Real-time construction progress optimization system based on bim collaborative design and industrial cloud platform

Abstract

The application relates to the technical field of construction optimization, in particular to a real-time construction progress optimization system based on BIM collaborative design and an industrial cloud platform, which realizes the spatio-temporal correlation and standardized management of multidimensional data such as process progress, resource cost and carbon emission by constructing a multi-source fusion construction scene causal data pool, provides comprehensive and accurate data support for construction progress optimization, meanwhile, the dynamic causal chain mining technology breaks through the limitations of traditional experience dependence or simple correlation analysis, can identify the root cause of affecting the construction progress in stages, quantifies the causal relationship strength, makes the progress optimization upgrade from passive response to active root tracing, the optimization strategy generation mechanism based on the causal diagram, combined with multi-objective collaborative optimization and digital twin verification, realizes the comprehensive balance of time, cost and carbon emission, quantifies the node combination effect, and ensures that the strategy has anti-interference ability and executability in the complex construction scene.

Description

Technical Field

[0001] This invention relates to the field of construction optimization technology, and more specifically, to a real-time construction progress optimization system based on BIM collaborative design and an industrial cloud platform. Background Technology

[0002] The construction management field currently faces the common pain points of data silos and low collaboration efficiency. Data generated during construction, such as process progress, resource consumption, and environmental parameters, are often scattered across different systems like BIM models, project management software, and IoT devices. The lack of a unified integration framework and standardized processing mechanism makes it difficult to achieve spatiotemporal correlation and in-depth data mining, hindering accurate progress analysis and optimization decisions. Furthermore, current progress optimization methods largely rely on experience-based judgment or simple parameter comparisons, failing to identify the root causes affecting progress and often resulting in passive adjustments to superficial phenomena, limiting optimization efficiency and accuracy.

[0003] The generation of optimization strategies lacks quantitative analysis and dynamic verification of the node combination effect. When faced with complex disturbances such as weather changes and resource fluctuations, the solutions are prone to failure, have weak anti-interference capabilities, and cannot provide reliable decision support for managers.

[0004] Based on the above, this application proposes a real-time construction progress optimization system based on BIM collaborative design and an industrial cloud platform. Summary of the Invention

[0005] In view of the shortcomings of existing technologies, the purpose of this invention is to provide a real-time construction progress optimization system based on BIM collaborative design and industrial cloud platform.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The real-time construction progress optimization system based on BIM collaborative design and industrial cloud platform is characterized by including a construction decision graph construction module, which constructs a causal-driven construction decision knowledge graph based on BIM collaboration and industrial cloud platform.

[0008] The causal relationship mining module is used to dynamically mine causal chains;

[0009] The construction progress optimization strategy generation module generates construction progress optimization strategies based on cause-effect graphs.

[0010] Furthermore, a causal-driven construction decision-making knowledge graph is constructed based on BIM collaboration and an industrial cloud platform:

[0011] S11: Define the core dimensions and specific metrics of the data pool;

[0012] S12: Integrate BIM with industrial cloud platforms to build multi-source data acquisition channels;

[0013] S13: Data cleaning and standardization;

[0014] S14: Data Association and Storage Architecture Design.

[0015] Furthermore, the core dimensions include process progress, resource costs, carbon emissions, resource status, environmental factors, and management behavior. Specific indicators for process progress include planned start / end time, actual start / end time, process delay duration, and critical path identification. Specific indicators for resource costs include material usage / unit price, equipment rental costs, labor hours / daily wages, and machinery fuel / electricity costs. Specific indicators for carbon emissions include carbon contained in materials, carbon consumed by equipment, and carbon from transportation. Specific indicators for resource status include real-time equipment load rate, material inventory levels, and number of workers present. Specific indicators for environmental factors include weather and construction site conditions. Specific indicators for management behavior include the number of design changes, visa approval time, and supplier delivery delay days.

[0016] Furthermore, dynamically mine the causal chain: S21: Decomposition of the construction variable system;

[0017] S22: Timing data alignment and preprocessing;

[0018] S23: Stage-by-stage causal relationship mining;

[0019] S24: Causal chain strength quantification.

