A construction engineering progress cost dynamic early warning method and device

Abstract

The application relates to the technical field of intelligent management of construction engineering, and provides a cost dynamic early warning method and device for construction engineering progress, which constructs a mapping relationship between BIM components and a bill of quantities and between bill of quantities data and resource consumption mapping data through real-time collection of multi-source data; a component completion state is recognized through a neural network, and an engineering quantity deviation, a construction progress deviation, a resource consumption deviation and a completed work mapping deviation are determined in combination with the mapping relationship, then a progress resource linkage risk index is calculated and early warning is generated. Therefore, traditional progress cost split management is upgraded to linkage analysis, and the real-time performance, refinement and intelligent level of the construction engineering progress cost analysis are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent management technology for construction projects, and in particular to a method and device for dynamic cost early warning of construction project progress. Background Technology

[0002] As construction projects expand in scale, they become increasingly longer, involve more stakeholders, and involve more complex data types. Management processes involve various information categories, including construction progress, quantity calculation, cost expenditure, material procurement, labor and machinery input, and contract milestones. Progress and cost management, as crucial aspects of construction project control, directly impact investment control, project deadlines, and economic benefits. However, current progress management typically relies on manual data entry by construction units, supervision units, or project management personnel through on-site inspections, construction logs, and weekly / monthly reports. Management personnel then compile and compare these reports against the planned schedule. While this method can reflect project progress to some extent, it suffers from issues such as delayed data collection, strong subjectivity, and inconsistent statistical standards. It struggles to accurately reflect the true progress status of the construction site, especially in scenarios involving multiple buildings, disciplines, and processes operating concurrently. Relying solely on manual statistics makes it difficult to precisely determine the actual completion status of specific components, floors, construction areas, or sub-items of the project.

[0003] In terms of cost management, existing methods mostly rely on phased accounting based on bills of quantities, contract prices, material procurement records, labor and machinery shifts, and financial expenditure data. This type of cost analysis typically occurs after construction has begun, exhibiting clear post-construction statistical characteristics. When project progress lags behind, resource input increases abnormally, material prices fluctuate, or construction changes are frequent, existing systems often struggle to promptly assess the impact of these factors on cost and output deviations. This leads to untimely detection of cost risks and hinders dynamic early warning and proactive control. Furthermore, in existing construction project management systems, progress management and cost management are typically independent. The progress system primarily focuses on planned completion rates, schedule milestones, and construction task status, while the cost system mainly focuses on budget, contract, procurement, and expenditure data. There is a lack of effective correlation and mapping mechanisms between the two. Even if project delays occur on-site, it is difficult to automatically translate them into corresponding quantity, output, and cost deviations; even if cost expenditures increase abnormally, it is difficult to quickly trace whether they are caused by factors such as construction delays, unreasonable resource input, project changes, or rework.

[0004] In recent years, BIM technology, Internet of Things technology, artificial intelligence technology and digital engineering management platforms have been gradually applied in the field of construction engineering. BIM models can represent the spatial positions, component attributes and engineering quantity information of building components, while artificial intelligence technology can identify and analyze data such as construction site images, videos and text records, providing a technical basis for automatic judgment of construction status. Relevant studies have also proposed that BIM plus AI digital twin technology can realize dynamic output value analysis and optimal resource allocation through BIM component-level semantic analysis, multimodal time-series prediction, and cross-modal association between construction progress and equipment status. However, most existing technologies focus on single functional modules, such as construction progress display, video recognition, safety early warning, cost statistics or BIM visual management, and have not fully solved the problem of automatic association between actual construction progress recognition results and BIM components, bill of quantities and bill of costs. It is difficult to form a continuous analysis chain from actual construction status to BIM components, engineering quantities, cost items and then to risk early warning, resulting in the lack of linkage analysis capability between schedule deviation and cost deviation. SUMMARY OF THE INVENTION

[0005] The present invention provides a dynamic cost early warning method and device for construction engineering progress, which establishes a linkage analysis mechanism among actual construction progress, BIM components, bill of quantities and bill of costs, thereby improving the accuracy and real-time performance of construction engineering progress cost analysis.

[0006] In a first aspect, the present invention provides a dynamic cost early warning method for construction engineering progress, comprising:

[0007] Acquiring construction site image data, on-site video data, construction log data, schedule plan data, BIM model data, bill of quantities data, preset resource consumption mapping data and resource input data in real time;

[0008] Parsing the BIM model data to obtain a BIM component set, and establishing a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data;

[0009] Processing the construction site image data, the on-site video data and the construction log data respectively by using a plurality of preset neural network models, extracting and fusing corresponding modal features to obtain a comprehensive construction completion state of each BIM component;

[0010] Determining an actual completed engineering quantity of each bill of quantities item based on the comprehensive construction completion state and the first mapping relationship, and comparing the actual completed engineering quantity with a planned completed engineering quantity in the schedule plan data to obtain engineering quantity deviation and construction progress deviation;

[0011] Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, calculate the completed work resource mapping value and the planned work resource mapping value for each list item to determine the resource consumption deviation and the completed work mapping deviation.

[0012] Based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, a progress-resource linkage risk index is calculated; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0013] Optionally, the BIM model data is parsed to obtain a set of BIM components, and a first mapping relationship is established between each BIM component in the set of BIM components and the bill of quantities data, and a second mapping relationship is established between the bill of quantities data and the preset resource consumption mapping data, including:

[0014] The BIM model data is parsed to extract the component number, component type, component spatial location, construction area to which the component belongs, and component quantity attributes of each component, and a BIM component set is generated.

[0015] Construct a first mapping matrix between the BIM component set and the bill of quantities item set. Iterate through each element in the first mapping matrix and assign a value of 1 to the matrix element corresponding to the BIM component belonging to the same bill of quantities item, otherwise assign a value of 0 to form the first mapping relationship.

[0016] Construct a second mapping matrix between the set of bill of quantities items and the preset resource consumption mapping item set. Iterate through each element in the second mapping matrix and assign the matrix element corresponding to the bill of quantities item belonging to the same resource consumption mapping item to the corresponding unit consumption association value; otherwise, assign the value to 0 to form the second mapping relationship.

[0017] Optionally, multiple preset neural network models are used to process the construction site image data, the site video data, and the construction log data respectively, extracting and fusing corresponding modal features to obtain the comprehensive construction completion status of each BIM component, including:

[0018] The YOLOv8 object detection model is used to extract the image feature vector of the construction site image data, the ResNet50 classification model is used to extract the temporal change feature vector of the site video data, and the bidirectional long short-term memory network model is used to extract the semantic feature vector of the construction log data.

[0019] The image feature vector, temporal change feature vector, and semantic feature vector are aligned by timestamp and concatenated according to preset fusion weights to obtain a fused feature vector. The fused feature vector is then input to the fully connected layer, and the comprehensive construction completion status of each BIM component is output after being mapped by an activation function.

[0020] Optionally, based on the comprehensive construction completion status and the first mapping relationship, the actual completed work quantity for each item in the bill of quantities is determined, and the actual completed work quantity is compared with the planned completed work quantity in the schedule data to obtain the work quantity deviation and construction schedule deviation, including:

[0021] Based on the first mapping relationship, determine the subset of BIM components associated with each of the bill of quantities items;

[0022] Obtain the quantity attributes of each BIM component in the BIM component subset and the completion percentage value of the corresponding component in the comprehensive construction completion status;

[0023] The quantity attributes of each BIM component are multiplied and summed with their corresponding completion values ​​to obtain the actual completed quantity of the corresponding bill of quantities item.

[0024] The actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation.

