and different construction schemes are classified using clustering algorithms. The construction schemes of each category are ranked according to factors such as construction requirements and site environment. For the construction plans ranked high, we use optimization algorithms to further optimize them, and obtain more economical, reasonable and feasible construction plans by adjusting construction technology, construction procedures and resource allocation parameters; Compare the optimized construction plan with the original design plan, analyze its improvement in construction feasibility and economic rationality, and modify and improve the design plan based on the analysis results; In the digital twin model, virtual construction simulation is performed again based on the optimized design plan to verify its feasibility and economy, ensuring that the design plan can meet the construction requirements and achieve the expected goals; The optimized design plan and construction plan are output to form complete construction drawings and construction plan documents, providing guidance and support for subsequent construction management and on-site construction, ensuring the smooth implementation and efficient management of the project.

6. The method according to claim 1, characterized in that The above mentioned coordination mechanism between the design and construction phases is established, and experienced construction management personnel and technical personnel are invited to participate in the design review, and suggestions and recommendations are put forward for the design scheme from the construction perspective, so as to help the design party fully understand the construction requirements and optimize the design scheme, including: According to the design plan in the design stage, natural language processing technology is used to extract key information from the design documents, including design goals, design priorities and technical parameters, to form a structured design information table; Through knowledge graph technology, the structured design information table is associated and matched with the pre-built construction knowledge base to identify the key nodes and risk points related to construction in the design plan; For the identified key nodes and risk points, the rule-based reasoning engine is used in combination with the empirical rules in the construction knowledge base to automatically generate targeted construction suggestions and optimization plans; Compare and analyze the generated construction suggestions and optimization plans with the original design plans, use the text similarity algorithm to calculate the similarity between the two, and judge the feasibility of the optimization plan based on the set threshold; For optimization schemes with similarity below the threshold, machine learning algorithms, such as decision trees or support vector machines, are used to further optimize and adjust the schemes in combination with historical construction data to improve the feasibility of the schemes; Feedback optimized construction suggestions and optimization solutions to designers, and automatically generate visual reports to show the comparison effect before and after optimization, helping designers to intuitively understand the value of optimization; Establish a collaborative platform for design and construction, link the optimized design plan with the construction plan, realize information sharing and real-time updating of design and construction, and ensure that the construction process is consistent with the optimized design plan.

7. The method according to claim 1, characterized in that The construction schedule and resource demand analysis are carried out in the design stage, and the 4D construction schedule simulation technology is used to add the time dimension to the 3D model to simulate the construction process, optimize the construction schedule, and estimate the required resources based on the construction schedule to guide the optimization of the design plan, including: According to the design plan, a 3D model is constructed, and the time information of each construction stage is associated with the 3D model to form a 4D construction progress model; Through the 4D construction progress model, the construction process under different construction plans is simulated to obtain time parameters such as the start and end time and duration of each construction stage; Based on the construction progress data obtained through simulation, optimization algorithms such as critical path method and linear programming are used to optimize and adjust the construction schedule, shorten the construction period, and improve construction efficiency; According to the optimized construction schedule, combined with factors such as construction technology and construction intensity, estimate the amount of resources such as manpower, materials and equipment required for each construction stage; Link resource demand estimation results with the 3D model to visualize the resource demand distribution at each construction stage and identify resource demand peaks and bottlenecks; In response to problems such as uneven resource demand distribution and resource conflicts, adjust and optimize the construction schedule and resource allocation plan to achieve balanced utilization and reasonable allocation of resources; The optimized construction schedule and resource allocation plan will be fed back to the designers to guide the optimization and improvement of the design plan and improve the construction feasibility and economic rationality of the design.

8. The method according to claim 1, characterized in that The above-mentioned establishment of a design change management mechanism tracks and controls the entire process of design changes, compares and analyzes the design schemes before and after the changes, quantitatively evaluates the impact of design changes on construction progress, construction quality, project cost, etc., optimizes the design when necessary, and minimizes the adverse effects of design changes, including: Obtain design change applications, extract key information such as change content and reasons, and determine whether the changes comply with specification requirements. If not, the application will be rejected; if it is, the change assessment process will be initiated; According to the content of the change, obtain the design plans before and after the change, use BIM and other technologies to conduct 3D visual comparative analysis of the design plans, and identify the changed parts and the scope of impact; By analyzing the construction progress, quality and cost data of similar historical projects through big data, combined with the expert knowledge base, a change impact assessment model is established to quantitatively analyze the impact of design changes on construction period, quality and cost; If the impact of the change exceeds the preset threshold, the design optimization process is triggered; Using algorithms such as parametric design and topology optimization, we can optimize the design and reduce the impact of changes while meeting functional and performance requirements; Compare the optimized design with the original plan, calculate the optimization effect, and form a design change assessment report as the basis for change approval; If the change plan is approved, the change content will be synchronized to relevant documents such as construction drawings, schedules, material plans, and cost budgets to ensure coordinated updates in all links; During the construction process, we continuously track the implementation of changes, collect on-site data through intelligent information systems, compare and analyze the actual progress, quality and cost differences before and after the changes, dynamically evaluate the effect of the changes, and promptly warn of deviation risks to ensure that the change goals are achieved; It also includes: obtaining the design plans before and after the design changes, using comparative analysis techniques to determine the impact of the design changes on construction progress, construction quality, project cost, etc., and based on the quantitative evaluation results, determining whether design optimization is needed to obtain the optimized design plan.