[0020] Furthermore, the construction variable system is broken down as follows: core variables related to time, cost, and carbon in the construction scenario are defined. The core variables include target variables and influencing variables. Target variables include time, cost, and carbon dimensions, while influencing variables include process parameters, resource status, environmental factors, and management behavior.

[0021] Furthermore, a construction schedule optimization strategy based on cause-effect graphs is generated:

[0022] S31: Determine the construction phase. When any type of construction schedule delay occurs during the construction phase, generate a schedule delay log.

[0023] S32: Determine the direct causal nodes of the cause-effect graph based on the delayed process type in the schedule delay log, and perform source factor analysis on each direct causal node;

[0024] S33: Select the interventionable nodes for each source factor from the cause-effect diagram, determine the time reduction, cost change, and carbon emission change for each interventionable node, further generate multiple construction progress optimization strategies, determine the construction progress optimization index for each construction progress optimization strategy, and execute the construction progress optimization strategy with the largest construction progress optimization index value to optimize the construction progress.

[0025] Furthermore, the construction schedule optimization index of the construction schedule optimization strategy is calculated by analyzing the comprehensive time reduction, comprehensive cost change, comprehensive carbon emission change, and interaction impact value of a construction schedule optimization strategy.

[0026] Further, the steps for obtaining the interaction impact value of the construction schedule optimization strategy are as follows: identify all the interventionable nodes included in the construction schedule optimization strategy, obtain the total delay rate of each interventionable node in the digital twin environment, simultaneously obtain the total delay rate of the construction schedule optimization strategy, sum and average the total delay rates of all interventionable nodes to calculate the node average delay rate, and calculate the difference between the node average delay rate and the total delay rate of the construction schedule optimization strategy to calculate the interaction impact value.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The system of this invention constructs a multi-source fusion causal data pool for construction scenarios, realizing the spatiotemporal correlation and standardized management of multi-dimensional data such as process progress, resource costs, and carbon emissions. This provides comprehensive and accurate data support for construction progress optimization. At the same time, the dynamic causal chain mining technology breaks through the limitations of traditional experience-based or simple correlation analysis, enabling the identification of the root causes affecting construction progress in stages and quantifying the strength of causal relationships. This upgrades progress optimization from passively responding to phenomena to actively tracing the root causes. Based on the causal graph-based optimization strategy generation mechanism, combined with multi-objective collaborative optimization and digital twin verification, a comprehensive balance of time, cost, and carbon emissions is achieved, and the node combination effect is quantified to ensure that the strategy has anti-interference capabilities and executability in complex construction scenarios. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of a real-time construction progress optimization system based on BIM collaborative design and an industrial cloud platform.

[0030] Figure 2 A schematic diagram of the principle of industrial decision-making knowledge graph;

[0031] Figure 3 Flowchart for selecting strategies to optimize construction schedule. Detailed Implementation

[0032] Reference Figures 1 to 3 A real-time construction progress optimization system based on BIM collaborative design and industrial cloud platform, including a construction decision map construction module, a causal relationship mining module, and a construction progress optimization strategy generation module;

[0033] The construction decision graph construction module builds a causal-driven construction decision knowledge graph based on BIM collaboration and an industrial cloud platform.

[0034] S11: Define the core dimensions and specific indicators of the data pool; the core dimensions include process progress, resource cost, carbon emissions, resource status, environmental factors, and management behavior. Specific indicators for process progress include planned start / end time, actual start / end time, process delay duration, and critical path identification. Specific indicators for resource cost include material usage / unit price, equipment rental costs, labor hours / daily wages, and machinery fuel / electricity costs. Specific indicators for carbon emissions include carbon inherent in materials (e.g., 1.8t CO2 / t of steel reinforcement), carbon from equipment energy consumption (e.g., 15kWh of electricity consumed per hour by a tower crane → 12kg of carbon emissions), and carbon from transportation (e.g., 0.5kg of carbon emissions per kilometer of concrete transportation). Specific indicators for resource status include real-time equipment load rate (e.g., current lifting weight / rated lifting weight of a tower crane), material inventory balance, and number of workers present. Specific indicators for environmental factors include weather (sunny / rainy / temperature and humidity) and construction site conditions (e.g., site flatness, distance between material stacking areas). Specific indicators for management behavior include the number of design changes, visa approval time, and supplier delivery delay days.