[0025] Optionally, the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation, including:

[0026] Calculate the difference between the actual completed work volume and the planned completed work volume to obtain the single work volume deviation corresponding to each item in the bill of quantities;

[0027] The single quantity deviation of each item in the bill of quantities is weighted and summed according to the preset weight of each item in the bill of quantities to obtain the quantity deviation.

[0028] The actual completion rate is calculated as the ratio of the actual completed work volume to the total work volume for each item in the bill of quantities. The planned completion rate is calculated as the ratio of the planned completed work volume to the total work volume. The difference between the actual completion rate and the planned completion rate is taken as the construction progress deviation of the item.

[0029] The construction progress deviation is obtained by weighting and summing the construction progress deviations of each project according to the preset weights of each project.

[0030] Optionally, based on the actual completed work volume, the second mapping relationship, and the preset unit price mapping coefficient, the completed work resource mapping value and the planned work resource mapping value for each item in the bill of quantities are calculated to determine the resource consumption deviation and the completed work mapping deviation, including:

[0031] Obtain the preset unit price mapping coefficient for engineering quantities and the preset resource consumption mapping coefficient;

[0032] Multiply the actual completed work quantity of each item in the bill of quantities with the corresponding unit price mapping coefficient and sum them up to obtain the completed work resource mapping value. Multiply the planned completed work quantity of each item with the corresponding unit price mapping coefficient and sum them up to obtain the planned work resource mapping value. Calculate the difference between the completed work resource mapping value and the planned work resource mapping value to obtain the completed work mapping deviation.

[0033] The actual completed work volume of each item in the bill of quantities is multiplied by the corresponding resource consumption mapping coefficient and summed to obtain the actual resource consumption mapping value. The planned completed work volume of each item is multiplied by the corresponding resource consumption mapping coefficient and summed to obtain the planned resource consumption mapping value. The difference between the actual resource consumption mapping value and the planned resource consumption mapping value is calculated to obtain the resource consumption deviation.

[0034] Optionally, the resource input anomaly index includes a labor input anomaly index, a material consumption anomaly index, and a machinery usage anomaly index; based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, a progress-resource linkage risk index is calculated, including:

[0035] Based on the deviation of the ratio between the actual and planned labor input in the resource input data, a labor input anomaly index is determined;

[0036] Based on the deviation of the ratio between actual material consumption and planned material consumption in the resource input data, a material consumption anomaly index is determined;

[0037] Based on the deviation of the ratio between actual and planned machinery usage in the resource input data, a machinery usage anomaly index is determined;

[0038] The construction progress deviation, resource consumption deviation, completed work mapping deviation, abnormal labor input index, abnormal material consumption index, and abnormal machinery usage index are weighted and summed to obtain the progress-resource linkage risk index.

[0039] Optionally, after calculating the schedule-resource linkage risk index based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, the method further includes:

[0040] When the risk index reaches a preset medium or high risk level, an early warning signal is generated, specifically as follows:

[0041] If the deviation originates from a construction delay and a resource consumption mapping value that is ahead of schedule, a resource reconfiguration warning will be generated.

[0042] If the source of the deviation is abnormal material consumption, a supply plan adjustment warning will be generated.

[0043] If the deviation originates from a completed work mapping deviation being lower than a preset node value, a process optimization warning is generated.

[0044] If the deviation originates from the proximity of a preset schedule constraint node and insufficient actual completed work, a schedule performance risk warning will be generated.

[0045] Secondly, the present invention provides a dynamic cost early warning device for construction project progress, comprising:

[0046] The acquisition module is used to acquire construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data in real time.

[0047] The mapping relationship establishment module is used to parse the BIM model data to obtain a BIM component set, establish a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data.

[0048] The fusion module is used to process the construction site image data, the site video data and the construction log data using multiple preset neural network models, extract the corresponding modal features and fuse them to obtain the comprehensive construction completion status of each BIM component.

[0049] The first deviation determination module is used to determine the actual completed work volume of each bill of quantities item based on the comprehensive construction completion status and the first mapping relationship, and compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation.

[0050] The second deviation determination module is used to calculate the completed work resource mapping value and the planned work resource mapping value for each list item based on the actual completed work quantity, the second mapping relationship and the preset work quantity unit price mapping coefficient, so as to determine the resource consumption deviation and the completed work mapping deviation.

[0051] The risk index determination module is used to calculate the progress-resource linkage risk index based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0052] Thirdly, the present invention provides an electronic device including a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the steps of the method provided in the first aspect above.

[0053] Fourthly, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0054] Fifthly, the present invention provides a computer program product comprising a computer program that, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0055] As can be seen from the above technical solutions, the present invention has the following advantages:

[0056] This invention provides a method and apparatus for dynamic cost early warning of construction project progress. The method includes: real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data; parsing the BIM model data to obtain a set of BIM components, establishing a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data; processing the construction site image data, the on-site video data, and the construction log data using preset multiple neural network models, extracting and fusing corresponding modal features to obtain the comprehensive construction completion of each BIM component. The system is as follows: Based on the comprehensive construction completion status and the first mapping relationship, the actual completed work volume of each bill of quantities item is determined, and the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation; Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, the completed work resource mapping value and the planned work resource mapping value of each bill of quantities item are calculated to determine the resource consumption deviation and the completed work mapping deviation; According to the construction schedule deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, the schedule resource linkage risk index is calculated; The risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level. Through real-time collection of multi-source data, a mapping relationship is constructed between BIM components and the bill of quantities, and between the bill of quantities data and the resource consumption mapping data; The completion status of components is identified through neural networks, and the work volume deviation, construction schedule deviation, resource consumption deviation, and completed work mapping deviation are determined in combination with the mapping relationship, thereby calculating the schedule resource linkage risk index and generating an early warning. This upgrades the traditional fragmented management of schedule and cost to a linked analysis, significantly improving the real-time, refined, and intelligent level of schedule and cost analysis in construction projects. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart illustrating the steps of a method for dynamic cost early warning of construction project progress according to the present invention.

[0059] Figure 2 This is a flowchart illustrating the second embodiment of the cost dynamic early warning method for construction project progress according to the present invention.

[0060] Figure 3 This is a structural block diagram of an embodiment of a cost dynamic early warning device for construction project progress according to the present invention;

[0061] Figure 4 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0062] This invention provides a method and apparatus for dynamic cost early warning of construction project progress, establishing a linkage analysis mechanism between actual construction progress, BIM components, bill of quantities, and cost list, thereby improving the accuracy and real-time performance of construction project progress cost analysis.

[0063] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0064] Example 1

[0065] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of a method for dynamic cost early warning of construction project progress according to an embodiment of the present invention. The method includes:

[0066] Step S101: Real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data;

[0067] In this embodiment of the application, real-time data collection is performed on construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data, and the collected data is timestamped and preprocessed.

[0068] Step S102: Parse the BIM model data to obtain a BIM component set, establish a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data;

[0069] In this embodiment of the application, the BIM model is parsed to extract the component number, type, spatial location, region, and quantity attributes to generate a set of BIM components; a first mapping matrix between BIM components and bill of quantities items, and a second mapping matrix between bill of quantities items and cost list items are constructed to form an association chain.

[0070] Step S103: The construction site image data, the on-site video data and the construction log data are processed by using multiple preset neural network models respectively, the corresponding modal features are extracted and fused to obtain the comprehensive construction completion status of each BIM component;

[0071] In this embodiment, multiple preset neural network models are used to extract image features from construction site image data, video temporal features from on-site video data, and construction log semantic features from construction log data. The three modal features are then aligned by timestamps and fused according to preset weights. The comprehensive construction completion status of each BIM component is then output through a fully connected layer mapping.