9. The method according to claim 8, characterized in that The design schemes before and after the design change are obtained, and comparative analysis techniques are used to determine the impact of the design change on the construction progress, construction quality, project cost, etc. Based on the quantitative evaluation results, it is determined whether design optimization is needed to obtain the optimized design scheme, including: Obtain the design drawings before the change, extract the drawing information through optical character recognition technology, apply the pre-established 3D model library to generate the 3D model before the change, automatically calculate the bill of quantities before the change based on the 3D model before the change, and simultaneously capture and interpret the construction log file to determine the pre-change qualified rate, pre-change rework rate, pre-change material cost, pre-change labor cost, pre-change machinery cost, and then apply the regression analysis algorithm to obtain the total cost before the change; Obtain the design drawings after the change, extract the drawing information through optical character recognition technology, apply the pre-established 3D model library to generate the 3D model after the change, automatically calculate the bill of quantities after the change based on the 3D model after the change, and retrieve similar project data from the central database at the same time, and use the random forest algorithm to obtain the qualified rate after the change, the rework rate after the change, the material cost after the change, the labor cost after the change, and the machinery cost after the change, and calculate the total cost after the change; Compare the bill of quantities before and after the change to obtain the bill of quantities difference data, and compare the total cost before and after the change to obtain the total cost difference data; According to the three-dimensional models before and after the change, the graphic change data is obtained; If there are cost increase items in the bill of quantities difference data, and the sum of the cost increase items is greater than the preset cost threshold, the early warning information push module will be activated to issue a construction early warning prompt; At the same time, if the total cost after the change is higher than the total cost before the change, and the difference is greater than the preset price difference threshold, it will be pushed to the manual analysis stage; According to the graph change data, combined with the qualified rate before the change and the qualified rate after the change, the quality change data is compared; According to the comparison results of quality change data, if the qualified rate after the change is lower than the qualified rate before the change, the qualified rate reduction of all the constructions with design changes in the design scheme after the change is calculated, and the qualified rate reduction = (qualified rate before the change - qualified rate after the change) / qualified rate before the change 100%; Generate a list of modification suggestions based on a pre-established expert database; Determine whether there is a design change with a size modification of more than 20% in the graphic change data. If the size modification ratio exceeds the preset size ratio threshold, the computer-aided design software will automatically traverse multiple modification schemes and use comparative analysis technology to compare different design change schemes; Select a plan with the lowest cost and the size change closest to the expert's modification opinion from multiple modification plans; Through comprehensive cost calculation, combined with the qualified rate after change, the rework rate after change and the expert modification opinions, the design change optimization suggestions are determined, and the pre-trained support vector machine algorithm is used to determine whether the design scheme has reached a deliverable state. If it is deliverable, the results are output; if it is not deliverable, it returns to the modification opinion list interface.

10. The method according to claim 1, characterized in that The above mentioned use of big data analysis and machine learning algorithms to mine and analyze historical engineering project data, summarize the correlation rules and risk factors between design and construction, form a risk warning model for the design stage, conduct risk assessment on the design scheme, take targeted measures in advance, and improve the reliability and robustness of the design scheme, including: Collect multi-dimensional data of historical engineering projects, including engineering data, project type, construction unit, design unit, design parameters, number of changes, risk category, material type and service life, and establish a structured historical engineering database through data association, with each project data having a unique number; According to the risk categories included in the completed projects in the database, determine the completed project risk results corresponding to each unique number in the actual project, and count the number of different risk categories according to the risk results, sort the numbers from large to small, take out the top ten risk categories as specific risk categories, and determine the specific risk category set; According to each unique number in the database, a design parameter vector of each design scheme is constructed. The elements of the design parameter vector are composed of the characteristic values ​​of risk category, project type, construction unit, design unit, design parameter, number of changes and material type. The K-means clustering algorithm is used to divide the design parameter vector into different specific risk categories to determine the risk category of each design scheme. The risk level prediction model is obtained by training the structured historical engineering database through the LightGBM machine learning algorithm. The design schemes classified in the previous step, including the design parameters, project type, construction unit, design unit, number of changes and material type feature values, form the design scheme feature vector and input it into the risk level prediction model to determine the risk level prediction value. Use the XGBoost machine learning algorithm to train the design scheme robustness prediction model, determine the design parameters of each design scheme in the structured historical engineering database, including length, width and height, and multiply each design parameter by the disturbance coefficient to obtain each disturbance design parameter. If each disturbance design parameter is input into the risk level prediction model, the more disturbance parameter groups in the output result that are lower than the preset threshold, the lower the design robustness of this uniquely numbered scheme; Comprehensively analyze the risk category of the design solution, the risk level prediction value and the judgment result of the robustness, obtain the risk category of the design solution, and compare the current risk category with the specific risk category set to determine whether it belongs to the specific risk category. If it belongs to the specific risk category and the robustness is low, the unique number will be included in the key focus list; By focusing on each unique number in the list, historical corresponding project information can be obtained from the database, including the completion drawings and change notices of the corresponding projects. The design parameters with the most changes in similar projects can be obtained as key parameters, and adjustment suggestions can be made in advance.

Images