[0035] S12: Establish a multi-source data acquisition channel by integrating BIM with the industrial cloud platform; the multi-source data acquisition channel includes access to structured data from the BIM model (data sources: model files generated by BIM software such as Revit and Tekla (including IFC format), and schedules from construction simulation software (such as Navisworks). Acquisition method: Extract component-process-resource association data from the model through the API interface of the BIM platform (such as Revit API): for example, extract the corresponding process (reinforcement binding, formwork) from the column component (ID: COL-001) on the first basement floor. The data includes concrete pouring, required resources (Φ20 steel bars, wooden formwork, pump truck), and planned construction period (planned time for each process). The BIM WBS work breakdown structure (e.g., single project - unit project - sub-project - sub-item project) is converted into a process-level index for the data pool (e.g., WBS code: W-02-03-01 corresponding to main structure - second floor - beam pouring), serving as the core key value for subsequent data association. Real-time IoT data collection (data source: IoT devices deployed on the construction site (smart meters, fuel flow meters, GPS locators, AI cameras, etc.). Collection method: In...) The industrial cloud platform deploys edge computing gateways to connect devices via LoRa / NB-IoT protocols. For example, the smart meters on tower cranes upload real-time power (kW) every 15 minutes, the GPS locators on concrete mixer trucks upload their location every 5 minutes (for calculating transport distance), and AI cameras use computer vision to identify the number of workers (for counting manpower input in each process). Each piece of equipment and each process is bound to a unique IoT identifier ID and associated with the process ID in the BIM model. For example, a pump truck (equipment ID: P-002) is bound to the beam pouring process (WBS code: W-0). 2-03-01), ensuring energy consumption data is traceable to specific processes), importing historical project databases (data sources: enterprise ERP systems, project management platforms (such as Glodon, Luban) storing data on similar projects from the past 5 years (such as 30-story residential buildings, commercial complexes). Collection method: filter historical projects matching the current project type (such as consistent structural form, building area, climate zone), and extract their actual process time, resource consumption, and carbon emission records: for example, extract standard floor construction data from 10 30-story residential projects in the same area, including the average time for rebar tying (8 hours / 100m). 2 Average carbon emissions (50kgCO2 / 100m³) 2The historical data was anonymized: sensitive information such as project name and specific location was removed, retaining only technical parameters (e.g., Project A in 2021 was anonymized to historical Project H-001). External system data was integrated (data sources: supplier management system (material supply records), meteorological platform (historical + real-time weather), carbon emission database (e.g., the carbon factor database of the China Building Materials Academy). Data collection method: The supplier system was integrated via API to obtain material arrival time, quantity, and quality inspection reports. For example, a cement plant (supplier ID: S-005) uploaded the arrival time (2025-07-10), quantity (50m3), and EPD environmental product declaration (implicit carbon 280kgCO2 / m3) of C30 concrete. Integration with the national meteorological data platform was also performed to obtain historical + real-time weather data (sunny / rainy, temperature, wind force) for the project location, and stored by date-time period. For example, the period 2025-07-10 8:00-18:00 was recorded as a rainstorm (for analyzing the impact of weather on the process).

[0036] S13: Data Cleaning and Standardization; Data cleaning includes missing value handling (filling in key data (such as process time, equipment energy consumption) using interpolation: for example, if the energy consumption data for a process over 3 days is missing for day 2, it is estimated and supplemented by the average of the two days before and after (or the energy consumption trend of similar processes). For non-critical data (such as auxiliary usage of minor materials), the cause of the missing data is marked (e.g., not recorded, equipment failure) to avoid artificially fabricated data affecting causal analysis), and outlier identification and handling (using the 3σ principle + industry experience to screen outliers: for example, if the carbon emission of a process is 5 times the average of similar processes (far exceeding the 3σ range), and the abnormal energy consumption is confirmed by combining construction logs to be caused by equipment failure, it is marked as an outlier and stored separately (not included in causal analysis, but retained for subsequent troubleshooting). (Training of early warning models). For data on extreme weather (such as a typhoon causing a 3-day work stoppage), label it with a special operating condition tag: do not delete it directly (because weather is an important influencing factor), but classify it separately in causal analysis (such as atypical interference scenarios). Standardize the processing (unify units and formats: for example, unify carbon emissions to kgCO2, time to hours, cost to yuan; unify date format to YYYY-MM-DDHH:MM). Define a carbon factor standard library: for example, uniformly adopt the equipment carbon factor in the "Provincial Greenhouse Gas Inventory Compilation Guidelines" (such as diesel generator: 2.63 kgCO2 / L). If the supplier provides more accurate EPD data (such as the actual carbon factor of a certain brand of tower crane is 10% lower than the industry average), then the supplier's actual measurement will be used first and marked.