[0072] Step S104: Based on the comprehensive construction completion status and the first mapping relationship, determine the actual completed work volume of each bill of quantities item, and compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation.

[0073] In this embodiment of the application, based on the first mapping relationship, the quantity attributes of each BIM component are multiplied and summed with the completion value to obtain the actual completed quantity of each bill of quantities item; the actual completed quantity is compared with the planned completed quantity to calculate the single quantity deviation, and the overall quantity deviation is obtained by weighting and summing according to the weight of the bill of quantities item; at the same time, the difference between the actual completion rate and the planned completion rate of each item is calculated as the construction progress deviation, and then the overall construction progress deviation is obtained.

[0074] Step S105: Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, calculate the completed work resource mapping value and the planned work resource mapping value for each list item to determine the resource consumption deviation and the completed work mapping deviation.

[0075] In this embodiment of the application, the actual completed work volume and the planned completed work volume of each item in the list are multiplied by the corresponding comprehensive unit price of the contract, and the sum is used to obtain the resource mapping value of the completed work and the resource mapping value of the planned work. The difference between the two is the completed work mapping deviation. The actual completed work volume is multiplied by the actual unit cost to obtain the actual resource consumption mapping value, and the resource consumption deviation is obtained by comparing it with the planned resource consumption mapping value.

[0076] Step S106: Calculate the progress-resource linkage risk index based on the construction progress deviation, resource consumption deviation, completed work mapping deviation, and resource input anomaly index in the resource input data; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0077] In this embodiment, anomaly indices are calculated based on the deviation of the ratios of actual input of labor, materials, and machinery from planned input, and then weighted and summed to obtain a resource input anomaly index. The construction progress deviation, resource consumption deviation, completed work mapping deviation, and resource input anomaly index are weighted and summed to calculate a progress-resource linkage risk index. This index is compared with a preset threshold to generate corresponding early warning levels and dynamic early warning information.

[0078] This invention provides a method for dynamic cost early warning of construction project progress, comprising: real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data; parsing the BIM model data to obtain a set of BIM components, establishing a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data; processing the construction site image data, the on-site video data, and the construction log data using preset multiple neural network models, extracting and fusing corresponding modal features to obtain the comprehensive construction completion of each BIM component. The system is as follows: Based on the comprehensive construction completion status and the first mapping relationship, the actual completed work volume of each bill of quantities item is determined, and the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation; Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, the completed work resource mapping value and the planned work resource mapping value of each bill of quantities item are calculated to determine the resource consumption deviation and the completed work mapping deviation; According to the construction schedule deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, the schedule resource linkage risk index is calculated; The risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level. Through real-time collection of multi-source data, a mapping relationship is constructed between BIM components and the bill of quantities, and between the bill of quantities data and the resource consumption mapping data; The completion status of components is identified through neural networks, and the work volume deviation, construction schedule deviation, resource consumption deviation, and completed work mapping deviation are determined in combination with the mapping relationship, thereby calculating the schedule resource linkage risk index and generating an early warning. This upgrades the traditional fragmented management of schedule and cost to a linked analysis, significantly improving the real-time, refined, and intelligent level of schedule and cost analysis in construction projects.

[0079] Example 2

[0080] Please see Figure 2 , Figure 2 This is a flowchart illustrating a second embodiment of the cost dynamic early warning method for construction project progress according to the present invention. The steps include:

[0081] Step S201: Real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data;

[0082] This embodiment collects construction site image and video data using equipment such as tower crane cameras, floor-mounted cameras, and drone aerial photography. It simultaneously acquires schedule data, BIM model data, bill of quantities data, cost list (i.e., pre-defined resource consumption mapping data), and resource input data including material procurement ledgers, labor attendance ledgers, and machinery usage ledgers. All data undergoes preprocessing such as timestamp alignment, deduplication, ROI cropping, brightness normalization, log structuring, and standardization.

[0083] Step S202: Parse the BIM model data, extract the component number, component type, component spatial location, construction area to which the component belongs, and component quantity attributes of each component, and generate a BIM component set;

[0084] In this embodiment, the BIM component set is assumed to be... In the formula, For the first A BIM component, This represents the total number of BIM components. Each component... The attribute is represented as ,in Assign component numbers, For component type, For spatial location, Construction area For component quantities, This is the construction phase.

[0085] Step S203: Construct a first mapping matrix between the BIM component set and the bill of quantities item set; traverse each element in the first mapping matrix; assign a value of 1 to the matrix element corresponding to the BIM component belonging to the same bill of quantities item, otherwise assign a value of 0, thus forming the first mapping relationship.

[0086] In this embodiment of the application, the set of bill of quantities items is assumed to be... In the formula, For the first Each item in the bill of quantities; This represents the quantity of items in the bill of quantities. First mapping matrix. for Matrix, where =1 indicates the first The BIM component belongs to the first The first mapping matrix is ​​0 if it is a bill of quantities item. elements in Represented as:

[0087] .

[0088] This establishes the first mapping relationship between BIM components and bill of quantities items.

[0089] Step S204: Construct a second mapping matrix between the bill of quantities item set and the preset resource consumption mapping item set. Traverse each element in the second mapping matrix and assign the matrix element corresponding to the bill of quantities item belonging to the same resource consumption mapping item to the corresponding unit consumption association value; otherwise, assign the value to 0 to form the second mapping relationship.

[0090] In this embodiment of the application, the set of preset resource consumption mapping items (i.e., cost list items) is set as follows: The second mapping matrix for Matrix, where Indicates the first The bill of quantities item and the first The unit consumption associated value (usually the comprehensive unit price or unit cost) of each resource consumption mapping project. The second mapping matrix represents the number of items in the cost list. elements in Represented as:

[0091] ;

[0092] in, Indicates the first The bill of quantities item and the first The unit consumption correlation value (such as the comprehensive unit price or unit cost mapping coefficient) between resource consumption mapping items.

[0093] This establishes a complete association mapping relationship between BIM components and pre-defined resource consumption mapping projects:

[0094] ;

[0095] in, This establishes the indirect mapping relationship between BIM components and pre-defined resource consumption mapping items. Through the construction of the above two-level mapping matrix, the completion status of components identified on the construction site can be progressively transferred to the quantity of work and resource consumption mapping items.

[0096] Step S205: Use the YOLOv8 object detection model to extract the image feature vector of the construction site image data, use the ResNet50 classification model to extract the temporal change feature vector of the site video data, and use the bidirectional long short-term memory network model to extract the semantic feature vector of the construction log data.

[0097] In this embodiment, YOLOv8 takes image data as input, identifies targets such as templates, reinforcing bars, pump pipes, and pouring surfaces, and outputs the image completion rate of each component. The ResNet50 takes a component ROI image as input and identifies four construction states: unconstructed, rebar / formwork stage, pouring stage, and completed. It then outputs the video temporal completion score. The BiLSTM takes recent construction log text features, weather and calendar features as input, and outputs the log semantic completeness. .

[0098] Step S206: Align the image feature vector, temporal change feature vector, and semantic feature vector according to timestamps, and concatenate them according to preset fusion weights to obtain a fused feature vector. Input the fused feature vector into the fully connected layer, and output the comprehensive construction completion status of each BIM component through activation function mapping.

[0099] In this embodiment of the application, let the first Each BIM component at time The actual construction completion status is as follows:

[0100] ;

[0101] in, The first one obtained by the artificial intelligence model Each BIM component at time The actual completion status of the construction; For parameters Artificial intelligence recognition model; Indicates the relationship with the first On-site image features related to each component; In order to be with the first Video temporal features related to each component; Indicates the relationship with the first Features of construction log text related to each component.

[0102] Component integrated construction completion status satisfy Its value means: when When 0 = 0, it indicates that construction of the component has not yet begun; when 0 < When <1, it indicates that the component is under construction; when When the value is 1, it indicates that the construction of the component is complete.

[0103] For components with multiple construction stages, the overall construction completion status of the component can be further represented as a stage completion vector:

[0104] ;

[0105] in, For the first The first component The completion status of each construction phase, and the completion status of each phase also satisfies .

[0106] To improve the reliability of the recognition results, image recognition results, video recognition results, and construction log recognition results can be fused to obtain a comprehensive construction completion status:

[0107] ;

[0108] in, The completion status is based on on-site image recognition. The completion status is based on on-site video recognition. The completion status is obtained based on the text recognition of the construction log. , The fusion weights for image, video, and text data are respectively, satisfying... + + =1, and , ≥0.

[0109] Using the above method, the overall construction completion status of each BIM component constitutes a vector. .

[0110] Step S207: Based on the comprehensive construction completion status and the first mapping relationship, determine the actual completed work volume of each bill of quantities item, and compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation.

[0111] In this embodiment of the application, based on the first mapping relationship, a subset of BIM components associated with each of the bill of quantities items is determined; the quantity attributes of each BIM component in the subset of BIM components and the completion degree value of the corresponding component in the comprehensive construction completion status are obtained; the quantity attributes of each BIM component and the corresponding completion degree value are multiplied and summed to obtain the actual completed quantity of the corresponding bill of quantities item.

[0112] Simultaneously, the difference between the actual completed work volume and the planned completed work volume is calculated to obtain the single work volume deviation corresponding to each item in the bill of quantities; the single work volume deviations of each item are weighted and summed according to the preset weights of each item in the bill of quantities to obtain the work volume deviation; the ratio of the actual completed work volume to the total work volume of each item in the bill of quantities is calculated as the actual completion rate, and the ratio of its planned completed work volume to the total work volume is calculated as the planned completion rate; the difference between the actual completion rate and the planned completion rate is taken as the construction progress deviation of the item; the construction progress deviations of each item are weighted and summed according to the preset weights of each item to obtain the construction progress deviation.

[0113] In this embodiment of the application, based on the first mapping relationship , No. The actual completed quantity of each item in the bill of quantities at time tt for:

[0114] ;

[0115] in, For the first The quantity of work corresponding to each BIM component.

[0116] Based on the schedule data, obtain the first... Each item in the bill of quantities is at time Planned completion volume Then the first The quantity deviation for each item in the bill of quantities is:

[0117] ;

[0118] when When Δ < 0, it indicates that the actual completed work is less than the planned work, posing a risk of schedule delay; when A value greater than 0 indicates that the actual amount of work completed is higher than the planned amount of work completed, suggesting that the project is ahead of schedule or that resources have been invested ahead of schedule.

[0119] No. The completion rate of the quantities for each item in the bill of quantities is:

[0120] ;

[0121] in, For the first Each item in the bill of quantities is at the time The completion rate of the project. For the first The total quantity of work for each item in the bill of quantities.

[0122] Furthermore, the first Each item in the bill of quantities is at the time The planned progress completion rate and the actual progress completion rate are respectively:

[0123] ;

[0124] ;

[0125] in, To achieve the planned progress completion rate, This represents the completion rate of the planned schedule.

[0126] Then the first The construction progress deviation for each item on the list is:

[0127] ;

[0128] in, For the first Schedule deviations for each item in the bill of quantities. A value less than 0 indicates a delay in progress. =0 indicates that the progress is consistent. A value greater than 0 indicates that the progress is ahead of schedule.

[0129] Overall construction progress deviations are calculated based on the preset weights of each item in the bill of quantities. Weighted summation:

[0130] ;

[0131] in, This is due to a deviation in the construction schedule.

[0132] Step S208: Obtain the preset unit price mapping coefficient for engineering quantity and the preset resource consumption mapping coefficient;

[0133] Based on the actual completed work volume, planned completed work volume, and preset contract unit price for each item in the bill of quantities, calculate the resource mapping value for completed work, the resource mapping value for planned work, and the mapping deviation for completed work. Let the first... The comprehensive unit price of the contract for each item in the bill of quantities is: This unit price is a preset unit price mapping coefficient for the project quantity. Therefore, at time... , No. The actual completed output value of each item in the bill of quantities is:

[0134] ;

[0135] The planned output value is:

[0136] ;

[0137] The completed work resource mapping value is the sum of the actual completed output value of all listed items, that is:

[0138] ;

[0139] The planned work resource mapping value (planned value PV(t)) is the sum of the planned output values ​​of all listed items:

[0140] ;

[0141] The completed work mapping deviation is:

[0142] .

[0143] when When the value is less than 0, it indicates that the actual output value of the project is lower than the planned output value, which may indicate problems such as delayed progress, insufficient output value recognition, or low efficiency of resource input.

[0144] Step S209: Multiply the actual completed work quantity of each item in the bill of quantities with the corresponding resource consumption mapping coefficient and sum them up to obtain the actual resource consumption mapping value; multiply the planned completed work quantity of each item with the corresponding resource consumption mapping coefficient and sum them up to obtain the planned resource consumption mapping value; calculate the difference between the actual resource consumption mapping value and the planned resource consumption mapping value to obtain the resource consumption deviation.

[0145] In this embodiment of the application, let the first The unit cost of each resource consumption mapping item (cost list item) is: The unit cost is a preset resource consumption mapping coefficient. Based on the second mapping matrix... , No. Each cost item is at a certain time. The actual accrued costs are:

[0146] ;

[0147] The planned cost is:

[0148] ;

[0149] Taking into account fluctuations in material prices, changes in labor costs, and changes in machine operating costs, the unit cost can be dynamically adjusted as follows:

[0150] ;

[0151] The actual accrued cost at this point is:

[0152] ;

[0153] No. The resource consumption deviation (cost deviation) for each cost item is:

[0154] ;

[0155] The actual resource consumption mapping value is the sum of the actual accrued costs of all cost list items:

[0156] ;

[0157] The planned resource consumption mapping value is the sum of the planned costs of all cost list items:

[0158] ;

[0159] The overall resource consumption deviation of the project is:

[0160] .

[0161] when A value greater than 0 indicates a risk of overspending; when A value less than 0 indicates that the cost is lower than the planned cost.

[0162] Step S210: Calculate the progress-resource linkage risk index based on the construction progress deviation, resource consumption deviation, completed work mapping deviation, and resource input anomaly index in the resource input data; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0163] In this embodiment of the application, the abnormal resource input index includes the abnormal labor input index, the abnormal material consumption index, and the abnormal machinery usage index.

[0164] In specific implementation, a labor input anomaly index is determined based on the deviation of the ratio of actual labor input to planned labor input in the resource input data; a material consumption anomaly index is determined based on the deviation of the ratio of actual material consumption to planned material consumption in the resource input data; a machinery usage anomaly index is determined based on the deviation of the ratio of actual machinery usage to planned machinery usage in the resource input data; and the construction progress deviation, resource consumption deviation, completed work mapping deviation, labor input anomaly index, material consumption anomaly index, and machinery usage anomaly index are weighted and summed to obtain the progress-resource linkage risk index.

[0165] Based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, a progress-resource linkage risk index is calculated; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0166] Resource input anomaly index includes labor input anomaly index Abnormal Material Consumption Index and abnormal index of machinery use These are defined as the relative deviations between the actual input and the planned input, respectively:

[0167] ;

[0168] ;

[0169] ;

[0170] in, and These are the actual labor input and the planned labor input, respectively. and These are the actual material consumption and the planned material consumption, respectively. and These are the actual amount of machinery used and the planned amount of machinery used, respectively.