[0037] S14: Data Association and Storage Architecture Design (Constructing a spatiotemporal dual-index data association model to ensure cross-querying of multi-source data: Core association key design: Using WBS process code + timestamp as the primary index: For example, W-02-03-01 (beam pouring) + 2025-07-10 8:00-12:00 associates all data within this time period (rebar usage 5t, pump truck energy consumption 100kWh, carbon emissions 80kg, weather storm). Using BIM component ID + resource ID as the secondary index: For example, COL-001 (column component) + P-002 (pump truck) associates all pump truck consumption data during the construction of this component. Storage rack) Structure: A hybrid database architecture is adopted: a relational database (MySQL) stores structured data (such as BIM process information and historical project statistics), a time-series database (InfluxDB) stores real-time IoT data (such as energy consumption curves every 15 minutes), and a graph database (Neo4j) stores preliminary entity-relationships (such as the association between pump truck-P-002 and process-W-02-03-01). A data platform is deployed on the industrial cloud platform to provide a unified data access interface: for example, the BIM collaborative design end can query the historical carbon emission data of a certain process through the interface, and the construction management end can query the real-time energy consumption and causal relationship factors of a certain equipment.

[0038] The causal relationship mining module is used for dynamic causal chain mining.

[0039] S21: Decomposition of Construction Variable System; Define the core variables related to time, cost, and carbon in the construction scenario. The core variables include target variables and influencing variables. Target variables include the time dimension (including critical path duration, process delay duration, resource idle time, etc.), the cost dimension (including material procurement overruns, additional machinery rental costs, rework costs, etc.), and the carbon dimension (including process carbon emission increments, equipment idle carbon emissions, and material transportation carbon exceedances). Influencing variables include process parameters (such as the number of people tying rebar, concrete pouring speed, and formwork turnover), resource status (such as tower crane load rate, diesel generator start-up time, and on-time arrival rate of precast components), environmental factors (such as daily rainfall, wind speed, and temperature (affecting concrete curing)), and management behaviors (such as the number of design changes, safety inspection downtime, and pre-shift briefing completion rate).

[0040] S22: Time-series data alignment and preprocessing; Data source integration: IoT real-time data: smart meters (15 minutes / time), equipment GPS tracks (10 minutes / time), environmental sensors (1 hour / time); Management system data: BIM schedule (updated daily), cost ledger (updated weekly), quality acceptance records (updated by process); Historical experience data: database of process time consumption for similar projects, carbon emission index database (e.g., the daily average carbon emission benchmark value for a standard floor of a 30-story residential building). Preprocessing operations: Time alignment: Convert all data to hourly time granularity (e.g., break down daily rainfall by hour, and allocate material procurement costs by material usage period); Missing value handling: For missing data caused by equipment failure, fill with the average of the previous 3 hours of the same process + trend correction (e.g., if tower crane data is missing at 9 o'clock, use the average of 6-8 o'clock × 1.1 (the load rate is higher during the morning peak)); Outlier removal: Remove extreme values ​​using the 3σ principle (e.g., a single-day carbon emission of 1000t due to human error in recording, far exceeding 3 times the average of similar projects).