[0171] Resource input anomaly index The weighted sum of the three types of outlier indices:

[0172] ;

[0173] in , , These represent the weights of abnormal inputs in labor, materials, and machinery, respectively, satisfying... + + =1, and .

[0174] Schedule and resource linkage risk index for:

[0175] ;

[0176] in, This represents the absolute value of the overall project schedule deviation, reflecting the severity of the construction schedule delay. This is the normalized value of the resource consumption deviation, reflecting the relative extent of cost overruns; This is the normalized value of the mapping deviation of completed work, reflecting the relative magnitude of the output loss; An abnormal index of resource input; , , , Let be the weighting coefficient, satisfying + + + =1.

[0177] Step S211: When the risk index reaches the preset medium or high risk level, an early warning signal is generated.

[0178] In this embodiment of the application, if the source of the deviation is a delay in construction progress and an advance in resource consumption mapping value, a resource reconfiguration warning is generated; if the source of the deviation is abnormal material consumption, a supply plan adjustment warning is generated; if the source of the deviation is that the completed work mapping deviation is lower than the preset node value, a process optimization warning is generated; if the source of the deviation is that the preset construction period constraint node is approaching and the actual completed work volume is insufficient, a construction period performance risk warning is generated.

[0179] In the specific implementation, a warning signal is generated when the risk index reaches a preset medium or high risk level. A preset low-risk threshold is also included. Medium risk threshold High-risk threshold ,satisfy Warning level Defined as:

[0180] ;

[0181] When schedule cost risks exist, corresponding adjustment suggestions are generated based on the source of the deviation. If the source of the deviation is a construction schedule lag and a resource consumption mapping value that is ahead (i.e., and This generates a resource reconfiguration warning, suggesting that the shift personnel of the lagging processes be reassigned to the critical path, and that the backlog of workfaces be released first; if the source of the deviation is abnormal material consumption (i.e. If the deviation exceeds a preset threshold, a supply plan adjustment warning will be generated, suggesting a check of material inventory and procurement plans to avoid secondary shutdowns due to material shortages; if the deviation originates from a deviation in completed work that is lower than a preset node value (i.e., (If the deviation is severe and negative), a process optimization warning will be generated, suggesting that the construction process be re-examined, the flow operation be reasonably arranged, and the utilization rate of working hours be improved. If the source of the deviation is that the preset schedule constraint node is approaching and the actual amount of work completed is insufficient, a schedule performance risk warning will be generated, suggesting that the impact of the schedule delay be assessed in a timely manner, and measures to catch up on the work be formulated or the schedule adjustment be negotiated with the construction unit.

[0182] This invention provides a method for dynamic cost early warning of construction project progress, comprising: real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data; parsing the BIM model data to obtain a set of BIM components, establishing a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data; processing the construction site image data, the on-site video data, and the construction log data using preset multiple neural network models, extracting and fusing corresponding modal features to obtain the comprehensive construction completion of each BIM component. The project is divided into three phases: 1) A completion status is established; 2) Based on the comprehensive construction completion status and the first mapping relationship, the actual completed work volume of each bill of quantities item is determined, and the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction progress deviation; 3) Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, the completed work resource mapping value and the planned work resource mapping value of each bill of quantities item are calculated to determine the resource consumption deviation and the completed work mapping deviation; 4) According to the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, a progress resource linkage risk index is calculated; 5) The risk index is compared with a preset threshold to obtain dynamic early warning information of the corresponding level. 6) Based on the first mapping matrix, the completion degree is mapped level by level to the actual completed work volume of the bill of quantities, and the work volume deviation and construction progress deviation are obtained by comparing with the planned quantity; 7) Through the second mapping matrix and the comprehensive unit price, the completed work resource mapping value, the planned work resource mapping value, and the actual resource consumption mapping value are automatically calculated to obtain the completed work mapping deviation and resource consumption deviation. Simultaneously, based on the deviation between actual and planned inputs of labor, materials, and machinery, three types of abnormal indices are calculated. These indices are then weighted and integrated with schedule deviation, resource deviation, and completed work mapping deviation to obtain a schedule-resource linkage risk index, generating multi-level early warnings and differentiated solution strategies. This upgrades traditional fragmented schedule and cost management to a linked analysis, significantly improving the real-time, refined, and intelligent level of schedule and cost analysis in construction projects.

[0183] Example 3

[0184] This embodiment employs a method and system for dynamic cost early warning of construction project progress as described in this invention. It identifies the actual progress of the construction site and associates the identification results with BIM components, bill of quantities, and cost list to achieve dynamic analysis of progress deviations, cost deviations, and risk levels.

[0185] This example uses a high-rise residential project in Guangzhou as a case study. The project employs a shear wall structure, with one basement level and 18 floors above ground, totaling a building area of ​​24,860 m². The analysis focuses on the construction phase of the 6th floor's main structure, specifically including the walls, beams, and slabs of sections A and B on the 6th floor. This application case includes 6 BIM component groups, 3 bill of quantities items, and 3 cost list items.

[0186] (1) Data acquisition and preprocessing

[0187] The system first collects images, videos, construction logs, BIM models, material procurement ledgers, labor attendance data, and machinery usage ledgers from the construction site to form a multi-source data set required for the progress and cost analysis of the construction project.

[0188] The system uses 1080p video from tower crane cameras, extracting one frame every 10 minutes; four fixed cameras on each floor use four video feeds, extracting one frame every 15 minutes; construction logs are collected daily; the BIM model is synchronized weekly; and material, labor, and machinery ledgers are synchronized daily. The system aligns the above data with unified timestamps and performs deduplication, blurry frame removal, ROI clipping, brightness normalization, log structuring, unified component numbering, and unified cost data standards.

[0189] (2) Mapping of BIM components and bill of quantities

[0190] In this embodiment, the 6th floor main structure is divided into 6 BIM component groups, namely:

[0191] Shear wall in zone 6F-A, shear wall in zone 6F-B, beam in zone 6F-A, beam in zone 6F-B, slab in zone 6F-A, and slab in zone 6F-B.

[0192] The BIM statistical quantities for each BIM component group are as follows:

[0193]

[0194] Set the bill of quantities as follows:

[0195] ;

[0196] The mapping matrix between BIM components and the bill of quantities is as follows:

[0197] ;

[0198] Through the above mapping relationship, the system can map the component completion status identified by artificial intelligence to the completion status of engineering quantity and cost completion status step by step.

[0199] (3) Configuration of AI progress recognition model

[0200] This embodiment employs a combined structure of "YOLOv5s object detection model + ResNet50 stage classification model + BiLSTM temporal fusion model" to identify the completion status of components at construction sites. The model configuration provided in the document includes: YOLOv5s input of a 640×640 image for identifying targets such as formwork, rebar, pump pipes, pouring surfaces, workers, pump trucks, and material stacks; ResNet50 input of a 224×224 component ROI for identifying four construction states: not under construction, rebar / formwork stage, pouring stage, and completed; and BiLSTM input of the stage probabilities of the past 7 days, log vectors, and weather / calendar features for outputting the component completion status.

[0201] The component completion degree is calculated using the following multimodal fusion formula:

[0202] ;

[0203] In the formula, Indicates the first Each BIM component at time Overall completion rate; This indicates the degree of completion obtained from image recognition; This indicates the completion rate of video temporal recognition. This indicates the completion rate of the construction log text recognition. , , These represent the fusion weights for images, videos, and log text, respectively.

[0204] In this embodiment, the following is taken:

[0205] ;

[0206] With components For example, if the image recognition completion rate is 0.72, the video recognition completion rate is 0.68, and the log recognition completion rate is 0.70, then:

[0207] ;

[0208] With components For example, if the image recognition completion rate is 0.25, the video recognition completion rate is 0.35, and the log recognition completion rate is 0.30, then:

[0209] ;

[0210] After combining the 6 component groups, the component completion vector is obtained:

[0211] ;

[0212] The component quantity vector is:

[0213] ;

[0214] Therefore, the completed work volume for each component is:

[0215] ;

[0216] The above results indicate that: 60m³ of shear wall in zone 6F-A has been completed, 42m³ of shear wall in zone 6F-B has been completed, 28m³ of beams in zone 6F-A has been completed, 10.5m³ of beams in zone 6F-B has been completed, 32m³ of slabs in zone 6F-A has been completed, and slabs in zone 6F-B are not yet completed.

[0217] (4) Calculation of actual completed work volume and schedule deviation

[0218] Mapping the completed quantities of each component to the bill of quantities yields:

[0219] ;

[0220] Right now That is, the wall work volume is 6m³ ahead of schedule, the beam work volume is 10.5m³ behind, and the slab work volume is 64m³ behind.

[0221] Furthermore, the total workload of the three categories of items is as follows:

[0222] ;

[0223] The planned completion rate is:

[0224] ;

[0225] The actual completion rate was:

[0226] ;

[0227] Weights are determined based on the total amount of work completed.

[0228] ;

[0229] The overall schedule deviation is:

[0230] ;

[0231] Substituting the values ​​into the equation:

[0232] ;

[0233] Therefore, in this embodiment, the overall progress of the 6th floor main structure of the project is 19.57 percentage points behind schedule.

[0234] (5) Calculation of cost deviation and output value deviation

[0235] This embodiment uses a composite unit price to calculate the planned value and earned value. The benchmark prices given in the document include: HRB400 Φ12-25 steel bars at 3015 yuan / t, C30 ordinary pumped concrete at 401 yuan / m³, and steelworkers, carpenters, and concrete workers at 312.5 yuan / man-day, 325 yuan / man-day, and 290 yuan / man-day respectively, and a composite labor unit price is constructed with weights of 0.4, 0.4, and 0.2.

[0236] The unit price for composite labor is:

[0237] ;

[0238] The formula for calculating the comprehensive unit price of the list items is:

[0239] ;

[0240] in, Indicates the first Comprehensive unit price for each item in the list; Indicates the steel reinforcement consumption coefficient; Indicates the price of steel bars; Indicates the concrete consumption coefficient; Indicates the price of concrete; This indicates the cost of templates and reusable materials; Indicates the labor consumption coefficient; This indicates the composite labor unit price; This refers to equipment costs.

[0241] In this embodiment, the comprehensive unit prices for walls, beams, and slabs are as follows:

[0242]

[0243] The corresponding unit prices for walls, beams, and slabs in the document are 1750 yuan / m³, 1680 yuan / m³, and 1520 yuan / m³, respectively.

[0244] Planned value Earning value with completed work Calculate according to the following formulas:

[0245] ;

[0246] ;

[0247] Substituting the data, we get:

[0248]

[0249] Therefore, we can conclude that:

[0250] ;

[0251] ;

[0252] Cost deviation rate:

[0253] ;

[0254] Output deviation rate:

[0255] ;

[0256] Therefore, it can be seen that although the actual costs incurred The amount has not yet exceeded the planned value. The amount is 100 yuan, but it is already significantly higher than the value of the work already completed. The figure of 52,585.3 yuan indicates a cost imbalance where the project is behind schedule while costs are ahead of schedule. Traditional monthly financial statistics may not promptly alert to such situations, but this invention can identify the cost imbalance of 52,585.3 yuan through earned value comparison.

[0257] (6) Calculation of Resource Anomaly Index and Risk Index

[0258] This embodiment further incorporates abnormal inputs of manpower, materials, and machinery into the risk assessment.

[0259] The human error index compares actual workdays with theoretical workdays. Actual workdays are:

[0260] ;

[0261] Theoretical working days are:

[0262] ;

[0263] The artificial anomaly index is:

[0264] ;

[0265] The material anomaly index is:

[0266] ;

[0267] The mechanical anomaly index is:

[0268] ;

[0269] The resource anomaly index is defined as follows:

[0270] ;

[0271] Substituting the values ​​into the equation:

[0272] ;

[0273] The schedule cost risk index is defined as follows:

[0274] ;

[0275] Substituting the values ​​into the equation:

[0276] ;

[0277] Let the risk threshold be:

[0278] ;

[0279] because It can determine that the current project is in a high-risk state of level 3 and generate a high-risk warning.

[0280] (7) Output results and adjustment suggestions

[0281] Based on the risk identification results above, the system determines that the main risk stems from the delay in the quantity of slab-related work. The deviation in the quantity of slab-related work is:

[0282] ;

[0283] Meanwhile, the artificial anomaly index reached 0.739, indicating that there may be problems such as waiting for acceptance, waiting for materials, unreasonable process connection, or unbalanced team configuration during this construction stage.

[0284] Therefore, the system automatically generated the following adjustment suggestions:

[0285] First, eight carpenters from the wall construction team and six steelworkers from the beam construction team will be transferred to the sixth floor slab area for construction within the next two days.

[0286] Second, the concealed acceptance point will be moved forward by 0.5 days to prioritize the release of the slab area pouring work surface;

[0287] Third, the daily supply of C30 concrete will be increased from 35 m³ / day to 50 m³ / day, and one additional pumping shift will be added.

[0288] Fourth, the arrival time of HRB400 steel bars will be moved forward by one day to avoid a second work stoppage in the slab area.

[0289] To verify the application effect of the present invention, this embodiment is compared with the traditional "manual weekly report + monthly financial statistics" model. The traditional model usually reports the floor-level completion rate weekly, observes the cumulative cost at the end of the month based on the payment ledger, does not perform component-level BIM mapping, and does not calculate the earned value (EV) of completed work.

[0290] The comparison results are as follows:

[0291]

[0292] In this embodiment, under node W6, the traditional model only shows "cumulative actual cost of 344,405.3 yuan is less than the planned value of 396,240 yuan," which may be judged as not yet exceeding the budget. However, after adopting the present invention, the system can identify "cumulative actual cost of 344,405.3 yuan is greater than the earned value of completed work of 291,820 yuan," thus discovering a cost imbalance of 18.02%. This result shows that the present invention does not simply replace manual weekly reports, but rather, through the linkage of "actual construction status—BIM components—work volume—earned value / cost," it makes the schedule-cost imbalance that is difficult to identify in a timely manner by traditional methods explicit in advance.

[0293] One week after implementing the system recommendations, the overall schedule deviation decreased from 19.6 percentage points to 8.2 percentage points, the cost deviation rate decreased from 18.0% to 9.5%, and the risk index decreased from 0.238 to 0.126, indicating that the present invention can dynamically identify, warn, and assist in the control of schedule and cost risks in construction projects.

[0294] Example 4

[0295] Please see Figure 3 , Figure 3 This is a structural block diagram of an embodiment of a cost dynamic early warning device for construction project progress according to the present invention. The device includes:

[0296] The acquisition module 301 is used to acquire construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data in real time.

[0297] The mapping relationship establishment module 302 is used to parse the BIM model data to obtain a BIM component set, establish a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data.

[0298] The fusion module 303 is used to process the construction site image data, the site video data and the construction log data using multiple preset neural network models, extract the corresponding modal features and fuse them to obtain the comprehensive construction completion status of each BIM component.