[0041] S23: Phased Causal Relationship Mining; Phase Division Criteria: Based on the WBS (Work Breakdown Structure), the phases are divided into the foundation engineering phase → main structure phase → decoration and finishing phase → completion and acceptance phase, with a separate causal chain constructed for each phase. Causal Mining Algorithm Selection and Adaptation: An improved PC algorithm (suitable for multivariate causal discovery) is adopted and optimized for construction scenarios: Initial Correlation Matrix Calculation: The correlation between any two variables is calculated (e.g., the correlation coefficient between concrete transportation distance and transportation carbon emissions is r = 0.89), retaining strong correlation pairs with r > 0.3 (weak correlations are likely not causal); Conditional Independence Test: For strong correlation pairs, a third-party variable is controlled to determine whether it is a direct causal relationship. Example: To test whether rainy days and increased carbon emissions are a direct causal relationship—after controlling the tower crane waiting time, if the correlation between the two decreases from r = 0.7 to r = 0.2, it indicates that rainy days → tower crane waiting → increased carbon emissions is an indirect causal relationship, not a direct causal relationship; Causal Direction Determination: The direction is determined by time sequence + intervention experiments. Time sequence: For example, a material delay (t=5 days) must occur before the idle time in the process (t=6 days), the former being the cause; Intervention experiment: Forcefully change variable A in a simulated environment and observe whether variable B changes (e.g., artificially reducing tower crane idle time; if carbon emissions decrease significantly, then tower crane idle time is the cause, and increased carbon emissions are the effect). Output: Stage causal network; each stage generates a visual causal network diagram, nodes are variables, directed edges represent causal relationships, and edge thickness represents causal strength (e.g., in the causal diagram of the main structure stage, the edge for pump truck idling time → increased carbon emissions is thicker than the edge for worker number → increased carbon emissions, indicating that the former has a stronger impact).

[0042] S24: Causal chain strength quantification; Causal strength calculation: quantified using average impact amplitude + probability of occurrence dual indicators: Example 1: Pump truck idling for 1 hour during the main structure phase → carbon emissions increase by 25kg. The probability of this situation occurring in historical data is 78% (78 out of 100 idling times lead to increased carbon emissions), so the causal strength = 25kg × 78% = 19.5kg / hour (high strength); Example 2: Low worker skill level → extended formwork construction period, with an average extension of 0.5 days, but only 30% of low-skilled workers cause this result, so the causal strength = 0.5 days × 30% = 0.15 days (low strength).

[0043] The construction progress optimization strategy generation module generates construction progress optimization strategies based on cause-effect graphs.

[0044] S31: Determine the construction phase. When any type of construction schedule delay occurs during the construction phase (types of construction schedule delay include, but are not limited to, the following: actual time taken for a single process exceeds 10% of the planned time (e.g., planned formwork erection took 8 hours, but actually took 9 hours); critical path cumulative delay exceeds 2 days (e.g., the main structure critical path was originally planned for 30 days, but has now taken 32 days); non-critical path delays cause the total float to be exhausted (e.g., the decoration process, originally with a total float of 5 days, has been delayed by 6 days, starting to affect subsequent processes)), a schedule delay log is generated. It includes the type of delayed process, planned time, actual time, delay range, and related indicators; for example, the delayed process type of the progress delay log A is beam pouring process, the planned time is 8 hours (2025-07-15 8:00-16:00), the actual time is 11 hours (2025-07-15 8:00-19:00), the delay range is 3 hours (37.5%), and the related indicators are actual carbon emissions of 120kg (planned 100kg, exceeding 20%) and labor costs increased by 1500 yuan (overtime pay for an additional 3 hours).

[0045] S32: Based on the delayed process types in the schedule delay log, determine the direct causal nodes in the cause-effect diagram (e.g., the direct influencing factors of the beam pouring process in the cause-effect diagram include: concrete arrival time (causal strength: impact magnitude 2 hours / delay, probability of occurrence 80%); pump truck failure frequency (impact magnitude 1.5 hours / failure, probability of occurrence 30%); worker skill level (impact magnitude 0.5 hours / level difference, probability of occurrence 60%)). Perform source factor analysis on each direct causal node (e.g., source factor analysis for concrete arrival time delay (direct influencing factor)). The cause-and-effect graph shows that the direct cause of the delayed concrete arrival is the production scheduling conflict at the batching plant (directed edge: batching plant scheduling → concrete arrival); tracing back to the cause of the batching plant scheduling conflict: the cause-and-effect graph points to the delay in the contractor's submission of the demand order (directed edge: demand order submission → batching plant scheduling); verifying whether the delay in demand order submission is the root cause: checking the data pool records—the actual submission time of the demand order for this process was 2 days later than planned (consistent with the time sequence), and after submitting the demand order in advance in the simulation environment, the batching plant scheduling conflict disappeared (intervention experiment verification), therefore, the delay in demand order submission was determined to be the source factor).