[0299] The first deviation determination module 304 is used to determine the actual completed work volume of each bill of quantities item based on the comprehensive construction completion status and the first mapping relationship, and compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and the construction schedule deviation.

[0300] The second deviation determination module 305 is used to calculate the completed work resource mapping value and the planned work resource mapping value of each list item based on the actual completed work quantity, the second mapping relationship and the preset work quantity unit price mapping coefficient, so as to determine the resource consumption deviation and the completed work mapping deviation.

[0301] The risk index determination module 306 is used to calculate the progress-resource linkage risk index based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

[0302] In an optional embodiment, the mapping relationship establishment module 302 includes:

[0303] The parsing submodule is used to parse the BIM model data, extract the component number, component type, component spatial location, construction area to which the component belongs, and component quantity attributes of each component, and generate a set of BIM components;

[0304] The first mapping relationship construction submodule is used to construct a first mapping matrix between the BIM component set and the bill of quantities item set. It iterates through each element in the first mapping matrix, assigns a value of 1 to the matrix element corresponding to the BIM component belonging to the same bill of quantities item, and otherwise assigns a value of 0 to form the first mapping relationship.

[0305] The second mapping relationship construction submodule is used to construct a second mapping matrix between the bill of quantities item set and the preset resource consumption mapping item set. It iterates through each element in the second mapping matrix, assigns the matrix element corresponding to the bill of quantities item belonging to the same resource consumption mapping item to the corresponding unit consumption association value, otherwise assigns the value to 0, thus forming the second mapping relationship.

[0306] In an optional embodiment, the fusion module 303 includes:

[0307] The extraction submodule is used to extract the image feature vector of the construction site image data using the YOLOv8 object detection model, extract the temporal change feature vector of the site video data using the ResNet50 classification model, and extract the semantic feature vector of the construction log data using the bidirectional long short-term memory network model.

[0308] The stitching submodule is used to align the image feature vector, temporal change feature vector, and semantic feature vector according to timestamps, and then stitch them together in series according to preset fusion weights to obtain a fused feature vector. The fused feature vector is then input to the fully connected layer, and the integrated construction completion status of each BIM component is output after being mapped by an activation function.

[0309] In an optional embodiment, the first deviation determination module 304 includes:

[0310] A subset determination submodule is constructed to determine a subset of BIM components associated with each of the bill of quantities items based on the first mapping relationship;

[0311] The completion degree value acquisition submodule is used to acquire the engineering quantity attributes of each BIM component in the BIM component subset and the completion degree value of the corresponding component in the comprehensive construction completion status.

[0312] The actual completed work quantity determination submodule is used to multiply and sum the work quantity attributes of each BIM component with the corresponding completion degree value to obtain the actual completed work quantity of the corresponding bill of quantities item.

[0313] The deviation determination submodule is used to compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and the construction schedule deviation.

[0314] In an optional embodiment, the first deviation determination module 304 further includes:

[0315] The single quantity deviation calculation submodule is used to calculate the difference between the actual completed quantity of work and the planned completed quantity of work, and obtain the single quantity deviation corresponding to each item in the bill of quantities.

[0316] The quantity deviation calculation submodule is used to perform a weighted summation of the single quantity deviations of each item in the bill of quantities based on the preset weights of each item in the bill of quantities, so as to obtain the quantity deviation.

[0317] The construction progress deviation determination submodule is used to calculate the ratio of the actual completed work volume to the total work volume of each item in the bill of quantities as the actual completion rate, calculate the ratio of the planned completed work volume to the total work volume as the planned completion rate, and take the difference between the actual completion rate and the planned completion rate as the construction progress deviation of the item.

[0318] The construction progress deviation determination submodule is used to perform a weighted summation of the construction progress deviations of each project according to the preset weights of each project, so as to obtain the construction progress deviation.

[0319] In an optional embodiment, the second deviation determination module 305 includes:

[0320] The coefficient acquisition submodule is used to acquire the preset unit price mapping coefficient for engineering quantities and the preset resource consumption mapping coefficient;

[0321] The work mapping deviation determination submodule is used to multiply and sum the actual completed work quantity of each item in the bill of quantities with the corresponding unit price mapping coefficient to obtain the completed work resource mapping value, multiply and sum the planned completed work quantity of each item with the corresponding unit price mapping coefficient to obtain the planned work resource mapping value, and calculate the difference between the completed work resource mapping value and the planned work resource mapping value to obtain the completed work mapping deviation.

[0322] The resource consumption deviation determination submodule is used to multiply and sum the actual completed work volume of each item in the bill of quantities with the corresponding resource consumption mapping coefficient to obtain the actual resource consumption mapping value, multiply and sum the planned completed work volume of each item with the corresponding resource consumption mapping coefficient to obtain the planned resource consumption mapping value, and calculate the difference between the actual resource consumption mapping value and the planned resource consumption mapping value to obtain the resource consumption deviation.

[0323] In an optional embodiment, the resource input anomaly index includes a labor input anomaly index, a material consumption anomaly index, and a machinery usage anomaly index; the risk index determination module 306 includes:

[0324] The submodule for determining the abnormal index of labor input is used to determine the abnormal index of labor input based on the deviation of the ratio between the actual labor input and the planned labor input in the resource input data.

[0325] The material consumption anomaly index determination submodule is used to determine the material consumption anomaly index based on the deviation of the ratio between the actual material consumption and the planned material consumption in the resource input data.

[0326] The Machinery Usage Anomaly Index Determination Submodule is used to determine the machinery usage anomaly index based on the deviation of the ratio between the actual machinery usage and the planned machinery usage in the resource input data.

[0327] The progress-resource linkage risk index determination submodule is used to perform a weighted summation of the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, the abnormal index of labor input, the abnormal index of material consumption, and the abnormal index of machinery use to obtain the progress-resource linkage risk index.

[0328] In an optional embodiment, it further includes:

[0329] The early warning signal generation module is used to generate an early warning signal when the risk index reaches a preset medium or high risk level. Specifically, if the source of the deviation is a delay in construction progress and an overshoot in resource consumption mapping value, a resource reconfiguration early warning is generated; if the source of the deviation is abnormal material consumption, a supply plan adjustment early warning is generated; if the source of the deviation is a deviation in the completed work mapping value that is lower than a preset node value, a process optimization early warning is generated; if the source of the deviation is an approaching preset schedule constraint node and insufficient actual completed work, a schedule performance risk warning is generated.

[0330] Example 5

[0331] This invention also provides an electronic device. Figure 4 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 4 As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement a dynamic cost early warning method for construction project progress as described in any of the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0332] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0333] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0334] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0335] Example 6

[0336] This invention also provides a computer storage medium storing a computer program thereon, wherein the computer program, when executed by the processor, implements the steps of a method for dynamic cost early warning of construction project progress according to any embodiment.

[0337] Example 7

[0338] This invention also provides a computer program product storing a computer program, wherein when the computer program is executed by the processor, it implements the steps of a method for dynamic cost early warning of construction project progress according to any embodiment.

[0339] Those skilled in the art will clearly 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.

[0340] In the several embodiments provided in this application, it should be understood that the methods, apparatuses, electronic devices, and storage media disclosed in this invention 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 displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0341] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0342] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0343] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable 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 the present invention. The aforementioned readable 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.