[0046] S33: Select the interventionable nodes for each source factor from the cause-and-effect diagram (e.g., if there are three source factors: delayed order submission, mismatched pump truck maintenance cycle, and lack of worker training, the interventionable nodes for delayed order submission are online parallel approval and the green channel for emergency orders; the interventionable nodes for mismatched pump truck maintenance cycle are dynamic maintenance cycle and hydraulic pressure special testing; and the interventionable nodes for lack of worker training are VR vibration operation training and vibration operation guidance QR code). Determine the time reduction, cost change, and carbon emission change for each interventionable node (e.g., the time reduction for online parallel approval is 2 hours). The cost change is an investment of 5,000 yuan, and the carbon emission change is a reduction of 15 kg; if the time reduction of the green channel for emergency needs is 1 hour, the cost change is 0 yuan, and the carbon emission change is a reduction of 6 kg, then multiple construction progress optimization strategies are generated (based on the independent selection of the intervenable nodes in the source factors (the combination of intervenable nodes within a single source factor) and cross-combination (the combination of intervenable nodes between multiple source factors), the construction progress optimization index of each construction progress optimization strategy is determined, and the construction progress optimization strategy with the largest construction progress optimization index value is executed to optimize the construction progress.

[0047] Construction schedule optimization strategy construction schedule optimization index: Analyzing the overall time reduction (Compre) of a construction schedule optimization strategy. time Overall cost variation (Cost) changes Overall carbon emission change change and the interaction value Combinrate (For example, a construction schedule optimization strategy can be generated by combining online parallel approval with an emergency request green channel. The overall time reduction from this strategy would be 2 + 1 = 3 hours, the overall cost change would be 5000 + 0 = 5000 yuan, and the carbon emission change would be -15 - 6 = -21 kg.) The construction schedule optimization index Bga (the total number of calendar days (or hours) from the official commencement of the project to the completion and acceptance of the project, and the project carbon allowance is the upper limit of the total amount of carbon dioxide allowed to be emitted by the project throughout its entire life cycle (usually focusing on the construction phase)) is calculated. Among them, uk1 is the time reduction coefficient, uk2 is the cost variation coefficient, uk3 is the carbon emission variation coefficient, and uk4 is the interaction coefficient. The value of the time reduction coefficient can be 0.3, the value of the cost variation coefficient can be 0.3, the value of the carbon emission variation coefficient can be 0.3, and the value of the interaction coefficient can be 0.1.

[0048] Steps for obtaining the interaction impact value of the construction schedule optimization strategy: Identify all the interventionable nodes included in the construction schedule optimization strategy. In the digital twin environment, obtain the total delay rate of each interventionable node (e.g., a construction schedule optimization strategy is generated by combining online parallel approval and an emergency request green channel; in the digital twin environment, simulate the scenarios of online parallel approval and the emergency request green channel, respectively. For the online parallel approval scenario, input parameters: regular orders account for 70%, emergency orders account for 30%, result: regular order delay rate 15%, emergency order delay rate 30% due to lack of a green channel, total delay rate = 15% × 70% + 30% × 30% = 19.5%; for the emergency request green channel scenario...). Input parameters: Regular orders have no parallel approval, emergency orders use the green channel; Result: Regular order delay rate 40%, emergency order delay rate 10%, total delay rate = 40% × 70% + 10% × 30% = 31%). Simultaneously, obtain the total delay rate of the construction progress optimization strategy (Input parameters: Regular orders use online parallel approval, emergency orders use the green channel; Result: Regular order delay rate 12%, emergency order delay rate 8%, total delay rate = 12% × 70% + 8% × 30% = 10.8% (consistent with historical data)). Sum and average the total delay rates of all interventionable nodes to calculate the node average delay rate. Calculate the difference between the node average delay rate and the total delay rate of the construction progress optimization strategy to calculate the interaction value Combin. rate .

[0049] The aforementioned system, by constructing a multi-source integrated causal data pool for construction scenarios, achieves spatiotemporal correlation and standardized management of multi-dimensional data such as process progress, resource costs, and carbon emissions. This provides comprehensive and accurate data support for construction progress optimization. Meanwhile, the dynamic causal chain mining technology breaks through the limitations of traditional experience-based or simple correlation analysis, enabling the identification of the root causes affecting construction progress in stages and quantifying the strength of causal relationships. This upgrades progress optimization from passively responding to phenomena to actively tracing the root causes. Based on the causal graph-based optimization strategy generation mechanism, combined with multi-objective collaborative optimization and digital twin verification, it achieves a comprehensive balance of time, cost, and carbon emissions, and quantifies the node combination effect, ensuring that the strategy has anti-interference capabilities and executability in complex construction scenarios.

[0050] The above formulas are all dimensionless calculations, and the preset parameters in the formulas should be set by those skilled in the art according to the actual situation.

[0051] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0052] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0053] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0054] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0055] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0056] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0057] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A real-time construction progress optimization system based on BIM collaborative design and an industrial cloud platform, characterized in that, It includes a construction decision graph construction module, a causal relationship mining module, and a construction progress optimization strategy generation module; The construction decision graph construction module is used to build a causal-driven construction decision knowledge graph based on BIM collaboration and an industrial cloud platform, including: S11: Define the core dimensions and specific indicators of the data pool. The core dimensions include process progress, resource cost, carbon emissions, resource status, environmental factors, and management behavior. Specific indicators for process progress include planned start / end time, actual start / end time, process delay duration, and critical path identifier. Specific indicators for resource cost include material usage / unit price, equipment rental costs, labor hours / daily wages, and machinery fuel / electricity costs. Specific indicators for carbon emissions include carbon emissions inherent in materials, carbon emissions from equipment energy consumption, and carbon emissions from transportation. Specific indicators for resource status include real-time equipment load rate, material inventory levels, and number of workers present. Specific indicators for environmental factors include weather and construction site conditions. Specific indicators for management behavior include the number of design changes, visa approval time, and supplier delivery delay days. S12: Establish a multi-source data acquisition channel by combining BIM and the industrial cloud platform. The multi-source data acquisition channel includes BIM model structured data access, IoT real-time data acquisition, historical project database import, and external system data docking. Specifically, component-process-resource association data is extracted through the BIM platform interface, and the BIM's WBS work breakdown structure is transformed into a process-level index. IoT device data at the construction site is collected through the edge computing gateway of the industrial cloud platform, and a unique IoT identifier ID is bound to the device and process, so that the device data is associated with the process ID of the BIM model. S13: Clean and standardize the collected data, including filling missing values ​​for key data such as process time and equipment energy consumption, identifying and marking outliers, and unifying carbon emissions, time and cost into preset unit formats. S14: Construct a spatiotemporal dual-index data association model, using WBS process codes and timestamps as primary indexes, BIM component IDs and resource IDs as secondary indexes, and employing a relational database to store BIM process information and historical project statistics, a time-series database to store real-time IoT data, and a graph database to store entity-relationship data. The causal relationship mining module is used to dynamically mine causal chains, including: S21: Define the core variables related to time, cost, and carbon in the construction scenario. The core variables include target variables and influencing variables. The target variables include time, cost, and carbon dimensions. The influencing variables include process parameters, resource status, environmental factors, and management behavior. S22: Unify the conversion of real-time IoT data, management system data, and historical experience data into hourly time granularity, and perform missing value processing and outlier removal; S23: The construction process is divided into the foundation engineering period, main structure period, decoration and finishing period and completion and acceptance period according to the WBS work breakdown structure, and a causal chain is constructed separately for each stage; in each stage, the improved PC algorithm is used to mine causal relationships. First, the correlation between any two variables is calculated and strong correlation pairs with correlation coefficients greater than a preset threshold are retained. Then, the conditional independence test is performed on the strong correlation pairs by controlling