[0344] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic cost early warning of construction project progress, characterized in that, include: Real-time acquisition of construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data; The BIM model data is parsed to obtain a set of BIM components. A first mapping relationship is established between each BIM component in the set of BIM components and the bill of quantities data, and a second mapping relationship is established between the bill of quantities data and the preset resource consumption mapping data. The construction site image data, the on-site video data, and the construction log data are processed using multiple preset neural network models, and the corresponding modal features are extracted and fused to obtain the comprehensive construction completion status of each BIM component. Based on the comprehensive construction completion status and the first mapping relationship, the actual completed work volume of each bill of quantities item is determined, and the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation. Based on the actual completed work volume, the second mapping relationship, and the preset work volume unit price mapping coefficient, calculate the completed work resource mapping value and the planned work resource mapping value for each list item to determine the resource consumption deviation and the completed work mapping deviation. Based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data, a progress-resource linkage risk index is calculated; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

2. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, Parse the BIM model data to obtain a BIM component set, establish a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data, including: The BIM model data is parsed to extract the component number, component type, component spatial location, construction area to which the component belongs, and component quantity attributes of each component, and a BIM component set is generated. Construct a first mapping matrix between the BIM component set and the bill of quantities item set. Iterate through each element in the first mapping matrix and assign a value of 1 to the matrix element corresponding to the BIM component belonging to the same bill of quantities item, otherwise assign a value of 0 to form the first mapping relationship. Construct a second mapping matrix between the set of bill of quantities items and the preset resource consumption mapping item set. Iterate through each element in the second mapping matrix and assign the matrix element corresponding to the bill of quantities item belonging to the same resource consumption mapping item to the corresponding unit consumption association value; otherwise, assign the value to 0 to form the second mapping relationship.

3. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, Multiple preset neural network models are used to process the construction site image data, the site video data, and the construction log data, respectively, to extract and fuse the corresponding modal features, thereby obtaining the comprehensive construction completion status of each BIM component, including: The YOLOv8 object detection model is used to extract the image feature vector of the construction site image data, the ResNet50 classification model is used to extract the temporal change feature vector of the site video data, and the bidirectional long short-term memory network model is used to extract the semantic feature vector of the construction log data. The image feature vector, temporal change feature vector, and semantic feature vector are aligned by timestamp and concatenated according to preset fusion weights to obtain a fused feature vector. The fused feature vector is then input to the fully connected layer, and the comprehensive construction completion status of each BIM component is output after being mapped by an activation function.

4. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, Based on the comprehensive construction completion status and the first mapping relationship, the actual completed work volume of each bill of quantities item is determined, and the actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation, including: Based on the first mapping relationship, determine the subset of BIM components associated with each of the bill of quantities items; Obtain the quantity attributes of each BIM component in the BIM component subset and the completion percentage value of the corresponding component in the comprehensive construction completion status; The actual completed quantity of the corresponding bill of quantities item is obtained by multiplying and summing the quantity attributes of each BIM component with the corresponding completion value. The actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation.

5. The method for dynamic cost early warning of construction project progress according to claim 4, characterized in that, The actual completed work volume is compared with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation, including: Calculate the difference between the actual completed work volume and the planned completed work volume to obtain the single work volume deviation corresponding to each item in the bill of quantities; The single quantity deviation of each item in the bill of quantities is weighted and summed according to the preset weight of each item in the bill of quantities to obtain the quantity deviation. The actual completion rate is calculated as the ratio of the actual completed work volume to the total work volume for each item in the bill of quantities. The planned completion rate is calculated as the ratio of the planned completed work volume to the total work volume. The difference between the actual completion rate and the planned completion rate is taken as the construction progress deviation of the item. The construction progress deviations of each project are obtained by weighted summation based on the preset weights of each project.

6. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, Based on the actual completed work volume, the second mapping relationship, and the preset unit price mapping coefficient, the completed work resource mapping value and the planned work resource mapping value for each item in the bill of quantities are calculated to determine the resource consumption deviation and the completed work mapping deviation, including: Obtain the preset unit price mapping coefficient for engineering quantities and the preset resource consumption mapping coefficient; Multiply the actual completed work quantity of each item in the bill of quantities with the corresponding unit price mapping coefficient and sum them up to obtain the completed work resource mapping value. Multiply the planned completed work quantity of each item with the corresponding unit price mapping coefficient and sum them up to obtain the planned work resource mapping value. Calculate the difference between the completed work resource mapping value and the planned work resource mapping value to obtain the completed work mapping deviation. The actual completed work volume of each item in the bill of quantities is multiplied by the corresponding resource consumption mapping coefficient and summed to obtain the actual resource consumption mapping value. The planned completed work volume of each item is multiplied by the corresponding resource consumption mapping coefficient and summed to obtain the planned resource consumption mapping value. The difference between the actual resource consumption mapping value and the planned resource consumption mapping value is calculated to obtain the resource consumption deviation.

7. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, The abnormal resource input index includes an abnormal labor input index, an abnormal material consumption index, and an abnormal machinery usage index; based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the abnormal resource input index in the resource input data, a progress-resource linkage risk index is calculated, including: Based on the deviation of the ratio between the actual and planned labor input in the resource input data, a labor input anomaly index is determined; Based on the deviation of the ratio between actual material consumption and planned material consumption in the resource input data, a material consumption anomaly index is determined; Based on the deviation of the ratio between actual and planned machinery usage in the resource input data, a machinery usage anomaly index is determined; The construction progress deviation, resource consumption deviation, completed work mapping deviation, abnormal labor input index, abnormal material consumption index, and abnormal machinery usage index are weighted and summed to obtain the progress-resource linkage risk index.

8. The method for dynamic cost early warning of construction project progress according to claim 1, characterized in that, After calculating the schedule-resource linkage risk index based on the construction progress deviation, resource consumption deviation, completed work mapping deviation, and resource input anomaly index in the resource input data, the following steps are also included: When the risk index reaches a preset medium or high risk level, an early warning signal is generated, specifically as follows: If the deviation originates from a construction delay and a resource consumption mapping value that is ahead of schedule, a resource reconfiguration warning will be generated. If the deviation originates from abnormal material consumption, a supply plan adjustment warning will be generated. If the deviation originates from a completed work mapping deviation being lower than a preset node value, a process optimization warning is generated. If the deviation originates from the proximity of a preset schedule constraint node and insufficient actual completed work, a schedule performance risk warning will be generated.

9. A dynamic cost early warning device for construction project progress, characterized in that, include: The acquisition module is used to acquire construction site image data, on-site video data, construction log data, progress plan data, BIM model data, bill of quantities data, preset resource consumption mapping data, and resource input data in real time. The mapping relationship establishment module is used to parse the BIM model data to obtain a BIM component set, establish a first mapping relationship between each BIM component in the BIM component set and the bill of quantities data, and a second mapping relationship between the bill of quantities data and the preset resource consumption mapping data. The fusion module is used to process the construction site image data, the site video data and the construction log data using multiple preset neural network models, extract the corresponding modal features and fuse them to obtain the comprehensive construction completion status of each BIM component. The first deviation determination module is used to determine the actual completed work volume of each bill of quantities item based on the comprehensive construction completion status and the first mapping relationship, and compare the actual completed work volume with the planned completed work volume in the schedule data to obtain the work volume deviation and construction schedule deviation. The second deviation determination module is used to calculate the completed work resource mapping value and the planned work resource mapping value for each list item based on the actual completed work quantity, the second mapping relationship and the preset work quantity unit price mapping coefficient, so as to determine the resource consumption deviation and the completed work mapping deviation. The risk index determination module is used to calculate the progress-resource linkage risk index based on the construction progress deviation, the resource consumption deviation, the completed work mapping deviation, and the resource input anomaly index in the resource input data; the risk index is used to compare with a preset threshold to obtain dynamic early warning information of the corresponding level.

10. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the method as described in any one of claims 1-8.

Images