third-party variables to determine the direct causal relationship. Finally, the causal direction is determined by combining the chronological relationship of the occurrence of variables and the intervention experiment of forcibly changing the causal variable in the simulation environment and observing the change of the result variable. S24: Quantify the strength of the causal chain, where the causal strength is obtained by multiplying the average magnitude of the influence by the probability of occurrence; The construction progress optimization strategy generation module is used to generate a construction progress optimization strategy based on a cause-effect graph, including: S31: Determine the current construction phase. When any of the following construction progress delay types occur during the construction phase: the actual time of a single process exceeds the planned time by 10%, the cumulative delay of the critical path exceeds 2 days, or the delay of the non-critical path causes the total float to be exhausted, a progress delay log is generated. The progress delay log includes the delayed process type, planned time, actual time, delay magnitude, and related indicators. S32: Determine the direct causal nodes in the cause-effect graph based on the delayed process type in the progress delay log, and perform source factor analysis on each direct causal node; the source factor analysis includes tracing candidate source factors upstream along the directed edges of the cause-effect graph, verifying the temporal relationship between candidate source factors and direct causal nodes based on data pool records, and conducting intervention experiments on candidate source factors in a simulation environment. When the direct causal node disappears or changes after intervening in the candidate source factor, the candidate source factor is determined as the source factor. S33: Select the interventionable nodes for each source factor from the cause-effect diagram, determine the time reduction, cost change and carbon emission change for each interventionable node, and generate multiple construction progress optimization strategies based on the independent selection or combination of interventionable nodes within the same source factor and the cross combination of interventionable nodes between different source factors. For each construction schedule optimization strategy, analyze the overall time reduction achieved by that strategy. Changes in comprehensive costs Comprehensive carbon emission change and interaction impact value The construction schedule optimization index Bga of this construction schedule optimization strategy is calculated using the following formula: ; in, The overall time reduction achieved by this construction schedule optimization strategy is... The overall cost change for this construction schedule optimization strategy, The overall carbon emission change for this construction schedule optimization strategy. The interaction impact values ​​for this construction schedule optimization strategy are: uk1 is the time reduction coefficient, uk2 is the cost variation coefficient, uk3 is the carbon emission variation coefficient, and uk4 is the interaction impact coefficient. The interaction value The acquisition steps include: identifying all operable nodes included in the construction progress optimization strategy; establishing single-node simulation scenarios with only one operable node enabled in the digital twin environment, and obtaining the total latency rate of each single-node simulation scenario; establishing a combined simulation scenario in the same digital twin environment with all operable nodes in the construction progress optimization strategy enabled simultaneously, and obtaining the total latency rate of the combined simulation scenario; summing and averaging the total latency rates of all single-node simulation scenarios to obtain the node average latency rate; and calculating the difference between the node average latency rate and the total latency rate of the combined simulation scenario to obtain the interaction impact value. ; The construction schedule optimization strategy with the highest construction schedule optimization index (Bga) value is applied to optimize the construction schedule.

2. The real-time construction progress optimization system based on BIM collaborative design and industrial cloud platform according to claim 1, characterized in that, The core dimensions include process progress, resource cost, carbon emissions, resource status, environmental factors, and management behavior. Specific indicators for process progress include planned start / end time, actual start / end time, process delay duration, and critical path identification. Specific indicators for resource cost include material usage / unit price, equipment rental costs, labor hours / daily wages, and machinery fuel / electricity costs. Specific indicators for carbon emissions include carbon embedded in materials, carbon consumed by equipment, and carbon from transportation. Specific indicators for resource status include real-time equipment load rate, material inventory levels, and number of workers present. Specific indicators for environmental factors include weather and construction site conditions. Specific indicators for management behavior include the number of design changes, visa approval time, and supplier delivery delay days.

3. The real-time construction progress optimization system based on BIM collaborative design and industrial cloud platform according to claim 1, characterized in that, Construction variable system decomposition: Define the core variables related to time, cost, and carbon in the construction scenario. The core variables include target variables and influencing variables. Target variables include time dimension, cost dimension, and carbon dimension. Influencing variables include process parameters, resource status, environmental factors, and management behavior.

Images